Simcenter Nastran

User's Guide
Simcenter Femap 2406 Series
Unpublished work. © 2024 Siemens

This Documentation contains trade secrets or otherwise confidential information owned by Siemens Industry Software Inc. or
its affiliates (collectively, “Siemens”), or its licensors. Access to and use of this Documentation is strictly limited as set forth in
Customer’s applicable agreement(s) with Siemens. This Documentation may not be copied, distributed, or otherwise disclosed
by Customer without the express written permission of Siemens, and may not be used in any way not expressly authorized by
Siemens.

This Documentation is for information and instruction purposes. Siemens reserves the right to make changes in specifications
and other information contained in this Documentation without prior notice, and the reader should, in all cases, consult
Siemens to determine whether any changes have been made.
No representation or other affirmation of fact contained in this Documentation shall be deemed to be a warranty or give rise to
any liability of Siemens whatsoever.
If you have a signed license agreement with Siemens for the product with which this Documentation will be used, your use of
this Documentation is subject to the scope of license and the software protection and security provisions of that agreement.
If you do not have such a signed license agreement, your use is subject to the Siemens Universal Customer Agreement, which
may be viewed at https://www.sw.siemens.com/en-US/sw-terms/base/uca/, as supplemented by the product specific terms
which may be viewed at https://www.sw.siemens.com/en-US/sw-terms/supplements/.
SIEMENS MAKES NO WARRANTY OF ANY KIND WITH REGARD TO THIS DOCUMENTATION INCLUDING, BUT NOT LIMITED
TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE, AND NON-INFRINGEMENT OF
INTELLECTUAL PROPERTY. SIEMENS SHALL NOT BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, CONSEQUENTIAL OR
PUNITIVE DAMAGES, LOST DATA OR PROFITS, EVEN IF SUCH DAMAGES WERE FORESEEABLE, ARISING OUT OF OR RELATED
TO THIS DOCUMENTATION OR THE INFORMATION CONTAINED IN IT, EVEN IF SIEMENS HAS BEEN ADVISED OF THE POSSIBILITY
OF SUCH DAMAGES.

TRADEMARKS: The trademarks, logos, and service marks (collectively, "Marks") used herein are the property of Siemens or other
parties. No one is permitted to use these Marks without the prior written consent of Siemens or the owner of the Marks,
as applicable. The use herein of third party Marks is not an attempt to indicate Siemens as a source of a product, but is
intended to indicate a product from, or associated with, a particular third party. A list of Siemens’ Marks may be viewed at:
www.plm.automation.siemens.com/global/en/legal/trademarks.html. The registered trademark Linux® is used pursuant to a
sublicense from LMI, the exclusive licensee of Linus Torvalds, owner of the mark on a world-wide basis.

About Siemens Digital Industries Software

Siemens Digital Industries Software is a global leader in the growing field of product lifecycle management (PLM),
manufacturing operations management (MOM), and electronic design automation (EDA) software, hardware, and services.
Siemens works with more than 100,000 customers, leading the digitalization of their planning and manufacturing processes. At
Siemens Digital Industries Software, we blur the boundaries between industry domains by integrating the virtual and physical,
hardware and software, design and manufacturing worlds. With the rapid pace of innovation, digitalization is no longer
tomorrow’s idea. We take what the future promises tomorrow and make it real for our customers today. Where today meets
tomorrow. Our culture encourages creativity, welcomes fresh thinking and focuses on growth, so our people, our business, and
our customers can achieve their full potential.

Support Center: support.sw.siemens.com

Send Feedback on Documentation: support.sw.siemens.com/doc_feedback_form

 Contents

Overview of Simcenter Nastran

Overview and General Capabilities of Simcenter Nastran ────────
1-1
Modular Structure and DMAP ─────────────────────── 1-2
Pre-defined Solution Sequences ────────────────────── 1-2
Using DMAP for Programmability ────────────────────── 1-2
Overview of the Simcenter Nastran Analysis Process ────────── 1-2
Creating the Input File ────────────────────────── 1-2
Submitting the Input File for Analysis ──────────────────── 1-3
Reviewing the Results ────────────────────────── 1-4

Structure and Syntax for Input Data

Overview of Specifying Input Data ─────────────────── 2-1
General Input File Syntax Rules ──────────────────── 2-1
Understanding Types of Input Data ───────────────────── 2-1
Using Consistent Unit Systems ─────────────────────── 2-2
Specifying Valid Input File Characters ──────────────────── 2-4
Understanding Simcenter Nastran Input Records ──────────── 2-5
Understanding Free Field Format ────────────────────── 2-6
Understanding Small Field Format ───────────────────── 2-7
Understanding Large Field Format ───────────────────── 2-7
Specifying Continuations for Bulk Data ───────────────── 2-8
Manually Specifying a Continuation with Identifiers ────────────── 2-9
Automatically Specifying a Continuation ────────────────── 2-10
Replicating Bulk Data Entries ───────────────────── 2-11
Replicating Continuation Entries ───────────────────── 2-12
Replication with MSGMESH ──────────────────────── 2-12
Inserting External Files with INCLUDE ───────────────── 2-14
Using INCLUDE in Different Input File Sections ─────────────── 2-14
INCLUDE Statement Limitations ────────────────────── 2-15
Including Comments in the Input File ───────────────── 2-16

Understanding the Input Data File

Overview of the Simcenter Nastran Input File Sections ───────── 3-1
Example: Using Simcenter Nastran Input Files to Model a Truss ───────── 3-2
The NASTRAN Statement (Optional) ────────────────── 3-5
File Management Section (Optional) ────────────────── 3-5
Executive Control Section (Required) ───────────────── 3-6
Executive Control Format ───────────────────────── 3-6
Order Dependence for Certain DMAP-related Statements ─────────── 3-7
Commonly Used Executive Control Statements ──────────────── 3-7
Case Control Section (Required) ─────────────────── 3-10
Case Control Format ─────────────────────────── 3-11

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 3

© 2024 Siemens
 Using the SET Command to Limit Output ────────────────── 3-22
Bulk Data Section ────────────────────────── 3-24
Bulk Data Section Format Rules ────────────────────── 3-25
Sorting Bulk Data ──────────────────────────── 3-26
Storing Bulk Data in the Database for Restarts ──────────────── 3-26
Understanding the Relationship Between Bulk Data Entries and Case Control Commands
───────────────────────────────── 3-26
Using Parameters ────────────────────────── 3-27
Parameters in the Superelement Solution Sequences ───────────── 3-28

Simcenter Nastran Output Files

Overview of the Simcenter Nastran Output Files ──────────── 4-1
Understanding the .f06 File ────────────────────── 4-2
Controlling Printed Output ──────────────────────── 4-2
Understanding the General Analysis Output ───────────────── 4-6
Understanding Analysis Type Specific Output ───────────────── 4-8
Monitoring results ─────────────────────────── 4-11
Understanding the .f04 File ───────────────────── 4-58
General Information ─────────────────────────── 4-59
Summary of Physical File Information ─────────────────── 4-60
Memory Allocation Map ───────────────────────── 4-61
Day Log ──────────────────────────────── 4-62
Understanding the I/O MB Column ──────────────────── 4-63
User Information Messages 4157 and 6439 ───────────────── 4-63
Memory and Disk Usage Statistics ───────────────────── 4-64
Database Usage Statistics ──────────────────────── 4-65
Summary of Physical File I/O Activity ──────────────────── 4-66
Other Optional Diagnostic Information ─────────────────── 4-67
Comparing Different F04 Files with F04REPRT ──────────────── 4-68
Understanding the .log File ───────────────────── 4-69
Understanding the .plt and .pch Files ───────────────── 4-69
Output Listing Examples ─────────────────────── 4-69
Understanding the .DBALL and .MASTER Files ───────────── 4-96
Specifying Whether to Save .DBALL and .MASTER Files ──────────── 4-96
Manually Deleting .DBALL and .MASTER Files ──────────────── 4-96
Understanding the .xdb and .op2 Files ──────────────── 4-97
Understanding the .hdf File ───────────────────── 4-97

Defining Grid Points and Scalar Points

Overview of Grid Points and Scalar Points ─────────────── 5-1
Defining Grid Points ───────────────────────── 5-1
Format of the GRID Entry ───────────────────────── 5-2
Understanding the GRDSET Entry ────────────────────── 5-7
Specifying Grid Point Properties ──────────────────── 5-8
Defining Mass, Damping, and Stiffness through Direct Input Matrices ────── 5-8
Defining Mass through the Concentrated Mass Element ──────────── 5-8
Defining Temperatures at Grid Points ──────────────────── 5-9

4 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Contents

Assigning a Singularity Normal to a Degree-of-Freedom ──────────── 5-9

Understanding Scalar Points ───────────────────── 5-9
Comparing Grid Points to Scalar Points ─────────────────── 5-10
Defining Scalar Points ────────────────────────── 5-10
Using Scalar Points as Extra Points for Dynamics ─────────────── 5-10
Scalar Point Output Format ──────────────────────── 5-10

Understanding Coordinate Systems

Overview of Coordinate Systems in Simcenter Nastran ───────── 6-1
Understanding the Basic Coordinate System ────────────── 6-1
ID of the Basic Coordinate System ───────────────────── 6-1
Comparing the Basic and Global Coordinate Systems ───────────── 6-1
Defining a Local Coordinate System ────────────────── 6-4
Bulk Data Entries for Defining Coordinate Systems ────────────── 6-5
Defining a Rectangular Coordinate System ────────────────── 6-7
Defining Cylindrical Coordinate Systems ────────────────── 6-11
Example: Using Cylindrical and Spherical Coordinate Systems ───────── 6-17
Understanding the Element and Material Coordinate System ────── 6-19
Special Considerations for Grid Points on the Polar Axis ──────────── 6-19
Defining Element Offsets ───────────────────────── 6-19
Other Special Coordinate System Entries ─────────────── 6-20

Material Properties
Overview of Material Properties ──────────────────── 7-1
Material Properties for Linear Structural Analysis ──────────── 7-3
Isotropic Materials ──────────────────────────── 7-4
Orthotropic Materials ─────────────────────────── 7-6
Anisotropic Materials ────────────────────────── 7-18
Material Properties for Nonlinear Analysis ─────────────── 7-27
Understanding MATHP ────────────────────────── 7-28
Material Properties for Heat Transfer ───────────────── 7-29
Understanding MAT4 and MATT4 ───────────────────── 7-30
Understanding MAT5 and MATT5 ───────────────────── 7-31
Understanding RADM and RADMT ───────────────────── 7-31
Material Properties for Fluid Analysis ───────────────── 7-32
Understanding MAT10 ────────────────────────── 7-32
Understanding MFLUID ───────────────────────── 7-32

Applied Loads
Overview of Load Sets ──────────────────────── 8-1
Defining and Combining Subcases for Loading ───────────── 8-1
Example: Creating Two Load Subcases for the Truss ────────────── 8-1
Using REPCASE ───────────────────────────── 8-4
Using SUBCOM to Combine Subcases ──────────────────── 8-5
Defining Static Loads ───────────────────────── 8-7
Point Loads ───────────────────────────── 8-8

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 5

© 2024 Siemens
 FORCE, FORCE1, FORCE2 ───────────────────────── 8-8
Distributed Loads ────────────────────────── 8-15
PLOAD ──────────────────────────────── 8-16
PLOAD1 ──────────────────────────────── 8-17
PLOAD2 ──────────────────────────────── 8-26
PLOAD4 ──────────────────────────────── 8-27
PLOADE1 and PLOADX1 ───────────────────────── 8-33
Inertia Loads (Acceleration Loads) ────────────────── 8-33
Using GRAV ────────────────────────────── 8-33
Using RFORCE ───────────────────────────── 8-37
Defining Mass in Your Model ─────────────────────── 8-39
Inertia Load Example ────────────────────────── 8-40
Thermal Loads ─────────────────────────── 8-42
Defining Temperatures for Grid Points ─────────────────── 8-43
Defining Temperatures for Elements ──────────────────── 8-44
Understanding Coefficient of Thermal Expansion (CTE) ──────────── 8-44
Rigid Element Thermal Strain ─────────────────────── 8-48
Using Temperature Loads with Subcases ────────────────── 8-48
Enforced Element Deformation Loads ───────────────── 8-55
DEFORM Example ──────────────────────────── 8-57
Loads Due to Enforced Motion ──────────────────── 8-61
Statics ───────────────────────────────── 8-61
Dynamics ─────────────────────────────── 8-61
Using Residual Vectors (Modal Augmentation) in Modal Dynamic Response Analysis
───────────────────────────────── 8-63
Frequency-dependent Loads ───────────────────── 8-63
Time-dependent Loads ─────────────────────── 8-65
Using Static Loading Data in Dynamic Analysis ──────────── 8-67
Defining Loads for Cyclic Symmetry Analysis ───────────── 8-68
Combining Loads ────────────────────────── 8-69
Using LOAD ────────────────────────────── 8-70
Bolt Preload ──────────────────────────── 8-74
Bolt Preload Procedure ────────────────────────── 8-76
Bolt Preload Example — Line Element Model ──────────────── 8-81
Bar and Beam Bolt Preload Method ──────────────────── 8-85
Solid Bolt Preload Method ──────────────────────── 8-89
External Force Fields ──────────────────────── 8-90
EFPOINTS and EFFORCE Matrices ───────────────────── 8-91
Simcenter Nastran Input Requirements ────────────────── 8-91
Using the EFLOAD Case Control Command ───────────────── 8-95
Creating an EFPOINTS Matrix ─────────────────────── 8-96
Creating an EFFORCE Matrix ─────────────────────── 8-97
Linear Static and Normal Modes Example ────────────────── 8-98

Constraints
Overview of Constraints ─────────────────────── 9-1
Single-point Constraints ─────────────────────── 9-1

6 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Contents

Defining Single point Constraints in Simcenter Nastran ──────────── 9-2

Using GRID Entries ──────────────────────────── 9-3
Using SPC and SPC1 ─────────────────────────── 9-4
Automatically Applying Single-point Constraints ──────────── 9-7
Grid Point Singularity Processor ────────────────────── 9-7
Understanding the Grid Point Singularity Table ──────────────── 9-8
Enforced Displacements at Grid Points (SPCD, SPC) ─────────── 9-11
Multipoint Constraints ─────────────────────── 9-15
Uses of Multi-point Constraints ────────────────────── 9-15
Defining Multi-point Constraints in Simcenter Nastran ──────────── 9-16
Rules for Using Multi-point Constraints ─────────────────── 9-20
Rigid Body Supports ───────────────────────── 9-22
Linear Contact Using Constraints ─────────────────── 9-23
Introduction ────────────────────────────── 9-23
Input ───────────────────────────────── 9-24
Output ──────────────────────────────── 9-25
Limitations ─────────────────────────────── 9-25
EXAMPLE 1 (cd_0.dat) ────────────────────────── 9-26
EXAMPLE 2 (cd_1.dat) ────────────────────────── 9-30

Understanding Sets and Matrix Operations

Overview of Sets and Matrix Operations ─────────────── 10-1
Understanding the Global Displacement Set and General Solution Process
──────────────────────────────── 10-2
Understanding Basic Partitioning Operations ────────────────
10-3
Static Condensation (Guyan Reduction) ──────────────── 10-4
Using Static Condensation ──────────────────────── 10-6
Sets for Dynamic Reduction ───────────────────── 10-14
Sets for Aerodynamics ─────────────────────── 10-15
Output Selection Via Set Specification ──────────────── 10-16

Direct Matrix Input

Overview of Direct Matrix Input ─────────────────── 11-1
Using DMIG ──────────────────────────── 11-2
DMIG Bulk Data User Interface ────────────────────── 11-3
DMIG Case Control User Interface ───────────────────── 11-5
DMIG Examples ─────────────────────────── 11-6
Stiffness Matrix Using DMIG ─────────────────────── 11-6
Mass Matrix Using DMIG ──────────────────────── 11-14
Viscous Damping Matrix Using DMIG ─────────────────── 11-14
Structural Damping Matrix Using DMIG ────────────────── 11-15
Load Matrix Using DMIG ──────────────────────── 11-15

Element Data Recovery Resolved at Grid Points

Overview of Element Data Recovery ───────────────── 12-1
Stress Recovery at Grid Points ──────────────────── 12-2

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 7

© 2024 Siemens
 Description of Method ────────────────────────── 12-3
Topological Method – Plate Elements and Solid Elements ─────────── 12-4
Geometric Method – Surface Elements ─────────────────── 12-7
Treatment of Exception Points – “BRANCHing” ─────────────── 12-10
User Information ─────────────────────────── 12-12
Grid Point Stress Output Description ─────────────────── 12-13
General Remarks ─────────────────────────── 12-14
Mesh Stress Discontinuities at Grid Points ────────────── 12-15
Error Estimates for Grid Point Stress Data ───────────────── 12-15
Element and Grid Point Stress Discontinuity Output Requests ──────── 12-16
Discussion of Error Measures ────────────────────── 12-17
Grid Point Stresses and Mesh Stress Discontinuities ───────────── 12-18

Solution Sequences
Overview of Solution Sequences ─────────────────── 13-1
Obtaining DMAP Listings for Solution Sequences ────────────── 13-4
Using Alters to Modify Solution Sequences ───────────────── 13-4
Understanding Scheduled and Unscheduled Exits ────────────── 13-5
Restart Procedures ─────────────────────────── 13-5
Data Processing and Matrix Operations by Functional Module ────── 13-6
Geometry Processing in SubDMAP PHASEO ───────────────── 13-7
Matrix Assembly Operations in SubDMAP SEMG ─────────────── 13-9
Multipoint Constraint Operations in SubDMAPs SEKR and SEMR2 and SEMRB ── 13-11
Single Point Constraint Operations in SubDMAPs SEKR, SEMR2 and SEMRB ─── 13-16
Constraint and Mechanism Problem Identification in SubDMAP SEKR ───── 13-18
Static Condensation in SubDMAPs SEKR and SEMR2 ──────────── 13-22
Static Solutions in SubDMAP SEKRRS ─────────────────── 13-27
Static and Dynamic Load Generation ─────────────────── 13-32
Dynamic Reduction and Component Mode Synthesis in SubDMAP SEMR3 ─── 13-38
Formulation of Dynamic Equations in SubDMAP GMA ──────────── 13-44
Data Recovery Operations in SubDMAP SEDISP ─────────────── 13-50
Data Recovery Operations in SubDMAP SEDRCVR ────────────── 13-59
Summary of Solution Sequence Operations ───────────── 13-61
Static Solutions ──────────────────────────── 13-61
Real Eigenvalue Analysis in SubDMAPs SEMR3 and MODERS ───────── 13-70
Complex Eigenvalue Analysis in SubDMAP CEIGRS ───────────── 13-88
Frequency Response and Random Analysis in SubDMAP FREQRS ─────── 13-89
Transient Response in SubDMAPs SEDTRAN and SEMTRAN ────────── 13-90
Buckling Analysis in SubDMAP MODERS ────────────────── 13-91
Superelement Analysis ───────────────────────── 13-92
Nonlinear Heat Transfer Analysis ──────────────────── 13-105

Modeling Guidelines
Overview of Modeling Guidelines ────────────────── 14-1
Choosing the Right Element ───────────────────── 14-1
General Guidelines for Element Selection ────────────────── 14-2
Zero-dimensional Elements ─────────────────────── 14-2
One-dimensional Elements ──────────────────────── 14-2

8 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Contents

Two-dimensional Elements ──────────────────────── 14-3

Three-dimensional Elements ─────────────────────── 14-7
R-type Elements ──────────────────────────── 14-8
Determining Mesh Density ────────────────────── 14-8
h-elements ────────────────────────────── 14-15
Creating Mesh Transitions ───────────────────── 14-16
Transition from a Coarse Mesh to a Fine Mesh ─────────────── 14-16
Mesh Transition Between Dissimilar Element Types ───────────── 14-26
Shell-to-solid Transition Element (RSSCON) ──────────────── 14-28
Reviewing Grid Point Stresses ──────────────────── 14-42
Defining Consistent Loading ──────────────────── 14-44
Symmetry ───────────────────────────── 14-47
User Interface ───────────────────────────── 14-50

Model Verification
Overview of Model Verification ──────────────────── 15-1
Preprocessor Checks ──────────────────────── 15-1
Creating a Shrink Plot of the Elements ─────────────────── 15-1
Hidden Line/Shaded Plots ──────────────────────── 15-2
Free Edge/Face ───────────────────────────── 15-3
Zipper Effect ────────────────────────────── 15-5
CBAR/CBEAM Orientation and Offset Check ───────────────── 15-6
Duplicate Grid Points/Elements ────────────────────── 15-6
Properties/Material Plots ───────────────────────── 15-6
Using Consistent Normals ──────────────────────── 15-6
Strain Energy Output ──────────────────────── 15-9
Diagnostic Tools ────────────────────────── 15-16
Element Summary Output ─────────────────────── 15-17
Element Geometry Checks ─────────────────────── 15-19
Weight Center of Gravity and Moment of Inertia Check ─────────── 15-33
Mechanisms and Singularities ────────────────────── 15-35
Applied Loads Check ────────────────────────── 15-38
Verifying Reaction Loads ──────────────────────── 15-40
1g Load ─────────────────────────────── 15-42
Output of Maximum, Minimum Grid Point-based Responses ───────── 15-42
Performing an Unconstrained Equilibrium Check ────────────── 15-47
Automatic Identification of Unintentional Grounding ──────────── 15-48
Thermal Equilibrium Check ─────────────────────── 15-50
Grid Point Force Balance ──────────────────────── 15-51
Stress Error Estimators ─────────────────────── 15-57
Grid Point Stress Discontinuities ───────────────────── 15-58
Element Stress Discontinuities ────────────────────── 15-63
Discussion of Error Measures ────────────────────── 15-65
User Interface ───────────────────────────── 15-66
Postprocessor Checks ─────────────────────── 15-68
Examining Displacements ──────────────────────── 15-68
Examining Stress Contours ─────────────────────── 15-68
Creating Grid Point Stress and Stress Discontinuity Plots ─────────── 15-72

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 9

© 2024 Siemens
 Viewing Element Strain Energy Plots ─────────────────── 15-73

Restarts
Overview of Restarts ──────────────────────── 16-1
Types of Restarts ────────────────────────── 16-2
Understanding Restart Terminology ──────────────────── 16-3
Read-write Restarts ───────────────────────── 16-3
NASTRAN Statement Section ─────────────────────── 16-3
File Management Section (FMS) ───────────────────── 16-3
Executive Control Section ──────────────────────── 16-5
Case Control Section ────────────────────────── 16-5
Bulk Data Section ──────────────────────────── 16-5
Read-only Restarts ───────────────────────── 16-12
Mini-database Restarts ─────────────────────── 16-12
Determining the Version for a Restart ──────────────── 16-13

Database Concepts
Overview of Database Concepts ─────────────────── 17-1
Understanding DBsets ──────────────────────── 17-2
Delivery Database ───────────────────────── 17-5
Deleting DBsets ─────────────────────────── 17-5
Database Autoassignment ────────────────────── 17-5
Database Project and Version ──────────────────── 17-6
Migrating Databases ──────────────────────── 17-7
Database Archival, Compression, and Transfer ───────────── 17-7
Database Archival and Retrieval ────────────────────── 17-8
Database Compression ───────────────────────── 17-8
Transferring Databases Across Different Platforms ────────────── 17-9
Using the File Management Section with Databases ───────── 17-10
Using the INIT Statement ──────────────────────── 17-11
Using the ASSIGN Statement ────────────────────── 17-12
Using the EXPAND Statement ────────────────────── 17-14
Using the RESTART Statement ────────────────────── 17-15
Using the DBCLEAN Statement ────────────────────── 17-16
Using the DBDIR Statement ─────────────────────── 17-16
Using the INCLUDE Statement ────────────────────── 17-17
Guidelines for Large Problems ─────────────────── 17-18
Example I ─────────────────────────────── 17-18
Example J ─────────────────────────────── 17-19
Example K ────────────────────────────── 17-19
Example L ────────────────────────────── 17-20
BUFFSIZE ─────────────────────────────── 17-21

Inertia Relief in Linear Static Analysis

Overview of Inertia Relief ────────────────────── 18-1
Description of Inertia Relief Using PARAM,INREL,-1 ─────────── 18-1

10 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Contents

Implementation of Inertia Relief Using PARAM,INREL,-1 ──────── 18-2

Automatic Inertia Relief ────────────────────── 18-14
Limitations for the Automatic Inertia Relief Method ───────────── 18-16

Contact
Overview of Contact ──────────────────────── 19-1
Contact Regions ────────────────────────── 19-2
Contact Region Parameters ───────────────────── 19-3
Contact Pairs ──────────────────────────── 19-4
Contact Control Parameters ───────────────────── 19-7
Penalty Factor Adjustment and Contact Stiffness ─────────── 19-11
Contact Output Request ────────────────────── 19-15
Contact Stiffness Output ────────────────────── 19-15
Contact Preview Output ────────────────────── 19-16
Including CGAP Elements in a Contact Solution ──────────── 19-18
Contact conditions in modal, buckling, and dynamic solutions ─────
19-19
Contact Conditions in Linear Buckling (SOL 105) ─────────── 19-22
Non-converged Contact Option ─────────────────── 19-22
Contact Conditions with the Element Iterative Solver ───────── 19-23
Contact Conditions with Inertia Relief ──────────────── 19-24
Contact with Static Condensation ────────────────── 19-24
SOL 101 Contact Algorithm ───────────────────── 19-26
Kinematic Equations ────────────────────────── 19-26
Normal Contact Constraints ─────────────────────── 19-27
Coulomb Friction Contact Constraints ────────────────── 19-28
Finite Element Kinematic Equations ─────────────────── 19-28
Global Solution Strategy ──────────────────────── 19-30
Summary of Significant Aspects of Contact ──────────────── 19-31

Gluing Elements
Overview of Gluing Elements ──────────────────── 20-1
Glue Regions ──────────────────────────── 20-6
Defining and Selecting Glue Pairs ────────────────── 20-6
Glue Control Parameters ─────────────────────── 20-9
Glue Conditions in Acoustics Analysis ──────────────── 20-13
Glue Output Request ──────────────────────── 20-13
Glue stiffness output ──────────────────────── 20-14
Glue preview ─────────────────────────── 20-14
Edge-to-Surface Glue Stiffness ─────────────────── 20-16

Plotting
Overview of Plotting in Simcenter Nastran ──────────────
21-1
X-Y Plots ──────────────────────────────── 21-1
Flutter Analysis Plots ────────────────────────── 21-1
Superelement Plotting ─────────────────────── 21-1

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 11

© 2024 Siemens
 Post Processors ─────────────────────────── 21-6

Linear Buckling
Overview of Linear Buckling ───────────────────── 22-1
Finite Element Approach ────────────────────── 22-1
Eigenvalue Extraction Methods ─────────────────── 22-7
Enhanced Inverse Power Method (SINV) ────────────────── 22-8
Lanczos Method ──────────────────────────── 22-8
Comparison of Methods ───────────────────────── 22-8
User Interface ───────────────────────────── 22-9
Linear Buckling Assumptions and Limitations ───────────── 22-10
Buckling Examples ───────────────────────── 22-11
Example 1 – Classical Euler Beam Buckling ──────────────── 22-11
Example 2 – Lateral Buckling ────────────────────── 22-17
Example 3 – Planar Frame Buckling ──────────────────── 22-23
Example 4 – Buckling of a Stiffened Panel with Transverse Shear Flexibility ─── 22-27
Example 5 – Buckling of a Cylinder Under Uniform Axial Load ──────── 22-31
Theoretical Results for Example 5 ──────────────────── 22-35
Example 6 – Multiple Buckling Analyses in a Single Run ─────────── 22-36

Follower Stiffness
Overview of Follower Stiffness ──────────────────── 23-1
Benefits ────────────────────────────── 23-2
Input ─────────────────────────────── 23-2
Output ─────────────────────────────── 23-2
Linear Dynamic Analysis with Follower Stiffness (SOLs 108, 109, 111, and 112)
──────────────────────────────── 23-3

Laminates
Overview of Laminated Composite Materials ───────────── 24-1
Modeling Laminates in Simcenter Nastran ───────────────── 24-3
Classical Lamination Theory for Shell Composites (PCOMP,PCOMPG) ─── 24-8
Laminate Failure Indices and Strength Ratios ───────────── 24-25
Failure Theories for Shell Composites (PCOMP, PCOMPG) ────────── 24-26
Failure Theories for Solid Composites (PCOMPS) ────────────── 24-33
Bonding Material Failure Indices ───────────────────── 24-38
Ply Stress and Strain Calculation for Shell Composites (PCOMP,PCOMPG) 24-39
Calculating Interlaminar Shear Stress and Strains on Shell Composites
(PCOMP,PCOMPG) ────────────────────── 24-41
Examples: Defining Laminate Material Properties ─────────── 24-42
Example: Using PCOMP to Model a Honeycomb Plate ──────────── 24-42
Example: Using PSHELL to Model a Honeycomb Section ─────────── 24-49

Cyclic Symmetry
Overview of Cyclic Symmetry ──────────────────── 25-1
Supported Solution Sequences ────────────────────── 25-1

12 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Contents

Comparing Cyclic Symmetry and Traditional Reflective Symmetry Techniques ──

25-1
Understanding Types of Symmetry ────────────────── 25-2
Cyclic Symmetry Theory ─────────────────────── 25-6
Symmetrical Components ──────────────────────── 25-6
Understanding Rotational Symmetry ─────────────────── 25-8
Dihedral Symmetry ────────────────────────── 25-25
Reflective Symmetry ────────────────────────── 25-37
Axisymmetry ───────────────────────────── 25-42
Using Cyclic Symmetry ─────────────────────── 25-45
Overview of Performing a Cyclic Symmetry Analysis ──────────── 25-45
Bulk Data Entries Specific to Cyclic Symmetry ─────────────── 25-48
Case Control Commands Specific to Cyclic Symmetry ──────────── 25-55
Performing Axisymmetric Analysis ───────────────── 25-61
Solving Axisymmetric Problems ───────────────────── 25-61
Axisymmetric Example ───────────────────────── 25-62

Figures

3-1. Truss Example ───────────────────────── 3-2

3-2. Simcenter Nastran Input File Structure ────────────── 3-1
3-3. A Bulk Data Example: the GRID Entry ─────────────── 3-25
4-1. truss1.f06 Output File ────────────────────── 4-4
5-1. Displacement Components ──────────────────── 5-5
5-2. DOF Nomenclature ─────────────────────── 5-1
6-1. Coordinate System Definition ────────────────── 6-2
6-2. The Basic Coordinate System ─────────────────── 6-1
6-3. Model Change ───────────────────────── 6-6
6-4. Grid Point Locations Using a Local Coordinate System ─────── 6-8
6-5. CORDIR and CORD2R Definitions ───────────────── 6-10
6-6. CORD1C Definition ─────────────────────── 6-11
6-7. CORD2C Definition ─────────────────────── 6-13
6-8. Arch Structure ───────────────────────── 6-14
6-9. Definition of Local Cylindrical Coordinate System ───────── 6-14
6-10. CORD1S and CORD2S Definitions ──────────────── 6-17
6-11. Silo Model ────────────────────────── 6-18
6-12. Example of Local Coordinate Systems to Aid in Establishing Model
Geometry ─────────────────────────── 6-4
7-1. CQUADX4 Principal Material Coordinate System, XZ-Plane ───── 7-7
7-2. CQUADX4 Principal Material Coordinate System, XY-Plane ───── 7-8
7-3. Composite Tube with Orthotropic Properties ─────────── 7-13
7-4. Finite Element Model of the Orthotropic Tube ────────── 7-15
7-5. Pseudo-Anisotropic Model ─────────────────── 7-19
7-6. Verification Model Strain Output ───────────────── 7-23
7-7. Anisotropic Solids Model ──────────────────── 7-24
7-8. Output for Structure with Anisotropic Material ────────── 7-26
8-1. Truss Example with Two Loading Conditions ──────────── 8-1

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 13

© 2024 Siemens
 8-2. Edge Forces on a Plate Element ───────────────── 8-11
8-3. OLOAD Output ───────────────────────── 8-13
8-4. Distributed Load on CBAR Elements ─────────────── 8-20
8-5. CBARAO Output for the Center Element ───────────── 8-22
8-6. Shear and Moment Diagram Generated from CBARAO Output ─── 8-23
8-7. Planar Truss Structure with a Snow Loading ──────────── 8-24
8-8. SPC Forces Due to the Snow Loading ─────────────── 8-26
8-9. PLOAD1 Convention in CBAR or CBEAM Elements ───────── 8-18
8-10. Pressure Loads on the CQUAD4s ──────────────── 8-29
8-11. Displacements Due to Pressure Load ────────────── 8-32
8-12. RFORCE Vector at Grid Point Gi ───────────────── 8-37
8-13. Rotating Concentrated Masses ───────────────── 8-40
8-14. Axial Force in the Bar as a Function of Angular Velocity ────── 8-42
8-15. Umbrella Tent With Thermal Loads ─────────────── 8-51
8-16. Three-Bar Structure with Initial Strain ────────────── 8-57
8-17. Three-Bar Structures with Initial Strain ───────────── 8-58
8-18. Stress and Displacement Output of the Three Bar Structure ──── 8-60
8-19. Umbrella Tent Subjected to Gravity and Wind Loads ─────── 8-71
8-20. Load Combinations ────────────────────── 8-70
9-1. Pin-Connected Rods ─────────────────────── 9-2
9-2. Single Element with Vertical Load ──────────────── 9-9
9-3. Grid Point Singularity Table ─────────────────── 9-9
9-4. Single Element with a Vertical Load ─────────────── 9-10
9-5. Force and Displacement at Grid Point 2 with a Nonzero I2 ───── 9-10
9-6. Force and Displacement at Grid Point 2 with a Zero I2 ─────── 9-11
9-7. Flat Plate with Various Corner Constraints ──────────── 9-13
9-8. Using an MPC to Compute Relative Displacement ───────── 9-18
9-9. Abridged Output of Cantilever Beam with Stopper. ──────── 9-28
9-10. Undeformed Plot of the Thick Pad. ─────────────── 9-31
9-11. Deformed Plot of the Thick Pad. ──────────────── 9-31
9-12. Abridged Output of Thick Pad. ───────────────── 9-32
10-1. Cantilever Beam Using Guyan Reduction ──────────── 10-10
10-2. Results Comparison With and Without Guyan Reduction ───── 10-13
11-1. Results Comparison Between the Conventional Method Versus DMIG
─────────────────────────────── 11-13
11-2. Results Comparison Between Conventional Load Input Versus DMIG
─────────────────────────────── 11-17
11-3. Planer Cantilever Beam ──────────────────── 11-6
12-1. Three or More Elements Connected to the Same Line Segment ── 12-10
12-2. Slope Difference between Adjacent Elements Connected to the Same Line
─────────────────────────────── 12-11
13-1. Hierarchy in the Structural Solution Sequences ───────── 13-1
13-2. A Mechanism Type of Instability ──────────────── 13-20
13-3. Structured Solution Sequence Flow Chart ─────────── 13-96
13-4. Solution Sequence Flow Chart ──────────────── 13-101

14 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Contents

14-1. Unique Grid Point Normal ─────────────────── 14-6

14-2. Highly Skewed Element ──────────────────── 14-3
14-3. Element with High Aspect Ratio ──────────────── 14-3
14-4. Highly Warped Element ──────────────────── 14-4
14-5. CQUAD8 with Missing Midside Node ────────────── 14-4
14-6. Circular Disk with Different Meshes ─────────────── 14-9
14-7. Stresses with Different Elements and Meshes ────────── 14-10
14-8. Figure 9-7 Stresses with Different Elements and Meshes (continued)
─────────────────────────────── 14-11
14-9. Figure 9-7 Stresses with Different Elements and Meshes (continued)
─────────────────────────────── 14-12
14-10. Transition Methods ───────────────────── 14-16
14-11. Baseline Model ─────────────────────── 14-18
14-12. Transition with CTRIA3 ─────────────────── 14-19
14-13. Transition with RSPLINE ─────────────────── 14-20
14-14. Resultant Stresses for Membrane Loads ──────────── 14-21
14-15. Resultant Stresses for In-Plane Shear Loads ────────── 14-22
14-16. Resultant Stresses for In-Plane Bending Loads ───────── 14-23
14-17. Resultant Stresses for Vertical Loads ───────────── 14-24
14-18. Resultant Stresses for Twisting Loads ───────────── 14-25
14-19. Typical Transition Between Dissimilar Elements ──────── 14-27
14-20. Stress Distribution for CQUAD4 Model ──────────── 14-34
14-21. Stress Distribution for CQUAD4 Model ──────────── 14-39
14-22. Stress Distribution for CHEXA Model ───────────── 14-39
14-23. Stress Distribution for Solid-Plate ModelUsing MPCs for Transition
─────────────────────────────── 14-40
14-24. Stress Distribution for Solid-Plate ModelUsing RSSCONs for Transition
─────────────────────────────── 14-41
14-25. Clamped Connection of a Shell to a Solid ─────────── 14-28
14-26. ELEM and GRID Option on the RSSCON Bulk Data Entry ───── 14-29
14-27. Best Modeling Practice for RSSCON ────────────── 14-30
14-28. RSSCON in Meshes Where Solid and Shell Elements Share Grid Points
─────────────────────────────── 14-31
14-29. Modeling Options with RSSCON ─────────────── 14-31
14-30. RSSCON Modeling Practice Parameters ──────────── 14-33
14-31. Uniaxial Load for a 3 x 3 Mesh ──────────────── 14-42
14-32. Element Component Stresses ──────────────── 14-43
14-33. Element Component Stresses ──────────────── 14-43
14-34. Consistent Loads Due to Uniform In-Plane Element Edge Load ─ 14-45
14-35. Consistent Loads Due to Uniform Loads on a Solid Element Face 14-47
14-36. Results Comparison Using Symmetry Versus a Full Model ──── 14-52
14-37. Structures with Planes of Symmetry ───────────── 14-48
14-38. Decomposing a General Loading into Symmetric and Antisymmetric
Loadings ────────────────────────── 14-48
14-39. Symmetric Model ────────────────────── 14-49

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 15

© 2024 Siemens
 14-40. Antisymmetric Model ──────────────────── 14-50
15-1. Shrink Plot ────────────────────────── 15-1
15-2. Wireframe, Hiddenline, and Shaded Plots ──────────── 15-2
15-3. Wing Free Edge Check ───────────────────── 15-3
15-4. Wing Geometric Surfaces ─────────────────── 15-4
15-5. Possible Modeling Error as Indicated by Free Face ──────── 15-4
15-6. Model with Zipper Effect ──────────────────── 15-5
15-7. CQUAD4 Model ──────────────────────── 15-7
15-8. CQUAD4s with Pressure Loads ───────────────── 15-8
15-9. Load Deflection Curve ───────────────────── 15-9
15-10. Spring Sample Problem ─────────────────── 15-10
15-11. Cantilever Beam ────────────────────── 15-12
15-12. Element Strain Energy Output for a Cantilever Beam Model ─── 15-14
15-13. Element Strain Energy Contour Plot of an Automobile Model ── 15-15
15-15. Partial Output from the Grid Point Weight Generator ────── 15-33
15-16. OLOAD Resultant ────────────────────── 15-39
15-17. OLOAD Output ─────────────────────── 15-40
15-18. SPCFORCE RESULTANT ──────────────────── 15-40
15-19. Single-Point Constraint Forces ──────────────── 15-41
15-20. Clamped Plate with Stiffeners ──────────────── 15-51
15-21. Stress Contour Plots of a Cantilever Plate ─────────── 15-55
15-22. Grid Point Force Balance Output for a Model with Incorrect CBAR
Properties ────────────────────────── 15-55
15-23. Grid Point Force Balance Output for a Model with Correct CBAR
Properties ────────────────────────── 15-56
15-24. Partial Model for the Stress Discontinuity Calculations ───── 15-58
15-25. CQUAD4 Model ─────────────────────── 15-68
15-26. Consistent Connectivity ─────────────────── 15-70
15-27. Inconsistent Connectivity ────────────────── 15-71
18-1. Inertia Relief Analysis of a CBEAM Model ──────────── 18-3
18-2. Inertia Relief Output When Grid Point 1 Is the SUPORT Point ─── 18-4
18-3. Inertia Relief Output When Grid Point 1 Is the SUPORT Point ─── 18-5
18-4. Inertia Relief Output When Grid Point 3 Is the SUPORT Point ─── 18-8
18-5. Partial Output of the UACCEL Example ───────────── 18-13
18-6. Abridged Output Using the Automatic Inertia Relief Method ─── 18-17
19-1. Example 1 – Use the Radius of the Contacting Cylinder or the Radius plus
the block Thickness as the Length ─────────────── 19-12
19-2. Example 2 – Use the Radius of the Inner or Outer Cylinder as the Length
─────────────────────────────── 19-13
19-3. Example 3 – Use the Inner or Outer Radius of the Wheel as the Length
─────────────────────────────── 19-13
19-4. Example 4 – Use the Radius or Diameter of the Arch as the Length
─────────────────────────────── 19-14
19-5. Example 5 – Use the Length of One of the Cantilever Shells as the Length
─────────────────────────────── 19-14

16 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Contents

21-1. Phase 0 and I Superelement Plot Control In the Structured Solution

Sequences ────────────────────────── 21-2
21-2. Phase l Superelement Plot Control In the Unstructured Solution Sequences
──────────────────────────────── 21-3
21-3. Phase II, Ill, and IV Superelement Plot Control in All Solution Sequences
──────────────────────────────── 21-4
22-1. Single Planar Bar Element ─────────────────── 22-1
22-2. Eigenvalue Table for a Euler Beam ─────────────── 22-14
22-3. Eigenvectors for an Euler Beam ──────────────── 22-14
22-4. Mode Shapes for an Euler Beam ──────────────── 22-15
22-5. Euler Beam ───────────────────────── 22-11
22-6. Eigenvalue Table and Eigenvectors for Lateral Beam Buckling ── 22-20
22-7. Static and Buckling Shapes ────────────────── 22-21
22-8. Lateral Buckling ─────────────────────── 22-17
22-9. Planar Frame Buckling ──────────────────── 22-23
22-10. Frame Buckling ─────────────────────── 22-23
22-11. Eigenvalues for a Stiffened Panel ─────────────── 22-29
22-12. Stiffened Panel Buckling Mode ──────────────── 22-29
22-13. Buckling of a Stiffened Panel ───────────────── 22-27
22-14. Static Deflection of a Cylinder Due to a Compressive Load ─── 22-33
22-15. Buckling Shapes of a Cylinder with 20 Elements Along the z-Direction
─────────────────────────────── 22-34
22-16. Buckling Shape of a Cylinder with 40 Elements Along the z-Direction
─────────────────────────────── 22-34
22-17. Buckling of a Cylinder Under a Compressive Load ─────── 22-31
23-1. Follower Force Example ──────────────────── 23-1
24-1. Equivalent PSHELL and MAT2 Entries Are Generated ─────── 24-5
24-2. Laminae Arranged to Form a Laminate ───────────── 24-1
24-3. Exploded View of Three Cross-Ply Laminated Plates ─────── 24-2
24-4. Honeycomb Cantilever Plate ───────────────── 24-42
24-5. Composite Output of the Honeycomb Plate ─────────── 24-44
24-6. Typical Composite Layup ─────────────────── 24-46
24-7. Honeycomb Cantilever Plate Represented by PSHELL/MAT1 ─── 24-49
24-8. Displacement Outputs for Honeycomb Sandwich Using PCOMP ─ 24-55
24-9. Displacement Outputs for Honeycomb Sandwich Using PSHELL/MAT1
─────────────────────────────── 24-55
25-1. Types of Geometrical Symmetry ──────────────── 25-3
25-3. Example of Rotational Symmetry ─────────────── 25-21
25-4. Conforming Boundaries for the ROT Option ─────────── 25-8
25-7. Dihedral Symmetry ───────────────────── 25-26
25-8. Dihedral Symmetry ───────────────────── 25-26
25-9. Reflective Symmetry with One or Two Planes of Symmetry–A Special Case
of Dihedral Symmetry ──────────────────── 25-37
25-10. Reflective Symmetry ──────────────────── 25-41
25-11. Cyclic Symmetry Flowchart ───────────────── 25-46

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 17

© 2024 Siemens
 25-12. Cylinder Example ────────────────────── 25-57
25-13. Square Plate Example ──────────────────── 25-58
25-14. A Five Degrees Strip of a Cylinder Using Axisymmetry ───── 25-62
25-15. Abridged Output Using Axisymmetry ───────────── 25-67
25-16. Abridged Output Using Full Model ────────────── 25-67

Listings

3-1. Truss Example Input File ───────────────────── 3-3

3-2. Unsorted Listings ─────────────────────── 3-16
3-3. Sorted Listing ───────────────────────── 3-17
3-4. Controlling Output with Set Commands ───────────── 3-23
6-1. Truss Using CORD1R ─────────────────────── 6-9
6-2. Grain Silo Coordinate Systems ────────────────── 6-19
7-1. Orthotropic Tube Input File ─────────────────── 7-15
7-2. Anisotropic Verification Input File ──────────────── 7-21
7-3. Structure with Anisotropic Material Input File ─────────── 7-24
8-1. Truss Input File with Two Loadings ──────────────── 8-3
8-2. Truss Input File with the REPCASE Command ──────────── 8-4
8-3. SUBCOM Input File ─────────────────────── 8-5
8-4. Edge Loads ────────────────────────── 8-12
8-5. PLOAD1 Loading ──────────────────────── 8-21
8-6. Truss with Snow Loading ──────────────────── 8-25
8-7. CQUAD4 Elements with Pressure Loads ────────────── 8-31
8-8. Axial Force in the Rotating Arms ───────────────── 8-41
8-9. Using SUBCOMs with Temperature Loads ───────────── 8-51
8-10. Umbrella Tent Thermal Analysis ──────────────── 8-53
8-11. Temperature Distribution Generated with Solution 153 ────── 8-53
8-12. Initial Strain Input File ───────────────────── 8-59
8-13. Umbrella Tent with Wind and Gravity Loads ─────────── 8-73
9-1. The SPC/SPCD Combination ─────────────────── 9-14
9-2. Relative Displacement Using an MPC Entry ──────────── 9-19
9-3. Input file for Cantilever Beam with Stopper. ──────────── 9-27
9-4. Input File for Pad on Rigid Foundation. ────────────── 9-30
10-1. Input File for Common Bulk Data Entries ──────────── 10-11
10-2. Input File for Using Conventional Analysis ─────────── 10-11
10-3. Input File Specifying A-Set DOFs ──────────────── 10-12
10-4. Input File Specifying O-Set DOFs ──────────────── 10-13
11-1. Input File for Conventional Analysis ─────────────── 11-6
11-2. Input File for Reading a Stiffness Matrix Using DMIG with Small-Field
Format ─────────────────────────── 11-10
11-3. Input File for Reading a Stiffness Matrix with Large-Field Format ─ 11-12
11-4. Input File for Reading in a Load Vector Using DMIG ─────── 11-16
12-1. Grid Point Stress Output – Typical ─────────────── 12-13
12-2. Grid Point Stress Output with “BRANCHing” and TOLERANCE > 0 ─ 12-14

18 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Contents

12-3. Grid Point Stress Output with “BRANCHing” and TOLERANCE < 0 ─ 12-14
14-1. Input File for Solid-Plate Model Using MPCs for Transition ──── 14-36
14-2. Input File for Solid-Plate Model Using RSSCONS for Transition ── 14-38
14-3. Input File for Using Symmetric and Antisymmetric Boundary Conditions
─────────────────────────────── 14-51
15-1. Consistent CQUAD4 Connectivity ──────────────── 15-7
15-2. Inconsistent CQUAD4 Connectivity ─────────────── 15-8
15-3. Input File for the Strain Energy of a Cantilever Beam ────── 15-13
15-4. Input File for the Strain Energy of a Car Model ───────── 15-14
15-5. Input File for the Demonstration of Grid Point Force Balance ─── 15-52
15-6. Input File for Illustrating Diagnostic Checks ────────── 15-16
15-7. Input File for the Error Estimators for CQUAD4 ───────── 15-67
15-8. Input File for CQUAD4 Stress Contour Plotswith Consistent Connectivity
─────────────────────────────── 15-69
16-1. Input File for Printing/Punching a Listing of Bulk Data File from the
Database ─────────────────────────── 16-1
16-2. Input File for a Cold Start Run ───────────────── 16-5
16-3. Input File for a Restart Run with an Additional Load Case ───── 16-6
16-4. Input File for a Restart Run with Modified Material Property Entry ─ 16-8
16-5. Sorted Bulk Data Listing from a Cold Start Run ────────── 16-8
16-6. Input File for a BCELL16 Cold Start Run ───────────── 16-10
16-7. Input File for a BCELL16 Restart Run ────────────── 16-11
17-1. Memory Summary Table(mem = 2MW, buffsize = 2049) ───── 17-23
17-2. Memory Summary Table(mem = 4MW, buffsize = 2049) ───── 17-23
17-3. Memory Summary Table(mem = 4MW, buffsize = 17291) ──── 17-23
18-1. Three CBEAM Model with Inertia Relief ───────────── 18-3
18-2. Imposing a 10 g Acceleration Using the PARAM,UACCEL ───── 18-12
22-1. Input File for a Typical Buckling Analysis ──────────── 22-6
22-2. Input File for Euler Beam ─────────────────── 22-12
22-3. Input File for Lateral Buckling ───────────────── 22-18
22-4. Input File for Frame Buckling ───────────────── 22-24
22-5. Input File for the Buckling of Stiffened Panels ────────── 22-28
22-6. Input File for a Cylindrical Buckling Problem ────────── 22-31
22-7. Input File for Multiple Buckling Analyses ──────────── 22-36
24-1. Honeycomb Plate ────────────────────── 24-43
24-2. Honeycomb Sandwich Using PCOMP ────────────── 24-50
24-3. Honeycomb Sandwich Using PSHELL/MAT1 ─────────── 24-52
25-1. ROT or AXI Symmetry Type ────────────────── 25-59
25-2. DIH Symmetry Type ───────────────────── 25-59
25-3. Input File for Five Degrees Strip of Cylinder ─────────── 25-64

Tables

Determining Consistent Units ────────────────── 2-2

Determining Conversion Factors ───────────────── 2-2

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 19

© 2024 Siemens
 Consistent Sets of Units for Structural Analysis – Base Units of Force,
Length, and Time ─────────────────────── 2-3
ASCII Subset for Simcenter Nastran ──────────────── 2-4
Types of Files Output by Simcenter Nastran ──────────── 4-1
Summary of Material Bulk Entries ──────────────── 7-1
Elastic Constant Usage in Stiffness Formulation for Various Elements
──────────────────────────────── 7-5
Composite Tube Material Properties ─────────────── 7-14
Strains Due to Unit Loads on Pseudo Anisotropic Model ────── 7-20
Summary of Material Bulk Entries for Linear Structural Analysis ─── 7-3
Restrictions When Using the RFORCE Entry ──────────── 8-39
Bulk Data Entries Used for Temperature Definition on Elements ── 8-43
Normal Direction for Surface Patches ────────────── 8-96
Comparison of the SPC Versus the SPC/SPCD Combination ───── 9-14
Basic Set Partitioning Operations ──────────────── 10-3
Elements Types Supported in Element Data Resolved at Grid Point
Calculations (Hyperelastic Elements Not Included) ──────── 12-1
Evaluation of Average Grid Point Stresses for Plate and Shell Elements by
the Topological Method ──────────────────── 12-4
Evaluation of Average Grid Point Stress for Solid Elements by the
Topological Method ────────────────────── 12-7
Evaluation of Average Grid Point Stresses for Plate and Shell Elements by
the Geometric Method ──────────────────── 12-10
Structured Solution Sequences ───────────────── 13-2
Stresses Close to r = a for a Circular Disk ───────────── 14-13
Loadings Used for the Mesh Transition Test ─────────── 14-17
Tip Displacements and Maximum Stresses at the Clamped End ── 14-35
Cross-Sectional Properties of a Cantilever Beam ───────── 15-11
Contributions Used for the Grid Point Force Balance ─────── 15-51
The Values of δ g and ERROR ESTIMATE for Different Element Types(X =
28.5, Y = 0.0) ──────────────────────── 15-62
The Values of δ g and ERROR ESTIMATE for Different Element Types (X =
28.5, Y = 1.5) ──────────────────────── 15-62
Element Stresses for CQUAD4 ───────────────── 15-59
Grid Point Stresses for Surface 91 ──────────────── 15-59
The Values of δ e and ERROR ESTIMATE for Different Plate Element Types
─────────────────────────────── 15-65
The Values of δ e and ERROR ESTIMATE for Different Plate Element Types
─────────────────────────────── 15-65
BCLL16 Performance Summary ──────────────── 16-0
Number and Type of Roots Found with EIGRL Entry ──────── 22-9
Euler Beam Results Comparison ──────────────── 22-17
Lateral Buckling Results Comparison ────────────── 22-23
Frame Buckling Results Comparison ────────────── 22-26
Stiffened Panel Buckling Results Comparison ────────── 22-30

20 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Contents

Thin Cylinder Buckling Results Comparison ─────────── 22-35

Material Properties on the PSHELL Entry ───────────── 24-4
Honeycomb Material Properties ──────────────── 24-43
Types of Symmetry ─────────────────────── 25-5
Coordinate Systems for Half Segments ───────────── 25-28
Symmetrical Components for Reflective Symmetry Example ─── 25-31
Parameter C for the DIH Option ───────────────── 25-50
Static Load Definitions for Cyclic Symmetry ─────────── 25-53
Bulk Data Entries Unique to Cyclic Symmetry ────────── 25-48
Case Control Commands Unique to Cyclic Symmetry ─────── 25-55

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 21

© 2024 Siemens
22 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series
© 2024 Siemens
 1. Overview of Simcenter Nastran
Overview and General Capabilities of Simcenter Nastran
Simcenter Nastran is a general purpose finite element analysis computer program that lets you solve
a wide range of engineering problems. Simcenter Nastran is written primarily in FORTRAN and is
optimized to run efficiently and provide identical results on a wide variety of computers and operating
systems.

Simcenter Nastran contains the following analysis capabilities:

• Linear statics (including inertia relief)

• Normal modes and buckling

• Heat transfer (steady-state and transient)

• Transient response

• Frequency response

• Response spectrum and random response

• Geometric and material nonlinear static and transient response

• Design optimization and sensitivity (including dynamic and shape optimization)

• Composite materials

• Acoustic response

• Aeroelasticity

• Superelements

• Complex eigenanalysis

• Axisymmetric analysis

• Cyclic symmetry

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 1-1

© 2024 Siemens
 1. Overview of Simcenter Nastran

Modular Structure and DMAP

The Simcenter Nastran software is composed of a large number of building blocks called modules. Each
module is a collection of FORTRAN subroutines which is designed to perform a specific task, such as
processing model geometry, assembling matrices, applying constraints, solving matrix problems, and
calculating output quantities.

Within Simcenter Nastran, the modules are controlled by an internal language called the Direct Matrix
Abstraction Program (DMAP), which is a high-level programming language with its own compiler and
grammatical rules. A DMAP statement is like a subroutine call statement within FORTRAN, with both
input and output information.

Pre-defined Solution Sequences

Each type of analysis available in Simcenter Nastran is called a “solution sequence.” Each solution
sequence is a pre-defined collection of hundreds or thousands of DMAP commands. Once you select
a solution sequence, its particular set of DMAP commands sends instructions to the modules that are
needed to perform the requested solution.

Simcenter Nastran also includes DMAP (Direct Matrix Abstraction Programming), a high-level
programming language that allows you to build custom solution sequences or modify existing ones.

Using DMAP for Programmability

Although Simcenter Nastran contains a number of pre-defined solutions, you can use special DMAP
“alters” to either modify existing solution sequences or create new ones.

See also

• Simcenter Nastran DMAP Programmer's Guide

Overview of the Simcenter Nastran Analysis Process

To solve a finite element model in Simcenter Nastran, you must:

• Create the input file

• Use the nastran command to submit the input file for batch processing

• Examine your results

Creating the Input File

To perform an analysis with Simcenter Nastran, you must generate an input file that contains model
data (information about the finite element model, including geometry, elements, materials, and loads)

1-2 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Input File Sections

as well as analysis data (information about the type of analysis you want to perform, such as the analysis
type and the type of data you want output).

In Simcenter Nastran, the input file is an ASCII text file that you can create:

• Automatically using a preprocessor, such as Pre/Post

• Manually using a text editor

Input File Sections

A Simcenter Nastran input file contains the following five separate sections:

• NASTRAN statement

• File Management statement (FMS)

• Executive Control statement

• Case Control command

• Bulk Data entry

Documentation for Simcenter Nastran Input Options

Each section in the Simcenter Nastran input file is described in more detail later in this guide.
Comprehensive documentation for each File Management and Executive Control statement, Case
Control command, and Bulk Data entry is located in the Simcenter Nastran Quick Reference Guide.

Input File Naming

In general, Nastran input files are given a .dat extension. However, Simcenter Nastran tries to process
any file you submit, regardless of its file extensions. Other common Simcenter Nastran input file
extensions include .bdf or .blk.

Submitting the Input File for Analysis

Once you have created the input file, you submit the file for execution as a batch process in Simcenter
Nastran. Because Simcenter Nastran is not an interactive program, once you submit the input file, you
will not have any additional interaction with the software until the job completes, unless you need to
terminate the job.

To execute Simcenter Nastran, you typically use the nastran system command (in some cases, your
system administrator may assign a different name to the command) followed by the name of the input
file. For example:

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 1-3

© 2024 Siemens
 1. Overview of Simcenter Nastran

NASTRAN MODEL1A

The nastran command also allows you to specify certain keywords for controlling the job execution. The
format of the nastran command is:

nastran keyword1 = value1 keyword2 = value2 ...

See also

• “Using the nastran Command,” in the Simcenter Nastran Installation and Operations Guide

• “nastran Command and NASTRAN Statement,” in the Simcenter Nastran Quick Reference Guide

Reviewing the Results

When Simcenter Nastran completes the analysis, you can review your results. Simcenter Nastran
automatically creates a number of different types of files, which are summarized in the following table:

File type Description

.DBALL Contains permanent data for database runs for restart analyses
.f04 Contains database file information and a module execution summary, such as
the start and stop time for each module as well as the size of the database file
(if any)
.f06 Contains the results of your analysis, such as displacements and stresses, as
well as any diagnostic messages
.LOG Contains system information and any system level error messages
.MASTER Contains permanent data for database runs for restart analyses

In addition to these automatically generated files, you can manually request that Simcenter Nastran
create the following files:

File type Description

.pch Punch file
.plt Plot file
.op2 Output2 file
.xdb XDB file

Simcenter Nastran provides for direct support of interfaces to other products, such as Simcenter Nastran
Access.

For MSC.Patran, the support is provided through the DBC module, which creates a “graphics” database.
The interface to the DBC module is described in “POST,” in the Simcenter Nastran Quick Reference Guide.

1-4 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Reviewing the Results

The graphics database created by the DBC module can also be read by the Simcenter Nastran Access
library of object routines. This object library can be linked with a user-created program that extracts data
from the database. This is an indirect method for outside vendors to obtain Simcenter Nastran model
and results information. See the Simcenter Nastran Access Reference Manual for more information.

Communication with other software packages, such as I-DEAS Master FEM, is through the Simcenter
Nastran OUTPUT2 module, which creates a FORTRAN readable file that is converted by these outside
programs to their own particular data formats. The user interface and the supported data for these
outside programs are described in “POST,” in the Simcenter Nastran Quick Reference Guide (-1, -2, -4,
and -5).

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 1-5

© 2024 Siemens
 1. Overview of Simcenter Nastran

1-6 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 2. Structure and Syntax for Input Data
Overview of Specifying Input Data
In Simcenter Nastran, you submit all your data to the solver through the input file, which is sometimes
referred to an input “deck.” In most cases, you will automatically generate your Simcenter Nastran
input file by exporting an existing finite element model from a graphical pre-processing program.
However, because you may need to manually edit the contents of that input file, it is important that
you understand the rules and conventions of entering data into a Simcenter Nastran input file.

This chapter describes:

• The general syntax rules for input data in Simcenter Nastran

• Simcenter Nastran input records, including the different input data formats (free field, small field, and
large field)

• How to use the INCLUDE option to include external files in your Simcenter Nastran input file

• How to add comments to your Simcenter Nastran input file

The Simcenter Nastran Quick Reference Guide describes the input requirements and default values for
all Simcenter Nastran data file input.

General Input File Syntax Rules

The following sections provide the general rules and conventions for inputting data in Simcenter
Nastran.

Understanding Types of Input Data

In Nastran, you can input integer or real numbers, as well as character data.

• Integer numbers cannot contain a decimal point.

• Real numbers, including zero, must contain a decimal point.

You can enter real numbers in a variety of formats. For example, the following are all acceptable
versions of the real number, seven:

7.0 .7E1 0.7+1

.70+1 7.E+0 70.-1

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 2-1

© 2024 Siemens
 2. Structure and Syntax for Input Data

• Character data (sometimes called literal, or BCD—binary coded decimal data) can be alphanumeric,
but it must always begin with an alpha character and contain eight or fewer total characters.

Importantly, the different types of data are not interchangeable. If a particular field requires an integer
and you enter a real number instead, Simcenter Nastran issues an error message.

Using Consistent Unit Systems

Simcenter Nastran is unitless. Accordingly, the units for physical quantities defining the geometry,
material properties, and boundary conditions of a Simcenter Nastran model must be consistent. Because
Simcenter Nastran cannot detect inconsistent units, a warning message is not issued when inconsistent
units are used. The software simply calculates erroneous results.

As an example, suppose you want to construct a structural model with gram, millimeter, and second
as the base units. What units should you use in the Simcenter Nastran input file for force, density, and
elastic modulus?

Table 2-2 demonstrates how to determine the units for force, density, and elastic modulus that are
consistent with the base units of mm (millimeter), g (gram), and s (second).

Table. Determining Consistent Units

Physical Quantity Dimensions Consistent Unit

Force M·L/T2 g·mm/s2
Density M/L3 g/mm3
Elastic modulus M/L·T2 g/mm·s2
M, L, and T refer to mass, length, and time, respectively.

Now suppose the model includes an applied force of 5000 N, and a material having a density of 8000
kg/m3 and an elastic modulus of 200 GPa. What numerical values should you use in the Simcenter
Nastran input file?

You must convert N, kg/m3, and GPa to g·mm/s2, g/mm3, and g/mm·s2, respectively. Table 2-3 shows
how to calculate the appropriate conversion factors.

Table. Determining Conversion Factors

Physical Quantity Conversion Factor

Force 1 g·mm/s2 = (10-3 kg)(10-3 m)/s2 = 10-6 kg·m/s2 = 10-6 N
Density 1 g/mm3 = (10-3 kg)/(10-3 m)3 = 106 kg/m3
Elastic modulus 1 g/mm·s2 = (10-3 kg)/(10-3 m)s2 = 1 kg/m·s2 = 1 N/m2 = 1 Pa = 10-9 GPa

Using the conversion factors in Table 2-3 you can calculate the numerical values to use in the Simcenter
Nastran input file as follows:

2-2 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Using Consistent Unit Systems

• 5000 N = 5000 N (1 g·mm/s2/10-6 N) = 5 x 109 g·mm/s2 → Use 5.E9 in the input file.

• 8000 kg/m3 = 8000 kg/m3 (1 g/mm3/106 kg/m3) = 8 x 10-3 g/mm3 → Use 8.E-3 in the input file.

• 200 GPa = 200 GPa (1 g/mm·s2/10-9 GPa) = 2 x 1011 g/mm·s2 → Use 2.E11 in the input file.

Similarly, the results that Simcenter Nastran produces are in units consistent with the specified set of
base units. For example, suppose that Simcenter Nastran outputs the stress at a particular grid to be
1.E8 in the above example. You would interpret this result to be 1 x 108 g/mm·s2.

When solving mechanics problems, mass, length, and time are usually selected as the base units.
However, when solving mechanics problems using the finite element method, you typically use force,
length, and time as the base units. Force, length, and time are convenient to use because you typically
know:

• The physical dimensions of the component you are modeling.

• The forces that act on the component you are modeling.

Table 2-4 lists five sets of consistent units that are commonly used in structural analysis.

For information regarding consistent units for thermal analysis, see Consistent Units for Thermal
Analysis in the Thermal Analysis User’s Guide.

Table. Consistent Sets of Units for Structural Analysis – Base Units of Force, Length, and Time

Physical Quantity Dimensi English English SI SI SI

ons lbf-in-s lbf-ft-s mN-mm-s N-mm-s N-m-s

Base units
Force F lbf lbf mN N N
Length L in ft mm mm m
Time T s s s s s
Consistent units for typical inputs
Acceleration L/T2 in/s2 ft/s2 mm/s2 mm/s2 m/s2
Angular acceleration 1/T2 rad/s2 rad/s2 rad/s2 rad/s2 rad/s2
Angular velocity 1/T rad/s rad/s rad/s rad/s rad/s
Coordinate L in ft mm mm m
Density F·T2/L4 lbf·s2/in4 slug/ft3 kg/mm3 tonne/mm3 kg/m3
Displacement L in ft mm mm m
Elastic modulus F/L2 psi lbf/ft2 kPa MPa Pa
Force F lbf lbf mN N N

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 2-3

© 2024 Siemens
 2. Structure and Syntax for Input Data

Physical Quantity Dimensi English English SI SI SI

ons lbf-in-s lbf-ft-s mN-mm-s N-mm-s N-m-s

Mass F·T2/L lbf·s2/in slug kg tonne kg

Moment F·L lbf·in lbf·ft mN·mm N·mm N·m
Pressure F/L2 psi lbf/ft2 kPa MPa Pa
Velocity L/T in/s ft/s mm/s mm/s m/s
Consistent units for typical results
Displacement L in ft mm mm m
Force F lbf lbf mN N N
Strain energy F·L lbf·in lbf·ft μJ mJ J
Stress F/L2 psi lbf/ft2 kPa MPa Pa
1 slug = 1 lbf·s2/ft; 1 tonne = 1000 kg; 1 psi = 1 lbf/in2; 1 Pa = 1 N/m2; 1 J = 1 N·m
F, L, and T refer to force, length, and time, respectively.

Specifying Valid Input File Characters

Table 2-5 lists the valid characters for Simcenter Nastran.

Table. ASCII Subset for Simcenter Nastran

Character ASCII (HEX) Character ASCII (HEX)

HT (Horizontal Tabulation) 09 @ 40
A 41
SP (Space “blank”) 20 B 42
C 43
! 21 D 44
" 22 E 45
# 23 F 46
$ 24 G 47
% 25 H 48
& 26 I 49
‘ 27 J 4A
( 28 K 4B
) 29 L 4C
* 2A M 4D
+ 2B N 4E
' 2C O 4F

2-4 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Understanding Simcenter Nastran Input Records

Character ASCII (HEX) Character ASCII (HEX)

- 2D P 50
. 2E Q 51
/ 2F R 52
0 30 S 53
1 31 T 54
2 32 U 55
3 33 V 56
4 34 W 57
5 35 X 58
6 36 Y 59
7 37 Z 5A
8 38 [ 5B
9 39 \ 5C
: 3A ] 5D
; 3B ^ 5E
< 3C - 5F
= 3D
> 3E
? 3F

Understanding Simcenter Nastran Input Records

Input data records are 80 characters (or columns) in length on each line. Those records are then divided
into fields. How those records are divided depends upon the data format.

There are two basic categories of input data formats in Simcenter Nastran:

• “Free” format data, in which the data fields are simply separated by commas. This type of data is
known as free field data.

• “Fixed” format data, in which your data must be aligned in columns of specific width. There are two
subcategories of fixed format data that differ based on the size of the fixed column width:

• Small field format, in which a single line of data is divided into 10 fields that can contain eight
characters each.

• Large field format, in which a single line of input is expanded into two lines The first and last fields
on each line are eight columns wide, while the intermediate fields are sixteen columns wide. The
large field format is useful when you need greater numerical accuracy.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 2-5

© 2024 Siemens
 2. Structure and Syntax for Input Data

Which type of format data you use depends upon the section of the input file that you are working with.

• You must use the free field format for the NASTRAN statement and the File Management, Executive
Control, and Case Control sections.

• You can use the free field, small field, or large field formats with the Bulk Data section.

The rules for entering free field, small field, and large field format data are described in more detail
below.

Understanding Free Field Format

In free field format, you separate data fields by commas.

Free Field Format Rules

• You must begin free field data entries in column 1.

• You can use commas to skip fields.

• To skip one field, use two commas in succession.

• To skip two fields, use three commas in succession (and so on).

• Integer or character data is limited to eight characters. If you include an integer or character longer
than eight characters, Simcenter Nastran issues a fatal error.

• With free field format data, Simcenter Nastran rounds real numbers with more than eight characters.
Therefore, you may lose some precision. For example, an entry of 1.2345678+2 becomes 123.4568.
If you need more significant digits, use the large field format.

• Free field data cannot contain embedded blanks. An example of a free field embedded blank is shown
below:

• You use a dollar sign ($) to terminate the free field entry. You can follow the dollar sign with
comments.

Free Field Format Example

The following example shows the GRID Bulk Data entry in free field format:

2-6 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Understanding Small Field Format

Understanding Small Field Format

With the small field format, the software separates an 80–column line of Bulk Data entry into 10 equal
fields of eight characters each:

1 2 3 4 5 6 7 8 9 10
<-8 Char-> <-8 Char-> <-8 Char-> <-8 Char-> <-8 Char-> <-8 Char-> <-8 Char-> <-8 Char-> <-8 Char-> <-8 Char->

Small Field Format Rules

• Fields 1 and 10 must be left justified.

• Fields 2 through 9 do not need to be either right- or left-justified. However, aligning the data fields is a
good practice.

• You can't include any embedded blanks in small field format input data. An example of a small field
embedded blank is shown below:

Small Field Format Example

The following example shows the GRID entry in small field format:

1 2 3 4 5 6 7 8 9 10
GRID 2 1.0 -2.0 3.0 136

Understanding Large Field Format

A high degree of numerical accuracy is required in some Simcenter Nastran applications. You can use
large field format when small field format doesn't provide enough significant digits. For example, you
can use large field format when you need to input integers larger than eight characters, or you need to
enter real numbers with more significant digits than you can fit in an eight-character field.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 2-7

© 2024 Siemens
 2. Structure and Syntax for Input Data

Note: A minus sign, decimal point, and the “E” in scientific notation count as characters.

Large Field Format Rules

Large field format requires (at least) two lines for each entry:

• The first and last field of each line contains eight columns.

• The fields in between contain 16 columns.

Large field entries are denoted by an asterisk (*). The asterisk must:

• Immediately follow the character string in field 1A of the first line of the entry

• Immediately precede the character string in field 1B of the second line of the entry

Large Field Format Example

The following is an example of the GRID Bulk Data entry in large field format:

First Line: (Left half of single field)

Second Line: (Right half of single field)

Specifying Continuations for Bulk Data

Some Bulk Data entries require more than eight fields (72 columns) of data. In those cases, you need to
specify a continuation to continue the remainder of the data on one or more subsequent lines. To do
this, you follow the parent entry (the first line) with one or more continuation entries on subsequent
lines.

2-8 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Manually Specifying a Continuation with Identifiers

There are two ways to specify a continuation:

• Manually

• Automatically

Manually Specifying a Continuation with Identifiers

You can manually specify a continuation by using a continuation identifier. A continuation identifier
is a special character (+ or *) that indicates that the data continues on another line. Once you've
specified the continuation identifier, you then follow the parent entry (the first line) with one or more
continuation entries on subsequent lines.

Rules for Manually Specifying a Continuation

• You must use a + or * in column 1, field 1 of a continuation entry.

• The alphanumeric character strings you use as the continuation identifiers cannot contain the symbols
*, =, or $.

Note: The + symbol used in column 1 of field 10 is for clarity.

• Field 10 contains the continuation identifier that must be unique with respect to all other
continuation identifiers.

• Simcenter Nastran always ignores column 1 of field 10. It is not considered to be part of the identifier.

• The remaining contents in columns 2-8, field 1 of a continuation entry must be identical to the entry
in field 10 (columns 2-8) of either the parent entry or the preceding continuation entry.

• You can use small and large field continuation entries together to define a single data item entry.

Manual Continuation Example

For example, consider the following PBAR simple beam property entry:

Standard PBAR Entry

1 2 3 4 5 6 7 8 9 10
PBAR PID MID A I1 I2 J NSM
C1 C2 D1 D2 El E2 F1 F2
K1 K2 I12

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 2-9

© 2024 Siemens
 2. Structure and Syntax for Input Data

Continuation Example

PBAR 39 6 2.9 1.86 2.92 .48 +PB1

+PB1 0. 0. 0. 1. 1. 1. 1. 0. +PB2
+PB2 .86 .86

In this example:

• +PB1 in field 10 of the parent entry is an arbitrary (and unique) user-defined pointer to field 1 of the
second line of the entry.

• +PB2 in field 10 of the second line points to the third line.

Automatically Specifying a Continuation

The automatic continuation method allows you to avoid the continuation identifiers.

To have Simcenter Nastran automatically generate continuations:

• Leave both field 10 of the parent entry and field 1 of the continuation line blank.

• Place the continuation line directly following the parent entry.

The blank field indicates that it is a small-field continuation of the previous line.

You can mix small-field and large-field format continuations. However, it is generally not recommended
because it becomes difficult to discern the locations of the fields.

Rules for Automatically Generating Continuations

To have Simcenter Nastran automatically generate a continuation:

• The continuation line(s) must immediately follow the parent Bulk Data entry.

• You must leave fields 1 and 10 of the continuation line (or lines) blank.

• Your bulk data must be in sorted order.

Note: Setting SYSTEM CELL 357 to 1 tells Simcenter Nastran to ignore the continuation field. This
eliminates the concern of duplicate continuation identifiers.

Automatic Continuation Example

This process is illustrated in the following example:

2-10 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Replicating Bulk Data Entries

CHEXA, 1, 10, 3, 5, 7, 1, 15, 17,

, 19, 13, 4, 6, 8, 2, 10, 11,
, 12, 9, 16, 18, 20, 14

Input (.DAT) file

GEAR TOOTH EXAMPLE JULY 10, 2003 Simcenter Nastran

07/10/2003 PAGE 3

S O R T E D B U L K D A T A E C H
O

ENTRY

COUNT . 1 .. 2 .. 3 .. 4 .. 5 .. 6 .. 7 ..
8 .. 9 .. 10 .
1- CHEXA 1 10 3 5 7 1 15
17 +000001
2- ++00000119 13 4 6 8 2 10
11 +000002
3- ++00000212 9 16 18 20
14 +000003

Output (.f06) file

Note that column 1 of field 10 contains a + symbol. Since Simcenter Nastran ignores this column, the
choice of the + is arbitrary. However, using the + symbol improves readability and serves as a reminder
for the format of the continuation line.

Replicating Bulk Data Entries

To avoid having to input each bulk data entry individually, you can replicate the repetitive fields from
a single entry definition. You can use replication with bulk data that is in fixed-field or free-field format
for small field (that is, 8 character) input. Large field (that is, 16 character) input is not supported for
either fixed-field or free-field format, except in the case of bulk entries that have only the first four fields
specified. An example of this is GRID entries with the X3 and subsequent fields left blank.

• To replicate selected fields from the preceding entry, use the symbol =.

• To replicate all the remaining entries from the preceding entry, use the symbol ==.

• To generate an incremented value from the previous entry, use *x or *(x), where x is the value of the
increment. “x” should be a real number for real fields or an integer for integer fields.

• To repeatedly replicate a field, use =n or =(n), where n is the number of images to be generated using
the values of the increments on the preceding entry.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 2-11

© 2024 Siemens
 2. Structure and Syntax for Input Data

Replicating Continuation Entries

If you are manually creating continuation entries, you can replicate the continuation fields (fields one
and ten) using the following conventions:

• You can use only letters of the alphabet and integers. They are coded into a base 36 number. That is,
the sequence of numbers is 0,1, 2, ..., 8, 9, A, B, ...

• The software increments the continuation field by +1 regardless of the value you specify.

• The software does not increment the first character in fields 1 or 10.

• The software will not ever increase the number of characters in an incremented field. For example, if
the first field is “0", the thirty-seventh field will also be “0", resulting in an illegal entry. A method to
solve this problem is to start with a first field of “00". This provides thirty-six squared unique fields.

• At least one field in fields 2 through 8 of continuation entries must contain data.

Replication with MSGMESH

Replication is a limited data generation capability. You can use replication with bulk data that is either
fixed (large or small field) or free field format.

• The MSGMESH capability includes the capabilities described here, plus the following capabilities as
long as NASTRAN MESH is specified in the File Management Section.

• Continuation entry fields may be incremented or decremented.

• Repeated replication is indicated by coding =(n) in field 1, where n is number of entry images to be
generated using the values of increments from the current or preceding replication entry.

Entered entries:

GRID,101 ,17,1 .0,10.5,,17,3456

=(4), *(1),=,*(0.2),==$

Generated entries:

GRID 101 17 1.0 10.5 17 3456

GRID 102 17 1.2 10.5 17 3456
GRID 103 17 1.4 10.5 17 3456
GRID 104 17 1.6 10.5 17 3456
GRID 105 17 1.8 10.5 17 3456

2-12 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Replication with MSGMESH

• A blank in field 1 indicates immediate continuation entry replication. The default continuation entry
increment is 1. Example:

BSET1,123,1,2,3,4,5,6,7
,,*7,*7,*7,*7,*7,*7,*7
=(3)

Generated entries:

BSET1 123 1 2 3 4 5 6 7 +00001

++00001 8 9 10 11 12 13 14 +00002
++00002 15 16 17 18 19 20 21 +00003
++00003 22 23 24 25 26 27 28 +00004
++00004 29 30 31 32 33 34 35 +00005

• A “=(D)” in field 1 indicates delayed continuation entry replication. A maximum of nine entries may
be replicated as a group. The default continuation entry increment is 10. Example:

Entered entries:

CTRIA3,10,1,1,10,11/+C1
=(D),*(1),=,=,*(1),*(1)/*(20)
+C1,,,2.0,1.0,1.0
=(2),==

Generated entries:

CTRIA3 10 1 1 10 11 +C1
+C1 2.0 1.0 1.0
CTRIA3 11 1 1 11 12 +C21
+C21 2.0 1.0 1.0
CTRIA3 12 1 1 12 13 +C41
+C41 2.0 1.0 1.0

• Parentheses are optional on replication entries and an equal sign may replace an asterisk.

The following is an example of the use of replication, automatic continuation field generation, and
the free field format:

GRID,101 ,17,1.0,10.5,,17,3456
=,*1,=,*0.2, *(0.1), == $ COMMENTS MAY APPEAR AFTER $
=3
EIGR,13,LAN

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 2-13

© 2024 Siemens
 2. Structure and Syntax for Input Data

CBAR,1 ,1 ,101 ,102,0.,0.,1.,,+0

=,*1,=,*1,*1====*1
+0,56
*1,=$

The above free-field entries generate the following Bulk Data in the 8-column format, as seen in the
SORTED BULK DATA ECHO:

Note: A “,” should always be used after the “*1” for the continuation increment even if fixed field
format is being used.

CBAR 1 1 101 102 0. 0. 1. +0

+0 56
CBAR 2 1 102 103 0. 0. 1. +1
+1 56
EIGR 13 LAN
GRID 101 17 1.0 10.5 17 3456
GRID 102 17 1.2 10.6 17 3456
GRID 103 17 1.4 10.7 17 3456
GRID 104 17 1.6 10.8 17 3456
GRID 105 17 1.8 10.9 17 3456

The automatically generated continuation entries start with the number 1, are incremented by 1, and
are padded with zeros and plus signs as shown above. If this feature is used, it is the user's responsibility
not to enter continuation entries that also use this convention. In particular, data generated on another
run and then written to the PUNCH file with the ECHO=PUNCH, will cause problems when introduced
into other data with blank continuation fields.

Inserting External Files with INCLUDE

You can use the INCLUDE statement to insert external files into your Simcenter Nastran input file.
You can specify INCLUDE in any part of the input file as long as the data referenced by the INCLUDE
statement is appropriate for that section of the input file. This allows you to avoid repetitious input in the
different sections of the input file.

Using INCLUDE in Different Input File Sections

You can use INCLUDE in a variety of ways in the different sections of your input file.

• In the File Management section, you can use the INCLUDE statement to assign multiple database files
with one statement.

• In the Executive section, you can use the INCLUDE statement to include alters to a solution sequence.

2-14 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 INCLUDE Statement Limitations

• In the Case Control section, the INCLUDE file can be useful if you want to use the same subcase
structure and/or the same output requests for different models.

• In the Bulk Data section, the INCLUDE statement is very useful for working with superelement data.
You can use it to place each superelement of a superelement model in a different physical file. You
can then share the files with other users and analyze them separately. Since superelement models are
usually very large and complex, using the INCLUDE statement can greatly reduce the complexity of
the input file.

INCLUDE Statement Limitations

• Although you can nest INCLUDE statements within the external file, the nested depth level must not
be greater than 10.

• Each line has a 72 character limit. Multiple lines can be used when file names are long. The entire
string must begin and end with the single quote character.

For example,

D:\folder1\folder2\folder3\folder4\folder5\folder6\folder7\folder8\folder9\mydata.dat

can be defined on multiple lines:

INCLUDE
‘D:\folder1\folder2\folder3\folder4\folder5\folder6\folder7\fold
er8\folder9\mydata.dat'

The following input format is also supported:

INCLUDE ‘D:\folder1\folder2\folder3\
folder4\folder5\
folder6\folder7\
folder8\folder9\mydata.dat'

Example: Using the INCLUDE Statement

Below is an example of the use of the INCLUDE statement. Note that the syntax for the filenames is
machine dependent. The format of lowercase filenames within single quotes works on all computer
operating systems that support Simcenter Nastran.

INCLUDE 'nastran.dat'
INCLUDE 'database.dat'
SOL 101
INCLUDE 'sol101.alter'
CEND
TITLE = Complete Model
SUBTITLE = Three load cases

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 2-15

© 2024 Siemens
 2. Structure and Syntax for Input Data

INCLUDE 'output.dat'
SUBCASE 1
LOAD = 10
BEGIN BULK
INCLUDE 'se0.dat'
INCLUDE 'se10.dat'
INCLUDE 'se20.dat'
ENDDATA

Each of the INCLUDE files in the example contains different information. For this example, “nastran.dat”
contains NASTRAN system cells, “database.dat” contains FMS statements, “sol101.alter” contains a DMAP
Alter for SOL 101, “output.dat” contains Case Control commands, and “se*.dat” contains modeling
information for superelements 0, 10, and 20.

See also

• “INCLUDE,” in the Simcenter Nastran Quick Reference Guide

Including Comments in the Input File

You can insert comments in any part input file to help organize your input data. Comment entries begin
with a dollar ($) sign in column one followed by any characters out to column 80. Comments appear
only in the unsorted echo of the input file listing.

2-16 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 3. Understanding the Input Data File
Overview of the Simcenter Nastran Input File Sections
The input file is comprised of five distinct sections and three required one-line delimiter which must
follow the sequence shown in Figure 3-1. These delimiters and sections are described below.

Note: If any of the required delimiters are missing from your input file, Simcenter Nastran issues a fatal
error message when you submit your job.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 3-1

© 2024 Siemens
 3. Understanding the Input Data File

Figure 3-1. Simcenter Nastran Input File Structure

Example: Using Simcenter Nastran Input Files to Model a Truss

Throughout this chapter, a simple truss structure, shown in Figure 3-2, is used to illustrate different
aspects of input files.

3-2 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Example: Using Simcenter Nastran Input Files to Model a Truss

Figure 3-2. Truss Example

The truss model consists of four grid points that represent the joints of the structure. The locations of
these grid points are given in the default coordinate system for Simcenter Nastran, which is known as
the “basic coordinate system.” In this example, the origin is located at grid point 1.

The truss structure consists of five members, each with a cross-sectional area (A) of 4 in2 and a polar
moment of inertia (J) of 1.27 in4. The material is steel with a Young's modulus E of 30 × 106 psi and a
Poisson's ratio of 0.3. All the connections between the members are pin joints (i.e., they can transmit
forces but not moments). The goal of the analysis is to determine the displacement of the grid points
and the stresses and forces within the members.

Although the dimensions given in Figure 3-2 are in feet, the dimensions used for the grid point
locations in the input file (Listing 3-1) are in inches. For consistency, the units for the cross-sectional
area, torsional stiffness, and Young's modulus in the input file are also provided in inches.

Note: You must always ensure that you enter data in a consistent system of units.

Listing 3-1 shows the input file for the truss structure. Although this input file represents a fairly simple
structure, it is typical of all Simcenter Nastran input files.

$ FILENAME - TRUSS1.DAT Executive Control

$ section
ID LINEAR,TRUSS1
SOL 101
TIME 2
CEND
TITLE = LINEAR STATICS USER'S SAMPLE INPUT FILE Case Control section
SUBTITLE = TRUSS STRUCTURE
LABEL = POINT LOAD AT GRID POINT 4

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 3-3

© 2024 Siemens
 3. Understanding the Input Data File

LOAD = 10
SPC = 11
DISPLACEMENT = ALL
ELFORCE = ALL
ELSTRESS = ALL
BEGIN BULK Bulk Data section
$
$ THE GRID POINTS LOCATIONS
$ DESCRIBE THE GEOMETRY
$
GRID 1 0. 0. 0.
3456
GRID 2 0. 120. 0.
3456
GRID 3 600. 120. 0.
3456
GRID 4 600. 0. 0.
3456
$
$ MEMBERS ARE MODELED USING
$ ROD ELEMENTS
$
CROD 1 21 2 3
CROD 2 21 2 4
CROD 3 21 1 3
CROD 4 21 1 4
CROD 5 21 3 4
$
$ PROPERTIES OF ROD ELEMENTS
$
PROD 21 22 4. 1.27
$
$ MATERIAL PROPERTIES
$
MAT1 22 30.E6 .3
$
$ POINT LOAD
$
FORCE 10 4 1000. 0. -1. 0.
$
SPC1 11 123456 1 2
$
ENDDATA

Listing 3-1. Truss Example Input File

3-4 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 The NASTRAN Statement (Optional)

The NASTRAN Statement (Optional)

The NASTRAN statement lets you modify the default values for certain operational parameters (also
called system cells) at runtime to control internal solution processing or provide specific diagnostics. For
example, you can use the NASTRAN statement to control aspects of working memory, datablock size,
and datablock parameters.

• You can include more than one NASTRAN statement in a single input file.

• You can combine the NASTRAN statement with the File Management section.

• You can specify the NASTRAN statement in the Runtime Configuration (RC) files at the system, user,
and job level.

See also

• “Using the NASTRAN Statement” in the Simcenter Nastran Installation and Operations Guide

• “nastran Command and NASTRAN Statement” in the Simcenter Nastran Quick Reference Guide

• “BUFFSIZE” in the Simcenter Nastran User's Guide

File Management Section (Optional)

The File Management section (FMS) of the input file is optional and typically follows the NASTRAN
statement(s). It ends with the specification of an Executive Control statement. For example, the File
Management section lets you:

• Assign files

• Attach or initialize the Simcenter Nastran database sets (DBsets) and FORTRAN files

• Perform restart analyses

The initialization of a database includes specification of its maximum size, member names, and physical
filenames. The initialization of a FORTRAN file includes the specification of its filename, FORTRAN unit
numbers, and FORTRAN attributes.

The File Management section is commonly used in input files for larger analysis problems. For many
Simcenter Nastran problems, no File Management statements are required because a default File
Management section is executed by the software at the beginning of every run. For example, since
the truss model shown in the listing Truss Example Input File is small and the desired analysis is not a
restart, a File Management section isn't included in the input file.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 3-5

© 2024 Siemens
 3. Understanding the Input Data File

See also

• “File Management Statements” in the Simcenter Nastran Quick Reference Guide

• “Database Concepts” in the Simcenter Nastran User's Guide

• “Restarts” in the Simcenter Nastran User's Guide

Executive Control Section (Required)

The Executive Control section is the first required section in the Simcenter Nastran input file and provides
the overall control for the solution. For example, statements in the Executive Control section let you:

• Specify an optional ID statement to identify the job

• Select the type of analysis (solution sequence) you want to perform

• Specify an optional TIME statement to set limits on the maximum allowable CPU time for the run

• Request various diagnostics

• Include complete DMAP sequences, if you want to use DMAP in the analysis

For example, the Executive Control section for the truss example (the listing Truss Example Input File)
identifies the job, requests the static Structured Solution Sequence 101, and sets the maximum time
limit for the job to two minutes.

For the truss example, the Executive Control section specifies static solution SOL 101 and a TIME
statement that specifies a maximum of 2 CPU minutes for the run. The end of the Executive section is
denoted by the CEND statement. The ID statement is an optional statement to help document your input
file. If used, it can be located anywhere in the Executive Control section.

See also

• Executive Control Statements in the Simcenter Nastran Quick Reference Guide

Executive Control Format

The format of the Executive Control statements is free-field format between columns 1 and 72. The
statements may begin in any column (except for comments). The operands—that is, the input data
following the name—are separated from the name using a comma and/or one or more blanks. If more
than one operand is needed, they are also separated by a comma and/or one or more blanks.

You must use the CEND delimiter to mark the end of the Executive Control section.

3-6 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Order Dependence for Certain DMAP-related Statements

Order Dependence for Certain DMAP-related Statements

Most statements within the Executive Control section are order independent. However, the following
statements, all related to the use of DMAP, are order dependent:

• COMPILE

• COMPILER

• ALTER

• ENDALTER

• LINK

Notably:

• The LINK statement must appear after all COMPILE statements.

• The COMPILER statement (or equivalent DIAGs) must appear before all COMPILE statements. This
statement sets the default print control for subsequent COMPILE statements.

Commonly Used Executive Control Statements

Some commonly used Executive Control statements are:

• ID

• SOL

• ECHOON and ECHOOFF

• TIME

• DIAG

• CEND

The ID Statement (Optional)

You can use the ID statement to identify your job. Its format is:

ID i1, i2

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 3-7

© 2024 Siemens
 3. Understanding the Input Data File

where i1 and i2 are character strings. i1 may be one to eight characters in length. i2 may be of any
length. The first character of each string must be alphabetic.

You can use only one ID statement, although you can specify it anywhere in the Executive Control
section The format of the ID statement is the keyword “ID” followed by a delimiter and a comment. The
comment may be any ASCII characters. For example:

ID THIS IS RUN 2

See also

• “ID” in the Simcenter Nastran Quick Reference Guide

The SOL Statement (Required)

The SOL statement specifies the type of analysis (solution sequence) you want to perform. The format of
the SOL statement is:

SOL n

where n is a positive integer identifying the solution type or the character name of the solution
procedure.

See also

• “Understanding Solution Sequences” in the Simcenter Nastran User's Guide

• “SOL” in the Simcenter Nastran Quick Reference Guide

• “SOL 402” in the Simcenter Nastran Quick Reference Guide

ECHO Statement (Optional)

The ECHO statement lets you control whether the software prints (“echoes”) the Executive Control
section to the output file. ECHOON is the default.

• Use ECHOON (default) to print the section to the output file.

• Use ECHOOFF to prevent the software from printing the section to the output file.

In general, using ECHOOFF hinders the usability of your output file, especially if you need to review the
output file several months after your original analysis. However, the ECHOOFF statement can be useful
for security. For example, you can use ECHOOFF, to prevent Simcenter Nastran from printing proprietary
portions of the Executive Control section (such as user-written DMAP) to the output file.

You can use multiple ECHOON and ECHOOFF statements in an single input file.

3-8 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Time Statement (Optional)

See also

• “ECHO” in the Simcenter Nastran Quick Reference Guide

Time Statement (Optional)

The TIME statement allows you to set the maximum CPU time and I/O time for the job before Simcenter
Nastran terminates the run. Its format is

TIME t1, t2

where:

• t1 is the maximum allowed execution time in CPU minutes (real or integer; default = one minute,
which is adequate only for very small jobs).

• t2 is the maximum allowable I/O time in seconds (default is infinity).

Importantly, the time limits you can set with the TIME statement are different from the system time
limit imposed by your system administrator. In general, you should use the TIME statement to set the
Simcenter Nastran time limit to a value less than the system time limit. If your job exceeds the system
time limit during execution, the system aborts the job, which may corrupt the database. Once the
database is corrupted, it is difficult, if not impossible, to restart the job.

However if the required execution time exceeds the time specified on the Simcenter Nastran TIME
statement, Simcenter Nastran cleanly terminates the job. This allows you to restart the job later. Also,
Simcenter Nastran estimates the CPU time required to perform certain operations during execution.
If the time remaining is not sufficient to complete a particular operation, the job terminates without
wasting any additional computer resources.

See also

• “TIME” in the Simcenter Nastran Quick Reference Guide

DIAG Statement (Optional)

You can use the DIAG statement to request a variety of additional diagnostic information or output
and/or to modify the solution.

See also

• “DIAG” in the Simcenter Nastran Quick Reference Guide

• “Other Optional Diagnostic Information” in the Simcenter Nastran User's Guide

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 3-9

© 2024 Siemens
 3. Understanding the Input Data File

The CEND Statement (Required)

The CEND statement is a required statement that designates the end of the Executive Control section
(and the beginning of the Case Control section). Its format is simply:

CEND

For example:

ID SIMPLE,MODEL
SOL 101
TIME 5
CEND

See also

• “CEND” in the Simcenter Nastran Quick Reference Guide

Case Control Section (Required)

The Case Control section contains commands. It always follows the Executive Control section and
precedes the Bulk Data section. You use the Case Control section to:

• Define/select analysis subcases (multiple loadings in a single job execution)

• Define SETS to specify and control the type of analysis output produced (e.g., forces, stresses, and
displacements)

• Select output requests for printing, punching and plotting

• Define titles, subtitles and labels for documenting the analysis

For the truss example, a title, subtitle, and label are defined. These labels are printed on each page
of the .f06 output file. The LOAD = 10 command instructs Simcenter Nastran to apply the loading
defined by the FORCE entry with an ID of 10 in the Bulk Data section. The SPC = 11 command instructs
Simcenter Nastran to apply the constraints defined by the SPC1 entry with an ID of 11 in the Bulk Data
section. Printed displacements for all the grid points and the forces and stresses within each member are
requested with the DISPLACEMENT = ALL, FORCE = ALL, and STRESS = ALL commands. By default, these
output requests are printed to the .f06 output file.

See also

• “Case Control Commands” in the Simcenter Nastran Quick Reference Guide.

3-10 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Case Control Format

Case Control Format

The Case Control section must end with the BEGIN BULK delimiter.

Defining Subcases

In the case control section of an input file, you can create subcases. Subcases allow you to:

• Evaluate different combinations of loads, constraints, forcing frequencies, and so on for an FE model
from a single input file.

Without subcases, you would have to create and run a separate input file for each combination of
loads, constraints, forcing frequencies, and so on.

• Account for static preload in dynamic analysis and buckling analysis.

• Request output on a subcase-specific basis.

For example, you can obtain stress output from one subcase, but not from another.

You use the SUBCASE case control command to define subcases. Each SUBCASE case control command
must have a unique integer identification number such as:

SUBCASE 12

If you create an input file with multiple subcases, the identification numbers for each SUBCASE case
control command must be listed in ascending order. However, the identification numbers do not need
to be sequential. For example, identification numbers of 1, 14, 31, and 50 are valid, so long as they are
specified in that order.

If you do not create a subcase, the software assumes that the model contains a single subcase with an
identification number of 1.

Subcase Analysis Type

The type of analysis that the software performs at the subcase level is dictated by the SOL executive
control statement specification. For example:

• If SOL 101 is specified, the subcases are linear static subcases.

You can use the linear static subcases to examine various combinations of static loads and constraints.

• If SOL 103 is specified, the subcases are real eigenvalue subcases.

You can use the real eigenvalue subcases to evaluate different sets of eigenvalue extraction
parameters such as the frequency range over which the eigenvalues are extracted.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 3-11

© 2024 Siemens
 3. Understanding the Input Data File

• If SOL 108 is specified, the subcases are direct frequency response subcases.

You can use the direct frequency response subcases to examine various combinations of dynamic
loads and forcing frequencies.

The exception to this rule is when you use the STATSUB case control command to create dependent
subcases.

For more information, see Dependent Subcases.

Global versus Subcase Behavior

The case control commands that are listed prior to the first subcase in the input file are referred to as the
global case. The case control commands in the global case are inherited by all of the subcases. However,
if a subcase contains a case control command of the same name as one that is included in the global
case, the subcase specification of the case control command takes precedence. In other words, the two
case control specifications are not additive.

For example, suppose that an input file contains a LOAD specification in both the global case and a
subcase. For that subcase, the software uses the LOAD specification in the subcase and ignores the LOAD
specification in the global case.

Note:
Case control commands that are listed prior to the subcases in the input file are often referred to as
being above the subcase level.

With regard to global versus subcase behavior, single-point constraint sets (SPC) and multipoint
constraint sets (MPC) are treated as distinct constraint types. For example, if a single-point constraint set
is defined above the subcase level and a multipoint constraint set is defined in a subcase, the software
uses both the single-point constraint set and the multipoint constraint set for that subcase.

With regard to global versus subcase behavior, static load sets (LOAD) and dynamic load sets (DLOAD)
are also treated as distinct load types. However, subcases do not use both a static load set and a dynamic
load set. If the subcase is a static subcase, the software uses only a static load set. If the subcase is a
dynamic subcase, the software uses only the dynamic load set.

For example, if a static load set is defined above the subcase level and a dynamic load set is defined in a
subcase, if the subcase is a static subcase, the software uses the static load set and ignores the dynamic
load set. If the subcase is a dynamic subcase, the software uses the dynamic load set and ignores the
static load set.

Note:
For SOLs 401, 402, and 414, the behavior of the software may be different. For information on
static load sets and dynamic load sets in these solution sequences, see the Simcenter Nastran

3-12 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Defining Subcases

Multi-Step Nonlinear User's Guide (SOL 401 and SOL 402) and Simcenter Nastran SOL 414 Rotor
Dynamics User's Guide.

In the following example, the input file contains case control commands above the subcase level and in
the subcases. Assume that all three subcases are linear static subcases in a SOL 101 input file.

Case control commands above the subcase level

Case control commands within subcases

The following table shows the static load sets, single-point constraint sets, multipoint constraint sets,
and displacement output sets that the software uses for each subcase.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 3-13

© 2024 Siemens
 3. Understanding the Input Data File

Set Type Subcase 1 Subcase 2 Subcase 3

Static load LOAD=10 LOAD=10 LOAD=30
Single-point constraint SPC=25 SPC=25 SPC=25
Multipoint constraint MPC=5 MPC=5 MPC=15
Displacement output All grid points SET 35 (grid points 2, 4, and 6) SET 1 (grid points 1 through
50)

Dependent Subcases

When you use certain solution sequences to perform certain types of analysis, you can use the results
from one subcase in another subcase.

For example, suppose that you want to find the natural frequencies of a guitar string. You know that
the natural frequencies of the guitar string depend on the tension in the guitar string. To determine the
natural frequencies, you can create a SOL 103 input file that contains two subcases:

• In the first subcase, the software calculates the differential stiffness for the guitar string under tension.

• In the second subcase, the software adds the differential stiffness to the structural stiffness and then
calculates the natural frequencies.

In this example, you place a STATSUB(PRELOAD) case control command specification in the second
subcase that references the first subcase. The presence of the STATSUB(PRELOAD) specification directs
the software to:

1. Consider the first subcase as linear static and calculate the differential stiffness attributable to the
tension in the guitar string.

2. Add the differential stiffness to structural stiffness and compute the real modes in the second
subcase.

Dependent subcases are limited to specific solution sequences when you use them to perform
specific types of analysis. The details for the solution-specific dependent subcases are described in the
appropriate Simcenter Nastran user guide.

Defining Output

By default, Simcenter Nastran doesn't output any results. With Simcenter Nastran, you must use Case
Control commands to explicitly request the results you want. Simcenter Nastran output is requested in
the Case Control Section.

Echoing Bulk Data in the Output

A Simcenter Nastran model resides in a .DAT text file. The .DAT file is submitted to Simcenter Nastran and
an .f06 results file is produced.

3-14 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Echoing Bulk Data in the Output

The following Case Control command produces sorted and unsorted model file listings, called echoes, at
the beginning of the .f06 file:

ECHO=BOTH

The unsorted input file is an exact copy of the Executive Control, Case Control, and Bulk Data Sections
of the input (.DAT) file, including comment ($) entries. The sorted input file is a listing of the Bulk Data
Section with entries rearranged in alphabetical order and with comments removed. In addition, the
sorted Bulk Data is expanded to ten fields, each eight columns wide. Therefore, if the .DAT file is entered
in free field format, it appears in small field format in the sorted Bulk Data listing. This small field format
listing is especially helpful when reviewing the model's Bulk Data Section.

Examples of unsorted and sorted input listings are shown in Listing 3-2 and Listing 3-3.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 3-15

© 2024 Siemens
 3. Understanding the Input Data File

JULY 10, 2003 Simcenter Nastran

7/10/2003 PAGE 1
N A S T R A N E X E C U T I V E C O N T R O L D E C K E C H
O
ID MPM,EXAMPLE1
SOL
101
TIME
100
CEND

SIMPLY SUPPORTED BEAM JULY 10, 2003 Simcenter Nastran

07/10/2003 PAGE 2
WITH CONCENTRATED
FORCE
C A S E C O N T R O L D E C K E C H
O
CARD

COUNT

1
ECHO=BOTH
2
DISP=ALL
3
STRESS=ALL
4
FORCE=ALL
5
SPCF=ALL
6
SPC=100
7
LOAD=10
8 TITLE=SIMPLY SUPPORTED
BEAM
9 SUBTITLE=WITH CONCENTRATED
FORCE
10
$
11 BEGIN BULK

SIMPLY SUPPORTED BEAM JULY 10, 2001 Simcenter

Nastran 7/10/2003 PAGE 3
WITH CONCENTRATED
FORCE
I N P U T B U L K D A T A D E C K E
C H O
. 1 .. 2 .. 3 .. 4 .. 5 .. 6 .. 7 ..
8 .. 9 .. 10 .
$ DEFINE GRID
POINTS

3-16 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Echoing Bulk Data in the Output

GRID,1,,0.,0.,0.

GRID,2,,10.,0.,0.

GRID,3,,20.,0.,0.

GRID,4,,30.,0.,0.

$
$ DEFINE CBAR
ELEMENTS

CBAR,1,101,1,2,0.,1.,0.

CBAR,2,101,2,3,0.,1.,0.

CBAR,3,101,3,4,0.,1.,0.

$
$ DEFINE CBAR ELEMENT CROSS SECTIONAL
PROPERTIES

PBAR,101,201,2.,.667,.1667,.458,,,+PB1

+PB1,1.,.5,-1.,.5

$
$ DEFINE MATERIAL
PROPERTIES

MAT1,201,30.E6,,.3

$
$ DEFINE SPC CONSTRAINT
SET

SPC1,100,12345,1

SPC1,100,12345,4

$
$ DEFINE CONCENTRATED
FORCE

FORCE,10,3,,-100.,0.,1.,0.

$
ENDDATA
INPUT BULK DATA CARD COUNT = 26

Listing 3-2. Unsorted Listings

SIMPLY SUPPORTED BEAM JULY 10, 2003 Simcenter Nastran

7/10/2003 PAGE 4
WITH CONCENTRATED

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 3-17

© 2024 Siemens
 3. Understanding the Input Data File

FORCE
S O R T E D B U L K D A T A E C H
O
CARD
COUNT . 1 .. 2 .. 3 .. 4 .. 5 .. 6 .. 7 ..
8 .. 9 .. 10 .
1- CBAR 1 101 1 2 0. 1.
0.
2- CBAR 2 101 2 3 0. 1.
0.
3- CBAR 3 101 3 4 0. 1.
0.
4- FORCE 10 3 -100. 0. 1.
0.
5- GRID 1 0. 0.
0.
6- GRID 2 10. 0.
0.
7- GRID 3 20. 0.
0.
8- GRID 4 30. 0.
0.
9- MAT1 201 30.E6
.3
10- PBAR 101 201 2. .667 .1667
.458 +PB1
11- +PB1 1. .5 -1.
.5
12- SPC1 100 12345
1
13- SPC1 100 12345
4
ENDDATA
TOTAL COUNT= 14

Listing 3-3. Sorted Listing

Other options for the ECHO command include:

ECHO = SORT Prints only sorted Bulk Data (this is the default)
ECHO = UNSORT Prints only unsorted Bulk Data
ECHO = NONE Turns off the Bulk Data listing
ECHO = PUNCH Prints a sorted echo of the Bulk Data to a separate file

While learning to use Simcenter Nastran by using small models, you should set ECHO to BOTH. The
resulting listings in the .f06 file does not occupy very much space and having the model information
available can be useful. However, when the models become very large it may be best to use ECHO =
PUNCH or even ECHO = NONE.

Printing Bulk Data Using the ECHO Case Control Command

You can use the ECHO Case Control command to control how the Bulk Data section is printed to the
output file. For example, you can use ECHO to specify whether you want Simcenter Nastran to print the

3-18 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Types of Output

Bulk Data in sorted or unsorted format. For example, to echo the Bulk Data in the unsorted form, you
use the following ECHO command:

ECHO = UNSORT

The output requests, such as DISPLACEMENT, FORCE, STRESS (and so on), are needed whenever data
recovery quantities are to be computed, even if they are not printed. Such is the case when you use a
postprocessor to view the results. Typical output requests are as follows:

SET 1 = 5, 6, 7

SET 3 = 1, 5, 9

STRESS = ALL

DISP(PLOT) = 1 $ WHERE 1 IS THE ID OF A SET OF GRID POINTS

ELFORCE(PUNCH) = 3 $ WHERE 3 IS THE ID OF SET OF ELEMENTS

See also

• “ECHO” in the Simcenter Nastran Quick Reference Guide

Types of Output

There are two main types of output that you can request in Simcenter Nastran:

• Grid point output

• Element output

In a small or medium-sized model, it may be best to simply request output for all grid points or
elements. However, in a large model, you may want to be more selective.

Grid Point Output (DISP, SPCF, OLOAD, GPFORCE)

Requests for output quantities that occur at grid points include the following (n is a SET ID, and ALL
requests that all quantities be printed):

DISP Requests displacements for grid points. In general, you should use DISP=ALL as
displacement output takes up very little space.

SPCFORCE Requests forces of single point constraint (SPCFORCES). In general, you should
use SPCF = ALL so you can examine reaction forces during the model validation
process.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 3-19

© 2024 Siemens
 3. Understanding the Input Data File

OLOAD Requests the set of applied loads in static analysis.

GPFORCE Generates a table containing a grid point force balance at the selected grid
points. This is useful for determining load paths, contributions of applied loads
to element response, and the effects of initial thermal strain. Contributors to the
grid point force balance table include applied loads, SPC forces, and element
elastic forces.

Element Output (STRESS, FORCE, STRAIN, ESE)

Requests for element-based output quantities include the following:

STRESS Requests element stresses for a set of structural elements

FORCE Requests element forces to be calculated for a set of structural

elements

STRAIN Requests the strains for a set of plate elements

ESE Requests the strain energy for a set of elements

CQUAD4 stresses, strains, and forces are available at corner grid points, with output at the element
center the default. CQUAD4 center and corner output is obtained using the STRESS, STRAIN, and FORCE
Case Control commands as follows:

STRESS(CORNER) = {ALL or n}

STRAIN(CORNER) = {ALL or n}

FORCE(CORNER) = {ALL or n}

Where { } indicates that a choice of ALL or n is mandatory, but the braces are not included.

Only one type of element output (center or corner) is supported per run..

Selecting the Destination File for the Output

For each type of results you request, you can specify where you want Simcenter Nastran to print the
results.

• The PRINT option provides printed results in the .f06 output file. It is the default option for most result
types.

3-20 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Formatting Printed Output

• The PUNCH option prints the results in the ASCII format .pch file using a “punch” format (80 column
width per line). Writing results to the .pch file is useful if you want to export your results to other
programs.

• The PLOT option instructs Simcenter Nastran to calculate the requested results, but not print them.
In general, you use this option when you want to view the results as plots or examine them with
a post processing program. This option is also useful when you're working with large models where
the quantity of printed output would be excessive but you still need Simcenter Nastran to recover the
specified data for post processing.

Formatting Printed Output

You can include up to three title lines per page of Simcenter Nastran output with the TITLE, SUBTITLE,
and LABEL Case Control commands. You can use any character string. Titles are optional but highly
recommended as they can help you distinguish between your jobs.

Example

These Case Control commands

TITLE=SIMPLY SUPPORTED BEAM

SUBTITLE=WITH CONCENTRATED FORCE

LABEL=Simcenter Nastran USER'S GUIDE

produce the following headings on each page of output:

SIMPLY SUPPORTED BEAM JULY 10, 2001 Simcenter Nastran

7/10/2003 PAGE 12
WITH CONCENTRATED
FORCE
Simcenter Nastran GETTING STARTED USER'S
GUIDE
D I S P L A C E M E N T V E C T O R
POINT ID. TYPE T1 T2 T3 R1
R2 R3
1 G 0.0 0.0 0.0 0.0
0.0 -2.221112E-04
2 G 0.0 -1.943473E-03 0.0 0.0
0.0 -1.388195E-04
3 G 0.0 -2.221112E-03 0.0 0.0
0.0 1.110556E-04
4 G 0.0 0.0 0.0 0.0
0.0 2.776390E-04

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 3-21

© 2024 Siemens
 3. Understanding the Input Data File

Using the SET Command to Limit Output

In many cases, particularly in larger models, you may want to limit the output to a particular set of
grid points or elements. In Simcenter Nastran, you can use the SET Case Control command to identify a
group of grid points and/or elements as a specific set ID.

For example, suppose you want to output all of the grid point displacements for Subcase 1 and the
grid point displacements for grid points 3 and 4 for Subcase 2. In addition, you want the element force
output for element 3 for Subcase 1 and elements 3 and 4 for Subcase 2. A Case Control section that
meets these requirements is shown in Listing 3-4.

3-22 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Using the SET Command to Limit Output

$
$ FILENAME - TRUSS3.DAT
$
ID LINEAR,TRUSS3
SOL 101
TIME 2
CEND
TITLE = SET EXAMPLE
SUBTITLE = TRUSS STRUCTURE
LOAD = 10
SPC = 11
DISPLACEMENT = ALL
SET 1 = 3,4
SET 2 = 3
$
SUBCASE 1
LABEL = POINT LOAD AT GRID POINT 4 LOAD = 10
ELFORCE = 2
$
SUBCASE 2
LABEL = POINT LOAD AT GRID POINT 3
LOAD = 11
DISPLACEMENT =1
ELFORCE = 1
BEGIN BULK

Listing 3-4. Controlling Output with Set Commands

A Case Control set is a collection of grid point IDs or element IDs for use in output requests. Case Control
sets are used to obtain output for a selected portion of the model. Case Control sets are defined with the
SET command according to the following formats:

SET n = ALL

SET n = i1,i2,i3, ..., i12 THRU i28, i35, ..

where n is the set identification number and i1, i2, i3, etc., are entity identification numbers, e.g. grid
point numbers.

Example

Consider the Case Control Section shown below:

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 3-23

© 2024 Siemens
 3. Understanding the Input Data File

CEND
TITLE=OUTPUT SELECTION EXAMPLE
SUBTITLE=ILLUSTRATES USE OF SETS
LOAD=15
SET 1=3,4,7,9,11
SET 5=2,9,15 THRU 21,23
DISP=1
FORCE=1
STRESS=5
BEGIN BULK

DISP is a grid point output quantity, so displacements for grid points 3, 4, 7, 9, and 11 will be printed.
FORCE is an element output quantity, so forces for elements 3, 4, 7, 9, and 11 will be printed. Note that
grid quantities and element quantities can share the same set, since a set is simply a list of numbers.
STRESS is an element quantity, so the stresses in elements 2, 9, 15 through 21 (inclusive), and 23 will be
calculated and printed.

Bulk Data Section

The Bulk Data section always follows the Case Control section and begins with the BEGIN BULK delimiter.
In Simcenter Nastran, Bulk Data entries contain all the information necessary to define the finite element
model, such as:

• Geometry

• Coordinate systems

• Finite elements

• Element properties

• Loads

• Boundary conditions

• Material properties

The Bulk Data section generally constitutes the majority of the content of the input file. The order of Bulk
Data entries is not critical (continuations without continuation identifiers are an exception to this rule
and are discussed later).

Field 10 of the Bulk Data entry is used for two purposes. If the Bulk Data entry does not have a
continuation line, field 10 may be used as an optional comment field. If the Bulk Data entry has a
continuation line, field 10 is used for the continuation identifier. The continuation identifier must be
unique with respect to all the other identifiers in your Bulk Data Section.

3-24 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Bulk Data Section Format Rules

If you leave a data field blank, the software uses the appropriate default. Any applicable defaults for a
particular field are listed in the Simcenter Nastran Quick Reference Guide. There are certain situations
that allow either an integer or a real number in a data field. However, in these situations, whether you
specify an integer or a real number in a data field can affect the problem's solution.

Consider the definition for grid point 2 of the truss model shown in Figure 3-3. The name of the entry
is GRID, which begins in column 1 of field 1. The grid point ID (2 in this case) must be an integer (no
decimal point) greater than 0. Fields 3 and 7 represent coordinate system IDs and also must be integers.
Since these fields are blank, the default of 0 is used. Fields 4, 5, and 6 represent the physical location
of the grid point, and they must be entered as real numbers. Optional fields are field 8, which is used to
define permanently constrained degrees of freedom, and field 9, the superelement ID field. If they are
used, only integers are acceptable. Since the GRID entry does not have a continuation line, field 10 may
be used as a comment, if desired.

Figure 3-3. A Bulk Data Example: the GRID Entry

See also

• “Bulk Data Entries” in the Simcenter Nastran Quick Reference Guide

Bulk Data Section Format Rules

You must begin the Bulk Data section with the BEGIN BULK delimiter. The last entry must be the
ENDDATA delimiter. The BEGIN BULK entry is in free field format. The ENDDATA delimiter must begin in
Column 1 or Column 2. The ENDDATA delimiter also signifies the end of the Simcenter Nastran input file.

You must include both BEGIN BULK and ENDDATA in your input file, even if you are not introducing any
new Bulk Data into the problem or if all of the Bulk Data is coming from an alternate source, such as
user-generated input.

Simcenter Nastran Bulk Data contains 10 fields per input data entry:

• The first field contains the character name of the Bulk Data item, such as GRID, CBAR, MAT1.

• The second through ninth fields contain data input information for the Bulk Data entry.

• The tenth field never contains data. It is reserved for entry continuation information, if applicable.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 3-25

© 2024 Siemens
 3. Understanding the Input Data File

Sorting Bulk Data

You can enter Bulk Data entries in any order since Simcenter Nastran automatically sorts them (using the
XSORT module) prior to the execution of the Input File Processor (IFP) module.

However, for very large problems, this sorting process can be CPU intensive. If you anticipate submitting
the same input file multiple times, you should use the ECHO=PUNCH Case Control command to generate
a sorted Bulk Data file in the first analysis. Then, you can use the sorted Bulk Data input for subsequent
analyses.

Note: Alternatively, the sorted Bulk Data is stored on the database and is available using the Restart
feature.

See also

• “XSORT” in the Simcenter Nastran DMAP Programmer's Guide

• “IFP” in the Simcenter Nastran DMAP Programmer's Guide

Storing Bulk Data in the Database for Restarts

To minimize the handling of a large number of entries, Simcenter Nastran solution sequences let you
store the Bulk Data in the database. That way, you can easily modify it on subsequent runs.

For any initial analysis (also referred to as a “cold start”), you must submit the entire Bulk Data section
of the input file. For subsequent analyses, if you save the database from any previous run, the Bulk Data
exists in the database in sorted form where you can modify it and reuse it in a restart analysis.

When you perform a restart, the software adds the Bulk Data entries contained in the input file to the
Bulk Data stored in the database. You can use a ”/” entry to remove entries from the database.

See also

• “Restarts” in the Simcenter Nastran User's Guide

Understanding the Relationship Between Bulk Data Entries and Case Control
Commands

Most of the entries in the Bulk Data section do not need to be called out by a Case Control command to
be included in the model. However, for loads and constraints, the Bulk Data and Case Control sections of
the input file are carefully interrelated.

• In the Bulk Data section, you specify actual load and constraint entries.

3-26 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Using Parameters

• In the Case Control section, you specify which of the load and constraint entries you defined in the
Bulk Data section to use in a given analysis.

In other words, Simcenter Nastran includes only the load and constraint entries you define in the
Bulk Data section if they are specified in the Case Control section. Therefore, you can have load and
constraint entries in the Bulk Data section that are not specified in the Case Control. However, those
entries will not be used during the analysis.

Using Parameters
Each Simcenter Nastran solution sequence is designed with a specific set of default values and solution
paths. In the majority of the cases, the default values are the ones best suited for your model. There
are occasions, however, when you may want to deviate from the default values. Several methods are
available in Simcenter Nastran to perform this task: DMAP (Direct Matrix Abstraction Program) modules,
NASTRAN statements, and parameters.

Using parameters is the most straightforward method to change defaults in Simcenter Nastran.
Depending on the parameter, you can specify it either as a PARAM Bulk Data entry and/or a PARAM
Case Control command. The format of each is as follows:

PARAM,xname,xvalue

where xname and xvalue are the name and value of the PARAMeter, respectively.

PARAM,WTMASS,0.00259

For example, with PARAM,WTMASS, the software multiplies the terms of the structural mass matrix by
the value of this parameter.

In Simcenter Nastran, parameters are used extensively in the solution sequences for input of scalar
values and for requesting special features. You can specify values for parameters using either the PARAM
Bulk Data entry or the PARAM Case Control command.

See also

• “PARAM” in the Simcenter Nastran Quick Reference Guide (for the PARAM Case Control command)

• “PARAM” in the Simcenter Nastran Quick Reference Guide (for the PARAM bulk data entry)

• “Parameters and Parameter Applicability Tables” in the Simcenter Nastran Quick Reference Guide (for a
complete description of all parameters)

• Simcenter Nastran DMAP Programmer's Guide (for a description of the manner in which Simcenter
Nastran processes parameters)

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 3-27

© 2024 Siemens
 3. Understanding the Input Data File

Parameters in the Superelement Solution Sequences

The specification of PARAMeters differs between the rigid formats and the superelement solutions
sequences. These differences require different user action in placing the PARAM entries in the Simcenter
Nastran section, and in changing PARAMeter values on restarts.

The program module reads the Case Control section for PARAM entries. If PARAM entries exist, the VPS
table is updated with the values found in the Case Control section. The PARAM values are updated at the
beginning of the generation, assembly, reduction, and the data recovery loops of the superelement
solution sequences. This allows the values of some user-supplied parameters to change for each
superelement during the course of one run.

In the nonlinear solution sequences the PARAM values are updated at the beginning of the iteration
loop. This allows the values of some user-set parameters to vary between subcases. These parameters
may appear in either the Case Control and/or Bulk Data section. Other PARAMeters must appear in the
Bulk Data section. If they appear in the Case Control section, they will not cause a preface error, but
will be ignored. Other parameters may be reset to default values during the course of the run. This type
should be set only in the subcase of the Case Control section.

The PARAM Case Control command is treated in the same manner as constraint and output commands.
Parameters set above the subcase level are effective in every subcase. If there is also a parameter value
set in the subcase, it takes precedence. Only one value of a parameter may be current at any time during
the execution of the run. These rules imply an order of precedence if several forms are used in the same
run. The rules are illustrated in the following example.

A partial subcase structure is listed on the following page. The subcases are listed in ascending
superelement number order. PARAM, GRDPNT is varied between them.

See also

• The Simcenter Nastran Quick Reference Guide (for parameter applicability)

Subcase Structure, Sorted by SElD

SUBCASE 1
SUPER = 1
...
SUBCASE 2
SUPER = 2
PARAM,GRDPNT,100
...
SUBCASE 3
SUPER = 3
PARAM,GRDPNT,200
...
SUBCASE 4
SUPER = 4

3-28 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Subcase Structure, Sorted by SElD

...
SUBCASE 100 $ RESIDUAL STRUCTURE
PARAM,GRDPNT,-1
...
BEGIN BULK
PARAM,GRDPNT,0

Processing order Value of GRDPNT

SEID
1 0
2 100
3 200
4 0
0 –1

The Bulk Data entries are processed at the beginning of the run. Any PARAM entries in the Bulk Data
section set the value of that parameter until it is changed by a Case Control command.

• Superelement 1 is processed first. Since it has no PARAM,GRDPNT entry the value of 0, as set by the
Bulk Data entry PARAM,GRDPNT,0 is used.

• The value for superelement 2 is set at 100 by a Case Control command.

• Superelement 4 is processed next. Its subcase contains no PARAM entry, so that the value of the
parameter that remains is set by the bulk data.

• The values for superelement 3 and the residual structure are set by Case Control commands.

An alternate method of parameter specification is to set the value most used above the subcase level,
and exceptional values within the subcase. For example, if PARAM,GRDPNT,0 were placed above the
subcase level and the values of 100, 200, and -1 placed in subcases 2, 3 and 100, respectively, the value
used in each superelement will be the same as described above, regardless of processing order.

For the unstructured solution sequences, Simcenter Nastran does not store the VPS table in the
database. Therefore, it is not recovered on restarts. Therefore, parameters set by the user must be
in the Simcenter Nastran section if the default values are not wanted.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 3-29

© 2024 Siemens
 3. Understanding the Input Data File

3-30 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 4. Simcenter Nastran Output Files
Overview of the Simcenter Nastran Output Files
When you submit an input file to Simcenter Nastran for execution, the software generates a number
of different output files. Some files are generated automatically, while other files are generated only
by specific requests you make in the input file or on the command line when you submit the job. The
following table shows the different types of Simcenter Nastran output files.

Table. Types of Files Output by Simcenter Nastran

File File
Extension Format How Created Brief Description

.f06 ASCII Automatically Main output file with printed output, such as
displacements and stresses.

.f04 ASCII Automatically Provides history of assigned files, disk space

usage, and modules used during the analysis.

.log ASCII Automatically Gives a summary of the command line options

used and the execution links

.DBALL Binary Automatically A database containing the input file, assembled

matrices, and solutions. Used also for restarting
the run for additional analysis or output.

.MASTER Binary Automatically Contains the master directory of the files used by
the run and the physical location of those files on
your system. Needed for a restart.

.plt Binary By request Contains the plot information requested by the

plotter commands specified in input file.

.pch ASCII By request Contains the punch output as requested in the

input file.

.op2 Binary By request Results and geometry database file used for post-
processing.

.xdb Binary By request Results and geometry database file used for post-
processing.

.hdf Binary By request Results and geometry database file used for post-
processing.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 4-1

© 2024 Siemens
 4. Simcenter Nastran Output Files

In addition to the files listed above, the software also generates several temporary (scratch) files during
the analysis which are automatically deleted when the solve completes.

Note:
By default, the software writes the .f06, .f04, and .log files separately. However, you can use
the “append” keyword (append=yes) on the command line when you run your job to have the
software combine them into a single file with a .out extension.

The following sections describe the different output file types in more detail.

See also

• “Automatic Deletion of Scratch Data Blocks” in the Simcenter Nastran DMAP Programmer's Guide

• “Append” in the Simcenter Nastran Installation and Operations Guide

Understanding the .f06 File

The output file that you will use most frequently is the .f06 file. The software writes this file to FORTRAN
unit 6. The .f06 file contains all the requested results from your analysis, such as the displacements,
stresses, and element forces, as well as any diagnostic messages. The information in the .f06 file is
critical for model checkout and debugging.

The output from data recovery and plot modules is all optional and its selection is controlled by
commands in the Case Control section.

Although most of the solution sequence output is optional and requested in the Case Control Section,
some of the printer output is automatic or under control of DIAG statements in the Executive Control
Section.

See also

• “Flushing .f04 and .f06 Output to Disk” in the Simcenter Nastran Installation and Operations Guide

Controlling Printed Output

In Simcenter Nastran, there are two different formats in which the software prints results: SORT1 and
SORT2.

Understanding the SORT1 Format

With the SORT1 format, the software presents the analysis outputs as a tabular listing of the selected
output for each grid point in each individual subcase. With SORT1, the software prints the output for
each loading condition on any page. This is the default for static analysis, frequency response, steady
state heat transfer, real and complex eigenvalue analysis, flutter analysis, and buckling analysis.

4-2 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Understanding the SORT2 Format

If you select SORT1 in a transient solution for one or more of the commands listed below, then the
software outputs all remaining commands in the input file in SORT1 format:

ACCE, DISP, ENTH, FORC, HDOT, MPCF, OLOA, SPCF, STRA, STRE, VELO

Understanding the SORT2 Format

With the SORT2 format, the software presents the analysis output as a tabular listing of the subcases
for each selected output option. With SORT2, the software prints the output for each grid point or
element on a new page. Consequently, the SORT2 option can produce a large number of output pages.
Therefore, SORT2 should generally be used only when you're performing dynamic solutions.

The SORT2 format is the default output format in transient response analysis (structural and heat
transfer). The SORT2 option isn't available when you're performing real eigenvalue (including buckling)
analyses, complex eigenvalue analyses, or flutter analyses.

If you select SORT2 in a frequency response solution for one or more of commands listed below, then
the software outputs all remaining commands in the input file in SORT2 format:

ACCE, DISP, FORC, MPCF, OLOA, SPCF, STRA, STRE, VELO

Controlling General Output Options

The printer output is designed for 132 characters per line, with the lines per page controlled by the LINE
command in the Case Control Section. The default is LINE = 50 for 11-inch paper.

Optional titles are printed at the top of each page from information in the Case Control Section. You can
define these titles at the subcase level. All pages of the output are automatically dated and numbered.

A few printer output items are under the control of PARAM Bulk Data entries.

The following output is either automatically or optionally provided during execution:

• Simcenter Nastran title page – automatic

• File Management Section echo – automatic if statements are present

• Executive Control Section echo – automatic

• Case Control Section echo – automatic

• Unsorted Bulk Data Section echo – optional, selected in the Case Control Section

• Sorted Bulk Data Section echo – automatic, unless suppressed in the Case Control Section

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 4-3

© 2024 Siemens
 4. Simcenter Nastran Output Files

By default, the output file contains an echo (listing) of the input file with the Bulk Data sorted
alphabetically as shown in Figure 4-1.

4-4 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Controlling General Output Options

Figure 4-1. truss1.f06 Output File (Continued)

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 4-5

© 2024 Siemens
 4. Simcenter Nastran Output Files

Figure 4-1. truss1.f06 Output File

See also

• “Parameters and Parameter Applicability Tables” in the Simcenter Nastran Quick Reference Guide

Understanding the General Analysis Output

OLOAD Resultant

After the input file echo, the software prints the “OLOAD RESULTANT.” This shows the resultant of the
loads about the origin of the basic coordinate system in this case.

User Information Message 5293

Next is the USER INFORMATION MESSAGE (UIM) 5293 which shows the epsilon and external work for the
run. The epsilon is a measure of the numeric conditioning of the model, while the external work is the
work due to the applied loads. Both of these values are useful for model checkout.

In the truss example, output is requested for displacements, element forces, and element stresses. The
displacement request is the only grid point output. Element forces and stresses are element output.

The last summary is the maximum of the applied loads.

4-6 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Accuracy Diagnostic Information for Global Iterative Solver

After the summaries, the displacement for all of the grid points, the element forces, and the element
stresses is printed as requested in the Case Control Section.

For static analysis of the truss model, just one applied load is used. You can, however, specify
multiple loading conditions to be analyzed in a single input file if desired. If you do specify multiple
loading conditions, the OLOAD RESULTANT and EXTERNAL WORK are shown for each selected loading
conditions. The external work is calculated as 1/2 {Pl} T [Ul]. In the case of enforced displacement, the
equivalent load is calculated as the constraint force required to impose the enforced displacement. The
external work printout in this case is meaningless.

See also

• “Model Verification” in the Simcenter Nastran User's Guide

Accuracy Diagnostic Information for Global Iterative Solver

When you are using Simcenter Nastran's global iterative solver, the .f06 file also contains important
accuracy diagnostic information.

See also

• “Accuracy Diagnostics” in the Simcenter Nastran Numerical Methods User's Guide

Grid Point Singularity Table

If singularities remain in the stiffness matrix at the grid point level, the software automatically outputs
a Grid Point Singularity Table (GPST) after the Grid Point Singularity Processor (GPSP) executes. The
table lists singular degrees-of-freedom in the global coordinate system. You can have the software
automatically constrain singular degrees-of-freedom using the PARAM,AUTOSPC,YES option.

Note: You can print any of the matrices or tables that are prepared by the functional modules using
selected utility modules described in the Simcenter Nastran DMAP Programmer's Guide. These utility
modules can be scheduled at any point in a solution sequence by using the alter feature. In general, they
should be scheduled immediately after the functional module that generates the table or matrix to be
printed. However, you should check the calling sequence for the utility module, in order to be certain
that all required inputs have been generated prior to this point.

See also

• “Constraint and Mechanism Problem Identification in SubDMAP SEKR” in the Simcenter Nastran
User's Guide

• “Understanding the Grid Point Singularity Table” in the Simcenter Nastran User's Guide

• “AUTOSPC” in the Simcenter Nastran Quick Reference Guide

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 4-7

© 2024 Siemens
 4. Simcenter Nastran Output Files

Understanding Analysis Type Specific Output

The following descriptions relate to automatic output that is associated with particular types of solution
sequences.

Detailed information on the force and stress output available for each element type is given in the
Simcenter Nastran Element Library.

Real Eigenvalue Analysis

Each eigenvalue is identified with a mode number determined by sorting the eigenvalues by their
magnitude. The following summary is also automatically printed for all methods:

• Mode number

• Extraction order (this is irrelevant under the Lanczos method)

• Eigenvalue

• Frequency (radians)

• Frequency (Hz)

• Generalized mass

• Generalized stiffness

For all methods, if the orthogonality criterion fails, then the software issues User Warning Message
3034.

For the Lanczos method, the following additional summary content is provided:

• Block size used

• Number of decompositions

• Number of roots found

• Number of (block) solves required

• One of the following termination messages:

• Required number of eigenvalues found

• Not all the roots found within the range

4-8 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Complex Eigenvalue Analysis

• All eigenvalues found within the range

• Insufficient time to find more eigenvalues

Complex Eigenvalue Analysis

Each complex eigenvalue is identified with a root number determined by sorting the complex
eigenvalues by magnitude. The following summary of the complex eigenvalues extracted is
automatically printed for all methods:

• Root number

• Extraction order (this is irrelevant under the Lanczos method)

• Real and imaginary parts of the eigenvalue

• The coefficients f (frequency) and g (damping coefficient) in the following representation of the
eigenvalue (p = α + iω):

Equation 4-1.

Therefore,

Equation 4-2.

Equation 4-3.

If you use the complex Lanczos method, the software also issues User Information Messages 5444,
5445, and 5453.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 4-9

© 2024 Siemens
 4. Simcenter Nastran Output Files

If you use the Determinant method, the software also prints:

• Number of eigenvalues extracted

• Number of passes through starting points.

• Number of criteria changes

• Number of starting point moves

• Number of triangular decompositions

• Number of failures to iterate to a root

• Number of predictions outside region

• The reason for termination is indicated by one of the following integers:

Reason
Number Description

1 The number of roots desired have been found

2 All predictions for eigenvalues are outside the regions specified

3 Insufficient time to find another root

4 Matrix is singular at the first three starting points

• Swept determinant functions for each starting point

Superelement Analysis

• Superelement mapping information that lists the membership of grid points and elements in the
superelements, the order in which the superelements will be processed, and timing and space
estimates for each superelement

• A page header that identifies the superelement for which the printed information on the page
pertains

• Matrices wherein mechanisms are identified

• Resultants of external loads

• Maximum applied loads

4-10 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Monitor points for element results and grid point forces

• Maximum displacements

• Maximum SPC forces

For SOL 101, if PARAM,INREL,-1 or PARAM,INREL,-2, the following data is printed automatically:

INTERMEDIATE MATRIX...QRR

This is the rigid body mass matrix of the total structure, including superelements. It is presented as a 6
× 6 matrix, measured at the reference grid point, in its global coordinates. Masses on scalar points are
ignored.

INTERMEDIATE MATRIX...QRL

This is the resultant of the applied loads, measured at the reference point. There is one column for each
loading condition.

INTERMEDIATE MATRIX...URACCEL

This is the user-supplied rigid body acceleration, input on the DMIG,UACCEL entry. This output does not
appear if the entry is not present.

INTERMEDIATE MATRIX...URA

This is the rigid body acceleration matrix, computed from the applied loads.

Monitoring results

Monitor points for element results and grid point forces

For SOL 101 and SOL 103, Simcenter Nastran uses monitor points to output element results from stress,
strain, or force, and to provide summation of grid point forces at specified integrated load points in a
local coordinate system.

You use the bulk entry:

• MONPNT2 to output component of stress, strain, or force at a named element. The software outputs
the results in a table titled STRUCTURAL INTERNAL MONITOR POINT LOADS (MONPNT2).

Because MONPNT2 operates on the data block level, it can monitor an NDDL entry from the STRESS-
STRAIN (contained in the oes.dll file) or FORCE (contained in the oef.dll file) output data block.

• MONPNT3 to sum select grid point forces at a chosen integrated load point. You select a set of grids
to sum forces and moments about a specified point. The software outputs the results in a table titled
STRUCTURAL INTEGRATED FREE BODY MONITOR POINT LOADS (MONPNT3).

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 4-11

© 2024 Siemens
 4. Simcenter Nastran Output Files

Also, you can use the MONITOR case control command to control the output of MONPNT2 and
MONPNT3 results to the .f06 output file. You can place the MONITOR command above the subcase
level or in individual subcases.

Note:
You can optionally use the OIBULK parameter to request that the software writes all case control
commands, including the MONITOR3 command, to the CASECC data block in your .op2 file.
For more information, see PARAM, POST ,< 0.

For an example that demonstrates the use of MONITOR, MONPNT2, and MONPNT3, see Monitor points
example.

Monitor points example

Test problem description

In this test problem, the physical problem represents a beam model with a rectangular section. The
beam starts at 1 meter long, fixed at one end or two ends, and at –50° C is heated to 25° C.

The objective is to define an integrated load monitor point at a named element and output strain and
stress, and to sum grid point forces around a specific point.

Executive and case control

Two subcase runs with different restraints (one end fixed and two ends fixed) that use the same model
are used to investigate the effects of different supports. In both cases, the MONITOR case control
command requests monitor points information. In both subcases, MONITOR requests to print all monitor
results.

Bulk data

The key data in the Bulk Data section is as follows:

• SPC cards are used to restrain the model at one or two ends.

• A thermal load of 25° C is applied at all grid points of the bar model.

• MONPNT2 outputs EX1A (strain at end A at point C) and SX1A (bending stress at end A at the
recovery location C).

• MONPNT3 sums the grid point forces at the first grid point around all six component axes in the basic
coordinate system.

Results from Simcenter Nastran

4-12 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Monitor points example

Partial output shows Monitor point tables

Note:
REST. APPLIED stands for Resultant Applied Load. Each Resultant Applied Load value
represents a sum of all the loads at the specified location in a specific direction.

Input file

$*
$* Simcenter Nastran
$*
$* ANALYSIS TYPE: Structural
$* SOLUTION TYPE: SOL 101 Linear Statics - Subcase Constraints
$* SOLVER INPUT FILE: monitor-points.dat
$*
$*
$*
$* UNITS: SI-Meter (newton)
$* ... LENGTH : meter
$* ... TIME : sec
$* ... MASS : kilogram (kg)
$* ... FORCE : newton(N)
$* ... TEMPERATURE : deg Celsius
$*
$*$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$
$*
$* EXECUTIVE CONTROL
$*
$*$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$
$*
SOL 101
TIME 60

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 4-13

© 2024 Siemens
 4. Simcenter Nastran Output Files

CEND
$*
$*$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$
$*
$* CASE CONTROL
$*
$*$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$
$*
TITLE = monitor-points - Thermal Strain, Displacement, and Stress on
Heated Beam
ECHO = SORT
$*
TEMP(INIT) = 1
$*
SUBCASE = 1
$*
LABEL = One End Fixed
SPC = 1
TEMP(LOAD) = 10
DISPLACEMENT(PRINT,PUNCH) = ALL
SPCFORCES(PRINT,PUNCH) = ALL
FORCE(PRINT,PUNCH,CORNER) = ALL
STRESS = ALL
STRAIN(PRINT,PUNCH,CORNER,VONMISES,STRCUR) = ALL
ESE(PRINT,PUNCH) = ALL
MONITOR = ALL
$*
SUBCASE = 2
$*
LABEL = Both Ends Fixed
SPC = 2
TEMP(LOAD) = 10
DISPLACEMENT(PRINT,PUNCH) = ALL
SPCFORCES(PRINT,PUNCH) = ALL
FORCE(PRINT,PUNCH,CORNER) = ALL
STRESS = ALL
STRAIN(PRINT,PUNCH,CORNER,VONMISES,STRCUR) = ALL
ESE(PRINT,PUNCH) = ALL
MONITOR = ALL
$*
$*$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$
$*
$* BULK DATA
$*
$*$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$
$*
BEGIN BULK
$*
$* PARAM CARDS

4-14 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Monitor points example

$*
PARAM XFLAG 0
PARAM AUTOSPC YES
PARAM GRDPNT -1
PARAM POST -2
PARAM POSTEXT YES
$*
$* GRID CARDS
$*
GRID 1 0 0.0 0.0 0.0 0
GRID 2 0 .100000 0.0 0.0 0
GRID 3 0 .200000 0.0 0.0 0
GRID 4 0 .300000 0.0 0.0 0
GRID 5 0 .400000 0.0 0.0 0
GRID 6 0 .500000 0.0 0.0 0
GRID 7 0 .600000 0.0 0.0 0
GRID 8 0 .700000 0.0 0.0 0
GRID 9 0 .800000 0.0 0.0 0
GRID 10 0 .900000 0.0 0.0 0
GRID 11 0 1.00000 0.0 0.0 0
$*
$* ELEMENT CARDS
$*
CBAR 1 1 1 2 0.01.000000 0.0
CBAR 2 1 2 3 0.01.000000 0.0
CBAR 3 1 3 4 0.01.000000 0.0
CBAR 4 1 4 5 0.01.000000 0.0
CBAR 5 1 5 6 0.01.000000 0.0
CBAR 6 1 6 7 0.01.000000 0.0
CBAR 7 1 7 8 0.01.000000 0.0
CBAR 8 1 8 9 0.01.000000 0.0
CBAR 9 1 9 10 0.01.000000 0.0
CBAR 10 1 10 11 0.01.000000 0.0
$*
$* MATERIAL CARDS
$*
$* Material: 1 name: MATERIAL PROPERTY TABLE1
MAT1 1206.80+9 .29000007820.0001.2000-5
0.0+
+ 1.5000+91.5000+96.8000+7
$*
$* PROPERTY CARDS
$*
$*
$* Property: 1 name: PHYSICAL PROPERTY TABLE1
$* Fore Section: 2 name: RECTANGLE 0.1 X 0.1
PBAR 1 11.0000-28.3333-68.3333-61.4083-5
+
+

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 4-15

© 2024 Siemens
 4. Simcenter Nastran Output Files

.0500000.0500000.0500000-.050000-.050000-.050000-.050000.0500000+
+ .8333333.8333333
$*
$* RESTRAINT CARDS
$*
SPC 1 1 123456 0.0
$*
$* RESTRAINT CARDS
$*
SPC 2 1 123456 0.0
SPC 2 11 123456 0.0
$*
$* TEMPERATURE CARDS
$*
TEMP 10 125.00000 225.00000 325.00000
TEMP 10 425.00000 525.00000 625.00000
TEMP 10 725.00000 825.00000 925.00000
TEMP 10 1025.00000 1125.00000
TEMPD 1 -50.000
$*
$1.....12......23......34......45......56......67......78......89......9+
MONPNT2 LOCA testing1
STRAIN CBAR EX1A 1
STRESS CBAR SX1A 1
MONPNT3 A1
123456 1 0 0.0 0.0 0.0
0
$1.....12......23......34......45......56......67......78......89......9+
GROUP 1
GRID 1
ENDDATA

4-16 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 MONPNT2

MONPNT2

Integrated Load Monitor Point — Element Monitor Output Results Item

Defines an integrated load monitor point at a named element, and outputs component of stress, strain,
or force at that element for SOL 101 and SOL 103 (SORT1 format only).

FORMAT:

1 2 3 4 5 6 7 8 9 10
MONPNT2 NAME LABEL
TABLE TYPE COMP EID

EXAMPLE:

MONPNT2 MPT22 This is a MONPNT2 monitor point

STRAIN CBAR EX1A 1100082
STRESS CBAR SX1A 1100234

FIELDS:

Field Contents

NAME A unique character string that identifies the monitor point. (Character; 8 characters
maximum; No default)

LABEL A string that identifies and labels the monitor point. (Character; 56 characters maximum;
Optional)

TABLE Type of output to be monitored. (Character; "STRESS", "FORCE", or "STRAIN"; No default)

TYPE Element type. (Character; No default)

COMP Component of the element type to be monitored. This is the COMP label for
the particular table and element type. For a list of supported elements and their
components, see Stress monitor table and Force monitor table. (Character; No default)

EID Element ID. (Integer > 0; No default)

REMARKS:

1. For a list of supported elements and their components, see Stress monitor table and Force
monitor table.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 4-17

© 2024 Siemens
 4. Simcenter Nastran Output Files

2. Because of the name/descriptor field, free-field, and large field data formats are not supported for
this bulk entry.

3. For shell laminates, only the homogeneous stress/strain is output. The NOCOMPS parameter allows
you to control the shell laminate output. If this parameter:

= 0 or -1, The homogeneous stress/strain is output.

= 1, No homogeneous stress/strain is output.

Note:
Output for solid laminates and layer by layer output is not supported.

4. Output for superelements that are processed at the monitor point:

• For internal and partitioned superelements, MONPNT2 results can be output for the residual or
downstream superelements.

• For external superelements, MONPNT2 results are only output for the residual superelement.

Note:
The various elements defined for a MONPNT2 may not exist for all of the superelements in a
model. When this occurs, the MONPNT2 is not processed for the superelement.

5. Invalid Table, Type, COMP, or EID generates a warning message and is ignored.

6. The software prints the results in a table titled STRUCTURAL INTERNAL MONITOR POINT
LOADS (MONPNT2)

STRESS MONITOR TABLE

Element Name Real Stresses/Strains Complex Stresses or Strains

(Code)
NDDL Item NDDL Item Real/
entity entity Mag. or
Imag./
Phase
CBAR SX1A/ End A-Point C SX1AR/ End A-Point C RM
EX1A EX1AR
34
SX2A/ End A-Point D SX2AR/ End A-Point D RM
EX2A EX2AR
SX3A/ End A-Point E SX3AR/ End A-Point E RM
EX3A EX3AR

4-18 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 MONPNT2

SX4A/ End A-Point F SX4AR/ End A-Point F RM

EX4A EX4AR
AS/AE Axial ASR/AER Axial RM
BMAXA/ End A maximum SX1AI/ End A-Point C lP
EBMAXA EX1AI
BMINA/ End A minimum SX2AI/ End A-Point D IP
EBMINA EX2AI
MST/MST Safety margin in SX3AI/ End A-Point E IP
tension EX3AI
SX1B/EX1B End B-Point C SX4AI/ End A-Point F IP
EX4AI
SX2B/EX2B End B-Point D ASI/AEI Axial IP
SX3B/EX3B End B-Point E SX1BR/ End B-Point C RM
EX1BR
SX4B/EX4B End B-Point F SX2BR/ End B-Point D RM
EX2BR
BMAXB/ End B maximum SX3BR/ End B-Point E RM
EBMAXB EX3BR
BMINB/ End B minimum SX4BR/ End B-Point F RM
EBMINB EX4BR
MSC Safety margin in SX1BI/ End B-Point C IP
compression EX1BI
SX2BI/ End B-Point D IP
EX2BI
SX3BI/ End B-Point E IP
EX3BI
SX4BI/ End B-Point F IP
EX4BI

Element Name Real Stresses or Strains Complex Stresses or Strains

(Code)
NDDL Item NDDL Item Real/
entity entity Mag. or
Imag./
Phase
CBAR SD Station Distance/Length SD Station Distance/ RM
100 Length
Intermediate
Stations SXC/EXC Point C SXCr/EXCr Point C RM

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 4-19

© 2024 Siemens
 4. Simcenter Nastran Output Files

SXD/EXD Point D SXDr/ Point D RM

EXDr
SXE/EXE Point E SXEr/EXEr Point E RM
SXF/EXF Point F SXFr/EXFr Point F RM
AS/AE Axial ASr/AEr Axial RM
SMAX/ Maximum SMAXr/ Maximum RM
EMAX EMAXr
SMIN/ Minimum SMINr/ Minimum RM
EMIN EMINr
MS Margin of Safety SXCi/EXCi Point C IP
(NDDL entitys above are SXDi/EXDi Point D IP
given for End A. For
codes 2 through 10 SXEi/EXEi Point E IP
at intermediate stations SXFi/EXFi Point F IP
add (K-1)*9 where K
is the station number, ASi/AEi Axial IP
and for codes at End
B, K=number of stations SMAXi/ Maximum IP
plus 1.) EMAXi
SMINi/ Minimum IP
EMINi
(NDDL entitys above
are given for End
A. For codes 2
through 16 at
intermediate stations
add (K-1)*15 where
K is the station
number, and for
codes at End
B, K=number of
stations plus 1.)
Element Name Real Stresses or Strains Complex Stresses or Strains
(Code)
NDDL Item NDDL Item Real/
entity entity Mag. or
Imag./
Phase
CBEAM GRID External grid point ID GRID External grid point ID
2 SD Station distance/length SD Station distance/
Linear Format length

4-20 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 MONPNT2

SXC/EXC Long. Stress/Strain at SRCR/ Long. Stress/Strain at RM

Point C ERCR Point C
SXD/EXD Long. Stress/Strain at SXDR/ Long. Stress/Strain at RM
Point D EXDR Point D
SXE/EXE Long. Stress/Strain at SXER/ Long. Stress/Strain at RM
Point E EXER Point E
SXF/EXF Long. Stress/Strain at SXFR/ Long. Stress/Strain at RM
Point F EXFR Point F
SMAX/ Maximum Stress/Strain SXCI/EXCI Long. Stress/Strain at IP
EMAX Point C
SMIN/ Minimum Stress/Strain SXDI/EXDI Long. Stress/Strain at IP
EMIN Point D
MST Safety margin in SXEI/EXEI Long. Stress/Strain at IP
tension Point E
MSC Safety margin in SXFI/EXFI Long. Stress/Strain at IP
compression Point F
NDDL entitys are given NDDL entitys are
for End A. Addition of given for End A.
the quantity (K-1)*10 Addition of the
to the NDDL entity quantity (K-1)*10
points to the same to the NDDL entity
information for other points to the same
stations, where K is information for other
the station number. stations, where K is
K=11 for End B, and the station number.
K=2 through 10 for K=11 for End B,
intermediate stations. and K=2 through
10 for intermediate
stations.
Element Name Real Stresses or Strains Complex Stresses or Strains
(Code)
NDDL Item NDDL Item Real/
entity entity Mag. or
Imag./
Phase
CBEND GRID External grid point ID GRID External grid point ID
69 CA Circumferential angle CA Circumferential
angle
SC/EC Long. Stress/Strain at SCR/ECR Long. Stress/Strain at RM
Point C Point C

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 4-21

© 2024 Siemens
 4. Simcenter Nastran Output Files

SD/ED Long. Stress/Strain at SDR/EDR Long. Stress/Strain at RM

Point D Point D
SE/EE Long. Stress/Strain at SER/EER Long. Stress/Strain at RM
Point E Point E
SF/EF Long. Stress/Strain at SFR/EFR Long. Stress/Strain at RM
Point F Point F
SMAX/ Maximum Stress/Strain SCI/ECI Long. Stress/Strain at IP
EMAX Point C
SMIN/ Minimum Stress/Strain SDI/EDI Long. Stress/Strain at IP
EMIN Point D
MST Safety margin in SEI/EEI Long. Stress/Strain at IP
tension Point E
MSC Safety margin in SFI/EFI Long. Stress/Strain at IP
compression Point F
(NDDL entitys are given (NDDL entitys are
for End A. NDDL entitys given for End
12 through 21 point to A. NDDL entitys
the same information 12 through 21
for end B.) point to the same
information for End
B.)
Element Name Real Stresses or Strains Complex Stresses or Strains
(Code)
NDDL Item NDDL Item Real/
entity entity Mag. or
Imag./
Phase
CBUSH TX Translation-x TXR Translation-x R
102 TY Translation-y TYR Translation-y R
Linear Format TZ Translation-z TZR Translation-z R
RX Rotation-x RXR Rotation-x R
RY Rotation-y RYR Rotation-y R
RZ Rotation-z RZR Rotation-z R
TXI Translation-x I
TYI Translation-y I
TZI Translation-z I
RXI Rotation-x I

4-22 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 MONPNT2

RYI Rotation-y I
RZI Rotation-z I
Element Name Real Stresses or Strains Complex Stresses or Strains
(Code)
NDDL Item NDDL Item Real/
entity entity Mag. or
Imag./
Phase
CBUSH1D FE Axial force
40 UE Axial displacement
VE Axial velocity
AS Axial stress
AE Axial strain
EP
FAIL
Element Name Real Stresses or Strains Complex Stresses or Strains
(Code)
NDDL Item NDDL Item Real/
entity entity Mag. or
Imag./
Phase
CELAS1 S/E Stress/Strain SR/ER Stress/Strain RM
11
SI/EI Stress/Strain IP
CELAS2 S/E Stress/Strain SR/ER Stress/Strain RM
12
SI/EI Stress/Strain IP
CELAS3 S/E Stress/Strain SR/ER Stress/Strain RM
13
SI/EI Stress/Strain IP
Element Name Real Stresses or Strains Complex Stresses or Strains
(Code)
NDDL Item NDDL Item Real/
entity entity Mag. or
Imag./
Phase
CGAP CPX Normal x Not applicable
86 SHY Shear y
SHZ Shear z

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 4-23

© 2024 Siemens
 4. Simcenter Nastran Output Files

AU Axial u
SHV Shear v
SHW Shear w
SLV Slip v
SLP Slip w
Element Name Real Stresses or Strains Complex Stresses or Strains
(Code)
NDDL Item NDDL Item Real/
entity entity Mag. or
Imag./
Phase
CHEXA CID Stress coordinate CID Stress coordinate
system system
67
CTYPE Coordinate type CTYPE Coordinate type
Linear Format
(Character) (Character)
NODEF Number of active points NODEF Number of active
points
GRID External grid ID GRID External grid ID
(0=center) (0=center)
SX/EX Normal x SXR/EXR Normal x RM
TXY/ETXY Shear xy SYR/EYR Normal y RM
P1/EP1 First principal SZR/EZR Normal z RM
P1X/P1X First principal x cosine TXYR/ Shear xy RM
ETXYR
P2X/P2X Second principal x TYZR/ Shear yz RM
cosine ETYZR
P3X/P3X Third principal x cosine TZXR/ Shear zx RM
ETZXR
PR/EPR Mean pressure SXI/EXI Normal x lP
OCT/EOCT von Mises or octahedral SYI/EYI Normal y IP
shear stress
SY/EY Normal y SZI/EZI Normal z IP
TYZ/ETYZ Shear yz TXYI/ Shear xy IP
ETXYI
P2/EP2 Second principal TYZI/ Shear yz IP
ETYZI

4-24 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 MONPNT2

P1Y/P1Y First principal y cosine TZXI/ Shear zx IP

ETZXI
P2Y/P2Y Second principal y
cosine
P3Y/P3Y Third principal y cosine
SZ/EZ Normal z
TZX/ETZX Shear zx
P3/EP3 Third principal
P1Z/P1Z First principal z cosine
P2Z/P2Z Second principal z
cosine
P3Z/P3Z Third principal z cosine
Element Name Real Stresses or Strains Complex Stresses or Strains
(Code)
NDDL Item NDDL Item Real/
entity entity Mag. or
Imag./
Phase
CONROD AS/AE Axial Stress/Strain ASR/AER Axial Stress/Strain RM
10 MSA Axial safety margin ASI/AEI Axial Stress/Strain IP
Linear Format TS/TE Torsional Stress/Strain TSR/TER Torsional Stress/ RM
Strain
MST Torsional safety margin TSI/TEI Torsional Stress/ IP
Strain
Element Name Real Stresses or Strains Complex Stresses or Strains
(Code)
NDDL Item NDDL Item Real/
entity entity Mag. or
Imag./
Phase
CPENTA CID Stress coordinate CID Stress coordinate
system system
68
CTYPE Coordinate type CTYPE Coordinate type
Linear Format
(Character) (Character)
NODEF Number of active points NODEF Number of active
points
GRID External grid ID GRID External grid ID

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 4-25

© 2024 Siemens
 4. Simcenter Nastran Output Files

SX/EX Normal x SXR/EXR Normal x RM

TXY/ETXY Shear xy SYR/EYR Normal y RM
P1/EP1 First principal SZR/EZR Normal z RM
P1X/P1X First principal x cosine TXYR/ Shear xy RM
ETXYR
P2X/P2X Second principal x TYZR/ Shear yz RM
cosine ETYZR
P3X/P3X Third principal x cosine TZXR/ Shear zx RM
ETZXR
PR/EPR Mean pressure SXI/EXI Normal x IP
OCT/EOCT von Mises or SYI/EYI Normal y IP
Octahedral shear stress
SY/EY Normal y SZI/EZI Normal z IP
TYZ/ETYZ Shear yz TXYI/ Shear xy IP
ETXYI
P2/EP2 Second principal TYZI/ Shear yz IP
ETYZI
P1Y/P1Y First principal y cosine TZXI/ Shear zx IP
ETZXI
P2Y/P2Y Second principal y
cosine
P3Y/P3Y Third principal y cosine
SZ/EZ Normal z
TZX/ETZX Shear zx
P3/EP3 Third principal
P1Z/P1Z First principal z cosine
P2Z/P2Z Second principal z
cosine
P3Z/P3Z Third principal z cosine
Element Name Real Stresses or Strains Complex Stresses or Strains
(Code)
NDDL Item NDDL Item Real/
entity entity Mag. or
Imag./
Phase

4-26 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 MONPNT2

CPLSTN3 SX Normal stress in x SXR Normal stress in x RM

271 SY Normal stress in y SXI Normal stress in x IP
Triangle plane SZ Normal stress in z SYR Normal stress in y RM
strain
SP In-plane shear stress SYI Normal stress in y IP
Linear Format
SMAX Von Mises stress SZR Normal stress in z RM
(Center Only)
SZI Normal stress in z IP
SPR In-plane shear stress RM
SPI In-plane shear stress IP
Element Name Real Stresses or Strains Complex Stresses or Strains
(Code)
NDDL Item NDDL Item Real/
entity entity Mag. or
Imag./
Phase
CPLSTN4 GRID Grid ID, 0 for centroid GRID Grid ID, 0 for
centroid
272
SX Normal stress in x SXR Normal stress in x RM
Quadrilateral
plane strain SY Normal stress in y SXI Normal stress in x IP
Linear Format SZ Normal stress in z SYR Normal stress in y RM
(Center and SP In-plane shear stress SYI Normal stress in y IP
Corners)
SMAX Von Mises stress SZR Normal stress in z RM
SZI Normal stress in z IP
SPR In-plane shear stress RM
SPI In-plane shear stress IP
Element Name Real Stresses or Strains Complex Stresses or Strains
(Code)
NDDL Item NDDL Item Real/
entity entity Mag. or
Imag./
Phase
CPLSTN6 GRID Grid ID, 0 for centroid GRID Grid ID, 0 for
273 centroid
Triangle plane
SX Normal stress in x SXR Normal stress in x RM
strain
Linear Format SY Normal stress in y SXI Normal stress in x IP
(Center and
Corners) SZ Normal stress in z SYR Normal stress in y RM

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 4-27

© 2024 Siemens
 4. Simcenter Nastran Output Files

SP In-plane shear stress SYI Normal stress in y IP

SMAX Von Mises stress SZR Normal stress in z RM
SZI Normal stress in z IP
SPR In-plane shear stress RM
SPI In-plane shear stress IP
Element Name Real Stresses or Strains Complex Stresses or Strains
(Code)
NDDL Item NDDL Item Real/
entity entity Mag. or
Imag./
Phase
CPLSTN8 GRID Grid ID, 0 for centroid GRID Grid ID, 0 for
centroid
274
SX Normal stress in x SXR Normal stress in x RM
Quadrilateral
plane strain SY Normal stress in y SXI Normal stress in x IP
Linear Format SZ Normal stress in z SYR Normal stress in y RM
(Center and SP In-plane shear stress SYI Normal stress in y IP
Corners)
SMAX Von Mises stress SZR Normal stress in z RM
SZI Normal stress in z IP
SPR In-plane shear stress RM
SPI In-plane shear stress IP
Element Name Real Stresses or Strains Complex Stresses or Strains
(Code)
NDDL Item NDDL Item Real/
entity entity Mag. or
Imag./
Phase
CPLSTS3 SX Normal stress in x SXR Normal stress in x RM
275 SY Normal stress in y SXI Normal stress in x IP
Triangle plane SZ Normal stress in z SYR Normal stress in y RM
stress
SP In-plane shear stress SYI Normal stress in y IP
Linear Format
SMAX Von Mises stress SZR Normal stress in z RM
(Center Only)
SZI Normal stress in z IP
SPR In-plane shear stress RM

4-28 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 MONPNT2

SPI In-plane shear stress IP

Element Name Real Stresses or Strains Complex Stresses or Strains
(Code)
NDDL Item NDDL Item Real/
entity entity Mag. or
Imag./
Phase
CPLSTS4 GRID Grid ID, 0 for centroid GRID Grid ID, 0 for
centroid
276
SX Normal stress in x SXR Normal stress in x RM
Quadrilateral
plane stress SY Normal stress in y SXI Normal stress in x IP
Linear Format SZ Normal stress in z SYR Normal stress in y RM
(Center and SP In-plane shear stress SYI Normal stress in y IP
Corners)
SMAX Von Mises stress SZR Normal stress in z RM
SZI Normal stress in z IP
SPR In-plane shear stress RM
SPI In-plane shear stress IP
Element Name Real Stresses or Strains Complex Stresses or Strains
(Code)
NDDL Item NDDL Item Real/
entity entity Mag. or
Imag./
Phase
CPLSTS6 GRID Grid ID, 0 for centroid GRID Grid ID, 0 for
centroid
277
SX Normal stress in x SXR Normal stress in x RM
Triangle plane
stress SY Normal stress in y SXI Normal stress in x IP
Linear Format SZ Normal stress in z SYR Normal stress in y RM
(Center and SP In-plane shear stress SYI Normal stress in y IP
Corners)
SMAX Von Mises stress SZR Normal stress in z RM
SZI Normal stress in z IP
SPR In-plane shear stress RM
SPI In-plane shear stress IP
Element Name Real Stresses or Strains Complex Stresses or Strains
(Code)

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 4-29

© 2024 Siemens
 4. Simcenter Nastran Output Files

NDDL Item NDDL Item Real/

entity entity Mag. or
Imag./
Phase
CPLSTS8 GRID Grid ID, 0 for centroid GRID Grid ID, 0 for
centroid
278
SX Normal stress in x SXR Normal stress in x RM
Quadrilateral
plane stress SY Normal stress in y SXI Normal stress in x IP
Linear Format SZ Normal stress in z SYR Normal stress in y RM
(Center and SP In-plane shear stress SYI Normal stress in y IP
Corners)
SMAX Von Mises stress SZR Normal stress in z RM
SZI Normal stress in z IP
SPR In-plane shear stress RM
SPI In-plane shear stress IP
Element Name Real Stresses or Strains Complex Stresses or Strains
(Code)
NDDL Item NDDL Item Real/
entity entity Mag. or
Imag./
Phase
CPYRAM CID Stress/Strain coordinate CID Stress coordinate
system system
255
CTYPE Coordinate type CTYPE Coordinate type
Linear Format
(Character) (Character)
NODEF Number of active points NODEF Number of active
points
GRID External grid ID GRID External grid ID
(0=center) (0=center)
SX/EX Normal x SXR/EXR Normal x RM
TXY/ETXY Shear xy SYR/EYR Normal y RM
P1/EP1 First principal SZR/EZR Normal z RM
P1X/P1X First principal x cosine TXYR/ Shear xy RM
ETXYR
P2X/P2X Second principal x TYZR/ Shear yz RM
cosine ETYZR

4-30 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 MONPNT2

P3X/P3X Third principal x cosine TZXR/ Shear zx RM

ETZXR
PR/EPR Mean pressure SXI/EXI Normal x IP
OCT/EOCT von Mises or octahedral SYI/EYI Normal y IP
shear stress
SY/EY Normal y SZI/EZI Normal z IP
TYZ/ETYZ Shear yz TXYI/ Shear xy IP
ETXYI
P2/EP2 Second principal TYZI/ Shear yz IP
ETYZI
P1Y First principal y cosine TZXI/ Shear zx IP
ETZXI
P2Y Second principal y
cosine
P3Y Third principal y cosine
SZ/EZ Normal z
TZX/ETZX Shear zx
P3/EP3 Third principal
P1Z First principal z cosine
P2Z Second principal z
cosine
P3Z Third principal z cosine
Element Name Real Stresses or Strains Complex Stresses or Strains
(Code)
NDDL Item NDDL Item Real/
entity entity Mag. or
Imag./
Phase
CQUAD4 FD1 Z1=Fiber Distance 1 FD1 Z1=Fiber Distance 1
33 SX1/EX1 Normal x at Z1 SX1R/ Normal x at Z1 RM
EX1R
Linear Format
Center Only SY1/EY1 Normal y at Z1 SX1I/EX1I Normal x at Z1 IP
TXY1/EXY1 Shear xy at Z1 SY1R/ Normal y at Z1 RM
EY1R
SA1/EA1 Shear angle at Z1 SY1I/EY1I Normal y at Z1 IP

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 4-31

© 2024 Siemens
 4. Simcenter Nastran Output Files

SMJRP1/ Major principal at Z1 TXY1R/ Shear xy at Z1 RM

EMJRP1 EXY1R
SMNRP1/ Minor principal at Z1 TXY1I/ Shear xy at Z1 IP
EMNRP1 EXY1I
SMAX1/ von Mises or maximum FD2/FD2 Z2=Fiber distance 2
EMAX1 shear at Z1
FD2 Z2=Fiber distance 2 SX2R/ Normal x at Z2 RM
EX2R
SX2/EX2 Normal x at Z2 SX2I/EX2I Normal x at Z2 IP
SY2/EY2 Normal y at Z2 SY2R/ Normal y at Z2 RM
EY2R
TXY2/EXY2 Shear xy at Z2 SY2I/EY2I Normal y at Z2 IP
SA2/EA2 Shear angle at Z2 TXY2R/ Shear xy at Z2 RM
EXY2R
SMJRP2/ Major principal at Z2 TXY2I/ Shear xy at Z2 IP
EMJRP2 EXY2I
SMNRP2/ Minor principal at Z2
EMNRP2
SMAX2/ von Mises or maximum
EMAX2 shear at Z2
Element Name Real Stresses or Strains Complex Stresses or Strains
(Code)
NDDL Item NDDL Item Real/
entity entity Mag. or
Imag./
Phase
CQUAD144 TERM CEN/ TERM CEN/
144
Linear Format GRID 4 GRID 4
Center and FD1 Z1-Fiber distance FD1 Z1-Fiber distance
Corners
SX1/EX1 Normal x at Z1 SX1R/ Normal x at Z1 RM
EX1R
SY1/EY1 Normal y at Z1 SX1I/EX1I Normal x at Z1 IP
TXY1/ Shear xy at Z1 SY1R/ Normal y at Z1 RM
ETXY1 EY1R
A1 Shear angle at Z1 SY1I/EY1I Normal y at Z1 IP

4-32 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 MONPNT2

MJRP1/ Major principal at Z1 TXY1R/ Shear xy at Z1 RM

EMJRP1 ETXY1R
MNRP1/ Minor principal at Z1 TXY1I/ Shear xy at Z1 IP
EMNRP1 ETXY1I
TMAX1/ von Mises or maximum FD2 Z2-Fiber distance
ETMAX1 shear at Z1
FD2 Z2-Fiber distance SX2R/ Normal x at Z2 RM
EX2R
SX2/EX2 Normal x at Z2 SX2I/EX2I Normal x at Z2 IP
SY2/EY2 Normal y at Z2 SY2R/ Normal y at Z2 RM
EY2R
TXY2/ Shear xy at Z2 SY2I/EY2I Normal y at Z2 IP
ETXY2
A2 Shear angle at Z2 TXY2R/ Shear xy at Z2 RM
ETXY2R
MJRP2/ Major principal at Z2 TXY2I/ Shear xy at Z2 IP
EMJRP2 ETXY2I
MNRP2/ Minor principal at Z2
EMNRP2
TMAX2/ von Mises or maximum
ETMRP2 shear at Z2
Element Name Real Stresses or Strains Complex Stresses or Strains
(Code)
NDDL Item NDDL Item Real/
entity entity Mag. or
Imag./
Phase
CQUAD4 PLY Lamina Number PLY Lamina Number
95 SX1/EX1 Normal-1 SX1/EX1 Normal-1 RM
Composite SY1/EY1 Normal-2 SY1/EY1 Normal-2 RM
Center Only T1/ET1 Shear-12 T1/ET1 Shear-12 RM
SL1/EL1 Shear-1Z SL1/EL1 Shear-1Z RM
SL2/EL2 Shear-2Z SL2/EL2 Shear-2Z RM
A1 Shear angle SX1I/EX1I Normal-1 IP
MJRP1/ Major principal SY1I/EY1I Normal-2 IP
EMJRP1

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 4-33

© 2024 Siemens
 4. Simcenter Nastran Output Files

MNRP1/ Minor principal T1I/ET1I Shear-12 IP

EMNRP1
TMAX1/ Maximum shear SL1I/EL1I Shear-1Z IP
ETMAX1
SL2I/EL2I Shear-2Z IP
Element Name Real Stresses or Strains Complex Stresses or Strains
(Code)
NDDL Item NDDL Item Real/
entity entity Mag. or
Imag./
Phase
CQUAD8 TERM CEN/ TERM CEN/
64
Linear Format GRID 4 GRID 4
Center and FD1 Z1-Fiber distance FD1 Z1-Fiber distance
Corners
SX1/EX1 Normal x at Z1 SX1R/ Normal x at Z1 RM
EX1R
SY1/EY1 Normal y at Z1 SX1I/EX1I Normal x at Z1 IP
TXY1/ Shear xy at Z1 SY1R/ Normal y at Z1 RM
ETXY1 EY1R
A1 Shear angle at Z1 SY1I/EY1I Normal y at Z1 IP
SMJRP1/ Major principal at Z1 TXY1R/ Shear xy at Z1 RM
EMJRP1 ETXY1R
SMNRP1/ Minor principal at Z1 TXY1I/ Shear xy at Z1 IP
EMNRP1 ETXY1I
TMAX1/ von Mises or maximum FD2 Z2-Fiber distance
ETMAX1 shear at Z1
FD2 Z2-Fiber distance SX2R/ Normal x at Z2 RM
EX2R
SX2/EX2 Normal x at Z2 SX2I/EX2I Normal x at Z2 IP
SY2/EY2 Normal y at Z2 SY2R/ Normal y at Z2 RM
EY2R
TXY2/ Shear xy at Z2 SY2I/EY2I Normal y at Z2 IP
ETXY2
A2 Shear angle at Z2 TXY2R/ Shear xy at Z2 RM
ETXY2R

4-34 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 MONPNT2

SMJRP2/ Major principal at Z2 TXY2I/ Shear xy at Z2 IP

EMJRP2 ETXY2I
SMNRP2/ Minor principal at Z2
EMNRP2
TMAX2/ von Mises or maximum
ETMAX2 shear at Z2
Element Name Real Stresses or Strains Complex Stresses or Strains
(Code)
NDDL Item NDDL Item Real/
entity entity Mag. or
Imag./
Phase
CQUAD8 Same as Same as CQUAD4(95) Same as Same as
96 CQUAD4(9 CQUAD4( CQUAD4(95)
Composite 5) 95)
Center Only
Element Name Real Stresses or Strains Complex Stresses or Strains
(Code)
NDDL Item NDDL Item Real/
entity entity Mag. or
Imag./
Phase
CQUADR TERM CEN/ TERM CEN/
82
Center and GRID 4 GRID 4
Corners
FD1 Z1-Fiber distance FD1 Z1-Fiber distance
SX1/EX1 Normal x at Z1 SX1R/ Normal x at Z1 RM
EX1R
SY1/EY1 Normal y at Z1 SX1I/EX1I Normal x at Z1 IP
TXY1/ Shear xy at Z1 SY1R/ Normal y at Z1 RM
ETXY1 EY1R
A1 Shear angle at Z1 SY1I/EY1I Normal y at Z1 IP
MJRP1/ Major principal at Z1 TXY1R/ Shear xy at Z1 RM
EMJRP1 ETXY1R
MNRP1/ Minor principal at Z1 TXY1I/ Shear xy at Z1 IP
EMNRP1 ETXY1I
TMAX1/ von Mises or maximum FD2 Z2-Fiber distance
ETMAX1 shear at Z1

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 4-35

© 2024 Siemens
 4. Simcenter Nastran Output Files

FD2 Z2-Fiber distance SX2R/ Normal x at Z2 RM

EX2R
SX2/EX2 Normal x at Z2 SX2I/EX2I Normal x at Z2 IP
SY2/EY2 Normal y at Z2 SY2R/ Normal y at Z2 RM
EY2R
TXY2/ Shear xy at Z2 SY2I/EY2I Normal y at Z2 IP
ETXY2
A2 Shear angle at Z2 TXY2R/ Shear xy at Z2 RM
ETXY2R
MJRP2/ Major principal at Z2 TXY2I/ Shear xy at Z2 IP
EMJRP2 ETXY2I
MNRP2/ Minor principal at Z2
EMNRP2
TMAX2/ von Mises or maximum
ETMRP2 shear at Z2
Element Name Real Stresses or Strains Complex Stresses or Strains
(Code)
NDDL Item NDDL Item Real/
entity entity Mag. or
Imag./
Phase
CQUADR Same as Same as CQUAD4(33) Same as Same as
228 CQUAD4(3 CQUAD4( CQUAD4(33)
Center only 3) 33)
(not supported by
old formulation)
Element Name Real Stresses or Strains Complex Stresses or Strains
(Code)
NDDL Item NDDL Item Real/
entity entity Mag. or
Imag./
Phase
CQUADR Same as Same as CQUAD4(95) Same as Same as
232 CQUAD4(9 CQUAD4( CQUAD4(95)
Composite 5) 95)
Center Only
(not supported by
old formulation)
Element Name Real Stresses or Strains Complex Stresses or Strains
(Code)
NDDL Item NDDL Item Real/
entity entity Mag. or

4-36 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 MONPNT2

Imag./
Phase
CQUADX4 CTYPE 0 (Result at Center) CTYPE 0 (Result at Center)
243 LOC Location LOC Location
Linear Format SX Radial SXR Radial RM
Center and SY Azimuthal SXI Radial IP
Corners
SZ Axial SYR Azimuthal RM
Grid and Gauss
SS Shear stress SYI Azimuthal lP
MAXP Maximum principal SZR Axial RM
TMAX Maximum shear SZI Axial lP
OCTS von Mises or octahedral SSR Shear stress RM
SSI Shear stress IP
Element Name Real Stresses or Strains Complex Stresses or Strains
(Code)
NDDL Item NDDL Item Real/
entity entity Mag. or
Imag./
Phase
CQUADX8 Same as Same as CQUAD4(33) Same as Same as
245 CQUAD4(3 CQUAD4( CQUAD4(33)
Linear Format 3) 33)
Grid and Gauss
Center and
Corners
Element Name Real Stresses or Strains Complex Stresses or Strains
(Code)
NDDL Item NDDL Item Real/
entity entity Mag. or
Imag./
Phase
CROD Same as Same as CONROD(10) Same as Same as
1 CONROD(1 CONROD( CONROD(10)
Linear Format 0) 10)
Element Name Real Stresses or Strains Complex Stresses or Strains
(Code)
NDDL Item NDDL Item Real/
entity entity Mag. or
Imag./
Phase

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 4-37

© 2024 Siemens
 4. Simcenter Nastran Output Files

CSHEAR TMAX/ Maximum shear TMAXR/ Maximum shear RM

ETMAX ETMAXR
4
TAVG/ Average shear TMAXI/ Maximum shear lP
ETAVG ETMAXI
MS Safety margin TAVGR/ Average shear RM
ETAVGR
TAVGI/ Average shear IP
ETAVGI
Element Name Real Stresses or Strains Complex Stresses or Strains
(Code)
NDDL Item NDDL Item Real/
entity entity Mag. or
Imag./
Phase
CTETRA CID Stress coordinate CID Stress coordinate
system system
39
CTYPE Coordinate type CTYPE Coordinate type
Linear Format
(Character) (Character)
NODEF Number of active points NODEF Number of active
points
GRID External grid ID GRID External grid ID
(0=center) (0=center)
SX/EX Normal x SXR/EXR Normal x RM
TXY/ETXY Shear xy SYR/EYR Normal y RM
P1/EP1 First principal SZR/EZR Normal z RM
P1X First principal x cosine TXYR/ Shear xy RM
ETXYR
P2X Second principal x TYZR/ Shear yz RM
cosine ETYZR
P3X Third principal x cosine TZXR/ Shear zx RM
ETZXR
PR/EPR Mean pressure SXI/EXI Normal x IP
OCT/EOCT von Mises or octahedral SYI/EYI Normal y IP
shear stress
SY/EY Normal y SZI/EZI Normal z IP
TYZ/ETYZ Shear yz TXYI/ Shear xy IP
ETXYI

4-38 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 MONPNT2

P2/EP2 Second principal TYZI/ Shear yz IP

ETYZI
P1Y First principal y cosine TZXI/ Shear zx IP
ETZXI
P2Y Second principal y
cosine
P3Y Third principal y cosine
SZ/EZ Normal z
TZX/ETZX Shear zx
P3/EP3 Third principal
P1Z First principal z cosine
P2Z Second principal z
cosine
P3Z Third principal z cosine
Element Name Real Stresses or Strains Complex Stresses or Strains
(Code)
NDDL Item NDDL Item Real/
entity entity Mag. or
Imag./
Phase
CTRAX3 Same As Same As CQUADX4 Same As Same As CQUADX4
242 CQUADX4 (243) CQUADX4 (243)
Linear Format (243) (243)
Center and
Corners
Grid or Gauss
Element Name Real Stresses or Strains Complex Stresses or Strains
(Code)
NDDL Item NDDL Item Real/
entity entity Mag. or
Imag./
Phase
CTRAX6 Same As Same As CQUADX4 Same As Same As CQUADX4
244 CQUADX4 (243) CQUADX4 (243)
Linear Format (243) (243)
Grid and Gauss
Center and
Corners
Element Name Real Stresses or Strains Complex Stresses or Strains
(Code)

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 4-39

© 2024 Siemens
 4. Simcenter Nastran Output Files

NDDL Item NDDL Item Real/

entity entity Mag. or
Imag./
Phase
CTRIA3 Same as Same as CQUAD4(33) Same as Same as
74 CQUAD4(3 CQUAD4( CQUAD4(33)
Linear Format 3) 33)
Center Only
CTRIA3 Same as Same as CQUAD4(95) Same as Same as
97 CQUAD4(9 CQUAD4( CQUAD4(95)
Composite 5) 95)
Center Only
Element Name Real Stresses or Strains Complex Stresses or Strains
(Code)
NDDL Item NDDL Item Real/
entity entity Mag. or
Imag./
Phase
CTRIA6 Same as Same as CQUAD8(64) Same as Same as
75 CQUAD8(6 CQUAD8( CQUAD8(64)
Linear Format 4) 64)
Center and
Corners
CTRIA6 Same as Same as CQUAD4(95) Same as Same as
98 CQUAD4(9 CQUAD4( CQUAD4(95)
Composite 5) 95)
Center Only
Element Name Real Stresses or Strains Complex Stresses or Strains
(Code)
NDDL Item NDDL Item Real/
entity entity Mag. or
Imag./
Phase
CTRlAR TERM CEN/ TERM CEN/
70
Center and GRID 4 GRID 4
Corners
FD1 Z1-Fiber distance FD1 Z1-Fiber distance
SX1/EX1 Normal x at Z1 SX1R/ Normal x at Z1 RM
EX1R
SY1/EY1 Normal y at Z1 SX1I/EX1I Normal x at Z1 IP

4-40 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 MONPNT2

TXY1/ Shear xy at Z1 SY1R/ Normal y at Z1 RM

ETXY1 EY1R
A1 Shear angle at Z1 SY1I/EY1I Normal y at Z1 IP
MJRP1/ Major principal at Z1 TXY1R/ Shear xy at Z1 RM
EMJRP1 ETXY1R
MNRP1/ Minor principal at Z1 TXY1I/ Shear xy at Z1 IP
EMNRP1 ETXY1I
TMAX1/ von Mises or maximum FD2 Z2-Fiber distance
ETMAX1 shear at Z1
FD2 Z2-Fiber distance SX2R/ Normal x at Z2 RM
EX2R
SX2/EX2 Normal x at Z2 SX2I/EX2I Normal x at Z2 IP
SY2/EY2 Normal y at Z2 SY2R/ Normal y at Z2 RM
EY2R
TXY2/ Shear xy at Z2 SY2I/EY2I Normal y at Z2 IP
ETXY2
A2 Shear angle at Z2 TXY2R/ Shear xy at Z2 RM
ETXY2R
MJRP2/ Major principal at Z2 TXY2I/ Shear xy at Z2 IP
EMJRP2 ETXY2I
MNRP2/ Minor principal at Z2
EMNRP2
TMAX2/ von Mises or maximum
ETMAX2 shear at Z2
Element Name Real Stresses or Strains Complex Stresses or Strains
(Code)
NDDL Item NDDL Item Real/
entity entity Mag. or
Imag./
Phase
CTRlAR FD1 Z1=Fiber Distance 1 FD1 Z1=Fiber Distance 1
227 SX1/EX1 Normal x at Z1 SX1R/ Normal x at Z1 RM
EX1R
Center Only
(not supported by SY1/EY1 Normal y at Z1 SX1I/EX1I Normal x at Z1 IP
old formulation) TXY1/EXY1 Shear xy at Z1 SY1R/ Normal y at Z1 RM
EY1R
SA1/EA1 Shear angle at Z1 SY1I/EY1I Normal y at Z1 IP

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 4-41

© 2024 Siemens
 4. Simcenter Nastran Output Files

SMJRP1/ Major principal at Z1 TXY1R/ Shear xy at Z1 RM

EMJRP1 EXY1R
SMNRP1/ Minor principal at Z1 TXY1I/ Shear xy at Z1 IP
EMNRP1 EXY1I
SMAX1/ von Mises or maximum FD2/FD2 Z2=Fiber distance 2
EMAX1 shear at Z1
FD2 Z2=Fiber distance 2 SX2R/ Normal x at Z2 RM
EX2R
SX2/EX2 Normal x at Z2 SX2I/EX2I Normal x at Z2 IP
SY2/EY2 Normal y at Z2 SY2R/ Normal y at Z2 RM
EY2R
TXY2/EXY2 Shear xy at Z2 SY2I/EY2I Normal y at Z2 IP
SA2/EA2 Shear angle at Z2 TXY2R/ Shear xy at Z2 RM
EXY2R
SMJRP2/ Major principal at Z2 TXY2I/ Shear xy at Z2 IP
EMJRP2 EXY2I
SMNRP2/ Minor principal at Z2
EMNRP2
TMAX2/ von Mises or maximum
EMAX2 shear at Z2
Element Name Real Stresses or Strains Complex Stresses or Strains
(Code)
NDDL Item NDDL Item Real/
entity entity Mag. or
Imag./
Phase
CTRlAR Same as Same as CQUAD4(95) Same as Same as
233 CQUAD4(9 CQUAD4( CQUAD4(95)
Composite 5) 95)
Center Only
(not supported by
old formulation)
Element Name Real Stresses or Strains Complex Stresses or Strains
(Code)
NDDL Item NDDL Item Real/
entity entity Mag. or
Imag./
Phase
CTUBE AS Axial Stress
3
Linear Format SE Equivalent Stress

4-42 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 MONPNT2

TE Total Strain
EPS Eff Plastic Strain
ECS Eff Creep Strain
LTS Linear Torsional Stress

FORCE MONITOR TABLE

Element Name Real Element Forces Complex Element Forces

Code
NDDL Item NDDL Item Real/Mag.
entity entity or Imag/
Phase
CBAR BM1A Bending End A plane BM1AR Bending End A plane 1 RM
1
34
BM2A Bending End A plane BM2AR Bending End A plane 2 RM
2
BM1B Bending End B plane BM1BR Bending End B plane 1 RM
1
BM2B Bending End B plane BM2BR Bending End B plane 2 RM
2
TS1 Shear plane 1 TS1R Shear plane 1 RM
TS2 Shear plane 2 TS2R Shear plane 2 RM
AF Axial force AFR Axial force RM
TRQ Torque TRQR Torque RM
BM1AI Bending End A plane 1 IP
BM2AI Bending End A plane 2 IP
BM1BI Bending End B plane 1 IP
BM2BI Bending End B plane 2 IP
TS1I Shear plane 1 IP
TS2I Shear plane 2 IP
AFI Axial force lP
TRQI Torque IP
Element Name Real Element Forces Complex Element Forces
Code

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 4-43

© 2024 Siemens
 4. Simcenter Nastran Output Files

NDDL Item NDDL Item Real/Mag.

entity entity or Imag/
Phase
CBAR SD Station Distance/ SD Station Distance/ RM
Length Length
100
BM1 Bending Moment BM1r Bending Moment Plane RM
Plane 1 1
BM2 Bending Moment BM2r Bending Moment Plane RM
Plane 2 2
TS1 Shear Force Plane 1 TP1r Shear Force Plane 1 RM
TS2 Shear Force Plane 2 TP2r Shear Force Plane 2 RM
AF Axial AFr Axial RM
TRQ Torque TrQR Torque RM
NDDL entitys are BM1i Bending Moment Plane IP
given for End A. 1
Addition of the
quantity (K-1) * 7
to the NDDL entity
points to the same
information for other
stations, where K is
the station number.
K=8 for End B and
2 through 7 for
intermediate stations.
BM2i Bending Moment Plane IP
2
TS1i Shear Force Plane 1 IP
TS2i Shear Force Plane 2 IP
AFi Axial IP
TRQi Torque IP
(NDDL entitys above
are given for End A.
For codes 2 through 14
at intermediate stations
add (K-1) * 13 and K
is the station number,
and for codes at End B,
K+number of stations
plus 1.)

4-44 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 MONPNT2

Element Name Real Element Forces Complex Element Forces

Code
NDDL Item NDDL Item Real/Mag.
entity entity or Imag/
Phase
CBEAM GRID External grid point ID GRID External grid point ID
2 SD Station distance/ SD Station distance/length
length
BM1 Bending moment BM1R Bending moment plane RM
plane 1 1
BM2 Bending moment BM2R Bending moment plane RM
plane 2 2
TS1 Web shear plane 1 TS1R Web shear plane 1 RM
TS2 Web shear plane 2 TS2R Web shear plane 2 RM
AF Axial force AFR Axial force RM
TTRQ Total torque TTRQR Total torque RM
WTRQ Warping torque WTRQR Warping torque RM
(NDDL entitys are BM1I Bending moment plane IP
given for End A. 1
Addition of the
quantity (K-1) 9 to BM2I Bending moment plane IP
the NDDL entity 2
points to the same TS1I Web shear plane 1 lP
information for other
stations, where K is TS2I Web shear plane 2 IP
the station number.
AFI Axial force IP
K=11 for End B
and 2 through TTRQI Total torque IP
10 for intermediate
stations.) WTRQI Warping torque lP
(NDDL entitys are given
for End A. Addition
of the quantity (K-1)
16 to the NDDL entity
points to the same
information for other
stations, where K is the
station number. K=11
for End B and 2 through
10 for intermediate
stations.)

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 4-45

© 2024 Siemens
 4. Simcenter Nastran Output Files

Element Name Real Element Forces Complex Element Forces

Code
NDDL Item NDDL Item Real/Mag.
entity entity or Imag/
Phase
CBEAR TX Force-x TXR Force-x RM
280 TY Force-y TYR Force-y RM
TZ Force-z TZR Force-z RM
RX Moment-x RXR Moment-x RM
RY Moment-y RYR Moment-y RM
RZ Moment-z RZR Moment-z RM
TXI Force-x IP
TYI Force-y IP
TZI Force-z IP
RXI Moment-x IP
RYI Moment-y IP
RZI Moment-z IP
Element Name Real Element Forces Complex Element Forces
Code
NDDL Item NDDL Item Real/Mag.
entity entity or Imag/
Phase
CBEND GRID External grid point ID GRID External grid point ID
69 BM1 Bending moment BM1R Bending moment plane RM
plane 1 1
BM2 Bending moment BM2R Bending moment plane RM
plane 2 2
TS1 Shear plane 1 TS1R Shear plane 1 RM
TS2 Shear plane 2 TS2R Shear plane 2 RM
AF Axial force AFR Axial force RM
TRQ Torque TRQR Torque RM
(NDDL entitys are BM1I Bending moment plane IP
given for End A. 1
NDDL entitys 9
through 15 point to

4-46 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 MONPNT2

the same information

for End B.)
BM2I Bending moment plane IP
2
TS1I Shear plane 1 IP
TS2I Shear plane 2 IP
AFI Axial force lP
TRQi Torque IP
(NDDL entitys are given
for End A. NDDL entitys
15 through 27 point to
the same information
for End B.)
Element Name Real Element Forces Complex Element Forces
Code
NDDL Item NDDL Item Real/Mag.
entity entity or Imag/
Phase
CBUSH FX Force-x FXR Force-x RM
102 FY Force-y FYR Force-y RM
Linear Format FZ Force-z FZR Force-z RM
MX Moment-x MXR Moment-x RM
MY Moment-y MYR Moment-y RM
MZ Moment-z MZR Moment-z RM
FXI Force-x IP
FYI Force-y IP
FZI Force-z IP
MXI Moment-x IP
MYI Moment-y IP
MZI Moment-z IP
Element Name Real Element Forces Complex Element Forces
Code
NDDL Item NDDL Item Real/Mag.
entity entity or Imag/
Phase

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 4-47

© 2024 Siemens
 4. Simcenter Nastran Output Files

CDAMP1 Same as Same as CELAS1(11) Same as Same as CELAS1(11)

20 CELAS1( CELAS1(1
11) 1)
CDAMP2
21 Same as Same as CELAS1(11) Same as Same as CELAS1(11)
CELAS1( CELAS1(1
CDAMP3
11) 1)
22
CDAMP4 Same as Same as CELAS1(11) Same as Same as CELAS1(11)
23 CELAS1( CELAS1(1
11) 1)
Same as Same as CELAS1(11) Same as Same as CELAS1(11)
CELAS1( CELAS1(1
11) 1)
Element Name Real Element Forces Complex Element Forces
Code
NDDL Item NDDL Item Real/Mag.
entity entity or Imag/
Phase
CDUM3 Same as Same as CELAS1(11) Same as Same as CELAS1(11)
thru CELAS1( CELAS1(1
CDUM9 11) 1)
(55-61)
Element Name Real Element Forces Complex Element Forces
Code
NDDL Item NDDL Item Real/Mag.
entity entity or Imag/
Phase
CELAS1 F Force FR Force RM
11 FI Force IP
CELAS2 Same as CELAS1 Same as CELAS1
12
CELAS3 Same as CELAS1 Same as CELAS1
13
CELAS4 Same as CELAS1 Same as CELAS1
14
Element Name Real Element Forces Complex Element Forces
Code
NDDL Item NDDL Item Real/Mag.
entity entity or Imag/
Phase
CFAST BM1A mz bending End A BM1AR mz bending End A RM
119 plane 1 plane 1

4-48 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 MONPNT2

BM2A my bending End A BM2AR my bending End A RM

plane 2 plane 2
BM1B mz bending End B BM1BR mz bending End B RM
plane 1 plane 1
BM2B my bending End B BM2BR my bending End B RM
plane 2 plane 2
TS1 fy shear force plane 1 TS1R fy shear force plane 1 RM
TS2 fz shear force plane 2 TS2R fz shear force plane 2 RM
AF fx axial force AFR fx axial force RM
TRQ mx torque TRQR mx torque RM
BM1AI mz bending End A IP
plane 1
BM2AI my bending End A IP
plane 2
BM1BI mz bending End B IP
plane 1
BM2BI my bending End B IP
plane 2
TS1I fy shear force plane 1 IP
TS2I fz shear force plane 2 IP
AFI fx axial force IP
TRQI mx torque IP
Element Name Real Element Forces Complex Element Forces
Code
NDDL Item NDDL Item Real/Mag.
entity entity or Imag/
Phase
CONROD AF Axial force AFR Axial force RM
10 TRQ Torque AFI Axial force IP
TRQR Torque RM
TRQI Torque lP
Element Name Real Element Forces Complex Element Forces
Code
NDDL Item NDDL Item Real/Mag.
entity entity or Imag/
Phase

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 4-49

© 2024 Siemens
 4. Simcenter Nastran Output Files

CQUAD4 MX Membrane force x MXR Membrane force x RM

33 MY Membrane force y MYR Membrane force y RM
Center Only MXY Membrane force xy MXYR Membrane force xy RM
BMX Bending moment x BMXR Bending moment x RM
BMY Bending moment y BMYR Bending moment y RM
BMXY Bending moment xy BMXYR Bending moment xy RM
TX Shear x TXR Shear x RM
TY Shear y TYR Shear y RM
MXI Membrane force x IP
MYI Membrane force y IP
MXYI Membrane force xy IP
BMXI Bending moment x IP
BMYI Bending moment y IP
BMXYI Bending moment xy IP
TXI Shear x IP
TYI Shear y IP
Element Name Real Element Forces Complex Element Forces
Code
NDDL Item NDDL Item Real/Mag.
entity entity or Imag/
Phase
CQUAD4 TERM CEN/ TERM CEN/
144 GRID 4 GRID 4
Center and MX Membrane x MXR Membrane x RM
Corners
MY Membrane y MYR Membrane y RM
MXY Membrane xy MXYR Membrane xy RM
BMX Bending x BMXR Bending x RM
BMY Bending y BMYR Bending y RM
BMXY Bending xy BMXYR Bending xy RM
TX Shear x TXR Shear x RM
TY Shear y TYR Shear y RM
MXI Membrane x IP

4-50 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 MONPNT2

MYI Membrane y IP
MXYI Membrane xy IP
BMXI Bending x IP
BMYI Bending y IP
BMXYI Bending xy IP
TXI Shear x IP
TYI Shear y IP
Element Name Real Element Forces Complex Element Forces
Code
NDDL Item NDDL Item Real/Mag.
entity entity or Imag/
Phase
CQUAD8 Same as Same as Same as Same as
64 CQUAD1 CQUAD144(144) CQUAD14 CQUAD144(144)
Center and 44(144) 4(144)
Corners
Element Name Real Element Forces Complex Element Forces
Code
NDDL Item NDDL Item Real/Mag.
entity entity or Imag/
Phase
CQUADR Same as Same as Same as Same as
82 CQUAD1 CQUAD144(144) CQUAD14 CQUAD144(144)
Center and 44(144) 4(144)
Corners
CQUADR Same as Same as CQUAD4(33) Same as Same as CQUAD4(33)
228 CQUAD4( CQUAD4(3
Center Only 33) 3)
(not supported
by old
formulation)
Element Name Real Element Forces Complex Element Forces
Code
NDDL Item NDDL Item Real/Mag.
entity entity or Imag/
Phase
CROD Same as Same as CONROD(10) Same as Same as CONROD(10)
1 CONROD( CONROD(1
10) 0)
Element Name Real Element Forces Complex Element Forces
Code

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 4-51

© 2024 Siemens
 4. Simcenter Nastran Output Files

NDDL Item NDDL Item Real/Mag.

entity entity or Imag/
Phase
CSHEAR F41 Force 4 to 1 F41R Force 4 to 1 RM
4 F21 Force 2 to 1 F21R Force 2 to 1 RM
F12 Force 1 to 2 F12R Force 1 to 2 RM
F32 Force 3 to 2 F32R Force 3 to 2 RM
F23 Force 2 to 3 F23R Force 2 to 3 RM
F43 Force 4 to 3 F43R Force 4 to 3 RM
F34 Force 3 to 4 F34R Force 3 to 4 RM
F14 Force 1 to 4 F14R Force 1 to 4 RM
KF1 Kick force on 1 F41I Force 4 to 1 IP
S12 Shear 12 F21I Force 2 to 1 lP
KF2 Kick force on 2 F12I Force 1 to 2 IP
S23 Shear 23 F32I Force 3 to 2 IP
KF3 Kick force on 3 F23I Force 2 to 3 lP
S34 Shear 34 F43I Force 4 to 3 IP
KF4 Kick force on 4 F34I Force 3 to 4 IP
S41 Shear 41 F14I Force 1 to 4 IP
KF1R Kick force on 1 RM
S12R Shear 12 RM
KF2R Kick force on 2 RM
S23R Shear 23 RM
KF3R Kick force on 3 RM
S34R Shear 34 RM
KF4R Kick force on 4 RM
S41R Shear 41 RM
KF1I Kick force on 1 IP
S12I Shear 12 lP
KF2I Kick force on 2 IP
S23I Shear 23 IP

4-52 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 MONPNT2

KF3I Kick force on 3 IP

S34I Shear 34 IP
KF4I Kick force on 4 IP
S41I Shear 41 IP
Element Name Real Element Forces Complex Element Forces
Code
NDDL Item NDDL Item Real/Mag.
entity entity or Imag/
Phase
CTRIA3 Same as Same as CQUAD4(33) Same as Same as CQUAD4(33)
74 CQUAD4( CQUAD4(3
Center Only 33) 3)
CTRlA6 Same as Same as Same as Same as
75 CQUAD1 CQUAD144(144) CQUAD14 CQUAD144(144)
Center and 44(144) 4(144)
Corners
Element Name Real Element Forces Complex Element Forces
Code
NDDL Item NDDL Item Real/Mag.
entity entity or Imag/
Phase
CTRIAR Same as Same as Same as Same as
70 CQUAD1 CQUAD144(144) CQUAD14 CQUAD144(144)
Center and 44(144) 4(144)
Corners
CTRIAR Same as Same as CQUAD4(33) Same as Same as CQUAD4(33)
227 CQUAD4( CQUAD4(3
Center Only 33) 3)
(not supported
by old
formulation)
Element Name Real Element Forces Complex Element Forces
Code
NDDL Item NDDL Item Real/Mag.
entity entity or Imag/
Phase
CTUBE AF Axial force AFR Axial force RM
3 TRQ Torque AFI Axial force IP
TRQR Torque RM
TEQI Torque lP

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 4-53

© 2024 Siemens
 4. Simcenter Nastran Output Files

Element Name Real Element Forces Complex Element Forces

Code
NDDL Item NDDL Item Real/Mag.
entity entity or Imag/
Phase
CVlSC Same as Same as CONROD(10) Same as Same as CONROD(10)
24 CONROD( CONROD(1
10) 0)
Element Name Real Element Forces Complex Element Forces
Code
NDDL Item NDDL Item Real/Mag.
entity entity or Imag/
Phase
CWELDP (118) BM1A mz bending End A BM1AR mz bending End A RM
ELPAT or plane 1 plane 1
PARTPAT
BM2A my bending End A BM2AR my bending End A RM
CWELDC (117) plane 2 plane 2
ELIMID or GRIDID
with MSET=OFF BM1B mz bending End B BM1BR mz bending End B RM
plane 1 plane 1
CWELD (200)
ELIMID or GRIDID BM2B my bending End B BM2BR my bending End B RM
with MSET=ON plane 2 plane 2
TS1 fy shear force plane 1 TS1R fy shear force plane 1 RM
TS2 fz shear force plane 2 TS2R fz shear force plane 2 RM
AF fx axial force AFR fx axial force RM
TRQ mx torque TRQR mx torque RM
BM1AI mz bending End A IP
plane 1
BM2AI my bending End A IP
plane 2
BM1BI mz bending End B IP
plane 1
BM2BI my bending End B IP
plane 2
TS1I fy shear force plane 1 IP
TS2I fz shear force plane 2 IP
AFI fx axial force IP
TRQI mx torque IP

4-54 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 MONPNT3

MONPNT3

Integrated Load Monitor Point — Sums Grid Point Forces

Defines an integrated load monitor point at a chosen point and sums select grid point forces for SOL 101
and SOL 103.

FORMAT:

1 2 3 4 5 6 7 8 9 10
MONPNT3 NAME LABEL
AXES GRIDGRP ELEMGRP CP X1 X2 X3 XFLAG
CD

EXAMPLE:

MONPNT3 MPT33 This is a MONPNT3 monitor point.

246 1001 2002 1.0 2.0 3.0 MD
1003

FIELDS:

Field Contents

NAME A unique character string that identifies the monitor point. (Character; 8 characters
maximum; No default)

LABEL A string that identifies and labels the monitor point. (Character; 56 characters
maximum; Optional)

AXES Component axes about which to sum. (Integer; Any unique combination of the integers
1 through 6 with no embedded blanks; No default)

GRIDGRP GROUP entry that has a list of grids to be included in the monitor point. (Integer; No
default). See Remark 1.

ELEMGRP GROUP entry that has a list of elements to process at the monitor point. (Integer;
Optional). See Remark 2.

CP The identification number of a coordinate system in which the X1, X2, and X3
coordinates are defined. (Integer; Default = 0; Optional). See Remark 5.

X1, X2, X3 The coordinates in the CP coordinate system about which the forces are to be summed.
(Real; Default = 0.0).

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 4-55

© 2024 Siemens
 4. Simcenter Nastran Output Files

Field Contents

XFLAG Exclusion flag excludes the indicated Grid Point Force types from summation at the
monitor point. (Default = blank (no type excluded)).
If XFLAG = "S", SPC forces are excluded.
If XFLAG = "M", MPC forces are excluded.
If XFLAG = "A", "L", or "P", applied loads, which includes thermal loads, are excluded.
If XFLAG = "D", DMIGs at the monitored point are excluded.

CD The identification number of a coordinate system in which the results are output.
(Integer ≥ 0; Default = Coordinate system specified by the CP field)

REMARKS:

1. Requirements for grids that are included in the monitor point:

• You select a set of grids to sum forces and moments about the specified point.

• You must use only GRIDs in the GROUP definition. MONPNT3 does not process SPOINTs, EPOINTs,
and fluid grid points when they are in the GROUP definition.

2. Requirements for elements that are processed at the monitor point:

• When ELEMGRP is defined for any MONPNT3 bulk entry, you must define the GPFORCE=ALL
case control command for all subcases, and the request must be for the SORT1 output format.
You must either define GPFORCE=ALL in each subcase, or above the subcases (globally). When
ELEMGRP is blank, no contributions are made from the set of elements attached to the grid. The
elements attached to the grid are excluded.

3. Output for superelements that are processed at the monitor point:

• For internal and partitioned superelements, MONPNT3 results can be output for the residual or
downstream superelements.

• For external superelements, MONPNT3 results are only output for the residual superelement.

Note:
The various GROUPs and CPs defined for a MONPNT3 may not exist for all of the
superelements in a model. When this occurs, the MONPNT3 is not processed for the
superelement. A warning message indicates that the MONPNT3 is not processed.

4. Invalid grids or elements do not generate error or warning messages.

4-56 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 MONPNT3

5. A CP of zero (the default) references the basic coordinate system.

6. In a structure, you can apply MONPNT3 to determine the internal shear, moment, torque, and
thermal loads.

Example:
A long and slender component is divided and all grids that reside on the division are placed
in the GRIDGRP group. Based on the values of the ELEMGRP and XFLAG fields, the software
outputs various internal load resultants.

• ELEMGRP contains all elements connected to the GRIDGRP on the outer part of the split:

• XFLAG is blank:
The software calculates the internal load with the elements connected to the GRIDGRP
that are not included in ELEMGRP. This produces resultants on the inside of the split and
contains any loads applied to the GRIDGRP from any other source.

• XFLAG = SMAD:
The software calculates the internal load with the elements listed in ELEMGRP. This
produces resultants on the outer part of the split and does not contain loads applied to
the GRIDGRP from any other source.

7. The software prints the results in a table titled STRUCTURAL INTEGRATED FREE BODY
MONITOR POINT LOADS (MONPNT3). The total loads in this table include the thermal loads
unless you set XFLAG = A.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 4-57

© 2024 Siemens
 4. Simcenter Nastran Output Files

MONITOR

Print Selection for Monitor Data

Requests the type of monitor data to be output to the .f06 file in SOL 101 and SOL 103.

FORMAT:

EXAMPLES:

MONITOR(NOPNT2) = ALL

DESCRIBERS:

Describer Meaning

NOPNT2 Do not include MONPNT2 results in the MONITOR point prints. (Default =
Provide these prints)

NOPNT3 Do not include MONPNT3 results in the MONITOR point prints. (Default =
Provide these prints)

ALL or YES Print all monitor point results except for the selected items.

NONE or NO Do not print the monitor point results. (Default)

REMARKS:

1. The MONITOR command is required to obtain MONITOR results in the printed output.

Understanding the .f04 File

The .f04 file contains information that helps you monitor and tune the performance of your analysis job.
For example, the .f04 file contains:

• A summary of physical file information that describes the files used for DBsets

• A map showing memory allocation

• A “day log” that provides a DMAP execution summary

4-58 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 General Information

• A listing of memory and disk usage statistics

• A listing of database usage statistics that summarizes the I/O activity for the DBsets

• A summary of physical file I/O activity for each database file

• Module execution and statistical information

• Other optional diagnostic information as requested by the DIAG executive control command in the
input file

If you're performing an analysis using Simcenter Nastran's global iterative solver, the .f04 file also
contains important performance diagnostic information about the solver.

The software writes the .f04 file to FORTRAN unit 4.

See also

• “Flushing .f04 and .f06 Output to Disk” in the Simcenter Nastran Installation and Operations Guide

• “Performance Diagnostics” in the Simcenter Nastran Numerical Methods User's Guide

General Information

The .f04 file begins with a section that lists general information about the run.

• DAY TIME indicates the time of day as the run progresses.

• ELAPSED indicates the number of elapsed “wall clock” minutes:seconds at the beginning on the job.
This number may vary greatly depending on the system load in a time-sharing environment.

• I/O SEC is an internally generated measure of I/O. This measure is a count of the number of BUFFSIZE
blocks transferred, divided by the value of SYSTEM(84) (the estimated rate of number of blocks
moved per second). This number doesn't include I/O for loading the executable file or I/O for FORTRAN
files.

• DEL I/O is the I/O seconds spent in the previous DMAP statement.

• CPU SEC shows the number of CPU seconds that had elapsed when this line was printed.

In general, you should examine the above entries and look for large jumps in the CPU SEC number.
For example, the cost for “formatting” the output (modules SDR2 and OFP) may be greater than
the cost of generating the numbers. In such situations, simple techniques such as reducing output
requests may save more time than other more well-known techniques such as sequencing, OMITs,
better matrix methods, etc. In short, you should be aware of the expensive operations on each large
run.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 4-59

© 2024 Siemens
 4. Simcenter Nastran Output Files

• DEL CPU is the CPU time spent in the previous DMAP statement.

Summary of Physical File Information

This summary table describes the physical files used for the DBSets. A sample of this table, located near
the top of the .f04 file, is shown below.

S U M M A R Y O F P H Y S I C A L F I L E I N F O R M A T I O N
ASSIGNED PHYSICAL FILE NAME RECL
(BYTES) MODE FLAGS
------------------------------------------------------------
------------ ---- -----
/tmp/65872_57.SCRATCH
8192 R/W L
/tmp/65872_57.OBJSCR
8192 R/W L
/tmp/65872_57.MASTER
8192 R/W L
/tmp/65872_57.DBALL
8192 R/W L
/tmp/65872_57.DBALL2
8192 R/W L
/tmp/65872_57.SCR300
8192 R/W L
/scnas/aix/SSS.MASTERA 8192
R/O L
/scnas/aix/SSS.MSCOBJ 8192
R/O L
FLAG VALUES ARE --
F FFIO INTERFACE USED TO PROCESS FILE
H HPIO INTERFACE USED TO PROCESS FILE
L FILE HAS BEEN LOCKED
M FILE MAPPING USED TO PROCESS FILE
R FILE BEING ACCESSED IN 'RAW' MODE
** PHYSICAL FILES LARGER THAN 2GB FILES ARE NOT SUPPORTED ON THIS
PLATFORM

In this summary:

• ASSIGNED PHYSICAL FILENAME is the physical FILENAME with any symbols translated.

• RECL is the record length in bytes.

• MODE is the file access mode.

• R/W is read-write mode.

4-60 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Memory Allocation Map

• R/O is read-only mode.

• The FLAGS column contains various letters depending on the capabilities of the platform and user
requests. The text below the table indicates possible flag values for your specific platform.

• BUFFSIZE lists the buffer size of each DBset. This buffer size may be defined by the INIT command for
each DBset (File Management Statement).

• CLUSTER SIZE lists the unit of allocation for GINO and executive blocks. This unit may be defined on
the INIT command.

• The TIME STAMP column lists the time the file was created. The format is YYMMDDHHMMSS.

In this example, an INIT statement was used to create the DBALL DBSet with two files using the logical
names DBALL and DBALL2.

Below the summary is a message that indicates whether large files are available on the platform on
which you're running.

Memory Allocation Map

The next section of the file is a map of memory allocation.

** MASTER DIRECTORIES ARE LOADED IN MEMORY.

USER OPENCORE (HICORE) = 3804612 WORDS
EXECUTIVE SYSTEM WORK AREA = 78605 WORDS
MASTER(RAM) = 30000 WORDS
SCRATCH(MEM) AREA = 204900 WORDS ( 100 BUFFERS)
BUFFER POOL AREA (GINO/EXEC) = 76183 WORDS ( 37 BUFFERS)
TOTAL Simcenter Nastran MEMORY LIMIT = 4194300 WORDS

In this memory map:

• USER OPENCORE is the amount of memory available to the module for computation purposes.

• EXECUTIVE SYSTEM WORK AREA is the space reserved for the executive system.

• MASTER(RAM) is the space reserved to cache datablocks from the MASTER DBSet.

• SCRATCH(MEM) AREA is the space reserved to cache datablocks from the SCRATCH and SCR300
DBSets.

• BUFFER POOL AREA is the space reserved for the buffer pool.

• TOTAL Simcenter Nastran MEMORY LIMIT is the total space allocated to Simcenter Nastran's open core
using the “memory” keyword.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 4-61

© 2024 Siemens
 4. Simcenter Nastran Output Files

See also

• “Managing Memory” in the Simcenter Nastran Installation and Operations Guide

Day Log

The Day Log portion of the .f04 is a DMAP execution summary. This log, in table format, contains the
vast majority of the information in the .f04. The beginning an example Day Log is shown below:

DAY TIME ELAPSED I/O MB DEL_MB CPU SEC DEL_CPU SUB_DMAP/

DMAP_MODULE MESSAGES
10:32:16 0:16 13.6 .3 .8 .0 SESTATIC 20
IFPL BEGN
10:32:16 0:16 13.7 .1 .8 .0 IFPL 29
IFP1 BEGN
10:32:16 0:16 13.7 .0 .8 .0 IFPL 39
XSORT BEGN

In the Day Log:

• DAY TIME is the time of day of the entry.

• ELAPSED is the elapsed time since the start of the job.

• I/O MB is the megabytes of I/O to the databases since the start of the job

• DEL_MB is the delta I/O since the previous entry.

• CPU SEC is the total CPU seconds since the start of the job.

• DEL_CPU is the delta CPU since the previous entry.

• SUB_DMAP/DMAP_MODULE” indicates the DMAP statement being executed.

• MESSAGES are any messages issued by the module.

• BEGN and END indicate the start and end of the module or subDMAP.

The number that precedes the module name on lines that terminate with BEGN or END is the DMAP
statement number. This number corresponds to the statement number in the DMAP listing for the
solution sequence being used. The DMAP sequence listing will be printed as a part of the output if you
include either DIAG 14 or COMPILER LIST in the Executive Control section of your input file.

4-62 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Understanding the I/O MB Column

Understanding the I/O MB Column

• The software computes the “I/O MB” value by multiplying SYSTEM(85), which is incremented by one
for each GINO I/O, by BUFFSIZE. This value loses accuracy if the DBSets don't have the same BUFFSIZE.

• If SYSTEM(84) is set to 0, the “I/O MB” column will be the number of GINO I/Os.

• The software scales the “I/O MB” column by gigabytes. It appends a “G” after each number if the value
is greater than or equal to 100 000.

User Information Messages 4157 and 6439

The UIM 4157 text provides decomposition estimates when the preface of the decomposition module
completes. The format of this message varies slightly depending on whether you selected either sparse
or parallel methods.

This counterpart to UIM 4157 is UIM 6439, which provides actual information from the completed
decomposition process. These two messages are interspersed within the Day Log at each decomposition.

The following is an example from a sparse decomposition:

*** USER INFORMATION MESSAGE 4157 (DFMSA) ---PARAMETERS FOR SPARSE DECOMPOSITION OF DATA
BLOCK KLL (TYPE=RDP) FOLLOW
MATRIX SIZE = 64 ROWS NUMBER OF NONZEROES = 260
TERMS
NUMBER OF ZERO COLUMNS = 0 NUMBER OF ZERO DIAGONAL TERMS = 0
CPU TIME ESTIMATE = 0 SEC I/O TIME ESTIMATE = 0 SEC
MINIMUM MEMORY REQUIREMENT = 20 K WORDS MEMORY AVAILABLE = 3804 K
WORDS
MEMORY REQR'D TO AVOID SPILL = 30 K WORDS
EST. INTEGER WORDS IN FACTOR = 1 K WORDS EST. NONZERO TERMS = 1 K
TERMS
ESTIMATED MAXIMUM FRONT SIZE = 11 TERMS RANK OF UPDATE = 16
*** USER INFORMATION MESSAGE 6439 (DFMSA) ---ACTUAL MEMORY AND DISK SPACE REQUIREMENTS FOR
SPARSE SYM. DECOMPOSITION
SPARSE DECOMP MEMORY USED = 30 K WORDS MAXIMUM FRONT SIZE = 11
TERMS
INTEGER WORDS IN FACTOR = 1 K WORDS NONZERO TERMS IN FACTOR = 1 K
TERMS
SPARSE DECOMP SUGGESTED MEMORY = 30 K WORDS

In UIM 4157:

• MINIMUM MEMORY REQUIREMENT is an estimate of the open core memory that will allow the
decomposition to run, but with heavy spilling to disk.

• MEMORY REQR'D TO AVOID SPILL is the estimate of memory necessary for the decomposition to run in
“in core”, without spilling to disk.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 4-63

© 2024 Siemens
 4. Simcenter Nastran Output Files

Note: These two values represent memory requirement extremes. The amount of memory necessary
for optimal CPU performance is somewhere between the two.

• ESTIMATED MAXIMUM FRONT SIZE is a function of the model and affects the memory estimates
described above.

• MINIMUM MEMORY REQUIREMENT is a function of the front size.

• MEMORY REQR'D TO AVOID SPILL is a function of the square of the front size.

• NUMBER OF NONZEROES is the size of the input matrix, multiply this value by 8 to estimate the size of
the input file in bytes.

• The sum of EST. INTEGER WORDS IN FACTOR and EST. NONZERO TERMS is the size of the output
matrix.

Note: Multiply the integer value by 4 and the nonzero value by 8 to estimate the size of the output file
in bytes.

• RANK OF UPDATE is the number of rows that will be simultaneously updated during the
decomposition. You can set by either the “rank” keyword or SYSTEM(205).

Note:
Setting SYSTEM(69)=64 causes Simcenter Nastran to terminate after printing UIM 4157. This
can be useful for determining a job's memory and disk space requirements.

In UIM 6439:

• SPARSE DECOMP MEMORY USED states the actual memory used in the decomposition process.

• SPARSE DECOMP SUGGESTED MEMORY is based on the execution of the module. Using this value will
result in optimal throughput performance.

Memory and Disk Usage Statistics

Simcenter Nastran writes the memory and disk usage statistic tables after the job completes. These
tables indicate the maximum memory used by any sparse numerical module and the maximum disk
used by any module during the job. For example:

+---------- SPARSE SOLUTION MODULES -----------+ +------------- MAXIMUM DISK USAGE

-------------+
HIWATER SUB_DMAP DMAP HIWATER
SUB_DMAP DMAP
(WORDS) DAY_TIME NAME MODULE (MB) DAY_TIME
NAME MODULE
517786 04:35:44 SEKRRS 18 DCMP 15.625 04:35:48 SESTATIC
186 EXIT

4-64 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Database Usage Statistics

In the sparse solution modules table:

• HIWATER WORDS is the maximum amount of open core used by certain sparse numerical modules.

• DAY_TIME is the time of day the module ran.

• SUB_DMAP NAME is the name of the SUBDmap.

• DMAP MODULE indicates the line number and module name that made the maximum request.

In the maximum disk usage table:

• HIWATER (MB) is the maximum amount of disk space used by any module.

• DAY_TIME is the time of day the module ran.

• SUB_DMAP NAME is the name of the SUBDmap.

• DMAP MODULE indicates the line number and module name that made the maximum request.

Database Usage Statistics

These statistics, provided in table format, summarize the I/O activity for the DBSets.

This statistical table contains two parallel tables:

• The LOGICAL DBSETS table lists each DBset

• The DBSET FILES tables lists the component files of the DBSet.

In these tables:

• DBSET is the name of the DBSet.

• ALLOCATED is the Simcenter Nastran DBSet size limit in blocks.

• BLOCKSIZE is BUFFSIZE of the DBSet minus one.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 4-65

© 2024 Siemens
 4. Simcenter Nastran Output Files

• USED (BLOCKS) and USED % are the number of GINO and executive blocks used at the end of the job
as well as the percent of the DBSet actually used.

• FILE is the file's logical name associated with the DBSet to the left.

• ALLOCATED is the number of blocks allocated by Simcenter Nastran to the file.

• HIWATER (BLOCKS) and HIWATER (MB) are the maximum number of GINO and executive blocks and
actually used during the execution.

• I/O TRANSFERRED is the amount of I/O to the file. The last line of this column lists the total I/O
transferred.

In this example, the MASTER and OBJSCR DBSets are each composed of one file. The DBALL DBSet is
composed of two files, DBALL and DBALL2; and the SCRATCH DBSet has three components, MEMFILE,
SCRATCH, and SCR300.

You can use this database statistics table to determine whether the DBSets and files are appropriately
sized and to assess the amount of I/O activity associated with each file. You can obtain the best elapsed
time performance can be obtained if the files with the greatest activity are on different physical devices
(and better yet, separate I/O controllers or busses).

Summary of Physical File I/O Activity

This section of the .f04 file describes the physical file I/O for each database file.

*** SUMMARY OF PHYSICAL FILE I/O ACTIVITY ***

ASSIGNED PHYSICAL FILENAME RECL (BYTES) READ/WRITE COUNT MAP WSIZE (NUM)
MAP COUNT
------------------------------------------------------- ----------- ---------------
-----------
/tmp/65872_57.SCRATCH 8192 1 128KB
( 4) 1
/tmp/65872_57.OBJSCR 8192 378 128KB
( 4) 24
/tmp/65872_57.MASTER 8192 1247 128KB
( 4) 11
/tmp/65872_57.DBALL 8192 26 128KB
( 4) 1
/tmp/65872_57.SCR300 8192 1 128KB
( 4) 1
/scnas/aix/SSS.MASTERA 8192 162
N/A N/A
/scnas/aix/SSS.MSCOBJ 8192 202
N/A N/A

In this summary:

• ASSIGNED PHYSICAL FILENAME, RECL, and MAP WSIZE and NUM are repeated from the “Summary of
Physical FIle Information” table.

4-66 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Other Optional Diagnostic Information

• READ/WRITE COUNT is the number of GINO reads and writes that were performed on the file.

• MAP COUNT is the number of times the map window had to be remapped (these columns are only
present on systems supporting mapped I/O).

You can use this summary to tune I/O performance. For mapped I/O systems, if the map count
approaches the number of reads and writes, you should increase the map size and/or the number of
maps.

• Increasing the number of maps is suggested if a module simultaneously accesses more data blocks or
matrices in a file than there are windows.

• Increasing the size of the windows is suggested if a file contains very large data blocks or matrices.

You can obtain the best elapsed time performance, with or without mapping, if the files with the
greatest activity are on different physical devices, and better yet, on separate I/O controllers or busses.

Other Optional Diagnostic Information

The DIAG Executive Control command has a number of different options that you can use to have
Simcenter Nastran print additional, diagnostic information to the .f04 file. Several of these options are
described in detail below.

See also

• “DIAG” in the Simcenter Nastran Quick Reference Guide

DIAG 8

If you include DIAG 8 in the Executive Control section of your input file, the software prints matrix
trailers as the matrices are generated. Terms in the DIAG 8 output include:

• The number of GINO blocks in the matrix (BLOCKS);

• The average number of terms per string (STRL)

• The number of strings in a matrix data block (NBRSTR)

• The number of bank terms in a matrix data block on the average (BNDL)

• The number of banks in a matrix data block (NBRBND)

• The average first nonzero row position (ROWL)

• The average of full bank width (BNDAVG)

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 4-67

© 2024 Siemens
 4. Simcenter Nastran Output Files

• The maximum full bank width of all columns of data blocks (BNDMAX)

• The number of null columns in a data block (NULCOL)

DIAG 13

If you include DIAG 13 in your input file, the software includes the message REAL CORE IS xxxx
throughout the day log. It shows the value of the NASTRAN keyword REAL. The message only appears
after module executions which use the value. The open core (i.e., working space) for other modules is
set via HICORE

DIAG 19

If you include DIAG 19 in your input file, the software prints data for MPYAD and FBS method selection in
the Execution Summary table at the top of the .f04 file.

DIAG 49

If you include DIAG 49, the software includes a detailed analysis of timing data (statistical information
on module expected/actual CPU item) at the end of the .f04 file in three separate sections.

• The first section contains information about each module and submodule.

• The second section reports statistical information about the submodules (FBS, MPYAD, and DCMP)
grouped by specific methods. Instances where the CPU estimate doesn't pass the following criteria are
output:

• The third section contains information about the modules listed in alphabetical order.

Comparing Different F04 Files with F04REPRT

The Simcenter Nastran F04RERPT utility program contains a Perl script that allows you to summarize
and/or compare the contents of different .f04 files. This utility can be helpful if you want to compare the
relative performance of one or more jobs under various configurations.

See also

• “F04REPRT” in the Simcenter Nastran Installation and Operations Guide

4-68 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Understanding the .log File

Understanding the .log File

The .log file contains system information, such as the name of the computer on which the analysis was
run. It also contains any system errors encountered during the analysis.

Understanding the .plt and .pch Files

The .plt file is a binary file that contains the plotting information generated by the Simcenter Nastran
plotter. If the Simcenter Nastran plotter is not used, run1.plt is deleted following the completion of the
run.

The .pch file (“punch file”) is an ASCII file that contains “punched” output. It is written to FORTRAN unit 7.
In the Case Control section, you can specify whether you want the software to output specified results to
the punch file.

Output Listing Examples

Examples of most of the printed output formats that are used in linear static analysis are contained in
this section. Each example is annotated with comments and identified by title. These examples include:

Forces in Scalar Springs Forces in CQUAD8 Elements

Stresses in Scalar Springs Stresses in CQUAD8 Elements
Forces in CROD Elements Strains in CQUAD8 Elements
Stresses in CROD Elements Sresses in CHEXA Elements
Forces in CBAR Elements Strains in CHEXA Elements
Stresses in CBAR Elements Forces in CWELD Elements
Force Distribution in CBAR Elements (CBARAO) Grid Point Force Balance
Stress Distribution in CBAR Elements (CBARAO) Grid Point Stress Output
Forces in CBEAM Elements Grid Point Stress Discontinuities Output
Stresses in CBEAM Elements Element Stress Discontinuities Output
Forces in CBEND Elements Element Strain Energy
Stresses in CBEND Elements Sequence Processor Output
Forces in CBUSH Element Grid Point Singularity Table
Forces Acting on Shear Panel Elements User Information Message 5293
Stresses in Shear Panels OLOAD Resultant

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 4-69

© 2024 Siemens
 4. Simcenter Nastran Output Files

Forces in CQUAD4 Elements (Without Corner SPCFORCE Resultant

Option)
Stresses in CQUAD4 Elements (Without Corner Maximum SPCFORCES, DISPLACEMENT, and
Option) Applied Loads
Strains in CQUAD4 Elements (Without Corner Displacement Vectors
Option)
Forces in CQUAD4 Elements (With Corner Option) Load Vectors
Stresses in CQUAD4 Elements (With Corner Database Summary Table
Option)
Strains in CQUAD4 Elements (With Corner Option) Restartable Versions

Forces in Scalar Springs (C E L A S 2 )

• Standard FORCE output, requested in Case Control Section.

• This output is typical of all scalar elements (CELAS1, CELAS2, CELAS3, CELAS4).

Stresses in Scalar Springs (C E L A S 2)

• Standard STRESS output, requested in Case Control Section

• Stress output is obtained by multiplying the force in the spring by the stress coefficient entered in field
9 of the CELAS2 entry or field 9 of the PELAS entry.

• This output applies only to the CELAS1 and CELAS2 Bulk Data entries.

Forces in CROD Elements

4-70 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Output Listing Examples

• See “One-dimensional Elements” for a definition of positive torque.

• Positive axial force is tension.

Stresses in CROD Elements

Axial stress and margin-of-safety due to axial stress only

Torsional stress and margin-of-safety due to torsion stress only

• Positive stress is tension.

• Standard STRESS output, requested in Case Control Section.

• Definition of Margin-of-Safety (M.S.): The M.S. for axial stress is: M.S. = (ST/SA) – 1.0 if SA is greater
than zero or M.S. = (–SC/SA) – 1.0 if SA is less than zero, where ST and SC are the tension and
compression allowables given on the MAT1 entry and SA is the axial stress given in the output. M.S. is
computed and printed if and only if ST (or SC) is greater than zero and SA is not equal to zero.

The M.S. for torsional stress is: M.S. = (SS/TAU) – 1.0, where SS is the shear allowable given on the
MAT1 entry and TAU is the torsional stress given in the output. M.S. is computed and printed if and
only if SS is greater than zero and TAU is not equal to zero.

Forces in CBAR Elements

• Standard FORCE output, requested in the Case Control Section.

• See CBAR in the “Simcenter Nastran Element Library Reference” for definition of planes 1 and 2 and for
positive directions of bending moments, shears, axial force, and torque.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 4-71

© 2024 Siemens
 4. Simcenter Nastran Output Files

Stresses in CBAR Elements

Normal stresses due to bending only at ends A and B, points C, D, E, and F

Example: SA1 = stress at end A, point C. SB4 = stress at end B, point F.

Normal stresses due to axial loads only.

Normal stresses due to combined bending and axial loads. Maximum and minimum values
over points C, D, E, and F at ends A and B.

Margin-of-safety in tension (T) and compression (C). Minimum M.S. found over entire length
of element.

• Positive stress is tension.

• Standard STRESS output, requested in Case Control Section.

• Definition of Margin-of-Safety (M.S.):

The M.S. in tension is: M.S.–T = (ST / S-MAX) – 1.0, where ST is the tension allowable given on the
MAT1 entry and S-MAX is the maximum stress given in the output, S-MAX = max (SA-MAX, SB-MAX).
M.S. in tension is computed and printed if and only if ST is greater than zero and S-MAX is greater
than zero.

The M.S. in compression is: M.S.-C = (–SC / S-MIN) - 1.0, where SC is compression allowable given on
the MAT1 card and S-MIN is the minimum stress given in the output, S-MIN = min (SA-MIN, SB-MIN).
M.S. in compression is computed and printed if and only if SC is greater than zero and S-MIN is less
than zero.

4-72 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Output Listing Examples

Force Distribution in CBAR Elements (CBARAO)

Element ID

Distance from end A as a fraction of the element's length

• Optional output for the CBAR element using the CBARAO entry.

• See CBAR in the “Simcenter Nastran Element Library Reference” for definition of planes 1 and 2.

• Force output must be requested in Case Control Section.

Stress Distribution in CBAR Elements (CBARAO)

Distance from end A as a fraction of the element's length.

Normal stresses due to bending point at points C, D, E, and F.

Normal stresses due to axial loads only.

Normal stresses due to combined bending and axial loads. Maximum and minimum values
over points C, D, E, and F .

Margin-of-safety in tension (T) and compression (C). Minimum M.S. found over entire length
of element.

• The default position for points C, D, E and F is on the neutral axis.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 4-73

© 2024 Siemens
 4. Simcenter Nastran Output Files

• This is special output requested by including CBARA0 and/or PLOAD1 entries in the Bulk Data Section
along with a STRESS request in the Case Control Section.

• Definition of Margin-of-Safety (M.S.):

The M.S. in tension is: M.S.-T = (ST / S-MAX) - 1.0, where ST is the tension allowable given on the
MAT1 entry and S-MAX is the maximum stress given in the output. A M.S. in tension is computed and
printed if and only if ST is greater than zero and S-MAX is greater than zero.

The M.S. in compression is: M.S.-C = (-SC / S-MIN) - 1.0, where SC is compression allowable given
on the MAT1 entry and S-MIN is the minimum stress given in the output. An M.S. in compression is
computed and printed if and only if SC is greater than zero and S-MIN is less than zero.

Forces in CBEAM Elements

Element ID

Grid points at ends A and B

Distance from end A as a fraction of the element's length

Part of total torque due to cross-sectional warping

• See CBAR in the “Simcenter Nastran Element Library Reference” for definition of planes 1 and 2 and for
the positive directions of bending moments, shears, axial force, and torque.

• Standard FORCE output, requested in Case Control Section.

4-74 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Output Listing Examples

Stresses in CBEAM Elements

Element ID

Distance from end A as a fraction of the element's length

Normal stresses due to combined bending and axial loads at points C, D, E, and F

Maximum and minimum value over points C, D, E, and F

Minimum margin-of-safety in tension (T) and compression (C). Minimum M.S. found over
entire length of element

• Standard STRESS output, requested in Case Control Section.

• Positive stress is tension.

• Definition of Margin-of-Safety (M.S.):

The M.S. in tension is: M.S.–T = (ST / S-MAX) – 1.0, where ST is the tension allowable given on the
MAT1 entry and S-MAX is the maximum stress given in the output, S-MAX = max (SA-MAX, SB-MAX).
M.S. in tension is computed and printed if and only if ST is greater than zero and S-MAX is greater
than zero.

The M.S. in compression is: M.S.-C = (–SC / S-MIN) - 1.0, where SC is compression allowable given on
the MAT1 card and S-MIN is the minimum stress given in the output, S-MIN = min (SA-MIN, SB-MIN).
M.S. in compression is computed and printed if and only if SC is greater than zero and S-MIN is less
than zero.

Forces in CBEND Elements

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 4-75

© 2024 Siemens
 4. Simcenter Nastran Output Files

Element ID

Grid points at ends A and B

• See CBAR in the “Simcenter Nastran Element Library Reference” for definition of planes 1 and 2 and for
the positive directions of bending moments, shears, axial force, and torque.

• Standard FORCE output, requested in Case Control Section.

Stresses in CBEND Elements

Normal stresses due to combined bending and axial loads at points C, D, E, and F

Maximum and minimum value over points C, D, E, and F

Minimum margin-of-safety in tension (T) and compression (C). Minimum M.S. found over
entire length of element.

• Standard STRESS output, requested in Case Control Section.

• Positive stress is tension.

• Definition of Margin-of-Safety (M.S.):

The M.S. in tension is: M.S.–T = (ST / S-MAX) – 1.0, where ST is the tension allowable given on the
MAT1 entry and S-MAX is the maximum stress given in the output, S-MAX = max (SA-MAX, SB-MAX).
M.S. in tension is computed and printed if and only if ST is greater than zero and S-MAX is greater
than zero.

The M.S. in compression is: M.S.-C = (–SC / S-MIN) - 1.0, where SC is compression allowable given on
the MAT1 card and S-MIN is the minimum stress given in the output, S-MIN = min (SA-MIN, SB-MIN).
M.S. in compression is computed and printed if and only if SC is greater than zero and S-MIN is less
than zero.

4-76 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Output Listing Examples

Forces in CBUSH Elements

• The forces are in the CBUSH element coordinate system.

Forces Acting on Shear Panel Elements (CSHEAR)

• See 2D Elements in the “Simcenter Nastran Element Library Reference” for definitions of element
coordinate system and force components.

• Standard FORCE output, requested in the Case Control Section.

Stresses in Shear Panels (CSHEAR)

Maximum and average shear stresses and margin-of-safety due to shear only

• Standard STRESS output, requested in Case Control Section.

• Definition of Margin-of-Safety (M.S.): M.S. = (SS/MAX SHEAR) – 1.0, where SS is the shear allowable
given on the MAT1 entry and MAX SHEAR is the maximum shear stress given in the output. M.S. is
computed and printed if and only if SS is greater than zero and MAX SHEAR is not equal to zero.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 4-77

© 2024 Siemens
 4. Simcenter Nastran Output Files

Forces in CQUAD4 Elements without Corner Option

In-plane element forces in element coordinate system (force/length).

Element internal moments in element coordinate system (moment/length).

Traverse shear forces (force/length).

• Positive force is tension.

• See 2D Elements in the “Simcenter Nastran Element Library Reference” for definitions of element
coordinate system and positive shears and moments.

• This output is typical for all plate elements

• Standard FORCE output, requested in Case Control Section.

Stresses in CQUAD4 Elements without Corner Option

Fiber distance Z1 and Z2 specified on the PSHELL property entry

Stresses in element coordinate system

Angle in degrees from the x-axis of the element coordinate system to the direction of the
major (in-plane) principal stress

4-78 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Output Listing Examples

Note:
The positive sense of this angle is about the +z-axis of the element coordinate
system.

Major and minor (in-plane) principal stresses

Note:
For plane stress formulations, the third principal stress is always zero and the
corresponding principal direction is in the out-of-plane direction.

Maximum (in-plane) shear stress or Hencky-von Mises stress

Note:
If the major principal stress is tensile and the minor principal stress is compressive,
the maximum in-plane shear stress is the maximum shear stress.

• Standard STRESS output, requested in Case Control Section.

• This output is typical for all plate elements

• See 2D Elements in the “Simcenter Nastran Element Library Reference” for definitions of element
coordinate system.

Strains in CQUAD4 Elements without Corner Option

If fiber strains are requested (see Note 2) the fiber distances Z1 and Z2 will appear here. If
strain-curvature data is requested, this entry will appear as shown

Strains and curvatures in element coordinate system

Angle in degrees from the x-axis of the element coordinate system to the direction of the
major (in-plane) principal strain

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 4-79

© 2024 Siemens
 4. Simcenter Nastran Output Files

Major and minor (in-plane) principal strains

Maximum (in-plane) shear strain or Hencky-von Mises strain

• Strain output is available for plate elements and is requested using the Case Control command
STRAIN.

• Fiber strains at fiber distances Z1 and Z2 may be obtained by requesting for STRAIN (FIBER) in the Case
Control Section. Z1 and Z2 are specified on the property entry.

• This output is typical for all plate elements.

• See 2D Elements in the “Simcenter Nastran Element Library Reference” for definitions of element
coordinate system.

• Material coordinate systems are specified on MATi entries.

Forces in CQUAD4 Elements with Corner Option

Denotes CEN/4 for the center stress or the grid point ID for a corner stress

• Positive force is tension.

• See 2D Elements in the “Simcenter Nastran Element Library Reference” for definitions of element
coordinate system and definition of positive shears and moments.

• This output is typical for all shell elements.

• Standard FORCE output, requested in Case Control Section.

• Force output obtained when using the FORCE(CORNER) in Case Control Section.

4-80 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Output Listing Examples

• The transverse shear forces (QX and QY) are computed at the center only, and those same values are
displayed at the corners.

Stresses in CQUAD4 Elements with Corner Option

Denotes CEN/4 for the center stress or the grid point ID for a corner stress

• This is standard STRESS output requested in Case Control Section.

• Stress output obtained using the STRESS(CORNER) command in the Case Control Section.

• See 2D Elements in the “Simcenter Nastran Element Library Reference” for a definition of element
coordinate system.

Strains in CQUAD4 Elements with Corner Option

Denotes CEN/4 for the center strain and curvatures and a grid point ID for a corner strain
or curvature

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 4-81

© 2024 Siemens
 4. Simcenter Nastran Output Files

• Corner strain output is available for the plate elements and is requested using the Case Control
command STRAIN(CORNER).

• Fiber strains at fiber distances Z1 and Z2 may be obtained by requesting for STRAIN (FIBER) in the Case
Control Section. Z1 and Z2 are specified on the property entry.

• This output is typical for all plate elements.

• See 2D Elements in the “Simcenter Nastran Element Library Reference” for a definition of element
coordinate system.

• Material coordinate systems are specified on MATi entries.

Forces in CQUAD8 Elements

Denotes CEN/4 for the center forces or the grid point ID for a corner force

• Positive force is tension.

• See 2D Elements in the “Simcenter Nastran Element Library Reference” for a definition of element
coordinate system and a definition of positive shears and moments.

• This output is typical for all shell elements.

• Standard FORCE output, requested in Case Control Section.

4-82 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Output Listing Examples

Stresses in CQUAD8 Elements

Denotes CEN/4 for the center stress or the grid point ID for a corner stress

• This is standard STRESS output requested in Case Control Section.

• This output is typical for all plate elements.

• See 2D Elements in the “Simcenter Nastran Element Library Reference” for a definition of element
coordinate system.

Strains in CQUAD8 Elements

Denotes CEN/4 for the center strain and curvatures and a grid point ID for a corner strain
and curvature

• Strain output is available for plate elements and is requested using the Case Control command
STRAIN.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 4-83

© 2024 Siemens
 4. Simcenter Nastran Output Files

• Fiber strains at fiber distances Z1 and Z2 may be obtained by requesting for STRAIN (FIBER) in the Case
Control Section. Z1 and Z2 are specified on the property entry.

• This output is typical for all plate elements.

• See 2D Elements in the “Simcenter Nastran Element Library Reference” for a definition of element
coordinate system.

• Material coordinate systems are specified on MATi entries.

Stresses in CHEXA Elements

Stress coordinate system definition

0 = Basic Coordinate System
–1 = Element Coordinate System
+X = Material Coordinate System
Defined on PSOLID entry.

Location of Stresses
GRID = Stresses at center and vertex points
GAUSS = Stresses at center and Gauss points

Number of active grid points for the element

Center and corner point stresses in the stress coordinate system

Directional cosines of the principal stress with respect to the stress coordinate system

Hencky-von Mises stress or octahedral shear stress

4-84 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Output Listing Examples

• This output is typical for the CHEXA, CPENTA, CPYRAM, and CTETRA elements.

• Standard STRESS output, requested in Case Control Section.

• The convention for the principal stresses are such that ""(A ≥C ≥B). In the case where the principal
stresses are equal, the directional cosines are not unique, and the values of zero are output for the
directional cosines.

Strains in CHEXA Elements

• This output is typical for the CHEXA, CPENTA, CPYRAM and CTETRA elements.

• Standard STRAIN output, requested in Case Control Section.

Forces in CWELD Elements

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 4-85

© 2024 Siemens
 4. Simcenter Nastran Output Files

• Welded shell model, point-to-point option

Grid Point Force Balance

Forces and moments acting on the grid point from each source in the global coordinate
system

• Only the forces due to elements, SPCs and applied loads are considered. The totals will be nonzero
due to round-off error, or to the fact that forces due to MPCs, rigid elements, GENELs, DMIG, or inertia
loads (in dynamic analysis) are not included.

• Standard GPFORCE output, requested in Case Control Section.

Grid Point Stress Output

Grid point ID where stress averaging is performed

Surface or volume ID referenced on the Case Control SURFACE or VOLUME command

4-86 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Output Listing Examples

Referenced coordinate system ID

Fiber location as specified on the PSHELL entry

Average component stresses at grid points in reference coordinate system

Angle of principal stress in referenced coordinate system

Major and minor principal stresses

Maximum shear and von Mises stresses

• Stress surface output is available for all plate and shell elements (CQUAD4, CQUAD8, CTRIA3, CTRIA6,
CQUADR, and CTRIAR).

• Stress volume output is available for all solid elements (CHEXA, CPENTA, CPYRAM and CTETRA).

Grid Point Stress Discontinuities Output

Grid point ID where stress discontinuities are calculated

Surface or volume ID referenced on the Case Control SURFACE or VOLUME command

Referenced coordinate system ID

Fiber location as specified on the PSHELL entry

Probable error of component stresses δg

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 4-87

© 2024 Siemens
 4. Simcenter Nastran Output Files

Probable error of major principal, minor principal, maximum shear and von Mises stresses

Stress error measured at each grid point

• This output is obtained by using the STRFIELD, STRESS, GPSDCON, and SURFACE (or VOLUME) Case
Control command.

• Grid point stress discontinuity surface output is available for all plate and shell elements (CQUAD4,
CQUAD8, CTRIA3, CTRIA6, CQUADR, and CTRIAR).

• Grid point stress discontinuity volume output is available for all solid elements (CHEXA, CPENTA,
CPYRAM and CTETRA).

Element Stress Discontinuities Output

Element ID where element stress discontinuities are calculated

Surface or volume ID referenced on the Case Control SURFACE or VOLUME command

Referenced coordinate system ID

Fiber location as specified on the PSHELL entry

Probable error of component stresses δe

Probable error of major principal, minor principal, maximum shear and von Mises stresses

Stress error measured at each element

4-88 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Output Listing Examples

• This output is obtained by using the STRFIELD, STRESS, ELSDCON, and SURFACE (or VOLUME) Case
Control commands.

• Element stress discontinuity surface output is available for all plate and shell elements (CQUAD4,
CQUAD8, CTRIA3, CTRIA6, CQUADR, and CTRIAR).

• Element stress discontinuity volume output is available for all solid elements (CHEXA, CPENTA,
CPYRAM and CTETRA).

Element Strain Energy

Total strain energy for all elements of this type in the requested set

Strain energy in each element

Total strain energy in requested SET (SET –1 denotes ALL)

• This output is incorrect in the presence of thermal loads or element deformations.

• Standard output, requested with Case Control command ESE.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 4-89

© 2024 Siemens
 4. Simcenter Nastran Output Files

Sequence Processor Output

Connection data

Average, RMS, and maximum number of active columns

Resequencer method selected

• There are 20 points divided into one group.

• The average and RMS values depend on the manner in which the grid points are sequenced.

• This output is available only if the sequencer is turned on (param,oldseq,x).

• Further information can be found in the Simcenter Nastran Numerical Methods User’s Guide.

4-90 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Output Listing Examples

Grid Point Singularity Table

• This is automatically output from the Grid Point Singularity Processor (GPSP1) and is obtained anytime
a stiffness ratio is less than the parameter EPZERO (default = 10-8).

User Information Messages 5293 for Data Block KLL

Residual vector error

Where:
= Load on the l-set degrees of freedom

= Displacement of the l-set degrees of freedom

= Stiffness matrix of the l-set

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 4-91

© 2024 Siemens
 4. Simcenter Nastran Output Files

Work performed by the applied loads =

• Small epsilons are due to machine roundoff are acceptable. Large epsilons indicate a potential
modeling error.

• One line of output is printed for each static loading condition.

OLOAD Resultant

Subcase number

Resultant of the applied loads about the point specified on the “PARAM,GRDPNT,x”. The
default is at the origin of the basic coordinate system.

• One line is printed for each subcase.

• This output is printed by default. It can be deactivated by adding PARAM,PRTRESLT,NO in the bulk
data.

• The OLOAD output table sums all the applied loads to a single location. Fx, Fy, Fz, Mx, My, Mz are
relative to the applied loads in the basic coordinate system directions. T1, T2, T3, R1, R2, R3 are
the basic coordinate directions of the summed loads in the table. The origin of the summed loads
is specified with the PARAM,GRDPNT,x entry which is the origin of the basic coordinate system by
default.

4-92 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Output Listing Examples

SPCFORCE Resultant

Subcase number
Resultant of the applied loads about the point specified on the “PARAM,GRDPNT,x”. The
default is at the origin of the basic coordinate system

• One line is printed for each subcase.

• This output can be obtained by adding PARAM,PRTRESLT,YES in the bulk data file.

Maximum SPCFORCES, DISPLACEMENT, and Applied Loads

Subcase number

1. Quantities are transformed to the basic coordinate system.

2. Absolute value of transformed quantities are taken
3. Largest value for each component is found and printed

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 4-93

© 2024 Siemens
 4. Simcenter Nastran Output Files

• The largest magnitude of quantities transformed into the basic coordinate system is printed. The
maximums shown for T1 may not be at the same grid point as T2, etc.

• One line is printed for each subcase.

• This output can be obtained by adding PARAM,PRTMAXIM,YES in the bulk data file.

Displacement Vectors

G for grid point, S for scalar point

Displacements expressed in the global coordinate system

• Standard DISPLACEMENT output for a static analysis, requested in Case Control Section.

Load Vectors

Applied loads and moments expressed in the global coordinate system

• Standard OLOAD output, requested in Case Control Section.

4-94 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Output Listing Examples

Database Summary Table

Maximum disk space allocated in GINO blocks

Maximum disk space used in GINO blocks

Disk spaced used in megabytes

Amount of I/O transferred between each DBset and memory

Maximum disk space used by SCR300 in GINO blocks

• Standard OLOAD output, requested in Case Control Section.

Restartable Versions

There were seven versions created in the database. Only Versions 3 and 7 are, however,
restartable. The other versions denoted by a “**” are not restartable.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 4-95

© 2024 Siemens
 4. Simcenter Nastran Output Files

Understanding the .DBALL and .MASTER Files

The .DBALL and .MASTER files are database files (DBsets) that Simcenter Nastran generates automatically
during a solve. The files are generated in the scratch directory. These files are necessary if you want to
later perform a restart analysis.

• The .MASTER file is a directory of the database. It contains the names of all DBsets, DBset members
and their physical file names, and a directory of projects, versions, data blocks, parameters, DMAP
source and object files. It also contains the Nastran Data Definition Language (NDDL) scheme used to
describe the database. The default maximum size for .MASTER is 5,000 blocks.

• The .DBALL file contains all the DMAP data blocks you can save for reuse in a subsequent run. In the
solution sequences that don't allow restarts, such as SOL 101, this file is empty. The default size is
for .DBALL is 250,000 blocks.

Specifying Whether to Save .DBALL and .MASTER Files

The “scratch” option on the nastran command line allows you to specify whether you want the software
to save the files .DBALL and .MASTER files. By default, Simcenter Nastran deletes the files at the end of
the run.

• If you specify “scratch=yes,” the software deletes the files.

• If you specify “scratch=no,” the software keeps the files.

• If you specify “scratch=mini,” the software retains a reduced size database that you can use to perform
data recovery restarts.

See also

• “scratch” in the Simcenter Nastran Quick Reference Guide

• “Mini-Database Restarts” in the Simcenter Nastran User's Guide

• “Understanding DBsets” in the Simcenter Nastran User's Guide

Manually Deleting .DBALL and .MASTER Files

In some cases, you may need to manually delete the .DBALL and .MASTER files. For example:

• If your solve doesn't complete successfully, Simcenter Nastran may not automatically delete
the .DBALL and .MASTER files, even though you specified “scratch=yes.” In those cases, you must
delete the files manually before resubmitting your job.

4-96 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Understanding the .xdb and .op2 Files

• If you specified “scratch=no,” but you decide that you want to submit a .dat file a second time as
a cold start (you don't use any restart commands), you must first delete the corresponding .DBALL
and .MASTER files before you resubmit the job.

Understanding the .xdb and .op2 Files

The .xdb and .op2 files are binary files that contain Simcenter Nastran results for use with certain types
of postprocessors.

• The .xdb file is known as the “results database” file. It contains results that can be postprocessed by
certain postprocessors.

• The .op2 file is known as an “OUTPUT2” file. It is written to FORTRAN unit 12. Also, if an .op2 file
already exists for the run, it too should be deleted prior to resubmitting the run. It contains results
that can be postprocessed by certain postprocessors, such as Simcenter, and MSC Patran.

The setting for the PARAM,POST option in the Bulk Data section of the input file controls whether
Simcenter Nastran creates an .xdb or op2 file. For example, to have the software create an .xdb file, you
would specify PARAM,POST,0. Additionally, the options you specify for PARAM,POST control whether the
file contains geometry.

Note: Before you perform a restart analysis, you should delete the .xdb and .op2 files (if any) from your
original run.

See also

• “POST” in the Simcenter Nastran Quick Reference Guide

Understanding the .hdf File

Simcenter Nastran can write results and geometric data to output files in the open-source, binary HDF5
format. The file extension for HDF5 format output files is .hdf. The HDF5 format allows you to more
easily access and use Simcenter Nastran results in your own post-processing utilities.

For example, an airframe business might have a proprietary failure criteria that needs to be evaluate
with in-house software. To use its utility, the analyst needs to access the stress results from the
Simcenter Nastran run. For such a case, the HDF5 format offers the following advantages:

• It uses a hierarchical data format that is specifically designed to handle large amounts of data.

• Application programming interfaces (API) are readily available to read HDF5 files.

Note:
API are also available to create equivalent ASCII files from HDF5 files.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 4-97

© 2024 Siemens
 4. Simcenter Nastran Output Files

The standard OP2 file, although widely recognized by many post-processing tools, is not ideally suited
for such applications for the following reasons:

• OP2 files use a serial file format, which may force the utility to read through the entire OP2 file to
recognize the structure of the OP2 file.

• Although an API that reads OP2 files written by Simcenter Nastran is supplied with your Simcenter
Nastran installation, this API may not be able to read OP2 files written by other solvers.

Note:
For information on the API supplied with your Simcenter Nastran installation, see "Building and
Using TABTST" in the Simcenter Nastran Installation and Operations Guide.

Not all of the data blocks that the software writes to the OP2 file are written to the HDF5 file. The data
blocks that are written to the HDF5 file are as follows:

Data block Description

OEF1 Table of element forces in SORT1 format
OES1 Table of element stresses or strains in SORT1 format
OSTR1 Table of total strain in SORT1 format
OSTR1TH Table of thermal strain in SORT1 format
OSTR1EL Table of elastic strain in SORT1 format
OEE Output element energy (strain, kinetic, loss)
OUGV1 Table of displacement, velocity and acceleration in global
coordinate system in SORT1 format
BOUGV1 Table of displacement, velocity and acceleration in basic coordinate
system in SORT1 format
OQG1 Table of single-point or multi-point constraint forces in SORT1
format
OGF Table of grid point forces
GEOM1 Table of bulk entry images related to geometry
GEOM2 Table of bulk entry images related to element connectivity and
scalar points
CSTM Table of coordinate system transformation matrices
EPT Table of bulk entry images related to element properties
MPT Table of bulk entry images related to material properties
ICASEOP2 Echo of case control input data
IBULKOP2 Echo of bulk data input data

4-98 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Understanding the .hdf File

The variable names in an HDF5 file match those in the data definition language (DDL) schema. To find
descriptions of the variables, see the DMAP Programmer's Guide.

To request that the software write the results to an HDF5 format output file in addition to the standard
OP2 file, specify the following:

• SYSTEM(653) = 1

• PARAM,POST,-2

Note:
For PARAM,POST,-2, OMACHPR=NO is the default. However, the software only writes data blocks
in the current format to the HDF5 format output file, not the pre-MSC Nastran Version 69
or pre-MSC Nastran 2001 formats. Normally, you need to specify OMACHPR=YES to have the
software write the data blocks in the current format when PARAM,POST,-2.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 4-99

© 2024 Siemens
 4. Simcenter Nastran Output Files

4-100 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 5. Defining Grid Points and Scalar Points
Overview of Grid Points and Scalar Points
Whenever you analyze a finite element model using Simcenter Nastran, you must generate a set of grid
points (nodes). Those grid points, together with the elements, define the size and shape of your model.
The physical behavior of your finite element model is represented by the elements connecting the grid
points as well as the associated constraints and their coordinate system.

When you create a finite element model of a structure, you're actually constructing a mathematical
model that represents your structure in matrix form. The unknowns in the matrix equation are the
displacements in the model. These displacements consist of:

• The six components for each of the grid points

• One component for each of the scalar points

In general, the displacements at the grid points and the scalar points are referred to as the model's
degrees of freedom.

This chapter describes grid points, grid point properties, and scalar points. Detailed information about
coordinate systems is provided in the next chapter.

For many practical modeling situations, you may have several different output coordinate systems in a
model.

See also

• “Understanding Coordinate Systems” in the Simcenter Nastran User's Guide

Defining Grid Points

Each grid point has six degrees of freedom (DOF): three translations and three rotations. The six degrees
of freedom are identified as 1, 2, 3, 4, 5, and 6, as shown below.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 5-1

© 2024 Siemens
 5. Defining Grid Points and Scalar Points

Figure 5-1. DOF Nomenclature

Other commonly used terms for the components of displacement at a grid point are

DOF 1 = T1 = U1 = translation in direction 1

DOF 2 = T2 = U2 = translation in direction 2

DOF 3 = T3 = U3 = translation in direction 3

DOF 4 = T4 = θ4 = rotation in direction 4

DOF 5 = T5 = θ5 = rotation in direction 5

DOF 6 = T6 = θ6 = rotation in direction 6

When you define a location of the grid point in space using these components, you are inherently
defining the point relative to a coordinate system.

In Simcenter Nastran, you use the GRID bulk data entry to define a grid point. GRID allows you to:

• Assign a unique ID to the grid point

• Specify the location of the grid point in space with respect to a reference coordinate system

• Assign permanent constraints to the grid point

• Define the directions of motions at the grid point

Format of the GRID Entry

The format of the grid point entry is as follows:

5-2 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Specifying the Grid Point ID (Field 1)

1 2 3 4 5 6 7 8 9 10
GRID ID CP X1 X2 X3 CD PS SEID

Field Contents

lD Grid point identification number.

CP Identification number of coordinate system in which the location of the grid

point is defined.

X1, X2, X3 Location of the grid point in coordinate system CP.

CD Identification number of coordinate system in which the displacements, degrees

of freedom, constraints, and solution vectors are defined at the grid point.

PS Permanent single-point constraints associated with the grid point.

SEID Superelement identification number.

See also

• “GRID” in the Simcenter Nastran Quick Reference Guide

Specifying the Grid Point ID (Field 1)

Field 1 lets you assign an ID to the grid point. That ID must be unique with respect to all other grid points
and scalar points in your model. In Simcenter Nastran, grid point IDs must begin at 1, however, they
don't have to be contiguous.

You will often find it helpful to assign groups of IDs to the different sections of your model. For example,
if you're working on an automotive FE model, you may assign IDs in the range from 1 to 1000 for the
grid points for the driver's door and assign IDs in the range from 5001 to 6000 for the grid points for the
hood. A clear and logical scheme for numbering grid points can be very helpful on complex models.

Defining Grid Point Coordinate Systems (Fields 3 and 5)

In Simcenter Nastran, each grid point refers to two coordinate systems, CP and CD:

• The CP system that you specify in field 3 is used to locate the grid point in space. This is known as the
“position” coordinate system. It defines the grid point's location in space.

• The CD system that you specify in field 7 is used to establish the grid point's displacement (output)
coordinate system. The CD system defines the direction of displacements, constraints, and other grid
point related quantities such as reaction forces. It is the system in which the software outputs the
results.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 5-3

© 2024 Siemens
 5. Defining Grid Points and Scalar Points

Simcenter Nastran has two main categories of coordinate system:

• The basic coordinate system (the default for the CP and CD fields)

• Local coordinate systems, which you define yourself. These can be rectangular, cylindrical, or spherical

With the CP and CD fields, you can use either the basic coordinate system or a local coordinate system.

Specifying the Location Coordinate System in the CP Field

To specify the coordinate system for the location of the grid point, enter the ID of that coordinate system
in the CP field (Field 3). If you leave the CP field blank, or enter a 0 in the CP field, the software defines
the grid point in the basic coordinate system.

X1, X2, and X3 in fields 4-6 let you define the location of the grid point in the CP coordinate system you
specified in field 3. They have the following meanings for different types of coordinate systems:

Type X1 X2 X3
Rectangular X Y Z
Cylindrical R θ (degrees) Z
Spherical R θ (degrees) φ (degrees)

Specifying the Output Coordinate System in the CD Field

To specify the coordinate system in which you want Simcenter Nastran to generate and output the
results at grid points (displacements, grid point forces, etc.), enter the ID of the coordinate system in
the CD field (Field 7). For many practical modeling situations, you may have several different output
coordinate systems in a single model.

Although the CD field is referred to as the displacement coordinate system, it is more than the
coordinate system used to output the results. It has a fundamental role in Simcenter Nastran. The
collection of all the coordinate systems defined on the CD field of all the grid points constitutes the
global coordinate system. It is important to note that the global coordinate system is not necessarily
a single coordinate system but rather the collection of all of your CD coordinate systems. Simcenter
Nastran assembles the global stiffness matrix in the global coordinate system.

See also

• “Comparing the Basic and Global Coordinate Systems” in the Simcenter Nastran User's Guide

Understanding How the CP and CD Fields Work Together

Although you can have the CP and CD fields reference the same coordinate system, you can also have
them reference different coordinate systems. For example, you may want to specify a local coordinate
system to define the location of the grid point in the CP field, and then leave the CD field blank or enter

5-4 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Defining the Directions of Displacement (Fields 4-6)

a 0 to output your results in the basic coordinate system. On the other hand, if you need to output
the displacement information in particular coordinate system, you enter the ID of that local coordinate
system in the CD field.

Defining the Directions of Displacement (Fields 4-6)

In Simcenter Nastran, each grid point has six degrees of freedom or “displacement components”; three
translational and three rotational. The degrees of freedom are denoted as u1, u2, u3, θ1, θ2, and θ3, or
as T1, T2, T3, R1, R2, and R3.

• T1, T2, and T3 are the three orthogonal components of translation parallel to the 1, 2, and 3
directions of the grid point's displacement coordinate system, respectively.

• R1, R2, and R3 are the components of rotation in the same directions. The center of rotation is at the
grid point location.

The coordinate system you specify in the CD field determines the orientation of these displacement
components.

• If the system you specify in the CD field is rectangular, the displacement components are parallel to
the local system as indicated in (a) below.

• If the system you specify in the CD field is cylindrical, the displacement components are in the radial,
tangential, and axial directions as indicated in (b) below.

• If the system you specify in the CD field is spherical, the displacement components are in the radial,
meridional, and azimuthal directions as indicated in (c) below.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 5-5

© 2024 Siemens
 5. Defining Grid Points and Scalar Points

Figure 5-2. Displacement Components

5-6 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Specifying Permanent Single Point Constraints (Field 8)

Specifying Permanent Single Point Constraints (Field 8)

You use field 8 (PS) of the GRID entry to specify permanent single point constraints to constrain any or
all the degrees of freedom associated with the grid point. In this field, you enter the integers (1-6) that
correspond to the degrees of freedom you want to constrain.

The constraints that you define in field 8 are considered permanent because you can't change them
during the run. Any degrees of freedom you specify in the PS field are constrained for all subcases.
Constraints specified on the GRID entry are usually restricted to those degrees-of-freedom that will not
be elastically constrained and hence must be removed from the model in order to avoid singularities in
the stiffness matrix.

In many situations, you want to analyze your structure using more than one set of constraints. For
these situations, don't specify the constraints on the GRID entry. Instead, specify the constraints using
an SPC bulk data entry (Single-Point Constraint). The only difference from using the GRID entry is that
constraints applied using an SPC entry must be selected through the Case Control section. Therefore,
you can apply different SPC entries for different subcases.

Simcenter Nastran applies the constraints you define in field 8 in the output coordinate system (CD),
which you defined in field 7. This gives you greater flexibility for modeling a variety of joints and
boundary conditions.

Note: Simcenter Nastran applies all constraints, whether you define them with an SPC bulk data entry or
using the PS field of the GRID entry, in the output coordinate system.

See also

• “SPC” in the Simcenter Nastran Quick Reference Guide

• “Single-Point Constraints” in the Simcenter Nastran User's Guide

Defining the Superelement ID (Field 9)

The SEID option in field 9 lets you define superelements within the bulk data section of your input file.

See also

• “Superelement Definition” in the Simcenter Nastran Superelement Users's Guide

Understanding the GRDSET Entry

If many or all of the GRID entries in your model have the same values in the location coordinate system
(CP), displacement coordinate system (CD), single point constraint (PS) fields, or superelement ID (SEID),
you can use the GRDSET bulk data entry to avoid having to repeatedly specify the same information. If
you specify a value for any of these fields on the GRDSET entry, the software then uses those values to
replace corresponding blank fields on any GRID entry in your model.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 5-7

© 2024 Siemens
 5. Defining Grid Points and Scalar Points

GRDSET is useful in the case of such problems as space trusses where you want to remove all the
rotational degrees-of-freedom. It is also helpful with plane structures where you want to remove all
out-of-plane or all in-plane motions.

Note: You can only include a single GRDSET entry in the bulk data section of your input file.

See also

• “GRDSET” in the Simcenter Nastran Quick Reference Guide

Specifying Grid Point Properties

In Simcenter Nastran, you specify some of the characteristics of the model as properties of grid points,
rather than as properties of elements.

Defining Mass, Damping, and Stiffness through Direct Input Matrices

You use direct input matrices to define mass, damping, and stiffness properties either in part or entirely,
as properties of grid points.

• You can use the DMIG bulk data entry define direct input matrices. You can associate these matrices
with components of geometric grid points, scalar points or extra points introduced for dynamic
analysis.

• You can use the TF bulk data entry define transfer functions that Simcenter Nastran automatically
converts to direct matrix input.

See also

• DMIG in the Simcenter Nastran Quick Reference Guide

• TF in the Simcenter Nastran Quick Reference Guide

• Defining the Fluid in the Simcenter Nastran Acoustics Simcenter Nastran User's Guide

Defining Mass through the Concentrated Mass Element

You can also use the concentrated mass element to input mass properties as properties of grid points.

• You can use the CONM1 entry to define a 6 × 6 matrix of mass coefficients at a geometric grid point in
any selected coordinate system.

• You can use the CONM2 entry to define a concentrated mass at a geometric grid point in terms of its
mass, the three coordinates of its center of gravity, the three moments of inertia about its center of
gravity, and its three products of inertia, referred to any selected coordinate system.

5-8 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Defining Temperatures at Grid Points

See also

• “CONM1” in the Simcenter Nastran Quick Reference Guide

• “CONM2” in the Simcenter Nastran Quick Reference Guide

Defining Temperatures at Grid Points

You can define thermal fields by specifying the temperatures at grid points.

• You use TEMP bulk data entry to specify the temperature at grid points for use in connection with
thermal loading and temperature-dependent material properties.

• You use the TEMPD bulk data entry to specify a default temperature to avoid a large number of
duplicate entries on a TEMP entry when the temperature is uniform over a large portion of the
structure.

See also

• “TEMP” in the Simcenter Nastran Quick Reference Guide

• “TEMPD” in the Simcenter Nastran Quick Reference Guide

Assigning a Singularity Normal to a Degree-of-Freedom

The SNORM Bulk Data entry and/or the PARAM,SNORM assigns the singularity normal of a shell element
to a degree-of-freedom. Simcenter Nastran then uses the AUTOSPC capability to remove this singularity.

See also

• “SNORM” in the Simcenter Nastran Quick Reference Guide

• “Automatically Applying Single-Point Constraints” in the Simcenter Nastran User's Guide

Understanding Scalar Points

In Simcenter Nastran, a scalar point has only one degree of freedom. A scalar point is similar to a
grid point in the way that Simcenter Nastran assembles it into the matrix equation. However, a scalar
point has no spatial orientation, so it does not require a coordinate system. It is simply an additional
degree of freedom that you can define in your model. Scalar points are useful for modeling with scalar
elements, multipoint constraints (MPCs), computing relative and average motions, and defining the
warping coefficients of a thin-walled beam.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 5-9

© 2024 Siemens
 5. Defining Grid Points and Scalar Points

Comparing Grid Points to Scalar Points

While you can use a grid point as a scalar point, a grid point requires more input and requires you to
constrain five of the six degrees of freedom. Consequently, using scalar points instead of a grid point is
more efficient, both computationally and in terms of the actual input required.

Defining Scalar Points

In Simcenter Nastran, you define use the SPOINT bulk data entry to define a scalar point. You can specify
up to eight scalar points on a single SPOINT entry. You can also implicitly define scalar points on a
connection entry for a scalar element SPOINT entries are used primarily to define scalar points appearing
in constraint equations but to which no structural elements are connected.

See also

• “SPOINT” in the Simcenter Nastran Quick Reference Guide

Using Scalar Points as Extra Points for Dynamics

You can also introduce special scalar points, called “extra points,” for dynamic analyses. Extra points
are used in connection with transfer functions and other forms of direct matrix input used in dynamic
analyses and are defined on EPOINT entries.

See also

• “DMIGs, Extra Points, and Transfer Functions” in the Simcenter Nastran Advanced Dynamic Analysis
User's Guide

Scalar Point Output Format

Scalar point output is formatted differently in f06 files as compared to punch files. There are four scalar
points in the example below labeled 101, 102, 103 and 104 to demonstrate these formats.

Scalar point results from an *.f06 file:

D I S P L A C E M E N T V E C T O R
POINT ID. TYPE T1 T2 T3 R1
R2 R3
1 G 0.0 0.0 0.0 0.0
0.0 0.0
2 G 0.0 0.0 1.228000E-04 0.0
-2.333333E-05 0.0
3 G 0.0 0.0 4.456000E-04 0.0
-4.000000E-05 0.0
40 G 0.0 0.0 9.017334E-04 0.0
-5.000000E-05 0.0
50 G 0.0 0.0 1.424533E-03 0.0

5-10 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Scalar Point Output Format

-5.333333E-05 0.0
100 G 1.024533E-03 1.870133E-03 0.0 0.0
0.0 0.0
101 S 1.024533E-03 1.870133E-03 1.024533E-03 1.870133E-03

In the f06 displacement results shown above, the “TYPE” column determines if the data is for a grid point
“G” or a scalar point “S”. Grid points have six degrees of freedom, whereas the scalar points have only
one. To reduce the f06 file, Simcenter Nastran prints up to six scalar point outputs per line. The ID at the
beginning of a row of scalar point values corresponds to the first scalar point output listed at the start of
that line. In the above example, scalar point “101” corresponds to the value listed at the start of the line,
then the subsequent values correspond to the single output for scalar points 102, 103 and 104.

Scalar point results from a punch file:

$DISPLACEMENTS 4
$REAL OUTPUT 5
$SUBCASE ID = 1 6
1 G 0.000000E+00 0.000000E+00 0.000000E+00 7
-CONT- 0.000000E+00 0.000000E+00 0.000000E+00 8
2 G 0.000000E+00 0.000000E+00 1.228000E-04 9
-CONT- 0.000000E+00 -2.333333E-05 0.000000E+00 10
3 G 0.000000E+00 0.000000E+00 4.456000E-04 11
-CONT- 0.000000E+00 -4.000000E-05 0.000000E+00 12
40 G 0.000000E+00 0.000000E+00 9.017334E-04 13
-CONT- 0.000000E+00 -5.000000E-05 0.000000E+00 14
50 G 0.000000E+00 0.000000E+00 1.424533E-03 15
-CONT- 0.000000E+00 -5.333333E-05 0.000000E+00 16
100 G 1.024533E-03 1.870133E-03 0.000000E+00 17
-CONT- 0.000000E+00 0.000000E+00 0.000000E+00 18
101 S 1.024533E-03 0.000000E+00 0.000000E+00 19
-CONT- 0.000000E+00 0.000000E+00 0.000000E+00 20
102 S 1.870133E-03 0.000000E+00 0.000000E+00 21
-CONT- 0.000000E+00 0.000000E+00 0.000000E+00 22
103 S 1.024533E-03 0.000000E+00 0.000000E+00 23
-CONT- 0.000000E+00 0.000000E+00 0.000000E+00 24
104 S 1.870133E-03 0.000000E+00 0.000000E+00 25
-CONT- 0.000000E+00 0.000000E+00 0.000000E+00 26

As shown in the punch file example above, a new row begins for each scalar point output beginning
with the ID. Since a scalar point has one degree of freedom, only the first value is significant and the
remaining 5 zero values can be ignored.

The modal degree of freedom output from a dynamic response solution is also a scalar quantity and has
the same format as the scalar point output described above, except it will have “M” under the “TYPE”
column.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 5-11

© 2024 Siemens
 5. Defining Grid Points and Scalar Points

5-12 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 6. Understanding Coordinate Systems
Overview of Coordinate Systems in Simcenter Nastran
In Simcenter Nastran, there are two general types of coordinate systems:

• The “basic” coordinate system, which is an implicitly defined reference coordinate system

• “Local” coordinate systems, which you define explicitly

Both of these different types of coordinate system is described in more detail below.

Understanding the Basic Coordinate System

Simcenter Nastran has a built-in Cartesian system called the basic coordinate system. It is also called
the default coordinate system. The basic system is a rectangular system, except when you're using
axisymmetric elements.

Figure 6-1. The Basic Coordinate System

ID of the Basic Coordinate System

In Simcenter Nastran, all coordinate systems have a coordinate system identification number (CID). The
basic coordinate system's identification number is zero (0) or blank.

Comparing the Basic and Global Coordinate Systems

Many finite element textbooks refer to a “global coordinate system.” However, this term means
something different in Simcenter Nastran. It is important to understand the Simcenter Nastran definition
of the global coordinate system as it differs from the classical textbook approach. These differences in
terminology can often be confusing to new Simcenter Nastran users.

In the classical approach, individual element stiffness matrices are generated for all the elements. Those
individual elemental matrices are then transformed into a single stiffness matrix commonly referred to

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 6-1

© 2024 Siemens
 6. Understanding Coordinate Systems

as the global stiffness matrix. Each row and column of this global stiffness matrix represents a degree of
freedom of the structural model. The coordinate system associated with each one of these degrees of
freedom is the same and is known as the global coordinate system.

In Simcenter Nastran, the software generates the element stiffness matrix similarly to the classical
approach. Simcenter Nastran assembles all the individual matrices into the global stiffness matrix.
However, instead of transforming the element stiffness matrices into a unique coordinate system,
Simcenter Nastran transforms these matrices into the coordinate system that you specified in the CD
field of the GRID entry for the grid points to which each element is attached. Simcenter Nastran then
assembles the individual elemental stiffness matrices to form the global stiffness matrix.

Therefore, in Simcenter Nastran, the “global” coordinate system isn't a unique coordinate system. It is
simply the collection of all the CD coordinate systems specified on all of the GRID entries. This approach
facilitates a variety of elegant modeling techniques. If you prefer to have the global coordinate system
be the same as the basic coordinate system, which is a unique system, you can leave the CD field blank
for all the grid points in your model.

Figure 6-2 shows the coordinate system definition where:

6-2 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Comparing the Basic and Global Coordinate Systems

Figure 6-2. Coordinate System Definition

(i, j, k) = local (user defined) coordinate system (x, y, z) or (r, θ, z) or (r, θ, φ)

(xb, yy, zy) = basic coordinate system
=
normalized

=
normalized

Each grid point you define with the GRID bulk entry includes six displacement components (three
translations and three rotations). You have the option to select a location coordinate system with the CP
field and a displacement coordinate system with the CD field on the GRID entry.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 6-3

© 2024 Siemens
 6. Understanding Coordinate Systems

See also

• “Defining Grid Points” in the Simcenter Nastran User's Guide

Defining a Local Coordinate System

Because you may need to create your model, or portions of your model, in a coordinate system
other than the basic coordinate system, Simcenter Nastran also lets you construct alternate coordinate
systems. In Simcenter Nastran, such coordinate systems are known as “local” coordinate systems. Local
coordinate systems may be:

• Rectangular

• Cylindrical

• Spherical

You can relate local coordinate systems either directly or indirectly to the basic coordinate system. Local
coordinate systems play an important role in generating your models, solving for the solution, and
displaying the results.

For example, consider the farm silo in Figure 6-3. It is cylindrical with a hemispherical dome. It is offset
from the origin of the basic coordinate system. Creating local cylindrical (r, θ, z) and spherical (r, θ, φ)
systems makes it easier to define model geometry or review displacement results.

6-4 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Bulk Data Entries for Defining Coordinate Systems

Figure 6-3. Example of Local Coordinate Systems to Aid in Establishing Model Geometry

Note: In cylindrical and spherical coordinate systems, angular input is in degrees. For example, when
defining the position of a grid point with a cylindrical coordinate system, the X2 field on the GRID entry
is entered in degrees. Although, rotational displacements defined with the SPCD entry and rotational
displacement output relative to any coordinate system are in radians.

Bulk Data Entries for Defining Coordinate Systems

In Simcenter Nastran, there are six different bulk data entries that you can use to define local coordinate
systems:

With these different bulk data entries, you can define coordinate systems by referencing:

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 6-5

© 2024 Siemens
 6. Understanding Coordinate Systems

• Three existing grid points

• Three existing grid points in an existing reference coordinate system

Defining a Coordinate System by Referencing Existing Grid Points

The CORD1R, CORD1C, and CORD1S entries let you define a local coordinate system by referencing three
defined grid points. This is also known as the “Method 1” approach of coordinate system definition in
Simcenter Nastran.

The method for defining a coordinate system with CORD1R, CORD1C, or CORD1S is very similar. With
each of these bulk data entries, you must:

• Specify a unique coordinate system ID (CID) in field 2 of the entry.

• Specify the IDs for each of the grid points that you want to reference. The three points must be
noncolinear and can't be coincident. Simcenter Nastran's geometry processor checks the points to
ensure that they're not colinear.

Additionally, with the CORD1R and CORD1C entries, you can define up to two coordinate systems on a
single entry.

Understanding the Importance of Using Fully Constrained, Separate Grid Points

Importantly, with Method 1s, if your model is later modified and any of the reference grid point locations
change, your local coordinate system orientation will also change. To avoid this type of problem when
using Method 1, you should always define three fully constrained grid points that are not part of the
structure. Then, if your model is modified, your local coordinate system remains unchanged.

This example illustrates the problems you can encounter if you use Method 1 to define a coordinate
system with the CORD1R entry using three existing grid points instead of defining three new ones.

Consider the truss structure shown in Figure 6-4. Suppose you want the positive T1 component of the
displacements for all the grid points to be in the negative Xb-direction as shown in (a).

In this case, we want to define CORD1R using existing grid points 2, 4, and 1 and assigning it an ID of 11.
You enter coordinate system 11 in the CD field of the GRID entry for all the grid points. The model solves
without problems and all results are in the correct direction.

Problems will occur if you need to subsequently modify the model. For example, suppose you need to
move the location of grid point 4 as shown in (b). However, in the process of moving grid point 4, you
also inadvertently rotate the local coordinate system 11.

While the deformed shape may look accurate if you post process the model, if you examine the results in
the output file, you will notice that they are not in the correct direction.

6-6 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Defining a Coordinate System By Referencing an Existing Coordinate System

Figure 6-4. Model Change

Defining a Coordinate System By Referencing an Existing Coordinate System

The CORD2R, CORD2C, and CORD2S entries let you define a local coordinate system by specifying the
location of three points in a previously defined reference coordinate system. This is also known as the
“Method 2” approach of coordinate system definition in Simcenter Nastran. This reference coordinate
system can either be the basic coordinate system or another local coordinate system. Importantly, if this
reference coordinate system is modified, the associated local coordinate system will also change.

The method for defining a coordinate system with CORD2R, CORD2C, or CORD2S is very similar. With
each of these bulk data entries, you must:

• Specify a unique coordinate system ID (CID) in field 2 of the entry.

• Specify the coordinate system ID (RID) in field 3 of the reference coordinate system you're using. If you
enter a 0 or a blank in this field, Simcenter Nastran uses the basic coordinate system for the reference
coordinate system.

• Enter the coordinates of three grid points in the reference coordinate system you defined in field 3.

Defining a Rectangular Coordinate System

You use the CORD1R and CORD2R Bulk Data entries to define a rectangular coordinate system.

Understanding CORD1R

When you define a rectangular coordinate with CORD1R, you must specify three reference grid points:
G1, G2, and G3.

• G1 defines the origin of the coordinate system.

• A vector taken from G1 to G2 defines the Z-axis

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 6-7

© 2024 Siemens
 6. Understanding Coordinate Systems

• G3, together with the Z-axis, define the XZ plane.

• The X-axis is defined to be in this XZ plane and is perpendicular to the Z-axis.

• The Y-axis is generated from the X- and Z-axes using the right-hand rule.

CORD1R Format

1 2 3 4 5 6 7 8 9 10
CORD1R CIDA G1A G2A G3A CIDB G1B G2B G3B

Field Contents

CIDA, CIDB Coordinate system identification number.

GiA, GiB Grid point identification numbers.

See also

• “CORD1R” in the Simcenter Nastran Quick Reference Guide

Example: Adding Three Additional Grid Points to Define CORD1R

To better understand the use of a CORD1R entry, consider the truss structure introduced earlier. The
input file has been modified so that the location of grid points 1 through 4 are defined in a rectangular
coordinate system with an ID of 11 as shown in Figure 6-5.

6-8 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Example: Adding Three Additional Grid Points to Define CORD1R

Figure 6-5. Grid Point Locations Using a Local Coordinate System

The new rectangular coordinate system with an ID of 11 has an origin located at the center of the truss
and is rotated 180 degrees about the basic Z-axis.

The corresponding input file for the structure is given in Listing 6-1. Note that it is necessary to add
three additional grid points to the input file. Importantly, these additional grid points don't affect the
overall stiffness of the structure.

• Grid point 11 is located at the center of the truss structure, but it isn't attached to the structure. It is
fully constrained.

• Grid point 12 is located 10 inches above grid point 11 in the basic Z-direction. The Z-axis for the new
coordinate system is therefore parallel to the basic Z-axis. The choice of 10 inches is arbitrary since
any positive value produces the same results.

• Grid point 13 defines the local XZ plane from which the X- and Y-axes of coordinate system 11 are
determined. The choice of 0 inches for the X location of grid point 13 is also somewhat arbitrary
since it is used only to define the XZ plane. Any real number less than 300 inches produces the same
results. If the X-component of grid point 13 is greater than 300 inches, then the X-axis will be in the
basic X-direction instead of the direction shown.

$
$ FILENAME - TRUSS6.DAT
$
BEGIN BULK
$
$ LOCAL COORDINATE SYSTEMS ARE DEFINED IN TERMS
$ GRID POINTS FOR METHOD 1

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 6-9

© 2024 Siemens
 6. Understanding Coordinate Systems

$
CORD1R 5 11 12 13
$
$ THE GRID POINTS LOCATIONS
$ DESCRIBE THE GEOMETRY IN CP FIELD COORDINATE SYSTEMS
$
GRID 1 5 -300. -60. 0. 3456
GRID 2 5 -300. 60. 0. 3456
GRID 3 5 300. 60. 0. 3456
GRID 4 5 300. -60. 0. 3456
$
$ THE FOLLOWING GRID POINTS ARE ADDED ONLY FOR DEFINING
$ LOCAL COORDINATE SYSTEM 5 AND ARE FULLY CONSTRAINED
$
GRID 11 300. 60. 0. 123456
GRID 12 300. 60. 10. 123456
GRID 13 0. 60. 5. 123456
$
.
.
.
ENDDATA
Listing 6-1. Truss Using CORD1R

Understanding CORD2R

With CORD2R, you define the locations of points A, B, and C in the same manner as with the CORD1R
entry. You define these points in space using coordinates relative to a reference coordinate system that
you specify in Field 3. You must define the reference coordinate system independently.

CORD2R Format

1 2 3 4 5 6 7 8 9 10
CORD2R CID RID A1 A2 A3 B1 B2 B3
C1 C2 C3

Field Contents

CID Coordinate system identification number.

RID Identification number of a coordinate system that is defined independently

from the CID coordinate system.

Ai, Bi, Ci Coordinates of three points in coordinate system defined in field 3.

6-10 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Defining Cylindrical Coordinate Systems

Figure 6-6. CORDIR and CORD2R Definitions

See also

• “CORD2R” in the Simcenter Nastran Quick Reference Guide

Defining Cylindrical Coordinate Systems

The CORD1C and CORD2C bulk data entries allow you to define cylindrical coordinate systems.

Understanding CORD1C

The CORD1C entry is used to define a cylindrical coordinate system using three grid points to define its
location and orientation in space as shown in Figure 6-7.

• Grid point G1 defines the origin of the cylindrical system.

• A vector from grid point G1 to grid point G2 defines the Z-axis.

• Grid point G3 together with the Z-axis defines the θ = 0 plane.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 6-11

© 2024 Siemens
 6. Understanding Coordinate Systems

Figure 6-7. CORD1C Definition

CORD1C Format

1 2 3 4 5 6 7 8 9 10
CORD1C CIDA G1A G2A G3A CIDB G1B G2B G3B

Field Contents

CIDA, CIDB Coordinate system identification number.

GiA, GiB Grid point identification numbers.

See also

• “CORD1C” in the Simcenter Nastran Quick Reference Guide

Understanding CORD2C

The CORD2C entry uses the location of three points A, B, and C in the same manner as the three grid
points are used for the CORD1C entry. These points are defined in space using coordinates relative to the
reference coordinate system that you specify in Field 3.

• Point A defines the origin.

• Point B defines the direction of the z-axis.

• Point C lies in the plane of the azimuthal origin.

You must independently define the reference coordinate system (RID).

6-12 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 CORD2C Format

The location of a grid point (P in the figure CORD2C Definition) in this coordinate system is given by
(r, θ,z) where θ is measured in degrees. If this coordinate system is used to specify the displacement
directions they are given by ur, uθ and uz (the displacements are ur, uθ and uz).

CORD2C Format

1 2 3 4 5 6 7 8 9 10
CORD2C CID RID A1 A2 A3 B1 B2 B3
C1 C2 C3

Field Contents

CID Coordinate system identification number. (Integer > 0)

RID Identification number of a coordinate system that is defined independently

from this coordinate system. (Integer ≥ 0; Default (zero) is the basic coordinate
system)

Ai, Bi, Ci Coordinates of three points in coordinate system defined in field 3. (Real)

Figure 6-8. CORD2C Definition

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 6-13

© 2024 Siemens
 6. Understanding Coordinate Systems

See also

• “CORD2C” in the Simcenter Nastran Quick Reference Guide

Example: Using the CORD2C Entry

The following example illustrates the use of the CORD2C entry. An arch with a semicircular top is shown
in Figure 6-9. To facilitate grid point input, we want to establish a local cylindrical coordinate system for
grid points 3 through 7.

Figure 6-9. Arch Structure

A cylindrical coordinate system with an ID of 100 is defined as shown in Figure 6-10.

6-14 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Example: Using the CORD2C Entry

Figure 6-10. Definition of Local Cylindrical Coordinate System

The local cylindrical coordinate system is defined by referencing the basic coordinate system. Three
points are required to define the local system:

• Point A is at the origin

• Point B lies on the z-axis of the new system.

• Point C defines the reference axis at which θ = 0°.

This coordinate system has an identification number in field CID which will be referenced by other
entities, such as the CP and CD fields of a GRID entry. In this example, the coordinates of points A, B, and
C are in the basic coordinate system.

1 2 3 4 5 6 7 8 9 10
CORD2C CID RID A1 A2 A3 B1 B2 B3
C1 C2 C3
CORD2C 100 15. 50. 0. 15. 50. 1.
16. 50. 0.

The associated GRID entries are defined as follows:

1 2 3 4 5 6 7 8 9 10
GRID ID CP X1 X2 X3 CD PS SEID
GRID 1 0. 0. 0.
GRID 2 0. 50. 0.
GRID 3 100 15. 150. 0.
GRID 4 100 15. 120. 0.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 6-15

© 2024 Siemens
 6. Understanding Coordinate Systems

1 2 3 4 5 6 7 8 9 10
GRID 5 100 15. 90. 0.
GRID 6 100 15. 60. 0.
GRID 7 100 15. 30. 0.
GRID 8 30. 50. 0.
GRID 9 30. 0. 0.

Grid points 3 through 7 use (r, θ, z) coordinates with r = 15.0 inches and θ varying from 30 degrees
(GRID 7) to 150 degrees (GRID 3). The output for all grid points is in the basic rectangular coordinate
system since field 7 is left blank.

Understanding CORD1S

The CORD1S entry is similar to the CORD1C entry since three grid points (G1, G2, and G3) are used to
define the new coordinate system. As shown in the figure CORD1S and CORD2S Definitions:

• Grid point G1 defines the origin of the spherical system.

• The vector starting at grid point G1 and passing through grid point G2 is the θ = 0 axis.

• This axis, together with grid point G3, determines the φ = 0 plane.

• The φ = 0 axis is in this plane and is oriented using the right-hand rule.

CORD1S Format

1 2 3 4 5 6 7 8 9 10
CORD1S CIDA G1A G2A G3A CIDB G1B G2B G3B

Field Contents

CIDA, CIDB Coordinate system identification numbers.

GiA, GiB Grid point identification numbers.

See also

• “CORD1S” in the Simcenter Nastran Quick Reference Guide

Understanding CORD2S

The CORD2S entry uses the location of three points A, B, and C in the same manner as the three grid
points are used for the CORD1S entry. These points are defined in space using coordinates relative to the
reference coordinate system (Field 3). The reference coordinate system must be independently defined.

6-16 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 CORD2S Format

CORD2S Format

1 2 3 4 5 6 7 8 9 10
CORD2S CID RID A1 A2 A3 B1 B2 B3
C1 C2 C3

Field Contents

CID Coordinate system identification number.

RID Identification number of a coordinate system that is defined independently

from this coordinate system.

Ai, Bi, Ci Coordinates of three points in coordinate system defined in field 3.

Figure 6-11. CORD1S and CORD2S Definitions

Note:
θ for the spherical coordinate system is different than the θ in the cylindrical coordinate system.

Example: Using Cylindrical and Spherical Coordinate Systems

To see the use of cylindrical and spherical coordinate systems, consider the grain silo shown below.

• The cylindrical portion of the silo is modeled using a cylindrical coordinate system.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 6-17

© 2024 Siemens
 6. Understanding Coordinate Systems

• The roof of the silo is modeled using a spherical coordinate system.

Figure 6-12. Silo Model

The cylindrical coordinate system has an ID of 1 and is defined with the CORD2C entry. It is defined with
respect to the basic coordinate system.

• The origin is given by point A1, and located by the vector (100,0,0) as shown in Figure 6-12.

• The Z-axis is defined by the line extending from point A1 to point B1, which is located at (100,1,0).

• The Z-axis along with point C1, which is located at (101,1,0), defines the θ = 0 plane.

• The R-axis is defined in this plane and is perpendicular to the Z-axis.

• The direction of the θ-axis is determined using the right-hand rule.

The spherical coordinate system has an ID of 2 is defined with the CORD2S entry. It is defined with
respect to the reference coordinate system 1.

• The location of origin of the spherical coordinate system 2 is given by point A2 at (0,0,50). This
location is with respect to coordinate system 1.

• The θ = 0 axis is the line extending from A2 to B2, which is located at (0,0,51).

• The φ = 0 plane is located by the φ = 0 axis and point C2, which is located at (1,0,51).

Since coordinate system 1 is defined in terms of the basic coordinate system, coordinate system 2
is also indirectly defined in terms of the basic coordinate system. You could also directly define both
coordinate systems in terms of the basic coordinate system. However, since coordinate system 2 is linked

6-18 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Understanding the Element and Material Coordinate System

to coordinate system 1, adjusting the position of the silo (for example, due to a change in the module's
geometry) is only a matter of changing the locations of the grid points on the CORD2C entry.

BEGIN BULK
$
$ THE CYLINDRICAL SYSTEM USED FOR THE SILO BODY
$
CORD2C 1 0 100. 0. 0. 100. 1. 0. +COR1
+COR1 101. 1. 0.
$
$ THE SPHERICAL SYSTEM USED FOR THE SILO CAP
$
CORD2S 2 1 0. 0. 50. 0. 0. 51. +COR2
+COR2 1. 0. 51.
.
.
.
ENDDATA
Listing 6-2. Grain Silo Coordinate Systems

Understanding the Element and Material Coordinate System

Every element in the Simcenter Nastran library has its own unique element coordinate system that
is used to output the element forces, moments, and stresses as well as to orient section properties.
For some elements, such as the CROD element, the element coordinate systems are implicitly defined
based on the element connectivity. For other elements, you must explicitly define the orientation of the
element coordinate system. For further discussion of element coordinate systems, refer to the Simcenter
Nastran Element Library.

In addition to the element coordinate system, an element may also have an optional material coordinate
system that can be used to define the orientation of orthotropic or anisotropic material.

Special Considerations for Grid Points on the Polar Axis

If the displacements of a grid point are defined to be in a spherical or cylindrical coordinate system and
the grid point is located on, or relatively close to, the polar axis, then it will be replaced by the “defining”
rectangular coordinate system. The z-axis of the “defining” rectangular coordinate system is coincident
with the polar axis and the x-axis is coincident with θ = 0 in cylindrical system, or θ = 0 and φ = 0 in the
spherical system. In this case, you should explicitly specify a rectangular coordinate system.

Defining Element Offsets

Element offsets are defined by a Cartesian coordinate system, which is also parallel to the displacement
coordinate system at the connecting grid point. However, the components of the offsets are always
defined in units of translation, even if the displacement coordinate system is cylindrical or spherical.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 6-19

© 2024 Siemens
 6. Understanding Coordinate Systems

Other Special Coordinate System Entries

CORD3G lets you define a general coordinate system using three rotational angles as functions of
coordinate values in the reference coordinate system. You use the CORD3G entry with the MAT9 entry to
orient the material principal axes for 3D composite analyses.

See also

• “CORD3G” in the Simcenter Nastran Quick Reference Guide.

6-20 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 7. Material Properties
Overview of Material Properties
Simcenter Nastran allows you to define material properties for solids and fluids. You use material bulk
entries to specify properties like Young’s modulus, Poisson’s ratio, density, thermal conductivity, and
so on. In some cases, material bulk entries are also used to specify parameters that are not normally
considered to be material properties. For example, in heat transfer analysis, the heat transfer coefficient
can be specified using a MAT4 bulk entry.

Generally, material properties are assigned to elements indirectly with each element bulk entry
referencing a property bulk entry and each property bulk entry referencing a material bulk entry.
Although infrequent, the material properties for some elements are also defined:

• Directly on the element bulk entry. CELAS2 is an example of this.

• Indirectly on a property bulk entry. CELAS1 is an example of this.

• Indirectly on a material bulk entry referenced by the element bulk entry. CONROD is an example of
this.

Table 7-1 summarizes the material bulk entries that are available to you. In Table 7-1, the material bulk
entries are categorized as primary or supplemental. The material bulk entries categorized as primary
are stand-alone bulk entries, whereas those categorized as supplemental are only used in combination
with a primary material bulk entry. For example, the strain hardening characteristic of a material can be
defined using a MATS1 bulk entry. However, the MATS1 bulk entry must be referenced by either a MAT1,
MAT2, or MAT9 bulk entry.

Table. Summary of Material Bulk Entries

Material Bulk Entry Primary or Property Type Description
Supplemental (1)
MAT1 Primary Mechanical (2) Defines properties for isotropic, linearly
elastic materials.
MAT2 Primary Mechanical (2) Defines properties for anisotropic,
linearly elastic materials. Valid for shell
elements only.
MAT3 Primary Mechanical (2) Defines properties for orthotropic,
linearly elastic materials. Valid for
axisymmetric, plane stress, and plane
strain elements.
MAT4 Primary Thermal Defines properties for isotropic materials.
MAT5 Primary Thermal Defines properties for anisotropic
materials.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 7-1

© 2024 Siemens
 7. Material Properties

Material Bulk Entry Primary or Property Type Description

Supplemental (1)
MAT8 Primary Mechanical (2) Defines properties for orthotropic,
linearly elastic materials. Valid for shell
elements only.
MAT9 Primary Mechanical (2) Defines properties for anisotropic,
linearly elastic materials. Valid for solid
elements only.
MAT10 Primary Fluid Defines properties for fluid or absorber
elements in coupled fluid-structural
analysis.
MAT11 Primary Mechanical (2) Defines properties for orthotropic,
linearly elastic materials.
MAT1F Supplemental Mechanical Defines frequency-dependent material
properties on MAT1 entry fields for SOLs
108 and 111.
MATFT Supplemental Mechanical Defines failure criterion when used
in combination with MAT1, MAT9, or
MAT11 bulk entries.
MATHP Primary Mechanical (2) Defines properties for hyperelastic
materials.
MATPOR Primary Fluid Defines properties for porous materials
used as acoustic absorbers.
MATS1 Supplemental Mechanical Defines stress-dependent properties on
MAT1, MAT2, or MAT9 bulk entries.
MATSMA Primary Mechanical (2) Defines properties for shape-memory
alloys.
MATT1 Supplemental Mechanical Defines temperature dependency for
properties on MAT1 bulk entries.
MATT2 Supplemental Mechanical Defines temperature dependency for
properties on MAT2 bulk entries.
MATT3 Supplemental Mechanical Defines temperature dependency for
properties on MAT3 bulk entries.
MATT4 Supplemental Thermal Defines temperature dependency for
properties on MAT4 bulk entries.
MATT5 Supplemental Thermal Defines temperature dependency for
properties on MAT5 bulk entries.
MATT8 Supplemental Mechanical Defines temperature dependency for
properties on MAT8 bulk entries.
MATT9 Supplemental Mechanical Defines temperature dependency for
properties on MAT9 bulk entries.
MATT11 Supplemental Mechanical Defines temperature dependency for
properties on MAT11 bulk entries.

7-2 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Material Properties for Linear Structural Analysis

Material Bulk Entry Primary or Property Type Description

Supplemental (1)
MFLUID Primary FluidDefines properties of an incompressible
fluid for the purpose of representing a
fluid mass.
CREEP Supplemental Mechanical Defines creep properties when used in
combination with MAT1, MAT2, or MAT9
bulk entries.
(1) Primary material bulk entries are stand-alone, whereas supplemental material bulk entries are only

used in combination with a primary material bulk entry.

(2) Includes properties for thermal expansion.

Material Properties for Linear Structural Analysis

For linear structural analysis, you can use the MAT1, MAT2, MAT3, MAT8, MAT9, and MAT11 bulk entries
to define the mechanical properties, coefficient of thermal expansion, and reference temperature for a
linear elastic material. The choice of which material bulk entry to use depends on the type of element
used and the degree of anisotropy required. Table 7-2 is provided to simplify the selection process.

Table. Summary of Material Bulk Entries for Linear Structural Analysis

Material Bulk Types of Elements Constitutive Number of Independent

Entry Model Elastic Constants
MAT1 Line, Shell, Axisymmetric, 2D Isotropic 2
Solid, 3D Solid
MAT2 Shell, Axisymmetric Anisotropic (a) 6 (b)
MAT3 Axisymmetric, 2D Solid Orthotropic 7 (c)
MAT8 Shell Orthotropic 4 (d)
MAT9 Axisymmetric, 3D Solid Anisotropic (a) 21
MAT11 Axisymmetric, 2D Solid, 3D Orthotropic 9
Solid
(a) Also referred to as triclinic or aelotropic.
(b) The number of independent elastic constants required to define an anisotropic material is reduced
from 21 for a 3D stress state to 6 for a 2D stress state.
(c) The number of independent elastic constants required to define an orthotropic material is reduced

from 9 for a 3D stress state to 7 for an axisymmetric representation of a 3D stress state.

(d) The number of independent elastic constants required to define an orthotropic material is reduced

from 9 for a 3D stress state to 4 for a 2D stress state.

The material properties defined on any of the material bulk entries listed in Table 7-2 can be made
temperature dependent by defining a corresponding MATTi bulk entry. For example, to define an
isotropic material having a temperature dependent Young’s modulus, you can use a MATT1 bulk entry in
combination with a MAT1 bulk entry to define the temperature dependency.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 7-3

© 2024 Siemens
 7. Material Properties

Isotropic Materials

For an isotropic material, the properties are the same in all directions. Two independent elastic constants
are required to characterize the mechanical behavior of an isotropic material. The constitutive relation
for a linear elastic isotropic material subjected to a three-dimensional stress state is:

where:

σx σy σz Normal stresses
τxy τyz τzx Shear stresses
εx εy εz Normal strains
γxy γyz γzx Engineering shear strains
E Young's modulus
ν Poisson's ratio
A Coefficient of thermal expansion
(T − Tref) Temperature difference used to calculate thermal strain

Linear elastic isotropic materials can be defined directly using MAT1 bulk entries. On the MAT1 bulk
entry, there are three fields for entering elastic constants including fields for Young’s modulus, Poisson’s
ratio, and shear modulus even though only two independent elastic constants are required to define a

7-4 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Constitutive Relation for Plane Stress

linear elastic isotropic material. When you specify only two of these elastic constants, Simcenter Nastran
calculates the third from:

If you enter all three elastic constants and they do not satisfy this relationship, Simcenter Nastran prints
a warning message indicating that the isotropic relationship has been violated. However, depending
on the element type, Simcenter Nastran may use all three values to formulate the element stiffness
matrix even though the values of the elastic constants are incompatible. Because of this, we strongly
recommend that you input only two of the three elastic constants on a MAT1 bulk entry.

Which elastic constants are used in the element stiffness matrix formulation depends on the element
type. Table 7-3 shows which elastic constants are used to formulate the stiffness matrices for various
elements.

Table. Elastic Constant Usage in Stiffness Formulation for Various Elements

Type Element E ν G
1D CROD Axial and Bending Not Used Torsion and Transverse
Shear
CBEAM
CBAR
2D CQUADi Membrane and Bending Transverse Shear
CTRIAi
CRAC2D All Terms Not Used
CSHEAR Not Used Shear
Axisymmetric CQUADXi Radial, Axial, All Coupled Ratios Shear
Circumferential
CTRAXi
3D CHEXA All Terms Not Used
CPENTA
CPYRAM
CRAC3D
CTETRA

Constitutive Relation for Plane Stress

The stress state for 2D elements is modeled as either plane stress (the default) or plane strain. For a
plane stress state, σz = τyz = τzx = 0 and the constitutive relation for an isotropic material reduces to:

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 7-5

© 2024 Siemens
 7. Material Properties

Equation 7-1.

Constitutive Relation for Plane Strain

You can obtain plane strain behavior for a 2D element by specifying certain settings on the property bulk
entry. For a state of plane strain, εz = γyz = γzx = 0 and the constitutive relation for an isotropic material
reduces to:

Equation 7-2.

Constitutive Relation for Transverse Shear

For calculating transverse shear stresses in 2D elements, the transverse shear strains are related to the
transverse shear stresses through the shear modulus, G, as follows:

Equation 7-3.

Orthotropic Materials

Orthotropic materials have three planes of material property symmetry. Nine independent elastic
constants are required to fully describe three-dimensional orthotropic material behavior. For two-
dimensional stress states and axisymmetric representations of three-dimensional stress states, the

7-6 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Understanding the MAT3 Bulk Entry

number of independent elastic constants required to define orthotropic material behavior is reduced.
Four independent elastic constants are required for a two-dimensional stress state and seven
independent elastic constants are required for an axisymmetric stress state.

You can use MAT3, MAT8, and MAT11 bulk entries to define orthotropic material properties.

• MAT3 allows you to define orthotropic material properties for some axisymmetric elements, and all
plane stress and plane strain elements.

• MAT8 allows you to define orthotropic material properties for some shell elements.

• MAT11 allows you to define orthotropic material properties for all axisymmetric, plane stress, plane
strain, and some 3D solid elements.

Understanding the MAT3 Bulk Entry

The MAT3 entry allows you to define orthotropic material behavior for the axisymmetric elements
CQUADX4, CQUADX8, CTRAX3, CTRAX6, the plane strain elements CPLSTN3, CPLSTN4, CPLSTN6,
CPLSTN8, and the plane stress elements CPLSTS3, CPLSTS4, CPLSTS6, CPLSTS8. These elements can
be defined on either the XZ plane or the XY plane of the basic coordinate system.

Xm, Ym, and Zm on the MAT3 are the principal material coordinates. The principal material coordinate are
defined relative to the basic coordinates by angle THETA. For example, Figure 1 and Figure 2 show the
positive sense of THETA for the CQUADX4 element.

Elements defined on the XZ plane of the basic coordinate system have different material coordinates
than elements defined on the XY plane of the basic coordinate system. As a result, MAT3/MATT3 entries
defined for elements on one plane can not be used for elements on the other plane.

Note:
When axisymmetric elements are defined on the XZ plane, X is the radial direction, and Z is the
axial direction. When axisymmetric elements are defined on the XY plane, Y is the radial direction,
and X is the axial direction. See CQUADX4, CQUADX8, CTRAX3, and CTRAX6 in the Simcenter
Nastran Quick Reference Guide.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 7-7

© 2024 Siemens
 7. Material Properties

Figure 7-1. CQUADX4 Principal Material Coordinate System, XZ-Plane

Figure 7-2. CQUADX4 Principal Material Coordinate System, XY-Plane

Axisymmetry Elements with Orthotropic Materials

The strain-stress relation in material coordinates for axisymmetric elements oriented on the XZ plane is
given by:

7-8 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Understanding the MAT3 Bulk Entry

Equation 7-4.

where

Equation 7-5.

and

σx σy σz Normal stresses
τzx Shear stress
εx εy εz Normal strains
γzx Engineering shear strain
Ex Ey Ez Young's moduli
νyx νzx νxy νzy νxz νyz Poisson's ratios
Gzx Shear modulus when elements are modeled on the XZ plane of the basic
coordinate system
Ax Ay Az Coefficients of thermal expansion
(T − Tref) Temperature difference used to calculate thermal strain

For axisymmetric elements oriented on the XZ plane, you must define the elastic constants: Ex, Ey, Ez,
νxy, νyz, νzx, and Gzx.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 7-9

© 2024 Siemens
 7. Material Properties

For axisymmetric elements oriented on the XY plane, in the equations above, Y replaces X, X replaces Z,
Z replaces Y. The shear modulus for the XY orientation is Gxy.

Plane Stress Elements with Orthotropic Materials

The strain-stress relation in material coordinates for plane stress elements oriented on the XZ plane is
given by:

where

and

For plane stress elements oriented on the XZ plane, you must define the elastic constants: Ex, Ez, νxy, νyz,
νzx, and Gzx.

For plane stress elements oriented on the XY plane, in the equations above, Y replaces X, X replaces Z, Z
replaces Y. The shear modulus for the XY orientation is Gxy.

Plane Strain Elements with Orthotropic Materials

The strain-stress relation in material coordinates for plane strain elements oriented on the XZ plane is
given by:

7-10 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Understanding the MAT8 Entry

where

and

For plane strain elements oriented on the XZ plane, you must define the elastic constants: Ex, Ey, Ez, νxy,
νyz, νzx, and Gzx.

For plane strain elements oriented on the XY plane, in the equations above, Y replaces X, X replaces Z, Z
replaces Y. The shear modulus for the XY orientation is Gxy.

Understanding the MAT8 Entry

The MAT8 bulk entry is used to define orthotropic materials for a 2D stress state. For a 2D stress state, σ3
= τ23 = τ31 = 0. The constitutive relation expressed in terms of the principal material directions is given
by:

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 7-11

© 2024 Siemens
 7. Material Properties

Equation 7-6.

subject to:

Equation 7-7.

where:

σ1 σ2 Normal stresses
τ12 Shear stress
ε1 ε2 Normal strains
γ12 Engineering shear strain
E1 E2 Young's moduli
ν12 ν21 Poisson's ratios
G12 Shear modulus
A1 A2 Coefficients of thermal expansion
(T − Tref) Temperature difference used to calculate thermal strain

Five elastic constants are included in Equation 7-6. However, Equation 7-7 reduces the number of
independent elastic constants to four. Accordingly, the MAT8 bulk entry requires you to enter four elastic
constants: E1, E2, ν12, and G12. Using the elastic constants you define and Equation 7-7, Simcenter
Nastran calculates the remaining terms in Equation 7-6.

To account for transverse shear stresses, you can entering values for G1z and G2z on the MAT8 entry
where the z-direction is through the thickness of the 2D element. The transverse shear stresses are
related to the transverse shear strains by:

7-12 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 MAT8 Example

Equation 7-8.

where:

τ1z τ2z Transverse shear stresses

γ1z γ2z Engineering shear strains
G1z G2z Shear moduli

When G1z or G2z are input as zero, Simcenter Nastran neglects the corresponding effects of transverse
shear flexibility in the shell formulation.

You can use the MAT8 bulk entry in conjunction with either a PCOMP or PCOMPG property bulk entry to
define composite lamina properties for thin shell elements.

See also

• “PCOMP and PCOMPG” in the Simcenter Nastran Quick Reference Guide.

MAT8 Example

The composite tube structure shown in Figure 7-3 illustrates the modeling of an orthotropic material.
The longitudinal direction is the fiber direction and is oriented at an angle of 45 degrees relative to the
tube axes as shown. The tube is held fixed at one end, and a uniform axial load is applied at the other
end.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 7-13

© 2024 Siemens
 7. Material Properties

Figure 7-3. Composite Tube with Orthotropic Properties

A sample of the tube material has been tested, and the material properties are determined in the
longitudinal (Xmat) and transverse (Ymat) directions. The material properties are shown in Table 7-4.

Table. Composite Tube Material Properties

Longitudinal Direction Transverse Direction Shear

E –

ν 0.4 0.16 –
G12 – –

Listing 7-1 shows the input file. The geometry has been defined with a local cylindrical coordinate
system for convenience. This does not affect the orientation of the material coordinate system.

7-14 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 MAT8 Example

$
$ GEOMETRY DEFINED AND OUTPUT IN THE CYL COORD SYS 1
$
CORD2C 1 0 0.0 0.0 0.0 0.0 0.0 1
1. 0.0 0.0
$
GRID 110 1 15. 150. 30. 1
GRID 111 1 15. 150. 60. 1
GRID 114 1 15. -150. 30. 1
GRID 115 1 15. -150. 60. 1
$
CQUAD4 8 1 110 114 115 111 135.
$
PSHELL 1 1 3. 1
$
MAT8 1 250000. 100000. .4 80000. 80000. 80000.
$

Listing 7-1. Orthotropic Tube Input File

Note:
Only element 8 and its connected grid points are shown.

The finite element model for the tube is shown in Figure 7-4. Element 8 is highlighted to illustrate how
the material coordinate system is used. The material coordinate system is oriented at an angle of 135
degrees (field 8 of the CQUAD4 entry). The 135 degrees is with respect to a line drawn from G1 (which
has a grid ID of 110) to G2 (which has a grid ID of 114). The resulting material coordinate system is
therefore the same as that shown in Figure 7-3. If you look at the deformed model, it is interesting to
note that the tube rotates about the z-axis even though a uniform axial load is applied. This rotation is
due to the orthotropic material behavior and the fact that the fiber direction is oriented in a different
direction than the applied load.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 7-15

© 2024 Siemens
 7. Material Properties

Figure 7-4. Finite Element Model of the Orthotropic Tube

Understanding the MAT11 Entry

For the MAT11 bulk entry, the constitutive relation expressed in terms of the principal material directions
is given by:

7-16 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Understanding the MAT11 Entry

Equation 7-9.

subject to:

Equation 7-10.

where:

σ1 σ2 σ3 Normal stresses
τ12 τ23 τ31 Shear stresses
ε1 ε2 ε3 Normal strains
γ12, γ23, γ31 Engineering shear strains
E1 E2 E3 Young's moduli
ν12 ν21 ν13 ν31 ν23 ν32 Poisson's ratios
G12 G23 G31 Shear modulus
A1 A2 A3 Coefficients of thermal expansion
(T − Tref) Temperature difference used to calculate thermal strain

Twelve elastic constants are included in Equation 7-9. However, the three equalities in Equation 7-10
reduce the number of independent elastic constants to nine. Accordingly, the MAT11 bulk entry requires
you to enter nine elastic constants: E1, E2, E3, ν12, ν13, ν23, G12, G13, and G23. Using the elastic constants
you define and Equation 7-10, Simcenter Nastran calculates the remaining terms in Equation 7-9.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 7-17

© 2024 Siemens
 7. Material Properties

Anisotropic Materials

Materials that have no planes of symmetry for the material properties are termed anisotropic.
Frequently, they are also referred to as triclinic or aelotropic. You can define anisotropic materials using
the MAT2 and MAT9 bulk entries. You can also define other classes of material behavior like monoclinic,
orthotropic, transverse isotropic, and isotropic through proper specification of the elastic constants.

• MAT2 allows you to define anisotropic material properties for 2D elements.

• MAT9 allows you to define anisotropic material properties for 3D elements.

To characterize an anisotropic material for a three-dimensional stress state, 21 independent elastic

constants are required. For a two-dimensional stress state, the number of independent elastic constants
is reduced to six.

Understanding MAT9

The MAT9 bulk entry is used to define anisotropic material behavior for 3D (solid) elements. The
constitutive relation is:

Equation 7-11.

where:

σx σy σz Normal stresses
τxy τyz τzx Shear stresses
εx εy εz Normal strains
γxy γyz γzx Engineering shear strains
Cij Elements of the 6 × 6 symmetric material property matrix in the material
coordinate system

7-18 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 MAT9 Example

Ai Coefficients of thermal expansion

(T − Tref) Temperature difference used to calculate thermal strain

When you use the MAT9 entry, you should define a material coordinate system on the PSOLID entry
(using field 4). For solid elements, Simcenter Nastran outputs stresses in the material coordinate system,
which by default, is the basic coordinate system. In general, for solid elements, it is difficult to determine
the orientation of the element coordinate system.

MAT9 Example

The following example illustrates the use of a MAT9 bulk entry to define an anisotropic material. This
example consists of three steps:

1. Obtain the elastic constants using a pseudo-anisotropic model consisting of a single CHEXA
element and CBAR elements applied along the edges and across the diagonals of the CHEXA
element.

2. Verify that the constitutive relation developed in the previous step is correct.

3. Use the constitutive relation in a structural analysis.

This three-step approach is analogous to constructing an anisotropic material test coupon and then
testing that coupon to obtain a strain matrix. From the strain matrix, you can develop the constitutive
relation. Then, you can verify the constitutive relation by modeling the test coupon. Finally, you use the
newly developed constitutive relation in a structural analysis.

Step 1 – Develop the Material Matrix

To simulate a three-dimensional anisotropic material, this example uses a single CHEXA element with
CBAR elements of various cross-sectional areas and bending stiffness defined along the edges and across
the diagonals of the CHEXA element as shown in Figure 7-5(a). The cross sections of the bars are not
important, but if you are interested in the actual dimensions, refer to the input file “anis1.dat” in the
Test Problem Library located on the delivery media. The dimensions of the cube are 1 x 1 x 1 inches. To
constrain the cube, one of the corner grid points is fixed in all six component directions.

Six self-equilibrating load cases are used to represent each of the six stress components. Each direction
was applied as a separate subcase. A self-equilibrating normal load consists of applying equal and
opposite forces to opposite faces of the cube (i.e., forces F1, F2, and F3).

For the shear load, a self-equilibrating load consists of applying four forces around the cube in order to
place the cube in a state of pure shear. Note that for all self-equilibrating load cases, the net resultant
should be zero; hence, the SPC force should also be zero. Because of the dimensions chosen for the
CHEXA element, each load case represents a unit stress, and the resulting strains are the strains due to
unit stress.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 7-19

© 2024 Siemens
 7. Material Properties

Figure 7-5. Pseudo-Anisotropic Model

The strains at the center of the CHEXA were extracted from the output file and are summarized in Table
7-5.

Table. Strains Due to Unit Loads on Pseudo Anisotropic Model

Applied Unit Load

FX FY FZ FXY FYZ FZX
Resulting εx 0.155 -0.034 -0.056 -0.016 0.002 -0.011
Strains (10–6)
εy -0.034 0.072 -0.018 -0.005 0.001 -0.003
εz -0.056 -0.018 0.107 -0.008 0.003 -0.005
γxy -0.016 -0.005 -0.008 0.197 -0.011 0.009
γyz 0.002 0.001 0.003 -0.011 0.219 -0.021
γzx -0.011 -0.003 -0.005 0.009 -0.021 0.180

Table 7-5 represents a strain matrix that, when inverted, is the material constitutive matrix [C]. Note that
since the strain matrix is symmetric, the material constitutive matrix [C] is also symmetric.

The result is the material constitutive matrix [C] for the anisotropic material is given by Equation 7-12.

7-20 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Step 2 – Verify the Material Matrix

Equation 7-12.

Step 2 – Verify the Material Matrix

Now that the material constitutive matrix has been generated, the next step is to verify that the matrix is
entered on the MAT9 entry correctly. To do so, return to the 1 x 1 x 1 inch cube. However, this time the
analysis will not include the CBAR elements. The same six load cases are applied as in Step 1.

If the material constitutive matrix is correct, the strains produced should be numerically the same as
those shown in the table Strains Due to Unit Loads on Pseudo Anisotropic Model.

The input file for the verification model is shown in Listing 7-2.

$
$ FILENAME - ANIS2.DAT
$
ID LINEAR,ANIS2
SOL 101
TIME 5
CEND
TITLE = ANISOTOPIC TEST ELEMENT
SPC = 1
SET 1 =13
STRAIN = 1
SPCFORCE=ALL
SUBCASE 1
LOAD = 1
SUBCASE 2
LOAD = 2
SUBCASE 3
LOAD = 3
SUBCASE 4
LOAD = 4
SUBCASE 5
LOAD = 5
SUBCASE 6
LOAD = 6

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 7-21

© 2024 Siemens
 7. Material Properties

BEGIN BULK$
GRID 1 0.0 0.0 0.0
GRID 2 1. 0.0 0.0
GRID 3 0.0 1. 0.0
GRID 4 1. 1. 0.0
GRID 6 0.0 0.0 1.
GRID 8 1. 0.0 1.
GRID 10 0.0 1. 1.
GRID 12 1. 1. 1.
CHEXA 13 1 1 6 8 2 3 10 +
+ 12 4
$
PLOAD4 1 13 1. 2 12 +
+ 1.
PLOAD4 1 13 1. 10 1 +
+ -1.
PLOAD4 2 13 1. 10 4 +
+ 0.0 1.
PLOAD4 2 13 1. 6 2 +
+ 0.0 -1.
PLOAD4 3 13 1. 6 12 +
+ 0.0 0.0 1.
PLOAD4 3 13 1. 1 4 +
+ 0.0 0.0 -1.
PLOAD4 4 13 1. 10 4 +
+ 1. 0.0
PLOAD4 4 13 1. 6 2 +
+ -1. 0.0
PLOAD4 4 13 1. 12 2 +
+ 1.
PLOAD4 4 13 1. 1 10 +
+ -1.
PLOAD4 5 13 1. 10 4 +
+ 1.
PLOAD4 5 13 1. 6 2 +
+ -1.
PLOAD4 5 13 1. 6 12 +
+ 1.
PLOAD4 5 13 1. 1 4 +
+ -1.
PLOAD4 6 13 1. 6 12 +
+ 1.
PLOAD4 6 13 1. 1 4 +
+ -1.
PLOAD4 6 13 1. 2 12 +
+ 1.
PLOAD4 6 13 1. 1 10 +
+ -1.
$

7-22 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Step 2 – Verify the Material Matrix

PSOLID 1 2 0
$
MAT9 2 10.67+6 6.91+6 6.89+6 1.25+6 -0.09+6 0.86+6 19.03+6
6.90+6 1.27+6 -0.12+6 0.83+6 14.28+6 1.23+6 -0.15+6 0.82+6
5.27+6 0.21+6 -0.11+6 4.63+6 0.52+6 5.72+6
SPC1 1 123 6
SPC1 1 12 1
SPC1 1 2 2
ENDDATA

Listing 7-2. Anisotropic Verification Input File

In the verification model input file, the six loads are applied separately, each with their own subcase.
The constraints are applied to three separate grid points and are non-redundant. The non-redundant
constraint set allows the element to expand in all directions without imposing any constraint forces
while preventing rigid body motion. Also note that the basic coordinate system is used for the material
definition.

The resulting center strains for the verification model are shown in Figure 7-6. As can be seen, there is
good agreement between these values and those in the table Strains Due to Unit Loads on Pseudo
Anisotropic Model.

Figure 7-6. Verification Model Strain Output

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 7-23

© 2024 Siemens
 7. Material Properties

Step 3 - Use the Anisotropic Material in a Structural Analysis

The last step in the example is to use the material properties you found to model the structure depicted
in Figure 7-7. The model consists of four elements with their material coordinate system oriented 45
degrees with respect to the basic coordinate system.

Figure 7-7. Anisotropic Solids Model

The input file is shown in Listing 7-3. Note the local coordinate system used to define the material
orientation. The displacement results are shown in Figure 7-8.

$ FILENAME - ANIS3.DAT
$
ID LINEAR, ANIS3
SOL 101
TIME 5
CEND
TITLE = USING THE MATERIAL MATRIX
LOAD = 1
SPC = 1
DISP = ALL
BEGIN BULK
PARAM POST 0
PARAM AUTOSPC YES
$ MATERIAL COORDINATE SYSTEM
$
CORD2R 1 0 0.0 0.0 0.0 .70710680.0 .7071068+
.70710680.0 -.707107

7-24 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Step 3 - Use the Anisotropic Material in a Structural Analysis

GRID 51 0.0 5. 0.0

GRID 52 5. 5. 0.0
GRID 53 0.0 5. 5.
GRID 54 5. 5. 5.
GRID 55 0.0 0.0 0.0
GRID 56 5. 0.0 0.0
GRID 57 0.0 0.0 5.
GRID 58 5. 0.0 5.
GRID 59 0.0 5. 10.
GRID 60 5. 5. 10.
GRID 61 0.0 0.0 10.
GRID 62 5. 0.0 10.
GRID 63 0.0 5. 15.
GRID 64 5. 5. 15.
GRID 65 0.0 0.0 15.
GRID 66 5. 0.0 15.
GRID 67 10. 5. 10.
GRID 68 10. 5. 15.
GRID 69 10. 0.0 10.
GRID 70 10. 0.0 15.
$
CHEXA 1 1 51 55 56 52 53 57 +
+ 58 54
CHEXA 2 1 53 57 58 54 59 61 +
+ 62 60
CHEXA 3 1 59 61 62 60 63 65 +
+ 66 64
CHEXA 4 1 60 62 69 67 64 66 +
+ 70 68
$
$
FORCE 1 69 0 100. 1. 0.0 0.0
FORCE 1 67 0 100. 1. 0.0 0.0
FORCE 1 68 0 100. 1. 0.0 0.0
FORCE 1 70 0 100. 1. 0.0 0.0
$
PSOLID 1 2 1 GRID SMECH
$
MAT9 2 10.67+6 6.91+6 6.89+6 1.25+6 -0.09+6 0.86+6 19.03+6
6.90+6 1.27+6 -0.12+6 0.83+6 14.28+6 1.23+6 -0.15+6 0.82+6
5.27+6 0.21+6 -0.11+6 4.63+6 0.52+6 5.72+6
$
SPC 1 51 123456 0.0
SPC 1 52 123456 0.0
SPC 1 56 123456 0.0
SPC 1 55 123456 0.0

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 7-25

© 2024 Siemens
 7. Material Properties

$
ENDDATA

Listing 7-3. Structure with Anisotropic Material Input File

Figure 7-8. Output for Structure with Anisotropic Material

Understanding MAT2

The MAT2 bulk entry is used to define anisotropic material behavior for 2D (plate and shell) elements.
When using a MAT2 bulk entry, you define the in-plane material properties with respect to an element
material coordinate system. The constitutive relation is given by:

Equation 7-13.

where:

σ1 σ 2 Normal stresses
τ12 Shear stress
ε1 ε2 Normal strains

7-26 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Material Properties for Nonlinear Analysis

γ12 Engineering shear strain

Cij Elements of the 3 × 3 symmetric material property matrix in the material
coordinate system
A1 A2 A12 Coefficients of thermal expansion
(T − Tref) Temperature difference used to calculate thermal strain

You can also include material properties for transverse shear. When a MAT2 bulk entry is used for
transverse shear, the transverse shear constitutive relation is:

Equation 7-14.

C33 must be set to 0.

If the MAT2 entry is referenced by the MID3 field (transverse shear) on the PSHELL, then C13, C23, and
C33 must be blank.

Material Properties for Nonlinear Analysis

In Simcenter Nastran, there are several different types of nonlinear analysis:

• With a nonlinear elastic analysis, you can use the MAT1 bulk entry in conjunction with the MATS1 and
TABLES1 entries to define the stress-dependent material properties.

• With an elastic-plastic analysis, you can use the MAT1 bulk entry in conjunction with the MATS1 and
TABLES1 entries to define the stress-dependent material properties.

• With a viscoelastic analysis, you can define elastoplastic properties to consider plastic deformation
using the MATS1 bulk entry in conjunction with MAT1.

• With a temperature dependent analysis, you can specify temperature-dependent properties by using
the MATTi, TABLEST, and TABLES1 bulk entries.

• With a SOL 402 analysis, you can use the MATG entry to define gasket materials.

• With a hyperelastic material analysis, you can use the MATHP entry (described below) to define the
material properties.

With a geometric nonlinear analysis, you can use either MAT2 or MAT9 to define material properties.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 7-27

© 2024 Siemens
 7. Material Properties

See also

• “Material and Geometric Nonlinearity” in the Simcenter Nastran Basic Nonlinear Analysis User's Guide

Understanding MATHP

In Simcenter Nastran, you can use the MATHP bulk entry to specify material properties for use in a fully
nonlinear hyperelastic analysis.

The generalized Mooney-Rivlin strain energy function may be expressed as follows:

Equation 7-15.

and 2D1 = K and 2(A10 + A01) = G at small strains where:

First distortional strain invariant

Second distortional strain invariant

Aij Material constants related to shear deformation (distortion)

Di Material constants related to volumetric deformation
T Current temperature
T0 Initial temperature
K Bulk modulus
G Shear modulus

The model reduces to a Mooney-Rivlin material if Na = 1 and to a Neo-Hookean material if Na = 1 and

A01 = 0.

Using MATHP to Define Material Constants

The MATHP entry consists of five continuation entries that let you define the Aij and Di material constants
of the strain energy function. You obtain the values to enter in the Aij and Di fields from least squares
fitting of experimental data.

7-28 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Using MATHP to Model Mooney-Rivlin or Neo-Hookean Materials

If you do not know the material constants, enter one of the following strings in the sixth continuation
entry of MATHP to invoke Simcenter Nastran's curve-fitting algorithm:

TAB1, TAB2, TAB3, TAB4, and TABD

These TABi fields reference experimental data that is contained in TABLES1 bulk entries.

• You can use one or more of four experiments (TAB1 to TAB4 options) to obtain Aij.

• You can use pure volumetric compression data (TABD option) to obtain Di.

If ND = 10 (order of the volumetric strain energy polynomial function) and you either enter a nonzero
value for D1 or a nonzero value for D1 is calculated from experimental data in TABD, then the parameter
estimation of the material constants Aij takes compressibility into account in the cases of simple tension/
compression, equibiaxial tension, and general biaxial deformation. Otherwise, the software assumes full
incompressibility in estimating the material constants.

See also

• “Defining Hyperelastic Materials in Simcenter Nastran” in the Simcenter Nastran Basic Nonlinear
Analysis User's Guide

Using MATHP to Model Mooney-Rivlin or Neo-Hookean Materials

If you are modeling a Mooney-Rivlin or Neo-Hookean material, you do not need to specify a continuation
entry for MATHP. Conventional Mooney-Rivlin and Neo-Hookean materials are fully incompressible.
With the MATHP entry, you cannot define full incompressibility. However, you can simulate full
incompressibility by specifying a large enough value for the D1 entry (field 5). However, entering a
value of D1 greater than 103 · (A10 + A01) is not recommended.

See also

• “MATHP” in the Simcenter Nastran Quick Reference Guide

Material Properties for Heat Transfer

Material properties for heat transfer analysis are defined using MAT4, MAT5, MATT4, MATT5, RADM and
RADMT entries.

• The MAT4, MAT5, and RADM entries provide temperature independent (constant valued) material
properties.

• The MATT4, MATT5, and RADMT entries supply information about temperature-dependent properties
through reference to material tables (TABLEMi (i = 1, 2, 3, 4)).

With heat transfer material properties:

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 7-29

© 2024 Siemens
 7. Material Properties

• Although the surface characteristics for free convection (heat transfer coefficient) and radiation
(emissivity and absorptivity) are not normally considered to be material properties, they are included
in the bulk entries described here.

• Temperature dependence of the quantity of interest is ultimately defined on a TABLEMi entry. For
example, the table is connected to the MAT4 through the MATT4 entry. Most material properties are
directed toward a structural element and are referenced by the property entry for the element of
interest.

Understanding MAT4 and MATT4

The MAT4 bulk entry lets you define thermal properties for temperature-independent isotropic
materials .The format of the MAT4 bulk entry is:

1 2 3 4 5 6 7 8 9 10
MAT4 MID K CP ρ H μ HGEN REFENTH
TCH TDELTA QLAT

where:

Symbol Property Consistent Units

K(T) Thermal Conductivity* W/m °C
CP(T) Specific Heat* J/kg °C
ρ Density* kg/m3
H(T) Free Convection Heat Transfer Coefficient W/m2 °C
μ (T) Dynamic Viscosity* N sec/m2
HGEN(T) Volumetric Internal Heat Generation W/m3
REFENTH Reference Enthalpy J/kg
TCH Lower Temperature Limit for Phase Change °C
TDELTA Temperature Range for Phase Change °C
QLAT Latent Heat J/kg
* Thermal conductivity, specific heat, density, and dynamic viscosity cannot be time dependent.
Thermal heat flux loads can reference explicit functions of time. Thermal boundary conditions for
convection and radiation utilize control node techniques for specifying time dependent behavior.

The MATT4 entry lets you define temperature-dependent material properties in conjunction with
MAT4. Simcenter Nastran always multiplies the basic quantities you specify on the MAT4 entry by the
corresponding tabular function (TABLEMi bulk entries) that you reference with the MATT4 entry.

7-30 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Understanding MAT5 and MATT5

See also

• “Convection and Radiation Boundary Conditions” in the Simcenter Nastran Thermal Analysis User's
Guide.

• “MAT4” in the Simcenter Nastran Quick Reference Guide

• “MATT4” in the Simcenter Nastran Quick Reference Guide

Understanding MAT5 and MATT5

The MAT5 bulk entry lets you define thermal properties for temperature-independent anisotropic
materials. The MATT5 entry lets you define temperature-dependent material properties in conjunction
with MAT5. Simcenter Nastran always multiplies the basic quantities you specify on the MAT5 entry by
the corresponding tabular function (TABLEMi bulk entries) that you reference with the MATT5 entry.

See also

• “MAT5” in the Simcenter Nastran Quick Reference Guide

• “MATT5” in the Simcenter Nastran Quick Reference Guide

Understanding RADM and RADMT

The RADM bulk entry lets you define the radiation properties of a boundary element for heat transfer
analysis. The format of the RADM bulk entry is:

1 2 3 4 5 6 7 8 9 10
RADM RADMID ABSORP EMIS1 EMIS2 EMIS3 EMIS4 EMIS5 EMIS6
EMIS7 -etc.-

The following table describes the fields for the RADM bulk entry.

Field Description Unit

ABSORP(T) Absorptivity for Directional Heat Flux and Radiation Nondimensional
Boundary Condition (QVECT and RADBC)
EMIS(λ,T) Emissivity for Radiation Boundary Condition and Nondimensional
Enclosure Radiation (RADBC and RADSET)

The RADMT entry lets you specify table references for properties in a RADM entry that you want to be
temperature-dependent.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 7-31

© 2024 Siemens
 7. Material Properties

See also

• “RADM” in the Simcenter Nastran Quick Reference Guide

• “RADMT” in the Simcenter Nastran Quick Reference Guide

Material Properties for Fluid Analysis

In Simcenter Nastran, the MAT10 and MFLUID bulk entries allow you to define material properties for
different types of fluid analyses.

Understanding MAT10

The MAT10 bulk entry allows you to define material properties for compressible fluids used in analyses
that account for coupled fluid-structural interaction.

See also

• MAT10 in the Simcenter Nastran Simcenter Nastran Quick Reference Guide

• Defining porous or equivalent porous materials in the Simcenter Nastran Acoustics Simcenter Nastran
User's Guide

Understanding MFLUID

The MFLUID bulk entry allows you to define properties of an incompressible fluid used to generate a
virtual mass matrix.

See also

• “MFLUID” in the Simcenter Nastran Quick Reference Guide

• “Virtual Fluid Mass” in the Simcenter Nastran Advanced Dynamic Analysis User's Guide

7-32 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 8. Applied Loads
Overview of Load Sets
Load sets in Simcenter Nastran are collections, or lists, of loads. You define the loads using bulk data
entries. You then use case control commands to specify which of the loading entries you want to use in
a given analysis.

Defining and Combining Subcases for Loading

Simcenter Nastran allows you to efficiently analyze multiple combinations of loads and constraints in a
single analysis using the SUBCASE case control command. Subcases also allow you to request output
on a subcase-specific basis. For example, you can obtain stress output from one subcase, but not from
another.

For more information about subcases, see Defining Subcases.

Example: Creating Two Load Subcases for the Truss

For example, suppose we add a second load case to the truss structure as shown in Figure 8-1 so that
we can perform two separate analyses during a single run.

• Subcase 1 consists of the 1000-pound force acting at grid point 4 in the negative Y-direction.

• Subcase 2 consists of the 2000-pound force acting at grid point 3 in the X-direction.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 8-1

© 2024 Siemens
 8. Applied Loads

Figure 8-1. Truss Example with Two Loading Conditions

In the input file for this analysis (shown in Listing 8-1). In this input file, we placed a displacement
request (DISPLACEMENT - ALL) above the first subcase but not within either of the two subcases.
Therefore, Simcenter Nastran outputs displacements for all the grid points for both subcases. Since we
included an element force request only in Subcase 2, Simcenter Nastran doesn't generate any element
force output for Subcase 1. It only outputs the element forces for the grid points for Subcase 2.

8-2 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Example: Creating Two Load Subcases for the Truss

$ Filename - TRUSS2.DAT
ID LINEAR,TRUSS2
SOL 101
TIME 2
CEND
TITLE = LINEAR STATICS USER'S GUIDE - TWO SUBCASES
SUBTITLE = TRUSS STRUCTURE
SPC = 11
DISPLACEMENT = ALL
$
SUBCASE 1
LABEL = POINT LOAD AT GRID POINT 4
LOAD = 10
$
SUBCASE 2
LABEL = POINT LOAD AT GRID POINT 3
LOAD = 11
ELFORCE = ALL
$
BEGIN BULK
$
$ THE GRID POINTS LOCATIONS
$ DESCRIBE THE GEOMETRY
$
GRID 1 0. 0. 0. 3456
GRID 2 0. 120. 0. 3456
GRID 3 600. 120. 0. 3456
GRID 4 600. 0. 0. 3456
$
$ MEMBERS ARE MODELED USING
$ ROD ELEMENTS
$
CROD 1 21 2 3
CROD 2 21 2 4
CROD 3 21 1 3
CROD 4 21 1 4
CROD 5 21 3 4
$
$ PROPERTIES OF ROD ELEMENTS
$
PROD 21 22 4. 1.27
$
$ MATERIAL PROPERTIES
$
MAT1 22 30.E6 .3
$

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 8-3

© 2024 Siemens
 8. Applied Loads

SPC1 11 123456 1 2
$
$ POINT LOAD SUBCASE 1
FORCE 10 4 1000. 0. -1. 0.
$
$ POINT LOAD SUBCASE 2
FORCE 11 3 2000. 1. 0. 0.
ENDDATA
Listing 8-1. Truss Input File with Two Loadings

Using REPCASE

For large models, you may need to further partition the results in the output files in a certain manner
or repeat the same output in different subcases. You can do this withe the REPCASE Case Control
command. The REPCASE command lets you organize the output based on the element and/or grid point
ID numbers.

Listing 8-2 shows how you can use REPCASE separate the results for the truss example into three
different sections in the .f06 output file. In this example, the REPCASE 2 command provides for
additional output from the solution generated in SUBCASE 1 without performing any additional analysis.

For the Case Control shown in Listing 8-2, SUBCASE 1 output consists of the displacement for all of
the grid points and the element forces for elements 1 and 2. The output for REPCASE 2 consists of
the element force for elements 3 through 5. Note that the default output request for the grid point
displacements is overridden by the DISPLACEMENT = NONE command. SUBCASE 3 is a new loading
condition with its own output requests. A second static solution is performed for this subcase. Also note
that only two static solutions are performed for this input file.

$ FILENAME TRUSS4.DAT
$
ID LINEAR,TRUSS4
SOL 101
TIME 2
CEND
TITLE = LINEAR STATICS USER'S GUIDE - REPCASE EXAMPLE
SUBTITLE = TRUSS STRUCTURE
LOAD = 10
SPC = 11
DISPLACEMENT = ALL
SET 1 = 3,4
SET 2 = 1,2
SET 3 = 3 THRU 5
$
SUBCASE 1
LABEL = POINT LOAD AT GRID POINT 4
LOAD = 10

8-4 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Using SUBCOM to Combine Subcases

ELFORCE = 2
$
REPCASE 2
LABEL = POINT LOAD AT GRID POINT 4
ELFORCE = 3
DISPLACEMENT = NONE
$
SUBCASE 3
LABEL = POINT LOAD AT GRID POINT 3
LOAD = 11
DISPLACEMENT =1
ELFORCE = 1
BEGIN BULK

Listing 8-2. Truss Input File with the REPCASE Command

See also

• “REPCASE” in the Simcenter Nastran Quick Reference Guide

Using SUBCOM to Combine Subcases

When you are performing a linear statics analysis, you can use the SUBCOM command to create new
subcases by combining other subcases. When you use SUBCOM to combine subcases, you must use the
SUBSEQ command to specify the coefficients for combining the previous subcases. A SUBSEQ command
is required for each SUBCOM subcase.

The use of the SUBCASE and SUBSEQ commands is best illustrated with an example. Returning to the
truss structure, the two load cases were previously treated as separate loads, each with their own
subcase. Now, suppose a third load case is desired that is a linear combination of the first two subcases.
The input file shown in Listing 8-3 illustrates this.

$ FILENAME - TRUSS5.DAT
$
ID LINEAR,TRUSS5
SOL 101
TIME 2
CEND
TITLE = LINEAR STATICS USER'S GUIDE SUBCOM EXAMPLE
SUBTITLE = TRUSS STRUCTURE
LOAD = 10
SPC = 11
DISPLACEMENT = ALL
$
SUBCASE 1
LABEL = POINT LOAD AT GRID POINT 4

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 8-5

© 2024 Siemens
 8. Applied Loads

LOAD = 10
$
SUBCASE 2
LABEL = POINT LOAD AT GRID POINT 3
LOAD = 11
ELFORCE = ALL
$
SUBCOM 3
SUBSEQ 1.0,0.5
ELFORCE = ALL
$
SUBCOM 4
SUBSEQ 1.0,-0.5
ELFORCE =ALL
$
BEGIN BULK

Listing 8-3. SUBCOM Input File

For this Case Control section, the SUBCOM 3 output is as follows:

The output for SUBCOM 4 is as follows:

The coefficients 1.0, and 0.5 for SUBCOM 3 are input on a SUBSEQ command. These coefficients used
in the SUBSEQ command refer to the previous Subcases 1 and 2. The 1.0 value in SUBCOM 3 is the
coefficient corresponding to SUBCASE 1 and the 0.5 is the coefficient associated with SUBCOM 2.
SUBCOM 3 is not considered a subcase. Likewise, the coefficients 1.0 and –0.5 for SUBCOM 4 refer to
SUBCASES 1 and 2. If you specify a third SUBSEQ coefficient for either SUBCOM, the job will fail.

You can use SUBCOM commands to combine any number of previous subcases. However, you must
define a coefficient on the SUBSEQ command for each subcase. If you want to exclude a subcase from
the summation, set the coefficient to 0.0 for that particular subcase.

Note: You can define the combined loads in their own subcases and not use SUBCOMs. If you only want
to generate the output of the combined loads (so you only need two subcases), this is the preferred way.
However, using SUBCOMs is helpful if you want to review the structural response to the individual loads
in addition to the combined load. The SUBCOM command sums up the results of the previous subcases

8-6 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Using SYMCOM to Create a Symmetry Combination Subcase

multiplied by the appropriate scale factors. One of the best applications of the SUBCOM command is
combining the results generated with different constraints.

See also

• “SUBCOM” in the Simcenter Nastran Quick Reference Guide

Using SYMCOM to Create a Symmetry Combination Subcase

In a static analysis, you can use the SYMCOM case control command to create a symmetry combination
subcase. When you use the SYMCOM command to combine selected SUBCASES, you must also use a
SYMSEQ command to specify the coefficients for combining the symmetry subcases.

See also

• “SYMCOM” in the Simcenter Nastran Quick Reference Guide

• “SYMSEQ” in the Simcenter Nastran Quick Reference Guide

Temperature Loads and Enforced Deformations in Subcase Combinations (SUBCOMS)

You can use the SUBCOM and SYMCOM case control commands to combine the results of two or
more immediately preceding subcases or SYM cases respectively. In Simcenter Nastran, the results
for a combined loading condition are efficiently obtained by forming the linear combination of the
displacement vectors obtained in the previous subcases. Simcenter Nastran calculates the element
forces and stresses in the usual manner from the displacement vector for the combined loading
condition. This technique is quite straightforward when the several loading conditions are composed
of forces and moments.

However, if any of the subcases that you're combining contain thermal loads and/or enforced element
deformations, you must exercise some care to obtain correct element stresses and forces for the
combined loading condition.

See also

• “Using Temperature Loads with Subcases” in the Simcenter Nastran User's Guide

Defining Static Loads

In Simcenter Nastran, you can apply static loads to geometric and scalar points in a variety of ways,
including:

• Loads applied directly to grid points

• Pressure loads on surfaces

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 8-7

© 2024 Siemens
 8. Applied Loads

• Distributed and concentrated loads on CBAR, CBEAM, and CBEND elements

• Gravity loads

• Centrifugal loads due to steady rotation

• Tangential loads due to angular acceleration

• Loads resulting from thermal expansion

• Loads resulting from enforced deformations of a structural element

• Loads resulting from enforced displacements at a grid point

While you can define multiple load sets in the bulk data section of your input file, the software only uses
the sets that you explicitly select in the case control section to solve the model.

See also

• “Defining SUBCASES for Loading” in the Simcenter Nastran User's Guide

• Section 3.6 of The NASTRAN Theoretical Manual

Point Loads
This section shows how to apply concentrated forces and concentrated moments at grid points.

• You can use the FORCE, FORCE1, and FORCE2 entries to apply concentrated forces directly to grid
points.

• You can use the MOMENT, MOMENT1, and MOMENT2 entries to apply a apply concentrated force in
the translational degrees of freedom (T1, T2, and T3) at a grid point.

• Use SLOAD to define a concentrated static load at a scalar point.

FORCE, FORCE1, FORCE2

You can apply concentrated forces directly to grid points with the FORCE, FORCE1, and FORCE2 entries.

• Use a FORCE entry if you want to define a static, concentrated force at a grid point by specifying a
vector.

• Use a FORCE1 entry if the direction is determined by a vector connecting two grid points.

• Use a FORCE2 entry if the direction is specified by the cross product of two such vectors.

8-8 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Using FORCE

Using FORCE

The FORCE entry lets you specify the magnitude and direction of a force vector in any coordinate system
as shown below. You can use the FORCE entry to define a static load applied to a geometric grid point
in terms of components defined by a local coordinate system. The orientation of the load components
depends on the type of local coordinate system you used define the load. The directions of the load
components are the same as those indicated in Defining the Directions of Displacement (Fields 4-6)
for displacement components.

Consider a force F acting on a cantilever beam:

A concentrated force F is applied to a CBAR element connecting grid points 1 and 2 using the FORCE
entry. Its format is as follows:

1 2 3 4 5 6 7 8 9 10
FORCE SID G CID F N1 N2 N3

Field Contents

SID Load set identification number. (Integer > 0)

G Grid point identification number. (Integer > 0)

CID Coordinate system identification number. (Integer ≥ 0; Default = 0)

F Scale factor. (Real)

Ni Components of a vector measured in coordinate system defined by CID. (Real;

at least one Ni ≠ 0.0)

In our example, the FORCE entry may be written as:

FORCE 2 2 10. 0. -1. 0.

or, in free field format,

FORCE,2,2,,10.,0.,-1.,0.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 8-9

© 2024 Siemens
 8. Applied Loads

The load set identification number (SID in column 2) refers to a command defined in the Case Control
Section (LOAD=2 in this example; the integer value 2 is arbitrarily chosen). Leaving column 4 blank
means that the basic (default) coordinate system is used to specify the orientation of the load. The (0.,
-1., 0.) entries in columns 6, 7, and 8 refer to a vector in the -Y direction, defining the direction of
application of the load. The force applied to the grid point is , given by

Equation 8-1.

where = (N1,N2,N3)

Thus, the value of F in column 5 is the full value of the applied load of 10 lbf because vector (in this
example) is of unit length.

The magnitude of the applied force is the scale factor (field 5) times the magnitude of the vector defined
in fields 6 through 8. For example, the force applied with the following two FORCE entries is the same.

1 2 3 4 5 6 7 8 9 10
FORCE 1 3 0 -100. 1. 0. 2.

1 2 3 4 5 6 7 8 9 10
FORCE 1 3 0 2. -50. 0. -100.

See also

• “FORCE” in the Simcenter Nastran Quick Reference Guide

Using FORCE1 and FORCE2

• The FORCE1 entry lets you define a force by specifying a magnitude and two grid points (not
necessarily the loaded grid point) to determine the direction.

• The FORCE2 entry lets you define a force by specifying a magnitude with the direction defined by the
vector product of two other vectors.

The format of the FORCE1 and FORCE2 entries is as follows:

1 2 3 4 5 6 7 8 9 10
FORCE1 SID G F G1 G2

8-10 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 FORCE1 and FORCE2 Example

1 2 3 4 5 6 7 8 9 10
FORCE2 SID G F G1 G2 G3 G4

Field Contents

SID Load set identification number.

G Grid point identification number.

F Magnitude of the force.

Gi Grid point identification numbers.

FORCE1 and FORCE2 Example

While the FORCE1 and FORCE2 forms are not used extensively, they are useful for some modeling
situations as shown in the following example.

Suppose you wish to model a curved structure with edge loads as shown in Figure 8-2. The edge loads
consist of 10 N per element (acting normal to the elements) and 8 N per element along the edge of
each element (acting as a shear load). The FORCE1 and FORCE2 entries are the ideal choice to apply
these loads. You could use the FORCE entry, but it would be more difficult because you would need to
compute the normals for each of the loaded CQUAD4 elements. Using FORCE1 and FORCE2 entries for
the shear and normal loads, respectively, makes the task much easier.

The 10 N normal force for each element is divided equally between the two grid points along the edge.
The directions of the resulting 5 N forces are applied using the FORCE2 entry by taking the cross product
of the vectors defined by the edges to which the grid points are attached. The result is a 5 N load acting
normal to the plate. The 8 N shear force for each element is also divided up equally between the two
grid points along the edge. The resulting 4 N loads are applied with the FORCE1 entry. The directions of
these loads are determined by the line connecting the edge grid points.

The input file is shown in Listing 8-4. Note that the normal forces and the shear forces are applied as
one load condition. If you want the forces to act as separate loadings, you can use different IDs for the
FORCE1 and FORCE2 entries and call them out using two separate subcases in your input file.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 8-11

© 2024 Siemens
 8. Applied Loads

Figure 8-2. Edge Forces on a Plate Element

$ FILENAME - EDGE.DAT
$
OLOAD=ALL
LOAD=3
.
.
.
BEGIN BULK
$
$ UNITS ARE CM AND N
$
FORCE1 3 3 4. 3 6
FORCE1 3 6 4. 3 6
FORCE1 3 6 4. 6 9
FORCE1 3 9 4. 6 9
FORCE1 3 9 4. 9 12
FORCE1 3 12 4. 9 12
FORCE2 3 3 5. 3 6 3 2
FORCE2 3 6 5. 3 6 3 2
FORCE2 3 6 5. 6 9 6 5
FORCE2 3 9 5. 6 9 6 5
FORCE2 3 9 5. 9 12 9 8
FORCE2 3 12 5. 9 12 9 8
$
GRID 1 0.0 -60. -2.
GRID 2 60. -60. -2.
GRID 3 120. -60. -2.

8-12 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 FORCE1 and FORCE2 Example

GRID 4 0.0 -30. 0.0

GRID 5 60. -30. 0.0
GRID 6 120. -30. 0.0
GRID 7 0.0 30. 0.0
GRID 8 60. 30. 0.0
GRID 9 120. 30. 0.0
GRID 10 0.0 60. -2.
GRID 11 60. 60. -2.
GRID 12 120. 60. -2.
$
PSHELL 1 4 .5 4
MAT1 4 7.1E6 .3
$
CQUAD4 1 1 1 2 5 4
CQUAD4 2 1 2 5 6 3
CQUAD4 3 1 4 5 8 7
CQUAD4 4 1 5 6 9 8
CQUAD4 5 1 7 8 11 10
CQUAD4 6 1 8 9 12 11
$
SPC1 1 123456 1 4 7 10
$
ENDDATA

Listing 8-4. Edge Loads

Note that the two loads are applied at the same grid points. This is perfectly acceptable. The total force
acting at the grid points is the sum of all the applied loads. The OLOAD (applied load output request)
output, which you request through the Case Control section, shows the applied force resultant acting
at a set of user specified grid points. The OLOAD output is useful in verifying that the loads are applied
correctly. A partial listing of the output file showing the OLOAD output is shown in Figure 8-3.

Figure 8-3. OLOAD Output

See also

• “FORCE1” in the Simcenter Nastran Quick Reference Guide

• “FORCE2” in the Simcenter Nastran Quick Reference Guide

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 8-13

© 2024 Siemens
 8. Applied Loads

Using MOMENT, MOMENT1, MOMENT2

MOMENT, MOMENT1, and MOMENT2 are similar to the three FORCE entries. The only difference is that
you use force entries to apply concentrated force in the translational degrees of freedom (T1, T2, and
T3), while you use moment entries apply concentrated moments in the rotational degrees of freedom
(R1, R2, and R3).

• With MOMENT, you define a static, concentrated moment at a grid point by specifying a scale factor
and a vector that determines the direction

• With MOMENT1, the software uses two grid points to determine direction ( , where is a unit
vector parallel to the vector from grid 1 to grid 2).

• With MOMENT2, the software uses four grid points to determine direction ( , where is the
unit vector parallel to the cross product of the vectors from G1 to G2, and G3 to G4).

The application of concentrated moments is analogous to forces. MOMENT has the following format:

1 2 3 4 5 6 7 8 9 10
MOMENT SID G CID M N1 N2 N3

Field Contents

SID Load set identification number. (Integer > 0)

G Grid point identification number at which the moment is applied. (Integer > 0)

CID Coordinate system identification number. (Integer ≥ 0 or blank)

M Scale factor. (Real)

Ni Components of the vector measured in the coordinate system defined by CID.

(Real; at least one Ni ≠ 0.0)

Consider moment M acting about the basic Z axis of the simply supported beam shown below:

In this case, the MOMENT entry could be written as

8-14 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Using SLOAD

1 2 3 4 5 6 7 8 9 10
MOMENT 6 2 -18.6 0.0 0.0 1.

or, in free field format,

MOMENT,6,2,,-18.6,0.,0.,1.

The applied moment is given by

Equation 8-2.

where is the vector (N1, N2, N3). As was the case with FORCE entry, M is the full magnitude of the
moment since is a vector of unit length. The direction of the applied moment is given by the sign of
M according to the right-hand rule because = (0., 0., 1.) is a vector in the direction of the positive Z
axis direction. Note that specifying M = 18.6 and = (0., 0., −1.) would produce an equivalent result.

See also

• “MOMENT” in the Simcenter Nastran Quick Reference Guide

• “MOMENT1” in the Simcenter Nastran Quick Reference Guide

• “MOMENT2” in the Simcenter Nastran Quick Reference Guide

Using SLOAD

You can use the SLOAD entry to define a concentrated static load at a scalar point or a grid point. With
SLOAD, you can only specify the magnitude since only one component of motion exists at a scalar point.

See also

• “SLOAD” in the Simcenter Nastran Quick Reference Guide

Distributed Loads
There are three Bulk Data entries available to apply distributed loads to element surfaces.

• Use PLOAD define a uniform static pressure load on a triangular or quadrilateral surface.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 8-15

© 2024 Siemens
 8. Applied Loads

• Use PLOAD1 and PLOAD2 to apply a distributed load to the CQUAD4, CTRIA3, and CSHEAR elements
only.

• Use PLOAD4 to apply distributed loads on any of the two-dimensional elements and on the surfaces of
three-dimensional elements.

PLOAD

You can use the PLOAD entry to define a uniform normal static pressure load on triangular or
quadrilateral surfaces by using grid points. You can apply PLOAD to 2-D (surface) or 3-D (solid) elements.

The PLOAD entry is different from the other PLOADi entries because it references three or four grid
points rather than the element itself. When you use PLOAD to load a CTRIA3 element, the total force
acting on the element is divided evenly among the three corner points. When you use PLOAD to load
a CQUAD4 element, the total force acting on the element is distributed to corner grid points using the
geometric shape of the element. The resultant of the applied corner loads acts through the centroid
of the element. The direction of the pressure load is determined using the right-hand rule by the
numbering sequence of the grid points on the PLOAD entry.

The PLOAD entry has the following form:

1 2 3 4 5 6 7 8 9 10
PLOAD SID P G1 G2 G3 G4

Field Contents

SID Load set identification number. (Integer > 0)

P Pressure. (Real)

Gi Grid point identification numbers. (Integer > 0; G4 may be zero or blank)

Grid Points G1, G2, G3, and G4 define either a triangular or quadrilateral surface; if G4 is zero or blank,
the software assumes that the surface is triangular. The direction of the pressure load is determined by
applying the right-hand rule to the grid point ordering sequence of the surface.

8-16 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 PLOAD1

Pressure is applied in the opposite direction by making the value of P negative.

See also

• “PLOAD” in the Simcenter Nastran Quick Reference Guide

PLOAD1

You can use the PLOAD1 entry to apply a distributed load to a CBAR, CBEAM, or CBEND element. You can
use the PLOAD1 entry for both concentrated and linearly distributed forces.

• For the CBAR and CBEAM elements, you can apply the linearly distributed force between any two
locations on the element (or off the element if you wish.)

• For the CBEND element, you can only apply the linearly distributed loads to linearly varying forces and
moments between the end points.

You can use PLOAD1 entry to define either distributed or concentrated loads at intermediate points on
CBAR and CBEAM elements. You define applied loads at the end points of CBAR and CBEAM elements
with either FORCE or MOMENT entries. You can also use PLOAD1 to define only distributed loads for the
CBEND element. For the CBEND element, you can only use PLOAD1 to define

You can use the CBARAO entry is used to define intermediate points on CBAR elements where you want
to obtain stress and/or force output. You can also use the PLOAD1 to define only distributed loads for
the CBEND element. The distributed load is always applied along the entire length of the element and
does not define loads at intermediate points on the CBEND. The PLOAD1 entry cannot be used to define
concentrated loads at intermediate stations on the CBEND element.

You can apply the load can be applied along the entire element length, a segment of the length, or at a
point along the length. The form of the PLOAD1 entry is shown below. The meanings of X1, X2, P1, and
P2 are shown in Figure 8-4.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 8-17

© 2024 Siemens
 8. Applied Loads

Format

1 2 3 4 5 6 7 8 9 10
PLOAD SID EID TYPE SCALE X1 P1 X2 P2

Field Contents

SID Load set identification number. (Integer > 0)

EID CBAR, CBEAM, or CBEND element identification number.(Integer > 0)

TYPE Load type. (Character: “FX”, “FY”, “FZ”, “FXE”, “FYE”, “FZE”, “MX”, “MY”, “MZ”,
“MXE”, “MYE”, “MZE”)

SCALE Determines scale factor for X1, X2. (Character: “LE”, “FR”, “LEPR”, “FRPR”)

X1, X2 Distances along the CBAR, CBEAM, or CBEND element axis from end A. (Real;
X2 may be blank, 0 ≤ X1 ≤ X2)

P1, P2 Load factors at positions X1, X2. (Real or blank)

Figure 8-4. PLOAD1 Convention in CBAR or CBEAM Elements

The TYPE option in field 4 lets you specify the type of the load. The load can be a concentrated force
or moment in the basic or element coordinate system. If the applied load is to be a concentrated load,
leave fields 8 and 9 blank and the concentrated load will be applied at the X1 location. If both X1 and X2
are input, the load will be taken as a linearly varying load between X1 and X2.

8-18 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 PLOAD1 Examples

The SCALE option in field 5 lets you specify the location of the load on the CBAR and CBEAM elements
using field 5 of the PLOAD1 entry.

• With the “LE” or “LEPR” methods, you specify the actual start and end positions of the load as
measured from end A of the element. When using this method, the distances X1 and X2, as shown in
the figure Distributed Load on CBAR Elements, are in the same units as the dimensions used for the
model.

• With the “FR” or “FRPR” methods, you specify the percent (using “X/XB”) along the element where the
load starts and ends. A value of 0.0 defines end A, while a value of 1.0 defines end B. Again, the start
of the load is measured from end A.

For both methods of describing the location of the loads, you can also specify whether the applied load
is to be a direct load (scale “LE” and “FR”) or a projected load (scale “LEPR” or “FRPR”). For the projected
loads, distributed loads are entered in terms of the projected length of the element as shown in the
figure Distributed Load on CBAR Elements. Remember that if you input the loads in terms of the basic
coordinate system (“FX”, “FY”, “FZ”, “MX”, “MY”, or “MZ”), then the projected angle a is with respect to the
basic coordinate system, not the element coordinate system.

PLOAD1 Examples

This section contains a series of examples that illustrate different uses of the PLOAD1 entry.

PLOAD1 Example 1

In this example, we used PLOAD1 to apply a uniformly distributed load over the full length of a CBAR
element using fractional (normalized) scaling.

Note that P1 = P2 = 12.6 lbf/inch

X1 = 0.0 fractional scaling
X2 = 0.0

1 2 3 4 5 6 7 8 9 10
PLOAD1 SID EID TYPE SCALE X1 P1 X2 P2
PLOAD1 36 52 FY FR 0.0 -12.6 1.0 -12.6

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 8-19

© 2024 Siemens
 8. Applied Loads

Note: You can also use PLOAD1 to apply a uniformly distributed load over the full length of a CBAR
element using length scaling.

1 2 3 4 5 6 7 8 9 10
PLOAD1 SID EID TYPE SCALE X1 P1 X2 P2
PLOAD1 36 52 FY LE 0.0 -12.6 2.0 -12.6

PLOAD1 Example 2

In this example, we used PLOAD1 to apply a linearly varying distributed load to the interior of a CBAR
element using length scaling:

1 2 3 4 5 6 7 8 9 10
PLOAD1 SID EID TYPE SCALE X1 P1 X2 P2
PLOAD1 36 52 FY LE 1.0 -3.1 2.0 -8.4

PLOAD Example 3

The example demonstrating the use of the PLOAD1 entry consists of applying a direct linearly varying
load to the three-bar structure shown in Figure 8-5.

8-20 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 PLOAD Example 3

Figure 8-5. Distributed Load on CBAR Elements

The input file for this example is shown in Listing 8-5.

$
$ FILENAME - DISTRIB.DAT
$
ID LINEAR,DISTRIB
SOL 101
TIME 2
CEND
LOAD = 1
FORCE = ALL
BEGIN BULK
$
$ UNITS ARE MM AND N
$
GRID 1 0. 0.0 0.0 123456
GRID 2 1000. 0.0 0.0 345
GRID 3 2000. 0.0 0.0 345
GRID 4 3000. 0.0 0.0 123456
$
CBAR 10 1 1 2 1.0 0.0 1.0
CBAR 11 1 2 3 1.0 0.0 1.0
CBAR 12 1 3 4 1.0 0.0 1.0
CBARAO 11 FR 19 .05 .05
$

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 8-21

© 2024 Siemens
 8. Applied Loads

PBAR 1 1 100.0 100.0 100.0 100.0

$
MAT1 1 4.E4 .3
$
PLOAD1 1 11 FY LE 300. 23. 800. 45.
$
ENDDATA

Listing 8-5. PLOAD1 Loading

Note the that the starting location for the linearly varying load starts and ends at 0.3 m (300 mm) and
0.8 m (800 mm), respectively. This is the distance measured from end A. One of the most common
mistakes is to make these distances 1.3 m and 1.8 m, respectively, while thinking that it is the actual
distance measured in the basic coordinate system. It is not; it is the distance measured from end A of the
loaded element-in this case, grid point 2.

When you use the PLOAD1 entry to define a distributed load, the CBARAO (CBAR Additional Output)
entry is helpful. You use the CBARAO entry to request stress and force output at intermediate locations
along the CBAR element. The CBARAO output for element 2 is shown in Figure 8-6.

Figure 8-6. CBARAO Output for the Center Element

8-22 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 PLOAD Example 3

The CBARAO entry is one of the best tools available for model checkout when you're using linearly
varying loads on the CBAR elements. The CBARO output lets you easily generate a shear and moment
diagram as shown in Figure 8-7 below.

Figure 8-7. Shear and Moment Diagram Generated from CBARAO Output

For CBEAM elements, you can use the PBEAM entry to directly request intermediate output along the
element.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 8-23

© 2024 Siemens
 8. Applied Loads

PLOAD1 Example 4

You can also use PLOAD1 to apply a concentrated load at an interior point of a CBAR element using
fractional scaling.

1 2 3 4 5 6 7 8 9 10
PLOAD1 SID EID TYPE SCALE X1 P1 X2 P2
PLOAD1 36 52 FY FR 0.5 -1000.

PLOAD1 Example 5

Another feature of the PLOAD1 entry is the ability to apply projected loads. To apply a load as a projected
load, you choose a scale (field 5) of “LEPR” for the actual length from end A or “FRPR” for the fractional
distance from end A. This example shows illustrates how you can use PLOAD1 to define a projected load
using a snow load on a truss structure as shown in Figure 8-8.

8-24 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 PLOAD1 Example 5

Figure 8-8. Planar Truss Structure with a Snow Loading

The snow load on the truss is 500 N/m acting in the basic Y-direction. The input file containing the grid
points and PLOAD1 entries is shown in Listing 8-6.

$
$ FILENAME - PLOAD1.DAT
$
$ UNITS ARE MM AND N
$
GRID 1 0.0 0.0 0.0
345
GRID 6 13000. 0.0 0.0
345
GRID 8 3250. 1250. 0.0
345
GRID 9 6500. 2500. 0.0
345
GRID 11 9750. 1250. 0.0 345
$
CBAR 5 1 1 8 1.0 1. 0.0

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 8-25

© 2024 Siemens
 8. Applied Loads

CBAR 6 1 8 9 1.0 1. 0.0

CBAR 7 1 9 11 1.0 1. 0.0
CBAR 8 1 11 6 1.0 1. 0.0
$
PLOAD1 1 5 FY FRPR 0.0 -.5 1. -.5
PLOAD1 1 6 FY FRPR 0.0 -.5 1. -.5
PLOAD1 1 7 FY FRPR 0.0 -.5 1. -.5
PLOAD1 1 8 FY FRPR 0.0 -.5 1. -.5

Listing 8-6. Truss with Snow Loading

To see the effect of using projected forces, the resulting SPC forces are shown in Figure 8-9. The total
reaction of the loads obtained by adding the SPC forces is 6500 N in the Y-direction. The total reaction
load is equal to the projected length of 13 meters times the distributed load of 500 N/m. If the projected
option had not been used, the applied load would have been 6964.2 N, which is the total length of the
top members times the distributed load.

Figure 8-9. SPC Forces Due to the Snow Loading

Note: With PLOAD1, if you're using a concentrated load and don't need the additional features that the
CBEAM element offers, you should use the CBAR element because of the way the equivalent end loads
are generated.

• For the CBAR element, the equivalent end loads are generated explicitly.

• For the CBEAM element, the end loads are generated by numerical integration along the length of the
element, which may not be as accurate as the method used for the CBAR element.

A linearly varying load applied with a PLOAD1 entry is accurate when used with either element.

PLOAD2

You can use the PLOAD2 entry to apply a normal uniform pressure load to CQUAD4, CSHEAR, or CTRIA3
2-D (surface) elements using element IDs. The positive direction of the loading is determined by the
order of the grid points on the element connection entry, using the right hand rule. The software
computes the magnitude and direction of the load from the value of the pressure and the coordinates of
the connected grid points. Simcenter Nastran applies the load to the connected grid points.

The PLOAD2 entry has two forms:

8-26 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 PLOAD4

1 2 3 4 5 6 7 8 9 10
PLOAD2 SID P EID1 EID2 EID3 EID4 EID5 EID6

Alternate form:

PLOAD2 SID P EID1 “THRU” EID2

Field Contents

SID Load set identification number. (Integer > 0)

P Pressure value. (Real)

EIDi Element identification number. (Integer ≥ 0 or blank; for the “THRU” option,
EID1 < EID2)

The direction of the pressure is determined using the connected GRID points in the same right-hand rule
sense as the PLOAD entry (i.e., with respect to the positive element z axis). In addition, you use the
“THRU” option, all referenced elements must actually exist.

The PLOAD2 entry is similar to the PLOAD entry except that the PLOAD2 references the element ID
instead of the grid points where the element is attached. The PLOAD2 is usually preferred over the
PLOAD entry because it is easier to use. The PLOAD2 entry, similar to the PLOAD entry, is limited to
pressure acting normal to the element surface. The direction of the pressure load is determined by the
numbering sequence of the grid points on the connectivity entry (CQUAD4, CTRIA3, etc.).

See also

• “PLOAD2” in the Simcenter Nastran Quick Reference Guide

PLOAD4

The PLOAD4 entry lets you create the most general pressure definition. You can use PLOAD4 to
apply pressures and/or tractions to any of the two-dimensional elements and the surfaces of the three-
dimensional elements. With PLOAD4, you can create a pressure load that's either normal to the surface
or that contains a traction (not normal to the surface) component. In addition, you can define a different
value of pressure at each corner.

You can use PLOAD4 with CQUAD4, CQUAD8, CQUADR, CTRIA3, CTRIA6, and CTRIAR, CHEXA, CPENTA,
CPYRAM and CTETRA elements. Since the surface hyperelastic elements CQUAD4, CQUAD8, CQUAD,
CTRIA3, and CTRIA6 are plane strain elements, you can't apply pressure loads on them.

The software automatically computes the magnitude and direction of the equivalent grid point forces
using the isoparametric shape functions of the element to which the load has been applied.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 8-27

© 2024 Siemens
 8. Applied Loads

Load intensities P1, P2, P3, (and P4) act at corner points G1, G2, G3, (and G4) for triangular (and
quadrilateral) elements. The default direction of positive pressure for faces of solid elements is inward.

You can set the direction of the pressure can be set by one of two methods.

• By default, the software uses the element normal to determine the direction of the pressure. For
two-dimensional elements, the direction of positive pressure is in the direction of the outward normal
as is determined by applying the right-hand rule to the ordering sequence of the grid points on the
connectivity entry. For surfaces of solid elements, the direction of positive pressure is inward toward
the center of the element. The face of the solid to which you are applying the pressure is determined
by specifying the appropriate corner grid points.

• Alternatively, you can input the direction of the pressure defined by the PLOAD4 entry using an
optional coordinate system and a vector entered on the continuation line. Using a local coordinate
system, you can define a pressure acting at any angle to the surface.You can also apply loads acting
parallel to the surface (tractions).

Another feature unique to the PLOAD4 entry is the ability to apply a nonuniform pressure. You can enter
the pressure at each of the corner grid points to create a linearly varying pressure load. If you enter
the pressure for the first grid point only, the software assumes that the pressure is constant over the
element.

The format of the PLOAD4 entry is:

1 2 3 4 5 6 7 8 9 10
PLOAD4 SID EID P1 P2 P3 P4 G1 G3 or G4
CID N1 N2 N3

The alternate format is

1 2 3 4 5 6 7 8 9 10
PLOAD4 SID EID1 P1 P2 P3 P4 “THRU” EID2
CID N1 N2 N3

Field Contents

SID Load set identification number.

EID Element identification number.

EID1
EID2

P1, P2, P3, P4 Load per unit surface area (pressure) at the corners of the face of the
element.

8-28 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 PLOAD4 Example 1

Field Contents

G1 Identification number of a grid point connected to a corner of the face.

Required data for solid elements only.

G3 For CHEXA, CPYRAM, or CPENTA quadrilateral faces, G3 is the identification

number of a grid point connected to a corner diagonally opposite to G1.
Required for quadrilateral faces of CHEXA, CPYRAM and CPENTA elements
only.
For CPYRAM element triangle faces, G1 and G3 are adjacent corner nodes
on the quadrilateral face, and the load is applied on the triangular face
which includes those grids.
For CPENTA element triangle faces, G3 must be omitted.

G4 Identification number of the CTETRA grid point located at the corner; this
grid point may not reside on the face being loaded. This is required data
and is used for CTETRA elements only.

CID Coordinate system identification number.

N1, N2, N3 Components of the vector measured in coordinate system defined by CID.
Used to define the direction (but not the magnitude) of the load intensity.

If P2, P3, and P4 are blank fields, the load intensity is uniform and equal to P1; P4 is left blank for a
triangular face. In addition, for pressure that acts normal to the face, the continuation entry is not used.

See also

• “PLOAD4” in the Simcenter Nastran Quick Reference Guide

PLOAD4 Example 1

Consider the curved plate shown in Figure 8-10. PLOAD4 entries are to be used to apply a normal
pressure to each of the six CQUAD4 elements.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 8-29

© 2024 Siemens
 8. Applied Loads

Figure 8-10. Pressure Loads on the CQUAD4s

This input file is given in Listing 8-7. When using the PLOAD4 entry, you may specify a PLOAD4 entry for
every element or use the alternate method of specifying several elements that have the same pressure.
The alternate method is used for this example.

8-30 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 PLOAD4 Example 1

$ FILENAME - PRESS.DAT
$ UNITS ARE CM, N
PLOAD4 3 1 -.5 THRU 6
$
GRID 1 0.0 -60.
-2.
GRID 2 60. -60.
-2.
GRID 3 120. -60. -2.
GRID 4 0.0 -30.
0.0
GRID 5 60. -30. 0.0
GRID 6 120. -30.
0.0
GRID 7 0.0 30.
0.0
GRID 8 60. 30.
0.0
GRID 9 120. 30.
0.0
GRID 10 0.0 60.
-2.
GRID 11 60. 60.
-2.
GRID 12 120. 60. -2.
$
CQUAD4 1 1 1 2 5 4
CQUAD4 2 1 2 5 6 3
CQUAD4 3 1 4 5 8 7
CQUAD4 4 1 5 6 9 8
CQUAD4 5 1 7 8 11 10
CQUAD4 6 1 8 9 12 11
$
SPC1 1 123456 1 4 7 10
$
PSHELL 1 4 .5 4
MAT1 4 7.1E6 .3

Listing 8-7. CQUAD4 Elements with Pressure Loads

The resulting deflection is shown in Figure 8-11. The deflected shape is not what you would expect.
Element 2 appears to bend up. The problem occurs because the normal for element 2 is reversed (see
the ordering of the grid point IDs on the CQUAD4s). The pressure on this element is acting upward. This
type of mistake is quite common.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 8-31

© 2024 Siemens
 8. Applied Loads

Figure 8-11. Displacements Due to Pressure Load

PLOAD4 Example 2

Specify the PLOAD4 entry for a uniform normal pressure load applied to the CHEXA solid element shown
in the following figure:

1 2 3 4 5 6 7 8 9 10
PLOAD4 SID EID P1 P2 P3 P4 G1 G3 or G4
CID N1 N2 N3

PLOAD4 12 100 26.4 5 7

This load is selected in the Case Control Section with the command LOAD = 12. Leaving P2, P3, and P4
blank assigns a uniform pressure value of 26.4 lb/in2.

8-32 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 PLOADE1 and PLOADX1

PLOADE1 and PLOADX1

• You can use the PLOADE1 entry to define pressure loading on the edge of CPLSTN3, CPLSTN4,
CPLSTN6, and CPLSTN8 plane strain elements and CPLSTS3, CPLSTS4, CPLSTS6, and CPLSTS8 plane
stress elements.

For CPLSTN3, CPLSTN4, CPLSTN6, CPLSTN8, CPLSTS3, and CPLSTS4 elements, depending on how you
specify the PA and PB fields on the PLOADE1 entry, the pressure can be constant or vary linearly along
the element edge. For CPLSTS6 and CPLSTS8 elements, depending on how you specify the PA, PB,
and PM fields on the PLOADE1 entry, the pressure can be constant, vary linearly, or vary quadratically
along the element edge.

• You can use the PLOADX1 entry to define pressure loading on the edge of CTRAX3, CTRAX6,
CQUADX4, and CQUADX8 elements.

Depending on how you specify the PA and PB fields on the PLOADX1 entry, the pressure can be
constant or vary linearly along the element edge.

See also

• “PLOADE1” in the Simcenter Nastran Quick Reference Guide.

• “PLOADX1” in the Simcenter Nastran Quick Reference Guide.

Inertia Loads (Acceleration Loads)

This section describes how to apply “static” acceleration loads to your model, such as gravity (the
response of a structure to its own weight) and vehicle maneuver loads. The acceleration causes a static
load and not a dynamic one - the structural response is steady state (static) with the transient part
being zero. With an acceleration load, you must specify both the magnitude of the acceleration and the
direction in which it acts.

Using GRAV

You can use the GRAV entry to define the direction and magnitude of a gravity vector in any user-defined
coordinate system. The software multiplies components of the gravity vector by the mass matrix to
obtain the components of the gravity force at each grid point. Since the mass matrix is used to compute
the forces, you must have mass in your model. Typically, you would define that mass by density you
specify on a material bulk data entry. The software doesn't calculate the gravitational acceleration at
scalar points. You must introduce gravity loads at scalar points directly.

Gravity loading is one the best loadings to use when you're performing model checkout because the
software applies the force at every point in the model that has mass. If you have modeling errors where
elements or grids points aren't attached to the structure, their displacement is unbounded, and the job
fails. If this occurs, the software prints a diagnostic message indicating the grid point and component,
where the singularity is detected.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 8-33

© 2024 Siemens
 8. Applied Loads

The format of the GRAV Bulk Data entry is:

1 2 3 4 5 6 7 8 9 10
GRAV SID CID A N1 N2 N3 MB

Field Contents

SID Set identification number. (Integer > 0)

CID Coordinate system identification number. (Integer ≥ 0; Default = 0)

A Acceleration vector scale factor. (Real)

Ni Acceleration vector components measured in coordinate system CID. (Real;

at least one Ni ≠ 0.0)

MB Used only in superelement analysis.

The direction and magnitude of acceleration are given by where the vector = (N1, N2, N3) gives
the direction. The magnitude of is equal to A times the magnitude of . Note, for example, that
entering the value of A in in/sec2 indicates that other specifications of the model that involve length
units (element length, moments of inertia, modulus of elasticity, etc.) must also be in inches to preserve
unit consistency.

For models that have only GRAV loads, the load set ID (SID) is selected in the Case Control Section with a
LOAD = SID Case Control command. For models that also contain other types of loads, such as FORCE or
PLOAD entries, you must use the LOAD entry to combine them with the GRAV entry.

Finally, the model must include mass density information to use acceleration loads. You enter mass
density on a material property bulk data entry. For example, for a model of typical structural steel using
English units, the mass density might be

Thus, the MAT1 Bulk Data entry might look like the following:

1 2 3 4 5 6 7 8 9 10
MAT1 MID E G NU RHO A TREF GE
MAT1 12 30.E6 0.3 7.0E-4

See also

• “Combining Loads” in the Simcenter Nastran User's Guide

8-34 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 GRAV Example 1

GRAV Example 1

This example shows how to calculate the tip deflection of the cantilever beam due to its own weight.

First, notice that the force of gravity (g) acts in the -Y direction (the direction of the center of the earth).
Thus, the vector can be written as:

Equation 8-3.

The acceleration due to gravity on the Earth's surface is approximately 32.2 ft/sec2 (386.4 in/sec2) or 9.8
m/sec2. Units must be consistent throughout the model (this example uses inches).

Thus, the GRAV entry is as follows:

1 2 3 4 5 6 7 8 9 10
GRAV 15 386.4 0. –1. 0.

or, in free field format,

GRAV,15,,386.4,0.,-1.,0.

The Case Control command required to apply this load is LOAD = 15.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 8-35

© 2024 Siemens
 8. Applied Loads

GRAV Example 2

Next, we want to calculate the tip deflection of the cantilever beam due to its own weight and a
concentrated force. Assume that a concentrated force is added to the beam as shown:

To combine gravity loading with the concentrated force, the following approach must be used:

In Case Control: LOAD = 15

In Bulk Data:

Importantly, the set ID on a GRAV entry can't be the same as the Set ID on any other load entry.

GRAV Example 3

As a final example, suppose you are analyzing an instrument package subjected to inertial loads in
specific directions, say:

1.3 g in the x-direction (1.3 g = 1.3 • 9.8 m/sec2 = 12.7 m/sec2)

2.8 g in the x-direction (2.8 g = 2.8 • 9.8 m/sec2 = 27.4 m/sec2)

0.3 g in the x-direction (0.3 g = 0.3 • 9.8 m/sec2 = 2.9 m/sec2)

Three separate GRAV entries can be written as:

1 2 3 4 5 6 7 8 9 10
GRAV SID CID A N1 N2 N3 MB

8-36 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Using RFORCE

GRAV 15 12.7 1. 0. 0.
GRAV 15 27.4 0. 1. 0.
GRAV 15 2.9 0. 0. 1.

Using RFORCE

You can use the RFORCE entry to define a static loading condition due to angular velocity and/or
acceleration. These loads are specified by the designation of a grid point that lies on the axis of rotation
and by the components of rotational velocity or angular acceleration in any defined coordinate system.

There are two methods to compute these loads.

• In the first method, the mass matrix is regarded as pertaining to a set of distinct rigid bodies
connected to grid points. Deviations from this viewpoint, such as the use of scalar points or the
use of mass coupling between grid points, can result in errors.

• The second method allows mass coupling, but does not allow offsets. The second method is the only
method used for angular acceleration.

On the RFORCE entry, you input the components of a spin vector that are used internally to compute
centrifugal forces. Each component of the spin vector is multiplied by the same scale factor. The format
of the RFORCE entry is as follows:

1 2 3 4 5 6 7 8 9 10
RFORCE SID G CID A R1 R2 R3 METHOD
RACC

Field Contents

SID Load set identification number.

G Grid point identification number through which the rotation vector acts.

CID Cartesian coordinate system defining the components of the rotation vector.

A Scale factor of the angular velocity in revolutions per unit time.

R1, R2, R3
Rectangular components of rotation vector . The vector defined passes
through point G.

METHOD Method used to compute centrifugal forces due to angular velocity.

RACC Scale factor of the angular acceleration in revolutions per unit time squared.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 8-37

© 2024 Siemens
 8. Applied Loads

Figure 8-12. RFORCE Vector at Grid Point Gi

where:

angular velocity =
(radians/unit time)

angular acceleration =
(radians/unit time squared)

[m]i = 3 × 3 translational mass matrix at grid point Gi

Note:

The equation for has additional terms if the mass is offset and METHOD = 1 is selected.)

You must select one of two methods for the internal calculation of the loading vector (field 9).

• Method=1 yields correct results only when there is no coupling in the mass matrix. This occurs when
the lumped mass option is used with or without the ZOFFS option (see the CQUAD4 entry for a
description of ZOFFS).

• Method=2 yields correct results for lumped or coupled mass matrix only if the ZOFFS option is not
used. The acceleration terms due to the mass offset (X1, X2, X3) on the CONM2 entry are not
computed with method=2.

8-38 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Defining Mass in Your Model

All possible combinations of mass matrices and offset and the correct method to be used are shown in
Table 8-39.

Table. Restrictions When Using the RFORCE Entry

No Offset Offset
Lumped Method=1 or Method=2 Method=1
Coupled Method=2 Neither

In addition, for problems with elements that have edge grid points (CQUAD8, CTRIA6, CTRlAX6, CHEXA,
CPENTA, CPYRAM and CTETRA), the software only produces correct centrifugal loads if you include the
parameter PARAM,COUPMASS,x (where x is greater than 1) in the input file and you select Method 2.

See also

• “RFORCE” in the Simcenter Nastran Quick Reference Guide

• “COUPMASS” in the Simcenter Nastran Quick Reference Guide

Defining Mass in Your Model

There are several different ways that you can define mass in your model in Simcenter Nastran:

• The concentrated mass elements and scalar points (CMASSi and CONMi)

• The mass density option on the material entries (MAT1, MAT2, etc.)

• Nonstructural mass defined on the property entries

You define mass density on the material entries in terms of mass/unit volume. The mass unit must be
consistent with the other units in the model. For example, in the English system (in, lb, sec), the mass
density of steel is approximately 7.32 × 10−4 lb-sec2/in4.

The software adds the nonstructural mass you define on the property entries to the structure in addition
to the structural mass from the elements.

• For one-dimensional elements, the units are mass/unit length.

• For two-dimensional and three-dimensional elements, the units are mass/unit area and mass/unit
volume, respectively.

Using the WTMASS Parameter to Express Mass in Weight Units

In some cases, you may want to use the WTMASS parameter to express the mass in terms of weight
units instead of mass units. The WTMASS parameter multiplies the assembled mass matrix by the scale
factor you specify with the WTMASS parameter.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 8-39

© 2024 Siemens
 8. Applied Loads

For the steel example, you can enter the mass density a weight density of 0.283 lb/in3 with a WTMASS
parameter of 0.00259 (which is 1/386.4). However, if you enter any of the mass in terms of weight, you
must enter all the mass in terms of weight. The WTMASS multiplies all the mass in the model by the
same scale factor, with the only exception being the mass you enter in Direct Matrix Input.

See also

• “WTMASS” in the Simcenter Nastran Quick Reference Guide

Inertia Load Example

The following inertia load example includes two rotating masses as shown in Figure 8-13. The masses
are accelerated at constant angular acceleration of 20 rev/sec2 from 0 to 200 rev/min. The goal is to
determine the axial force in the CBAR as a function of angular velocity.

Figure 8-13. Rotating Concentrated Masses

The input file is shown in Listing 8-8. Multiple load cases are used to determine axial forces at six
rotational speeds. Using multiple load cases for this problem is much more efficient than submitting
separate runs because Simcenter Nastran doesn't have to regenerate the stiffness and mass matrices.
Another interesting note is that the model doesn't actually rotate; it is fixed at the center. If the
constraints permitted the model to rotate, the analysis would fail.

8-40 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Inertia Load Example

$
$ FILENAME ROTATE.DAT
$
ID LINEAR,ROTATE
SOL 101
TIME 2
CEND
$
FORCE = ALL
SUBCASE 1
LABEL = 0 REV/min
LOAD = 1
SUBCASE 2
LABEL = 40 REV/min
LOAD = 2
SUBCASE 3
LABEL = 80 REV/min
LOAD = 3
SUBCASE 4
LABEL = 120 REV/MIN
LOAD = 4
SUBCASE 5
LABEL = 160 REV/MIN
LOAD = 5
SUBCASE 6
LABEL = 200 REV/MIN
LOAD = 6
BEGIN BULK
PARAM GRDPNT 0
$
GRID 1 0.0 0.0 0.0 123456
GRID 2 10. 0.0 0.0
GRID 3 -10. 0.0 0.0
$
CBAR 10 1 1 2 1.0 0.0 1.0
CBAR 11 1 1 3 1.0 0.0 1.0
$
PBAR 1 1 1.0 1.0 1.0 1.0
$
MAT1 1 30.E6 .3
$
RFORCE 1 1 0. 0.0 0.0 1.0 1
20.
RFORCE 2 1 .666667 0.0 0.0 1.0 1
20.
RFORCE 3 1 1.33333 0.0 0.0 1.0 1

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 8-41

© 2024 Siemens
 8. Applied Loads

20.
RFORCE 4 1 2.0 0.0 0.0 1.0 1
20.
RFORCE 5 1 2.66667 0.0 0.0 1.0 1
20.
RFORCE 6 1 3.33333 0.0 0.0 1.0 1
20.
$
CONM2 101 2 .1
CONM2 102 3 .1
$
ENDDATA

Listing 8-8. Axial Force in the Rotating Arms

Figure 8-14 shows a plot summarizing the axial force. The plot isn't linear because centrifugal forces
aren't linear with angular velocity. Consequently, using a SUBCOM command to combine the subcases
with RFORCE entries can lead to misleading results. In other words, the element axial forces for Subcase
3 can't be obtained by merely doubling the element axial forces of Subcase 2 by using a SUBCOM/
SUBSEQ combination. This occurs because the axial force is a function of ω2, and not 2ω.

Figure 8-14. Axial Force in the Bar as a Function of Angular Velocity

Thermal Loads
You can define thermal loads on a structure to perform stress analysis or to determine thermal
expansion. In Simcenter Nastran, you must define a temperature distribution via TEMPij Bulk Data entries
and thermal expansion coefficients.

8-42 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Defining Temperatures for Grid Points

• Thermal expansion coefficients are specified on the material Bulk Data entries.

• Temperatures can be specified at grid points (TEMP and TEMPD entries) and interpolated to grid points
within elements.

You can also specify temperature data on an element-by-element basis as shown in Table 8-40.

Table. Bulk Data Entries Used for Temperature Definition on Elements

Elements Temperature Data Bulk Data Entry

CROD, CONROD, CTUBE Average temperature at ends A and B. TEMPRB
CBAR, CBEAM, CBEND Average temperature and cross-sectional temperature TEMPRB
gradients at ends A and B.
CQUAD4, Average temperature and gradient in the thickness TEMPP1
CTRIA3CQUAD8, CTRIA6 direction.

Any average temperatures you specify directly for an element take precedence over the temperatures
that the software interpolates from the element's connected grid points. Solid elements obtain their
temperatures only by interpolation from connected grid points. Note that interpolated grid point
temperatures provide temperature gradients over the neutral surface of shell elements, whereas the
TEMPPi entries do not.

Simcenter Nastran uses the temperature data and thermal expansion coefficients to calculate equivalent
forces and moments acting at the grid points.

The TEMP(INIT) and TEMP(LOAD) Case Control commands specify the initial temperature and applied
temperature, respectively. The TEMP(INIT) command must appear either above the first subcase or inside
the first subcase. As with other types of loads, Simcenter Nastran only applies a thermal load if you
specifically request it in the case control section.

Defining Temperatures for Grid Points

You can use the TEMP and TEMPD bulk data entries to define grid point temperatures.

• TEMP lets you define a scalar value of temperature at selected grid points.

• TEMPD lets you define a default temperature for all grid points whose temperatures aren't defined by a
TEMP entry.

See also

• “TEMP” in the Simcenter Nastran Quick Reference Guide

• “TEMPD” in the Simcenter Nastran Quick Reference Guide

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 8-43

© 2024 Siemens
 8. Applied Loads

Defining Temperatures for Elements

You can define temperatures for selected elements.

• You can specify temperatures for CROD, CBAR, CBEAM, CBEND, CONROD, or TUBE elements on the
TEMPRB bulk data entry. TEMPRB specifies the average temperature on both ends. With CBAR, CBEAM
and CBEND elements, you can use TEMPRB to define temperature gradients over the cross section.

• You can specify temperatures for two-dimensional plate and membrane elements on a TEMPP1,
TEMPP2, or TEMPP3 bulk data entry. The user-defined average temperature over the volume is used
to produce in-plane loads and stresses. You can use thermal gradients over the depth of the bending
elements, or the resulting moments to produce bending loads and stresses.

If you don't define any thermal element data for an element, the software averages temperatures of
the connected grid points from the TEMP or TEMPD entries to calculate an average temperature for the
element. The thermal expansion coefficients are defined on the material definition entries. Regardless of
the type of thermal data, if the material coefficients for an element are temperature dependent by use of
the MATTi entry, the software always calculates them from the “average” temperature of the element.

Understanding Coefficient of Thermal Expansion (CTE)

It is important to understand how the coefficient of thermal expansion (CTE) are defined and how
they are used in both linear and nonlinear solutions. There are two typical forms of thermal expansion
data, tangent data and secant data. The CTE value can vary considerably depending on which of these
definitions is used.

General Definition of CTE

The coefficient of thermal expansion is defined as the normalized change in length for a given change in
temperature:

Equation 8-4.

such that the thermal strain is given by:

Equation 8-5.

8-44 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Understanding Coefficient of Thermal Expansion (CTE)

Tangent CTE Data

The tangent data for the coefficient of thermal expansion is defined by the following derivative
approximation:

Equation 8-6.

such that the thermal strain is given by:

Equation 8-7.

Secant Data

ΔT for secant CTE data is the temperature range T – Tref:

Equation 8-8.

The secant data always begins at some reference temperature (Tref), which is assumed to be the zero
thermal strain condition. The axial strain is then measured at the consecutive temperatures, and divided
by the temperature change (T – Tref). For example, to calculate the secant CTE at temperatures T1, T2,
and T3 :

Equation 8-9.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 8-45

© 2024 Siemens
 8. Applied Loads

where ΔL1 is the axial deflection going from Tref to T1

Equation 8-10.

where ΔL2 is the axial deflection going from Tref to T2

Equation 8-11.

where ΔL3 is the axial deflection going from Tref to T3.

The secant data is then applied such that when a model has an initial temperature of T2 and a final
temperature of T3, the resulting thermal strain is equal to the thermal strain from “Tref to T3” minus the
thermal strain from “Tref to T2”. The thermal strain is accurately calculated from secant data using:

Equation 8-12.

CTE in Linear Solutions

The linear solution thermal strain calculation depends on the case control and the material definition.
The linear solutions assume a tangent CTE data described in equation 8-6. However, secant data can be
used as a good approximation if the variation of the CTEs as a function of temperature is small:

When neither TEMP(MAT) nor TEMP(INIT) is specified:

8-46 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Understanding Coefficient of Thermal Expansion (CTE)

Equation 8-13.

When using TEMP(INIT) with no MATTi:

Equation 8-14.

When using TEMP(MAT) with MATTi:

Equation 8-15.

where α(mat) is the CTE evaluated at TEMP(MAT).

When using TEMP(INIT) with MATTi:

Equation 8-16.

where α(init) is the CTE evaluated at TEMP(INIT).

CTE in Nonlinear Solutions

Nonlinear Solutions (SOLs 106, 401, 402, and 414) use the secant equation 8-12 when the CTEs are
temperature dependent:

where αload is the CTE evaluated at TEMP(LOAD) and αinit is the CTE evaluated at TEMP(INIT).

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 8-47

© 2024 Siemens
 8. Applied Loads

The nonlinear solutions require TEMP(INIT) when TEMP(LOAD) is defined. TEMP(MAT) is not supported
and will cause a fatal error if defined. If TEMP(LOAD) is not specified, the thermal load will default to
TEMP(INIT) and there will be no thermal strains, and temperature dependent material properties defined
with the MATTi entries are evaluated at TEMP(INIT). The reference temperature is defined on the MATi
entry or PCOMP entry for composites.

When the CTE is not temperature dependent (constant) in a nonlinear solution, equation 8-14 applies.

See the remarks on the TEMPERATURE case control command in the QRG for more information.

Rigid Element Thermal Strain

The RBAR, RBE1, RBE2, RBE3, RROD and RTRPLT elements support a thermal strain calculation when the
RIGID case control command is assigned to LAGRAN (default), and alpha is defined on the rigid element
bulk data entry.

The rigid element thermal strain is calculated as follows:

The TEMP(LOAD) and TEMP(INIT) value used on RBAR, RROD, and RTRPLT elements is an average
calculated from the grid point values. On RBE1, RBE2 and RBE3 elements, an average TEMP(LOAD)
and TEMP(INIT) is calculated for each leg of the element using the values on the independent/dependent
grid pairs such that each leg can have a different thermal strain if the temperatures vary at the grids.

The rigid element thermal strains are calculated by

εthermal = α(AVGTEMP(LOAD) – AVGTEMP(INIT)).

If TEMP(LOAD) or TEMP(INIT) are not defined, they are assumed to be zero.

See the RIGID case control command in the Simcenter Nastran Quick Reference Guide for more details.

Using Temperature Loads with Subcases

If temperature loads are present in one or more of the subcases in your input file, the software must take
thermal strain into account in the calculation of element stresses and forces. That is, a body subjected
to a temperature field will deform in a stress free manner in the absence of constraints. However,
if the deformation is resisted through boundary conditions and/or the presence of applied loads, the
total actual strain, at a point must be reduced by the thermal strain, α(T − To), to obtain the stress at
the point. For example, the stress-strain relation for a rod that undergoes thermal expansion and also
sustains an axial force is written in the form

8-48 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Using Temperature Loads with Subcases

Equation 8-17.

Simcenter Nastran calculates element stresses and forces on an element-by-element basis from the
displacement vector and the temperature field. Therefore, if you're using SUBCOM or SYMCOM to
combine selected subcases, you must supply a definition of the temperature field whenever element
stresses and forces are required.

Consider the following example where a temperature load is specified in SUBCASE 2.

.
.
.
SUBCASE 1
LOAD=2
SUBCASE 2
TEMP(LOAD)=4
SUBCASE 10
LOAD=10
SUBCOM 20
SUBSEQ=.5,1.,1.6
TEMP(LOAD)=4
.
.
.

Here, the multiplier for SUBCASE 2 is unity so that the temperature distribution required to calculate
element forces and stresses for the combined loading case is precisely TEMP(LOAD)=4. If, on the other
hand, the SUBSEQ entry is modified as indicated below, an additional TEMP(LOAD) must be defined.

.
.
.
SUBCASE 1
LOAD=1
SUBCASE 2
TEMP(LOAD)=4
SUBCASE 10
LOAD=10
SUBCOM 20 SUBSEQ=.5,.75,1.6
TEMP(LOAD)=40

Here, TEMP(LOAD)=40 must reflect a temperature distribution of the form

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 8-49

© 2024 Siemens
 8. Applied Loads

Equation 8-18.

Equation 8-19.

You must supply this new TEMP(LOAD) through standard Bulk Data entries such as TEMPD, TEMP,
TEMPP1, and TEMPP3 and them call them out in the Case Control section with the SUBCOM command.

In general, with n temperature loading conditions in the subcases immediately preceding a SUBCOM,
the temperature distribution that must be defined under the SUBCOM if you want element force or stress
output is of the form

Equation 8-20.

where the coefficients ai are defined on the SUBSEQ entry under the SUBCOM in question. The reference
temperature To, defined on MATi Bulk Data entries, can conveniently be defined as zero in order to
simplify the calculation of the temperature distribution for the combined loading condition of the
SUBCOM.

The same approach is used with combined loading condition defined through a SYMCOM Case Control
entry. However, the use of SYM subcases and SYMCOM entries implies the use of the concepts of
reflective symmetry, in which case you should use the DIH option of cyclic symmetry. Simcenter
Nastran internally generates the appropriate boundary conditions, and the you simply define the
actual temperature distribution on each portion of the structure without regard to decomposing the
temperatures and/or mechanical loads into symmetric and antisymmetric components.

This is best illustrated with the example problem shown in Listing 8-9.

8-50 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Using Temperature Loads with Subcases

.
.
.
CEND
$
TEMP(INIT) = 1
$
SUBCASE 1
TEMP(LOAD) = 10
$
SUBCASE 2
TEMP(LOAD) = 15
$ THE FOLLOWING SUBCOM WILL COMBINE
$ 100% OF THE RESULTS FROM SUBCASE 1
$ WITH 50% OF THE RESULTS FROM SUBCASE 2
SUBCOM 3
SUBSEQ = 1.0,0.5
TEMP(LOAD) = 20
$
BEGIN BULK
$
TEMPD,1, 100.
TEMPD,10,150.
TEMPD,15,250.
$ "TEMPD,20" IS CALCULATED AS FOLLOW:
$ 100 + 1.0*(150-100) + 0.5*( 250 - 100 ) = 225.
$
TEMPD,20,225.
$ REST OF THE BULK DATA ENTRIES
ENDDATA

Listing 8-9. Using SUBCOMs with Temperature Loads

As an example of thermal loads, consider the umbrella tent shown in Figure 8-15. The tent is subjected
to thermal radiation from partial exposure to the sun. The ground is held at 40 degrees Fahrenheit
and the air temperature is held at 50 degrees Fahrenheit. Can the tent withstand the thermal stresses
imposed by solar radiation?

This problem involves conduction, convection, and radiation, which is beyond the scope of this user's
guide, but the stress analysis of the resulting temperature profile is not. For further discussion on the
subject of heat transfer, refer to the Simcenter Nastran Thermal Analysis User's Guide.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 8-51

© 2024 Siemens
 8. Applied Loads

Figure 8-15. Umbrella Tent With Thermal Loads

To analyze this type of problem, you need to run the heat transfer Solution 153 to produce a
temperature profile for all of the grid points in the model. A Solution 153 input file is generated for
this model and is included in the Test Problem Library with the name ‘heat.dat'.

In Solution 153, you have the option of creating a temperature file that can be included directly into the
Solution 101 input file. The Solution 101 input file used for the stress analysis is shown in Listing 8-10.
The temperature file from the Solution 153 run is shown in Listing 8-11. Note the use of the INCLUDE
statement, which inserts the temperature input file at that location when you submit the job.

8-52 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Using Temperature Loads with Subcases

$ FILENAME - THERMAL.DAT
$
ID LINEAR,THERMAL
SOL 101
TIME 10
CEND
TITLE = UMBRELLA TENT WITH THERMAL LOADING
$ DEFINE INITIAL TEMPERATURE
TEMP(INIT) = 2
$ DEFINE THERMAL LOADING
TEMP = 1
STRESS = ALL
SPC = 15
BEGIN BULK
$
INCLUDE 'heat.pch'
$
TEMPD 2 40.0
$
MAT1 1 10.E6 .3 2.59E-3 21.E-6
25000. 25000. 15000.
MAT1 4 2.E6 .3 2.59E-5 21.E-6
25000. 25000. 15000.
$
$ GRID, CBAR, CQUAD4, PBAR, PSHELL, AND SPC ENTRIES
$ REMOVED TO SAVE SPACE
ENDDATA

Listing 8-10. Umbrella Tent Thermal Analysis

$
$FILENAME - HEAT.PCH
$
$TITLE = TENT THERMAL ANALYSIS SUBJECTED TO
RADIATION 1
$SUBTITLE=THE BASE TEMPERATURE AT 40 DEGREE
F 2
$LABEL
= 3
$DISPLACEMENTS
4
$REAL
OUTPUT 5

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 8-53

© 2024 Siemens
 8. Applied Loads

$SUBCASE ID =
1 6
TEMP* 1 1
4.011359E+01 7
TEMP* 1 2
5.999791E+01 8
TEMP* 1 3
1.194475E+02 9
TEMP* 1 4
4.001704E+01 10
TEMP* 1 5
4.705626E+01 11
TEMP* 1 6
6.816640E+01 12
TEMP* 1 7
4.000030E+01 13
TEMP* 1 8
4.004408E+01 14
TEMP* 1 9
6.814966E+01 15
TEMP* 1 10
4.005281E+01 16
TEMP* 1 11
4.012182E+01 17
TEMP* 1 12
1.193867E+02 18
TEMP* 1 99
4.000000E+01 19
TEMP* 1 101
7.500000E+01 20
TEMP* 1 103
7.500000E+01 21
TEMP* 1 110
7.500000E+01 22
TEMP* 1 112
7.500000E+01 23
TEMP* 1 999
0.000000E+00 24

Listing 8-11. Temperature Distribution Generated with Solution 153

The OLOAD output shows the applied loads generated by the imposed temperature. The thermal
stresses for this example are quite small, so the structure should have no difficulty surviving the thermal
loading.

See also

• “Understanding Cyclic Symmetry” in the Simcenter Nastran User's Guide

8-54 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Enforced Element Deformation Loads

Enforced Element Deformation Loads

Simcenter Nastran includes a limited capability to define enforced axial deformations. You can use the
DEFORM Bulk Data entry to apply an enforced extension (units of length) to CBAR, CBEAM, CONROD,
CROD, and CTUBE elements.

Enforced deformations, for example, are useful in the simulation of misfit or misalignment in
engineering structures. As in the case of thermal expansion, the software calculates the equivalent
loads by separate subroutines for each type of structural element and are applied to the connected grid
points. The magnitude of the axial deformation is specified on a DEFORM entry.

You can only use DEFORM in SOLs 101, 105, 114, and 200. To apply enforced displacements in other
solution sequences, use SPCD.

With DEFORM, the enforced deformation is analogous to a turnbuckle that can be adjusted to provide
a desired axial deformation in an element. The appropriate stress-strain relation for a rod element of
length that undergoes an enforced axial deformation, δ is given by the following equation:

Equation 8-21.

Consider this example:

.
.
.
SUBCASE 1
LOAD=1
SUBCASE 2
DEFORM=5
SUBCASE 10
LOAD=10
SUBCOM 20
SUBSEQ=.5,.75,1.6
DEFORM=50
.
.
.

Here DEFORM = 50 reflects a set of enforced deformations of the form

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 8-55

© 2024 Siemens
 8. Applied Loads

Equation 8-22.

For n DEFORM loading conditions in the subcases that immediately precede a SUBCOM, the DEFORM
loading condition that must be defined under the SUBCOM, if element force and stress output is desired,
is of the form

Equation 8-23.

The format of the DEFORM entry is shown below:

1 2 3 4 5 6 7 8 9 10
DEFORM SID EID1 D1 EID2 D2 EID3 D3

Field Contents

SID Deformation set identification number. (Integer > 0)

EIDi Element number. (Integer > 0)

Di Deformation. (Real; positive value represents elongation; negative value

represents contraction)

You must use the DEFORM=SID to select the DEFORM entry in the Case Control section.

When you use DEFORM, you must remember that you're not enforcing a strain or an actual extensional
length to the element. Rather, you're applying a force to the element that produces the specified
extension if the element is free to expand without internal forces being generated. The software adds
this computed force to the other forces in the model. Since most elements in your model aren't free to
expand, the extension value you specify may not be achieved. This feature is best illustrated with an
example.

8-56 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 DEFORM Example

See also

• “DEFORM” in the Simcenter Nastran Quick Reference Guide

• “SPCD” in the Simcenter Nastran Quick Reference Guide

DEFORM Example

Suppose you want to use the DEFORM entry to enforce a strain on the three-member structure shown in
Figure 8-16. The goal is to impose an initial strain equal to 100 μcm/cm in the center member.

Figure 8-16. Three-Bar Structure with Initial Strain

Consider: (1) the DEFORM entry can only impose a force, (2) from static equilibrium, if no forces are
imposed on grid points 2 and 3, the force in all three members must be the same. With this in mind, the
following procedure is used.

1. Determine the force required in member 11 that produces the 100 mcm/cm strain.

2. Since there are no external forces acting on the elements, the force of 1065 N is same in all the
members. Using the 1065 N force, the stresses computed in the other members.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 8-57

© 2024 Siemens
 8. Applied Loads

3. From the stresses found in part 2, the element strains are computed for elements 10 and 12.
Multiplying the strain within each element by its length yields the extensional change of the
element. These extensional changes are the deform values required to produce a 100 μcm/cm
strain in element 11.

Therefore,

These values are entered on the DEFORM Bulk Data entry as shown in Listing 8-12. To help understand
the implementation of the DEFORM entry, two configurations are being analyzed. The first configuration
(Subcase 1) specifies the boundary conditions as shown in Figure 8-17(a). For this case, the structure is
free to expand without inducing any element forces. The results of this subcase should be the extension
of the members as entered on the DEFORM entry. As mentioned previously, a force is applied to the
element that produces the specified extension if the element is free to expand.

The second subcase, shown in Figure 8-17(b), is the constrained configuration that is of interest. If the
calculations are correct, the strain in the center element should be the required 100 μcm/cm.

8-58 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 DEFORM Example

Figure 8-17. Three-Bar Structures with Initial Strain

$ FILENAME - STRAIN.DAT
ID LINEAR,STRAIN
SOL 101
TIME 2
CEND
$
DISP = ALL
STRESS = ALL
DEFORM = 1
SUBCASE 1
SUBCASE 2
SPC = 1
BEGIN BULK
$
SPC 1 4 1
$
GRID 1 0.0 0.0 0.0 123456
GRID 2 10. 0.0 0.0 23456
GRID 3 20. 0.0 0.0 23456
GRID 4 30. 0.0 0.0 23456
$
CBAR 10 1 1 2 1.0 0.0 1.0
CBAR 11 2 2 3 1.0 0.0 1.0
CBAR 12 3 3 4 1.0 0.0 1.0
$
PBAR 1 1 1.0

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 8-59

© 2024 Siemens
 8. Applied Loads

PBAR 2 1 1.5
PBAR 3 1 2.0
$
MAT1 1 7.1E6 .3
$
DEFORM 1 10 1.5E-3 11 1.E-3 12 .75E-3
ENDDATA

Listing 8-12. Initial Strain Input File

The stresses and displacements for the two subcases are shown in Figure 8-18. As can be seen for the
free end configuration, the extension of all of the members is consistent with the DEFORM entry.

Note that the extensions for the elements aren't the same as the values specified on the DEFORM entry.
The stress in member 11 is 710 N/cm2. Since E = 7.1 x 106, the strain in element 11 is equal to ε = E/σ =
100 μcm/cm strain. The stress output is requested for this example because strain output is not available
for one-dimensional elements.

Figure 8-18. Stress and Displacement Output of the Three Bar Structure

8-60 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Statics

Loads Due to Enforced Motion

Statics

You can specify enforced zero displacements on GRID, SPC, or SPC1 entries.

• If you use GRID entries, the constraints become part of the structural model and modifications cannot
be made at the subcase level.

• SPC or SPC1 entries are usually used for defining zero displacements, which result in nonzero forces of
constraint.

You can specify nonzero enforced displacements on SPC or SPCD entries.

• The SPC entry lets you specify both the component to be constrained and the magnitude of the
enforced displacement.

• The SPCD entry lets you specify only the magnitude of the enforced grid point displacement. When
you use an SPCD entry, you must specify the component to be constrained on either an SPC or SPC1
entry. Using SPCD avoids the decomposition of the stiffness matrix when changes are only made in
the magnitudes of the enforced displacements.

Simcenter Nastran calculates the equivalent loads resulting from enforced displacements of grid points
and adds them to the other applied loads. You specify the magnitudes of the enforced displacements on
SPC entries in the global coordinate system. Simcenter Nastran automatically applies the load when you
select the associated SPC set in the Case Control section.

See also

• “SPC” in the Simcenter Nastran Quick Reference Guide

• “SPCD” in the Simcenter Nastran Quick Reference Guide

Dynamics

In Simcenter Nastran, you can use the:

• TLOAD1 and TLOAD2 entries to define a time-dependent dynamic load or enforced motion for use in a
transient response analysis

• RLOAD1 and RLOAD2 entries to define a frequency-dependent dynamic load for use in a frequency
response analysis

You can use the TYPE field (field5 on TLOAD1/TLOAD2 and field 8 on RLOAD1 and RLOAD2) to specify the
type of dynamic excitation you're creating.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 8-61

© 2024 Siemens
 8. Applied Loads

• If the TYPE field on the TLOAD1/2 entries indicates an enforced motion, the software first assumes
that the EXCITEID field points to SPC-type data. If not present, Simcenter Nastran then assumes the
excitation is enforced motion using large masses and will then look for DAREA and various static
loading data, just as in the case of applied loads.

• With RLOAD1 and RLOAD2 entries, the software only look for SPC-type data in the case of enforced
motion.

For all entries (TLOAD1/2, RLOAD1/2), if the TYPE field indicates an applied load, the program will search
only for static loading data.

See also

• “TLOAD1” in the Simcenter Nastran Quick Reference Guide

• “TLOAD2” in the Simcenter Nastran Quick Reference Guide

• “RLOAD1” in the Simcenter Nastran Quick Reference Guide

• “RLOAD2” in the Simcenter Nastran Quick Reference Guide

Specifying Enforced Motion Data

In Simcenter Nastran, you specify enforced motion data on the SPC/SPC1/SPCD Bulk Data entries. (Note
that components specified in SPCD data must be also be referenced on SPC or SPC1 entries.)

You can define enforced motion with SPC data (without the use of SPCDs) alone. However, such usage
is discouraged in favor of SPC/SPC1 data in connection with SPCDs. In this way, you select SPCD entries
via the EXCITEID field in the dynamic load data, while the software uses the SPC/SPC1 sets to identify the
constrained displacement set in the Case Control section.

With SPC data alone, enforced motion is applied solely via the Case Control section. The entries
referenced by the EXCITEID field on the TLOAD1/2 or RLOAD1/2 entries do not have to exist. This method
of specifying enforced motion is less elegant and somewhat more difficult to interpret.

See also

• “SPC” in the Simcenter Nastran Quick Reference Guide

• “SPC1” in the Simcenter Nastran Quick Reference Guide

• “SPCD” in the Simcenter Nastran Quick Reference Guide

8-62 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Using Residual Vectors (Modal Augmentation) in Modal Dynamic Response Analysis

Using Residual Vectors (Modal Augmentation) in Modal Dynamic Response

Analysis

When enforced motion is employed in modal dynamic response analysis, using modal augmentation
vectors is highly recommended in order to improve solution accuracy. Without these vectors, your
obtained results may be unacceptable.

Qualitatively, the enforced motion shapes, or “modes,” may not necessarily be represented by
the component modes of the structure, leading to a truncation of the enforced motion shapes
during projection to the modal space. To obtain modal residual vectors from applied loads using
modal augmentation vectors, specify PARAM,RESVEC,YES in the Bulk Data section. Simcenter Nastran
automatically derives the necessary static loads from the enforced motion data; you don't need to
specify any additional loads.

In summary, the procedure for applying enforced motion in a dynamic analysis:

1. Specify the appropriate type of enforced motion to be applied via the TYPE field in RLOAD1/RLOAD2
(field 8) or TLOAD1/TLOAD2 (field 5) Bulk Data entry, as appropriate.

2. Define the desired enforced motion using the SPCD Bulk Data entry. The set IDs of these SPCD data
must match the IDs appearing in the EXCITEID fields of the corresponding dynamic load data in step
1.

3. Ensure that the components referenced in the SPCD Bulk Data entry above are also specified in the
SPC1 Bulk Data entry and this SPC1 is also selected in the Case Control Section.

4. Ensure PARAM,RESVEC is set to YES in the Bulk Data Section if a modal dynamic solution approach
is used.

See also

• “RESVEC” in the Simcenter Nastran Quick Reference Guide

Frequency-dependent Loads
You use the DLOAD entry to define linear combinations of frequency dependent loads that are defined
on RLOAD1, RLOAD2, or ACSRCE entries. The RLOAD1 entry defines a frequency dependent load of the
form

Equation 8-24.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 8-63

© 2024 Siemens
 8. Applied Loads

where A is defined on a load entry, for example DAREA, C (f) and D (f) are defined on TABLEDi entries, θ
is defined on a DPHASE entry, τ is defined on a DELAY entry, and f is defined on a FREQ, FREQ1 or FREQ2
entry.

The RLOAD2 entry defines a frequency-dependent load of the form

Equation 8-25.

where A is defined on a DAREA entry, B (f) and φ (f) are defined on TABLEDi entries, θ is defined on a
DPHASE entry, and τ is defined on a DELAY entry. DAREA, DELAY, and DPHASE entries may be different
for each loaded degree-of-freedom. The loads are applied to the specified components in the global
coordinate system.

The ACSRCE entry defines a frequency-dependent acoustic source of the form

where A is defined on a load entry, for example DAREA, P(f) is defined on TABLEDi entries, θ is optionally
defined on a DPHASE entry, τ is optionally defined on a DELAY entry, RHO and B are defined on the
ACSRCE entry, and f is defined on a FREQ, FREQ1 or FREQ2 entry.

The RANDPS entry defines load set power spectral density factors for use in random analysis of the form

Equation 8-26.

where G (f) is defined on a TABRNDi entry. The subscripts j and k define the subcase numbers of the
load definitions. If the applied loads are independent, only the diagonal terms (j = k) need be defined.
You use the RANDT1 entry to specify the time lag constants for use in the computation of the auto
correlation functions.

8-64 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Time-dependent Loads

See also

• “Frequency Response Analysis” in the Simcenter Nastran Basic Dynamic Analysis User's Guide (for a
discussion of frequency response calculations)

• Section 12.2 of The NASTRAN Theoretical Manual (for a discussion of random response calculations)

Time-dependent Loads
You can use the DLOAD entry to define linear combinations of time-dependent loads that are defined on
TLOAD1 and TLOAD2 entries. The TLOAD1 entry defines a time-dependent load of the form

Equation 8-27.

where A is defined on a DAREA entry, τ is defined on a DELAY entry, and F (t − τ) is defined on a TABLEDi
entry. The TLOAD2 entry defines a time-dependent load of the form

Equation 8-28.

where T1 and T2 are time constants, , A and τ are the same as on the TLOAD1 entry, B is the
growth coefficient, C is the exponential coefficient, F is the frequency, and P is the phase angle.

The coefficients on the DAREA and DELAY entries may be different for each loaded degree-of-freedom.
The loads are applied to the specified components in the global coordinate system.

Two methods are available to automatically create DAREA values from static load data.

• Simple static loads that supply load values directly to grid points (such as FORCE, MOMENT, SLOAD,
and PLOAD Bulk Data entries) will add to the DAREA factors when the load identification matches the
requested set.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 8-65

© 2024 Siemens
 8. Applied Loads

• More complicated loads may be generated using the LOADSET Case Control command. The requested
LSEQ Bulk Data entry matches DAREA sets to static loading sets. Any type of static load (e.g., LOAD,
GRAV, RFORCE, and SPCD Bulk Data entries) may be generated for either superelement or residual
only (non-superelement) models.

Nonlinear effects are treated as an additional applied load vector, for which the components are
functions of the degrees-of-freedom. This additional load vector is added to the right side of the
equations of motion and treated along with the applied load vector during numerical integration. It is
required that the points to which the nonlinear loads are applied and the degrees-of-freedom on which
they depend be members of the solution set; i.e., that they cannot be degrees-of-freedom eliminated
by constraints. It is further required that, if a modal formulation is used, the points referenced by the
nonlinear loads be members of the set of extra scalar points introduced for dynamic analysis.

There are four different types of nonlinear transient forcing functions:

• Arbitrary function generators

• Multiplies

• Positive power functions

• Negative power functions

The NOLIN1 entry defines a nonlinear load of the form

Equation 8-29.

where Pi is the load applied to ui, Si is a scale factor, T(uj) is a tabulated function defined with a TABLEDi
entry, and uj is any permissible displacement component or velocity component.

The NOLIN2 entry defines a nonlinear load of the form

Equation 8-30.

where uj and uk are any permissible pair of displacement components and/or velocity components. They
may be the same.

8-66 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Using Static Loading Data in Dynamic Analysis

The NOLIN3 entry defines a nonlinear load of the form

Equation 8-31.

where A is an exponent and uj is any permissible displacement component or velocity component.

The NOLIN4 entry defines a nonlinear load of the form

Equation 8-32.

where A is an exponent and uj is any permissible displacement component or velocity component.

See also

• “Transient Response Analysis” in the Simcenter Nastran Basic Dynamic Analysis User's Guide (for a
discussion of transient response calculations)

• Section 11.2 of The NASTRAN Theoretical Manual

• For a discussion of nonlinear forcing functions, see Section 11.2 of The NASTRAN Theoretical Manual.

Using Static Loading Data in Dynamic Analysis

To include static loading data in a dynamic analysis, all static loads and thermal loads that have the
same ID as the DAREA ID on the dynamic load entry are automatically selected, as shown in the example
below.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 8-67

© 2024 Siemens
 8. Applied Loads

.
CEND
.
$ THE FOLLOWING REQUEST SELECTS RLOAD1 ENTRY 150
DLOAD = 150
.
BEGIN BULK
$ THE FOLLOWING PLOAD4 IS AUTOMATICALLY SELECTED
$ BECAUSE ITS ID OF 100 MATCHES THE EXCITEID ID OF THE
$ SELECTED RLOAD1 ENTRY
PLOAD4,100,...
RLOAD1,150,100,...
$ THE ABOVE USAGE IS EQUIVALENT TO THE USER
$ SELECTING AN LSEQ BULK DATA OF THE FORM:-
$ LSEQ,SET_ID,100,100,100
.

The DAREA Bulk Data entry is used to specify point loads in dynamic analysis. In the case of grid points,
these loads are implicitly assumed to be in the displacement (or local) coordinate systems of those
points. The software automatically converts all DAREA Bulk Data entries for grid and scalar points into
the appropriate equivalent FORCE/MOMENT/SLOAD Bulk Data entries.

Two additional, important advantages of the above enhancement follow.

• When you perform dynamic analysis using the modal approach, you may want to employ residual
vectors to improve the quality of the solution. In addition to specifying PARAM,RESVEC,YES in the Bulk
Data Section, you must also specify static loads at those points that are dynamically excited. However,
with the above automatic conversion feature, it isn't necessary to explicitly specify static loads for the
purpose of residual vector calculations. Such loads are automatically generated by Simcenter Nastran.

• With the automatic conversion feature, you can use the DAREA entry to be used in static analysis
as well. This is particularly advantageous when you want to apply loads at grid points in the
displacement (or local) coordinate systems of those grid points.

When the software performs the automatic conversion, it issues you a message at the end of the Preface
module that contains an image of each converted DAREA Bulk Data entry as well as the corresponding
FORCE/MOMENT/SLOAD Bulk Data entry into which it has been converted.

See also

• “DAREA” in the Simcenter Nastran Quick Reference Guide

Defining Loads for Cyclic Symmetry Analysis

In Simcenter Nastran, there are several different bulk data entries that you can use to define loading for
cyclic symmetry analyses.

8-68 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Combining Loads

• You can use the LOADCYH entry to define the harmonic coefficients of a load in cyclic symmetry
analysis.

• You can use the LOADCYN entry to define a physical load in cyclic symmetry analysis.

• You can use the LOADCYT entry to specify a tabular load input for axisymmetric cyclic symmetry
problems.

See also

• “LOADCYH” in the Simcenter Nastran Quick Reference Guide

• “LOADCYN” in the Simcenter Nastran Quick Reference Guide

• “LOADCYT” in the Simcenter Nastran Quick Reference Guide

• “Understanding Cyclic Symmetry” in the Simcenter Nastran User's Guide

Combining Loads
You can use the LOAD case control command to combine the loads you specified on other entries to
form new loading conditions. You can use LOAD to define a static loading condition that's a linear
combination (superposition) of load sets consisting of loads applied directly to grid points, pressure
loads, gravity loads, and centrifugal forces.

Once you've used the loading bulk data entries to define the load, you can use case control commands
to specify which of the loading entries you want to use in a given analysis. The case control command
you use depends on the type of load or loads you're applying:

LOAD=n

DEFORM=n

TEMP(LOAD)=n

SPC=n

where n is a unique, user-defined set identification number (SID) in field 2 of the load Bulk Data entry.

For a given analysis, the total load the software applies is the sum of external (LOAD), thermal
(TEMP(LOAD)), element deformation (DEFORM), and displacement (SPC) loads.

See also

• “LOAD” in the Simcenter Nastran Quick Reference Guide

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 8-69

© 2024 Siemens
 8. Applied Loads

• “TEMPERATURE” in the Simcenter Nastran Quick Reference Guide

• “DEFORM” in the Simcenter Nastran Quick Reference Guide

• “SPC” in the Simcenter Nastran Quick Reference Guide

Using LOAD

You must use LOAD if you want to combine gravity loads with loads applied directly to grid points,
pressure loads, or centrifugal forces, even if there is only one loading condition.

You can define up to 300 (Si, Li) pairs with a single LOAD entry.

For any particular solution, the total static load will be the sum of the applied loads (grid point loading,
pressure loading, gravity loading, and centrifugal forces) and the equivalent loads.

The resulting combined load is determined by:

Equation 8-33.

Where:

• (S) is the overall scale factor you define on the LOAD entry

• (Si) is the scale factor you define on the LOAD entry for each individual load set

• {Li} is the applied load vector corresponding to load set ID Li

You request subcases, which may or may not include different loading conditions, in the Case Control
Section. Three Case Control commands (LOAD, TEMP(LOAD), and DEFORM) are used in selecting loading
conditions for subcases. Each of these commands identifies a unique set whose SID is the same as the
SID of one or more Bulk Data entries. Figure 8-19 shows the Bulk Data entries you can select with each
of the three Case Control commands.

8-70 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 LOAD Example 1

Figure 8-19. Load Combinations

LOAD Example 1

As an example of the LOAD entry combining two loads, consider the umbrella tent shown in Figure
8-20.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 8-71

© 2024 Siemens
 8. Applied Loads

Figure 8-20. Umbrella Tent Subjected to Gravity and Wind Loads

The input file containing the combined loading is shown in Listing 8-13. To apply the wind pressure in
the correct direction, a local coordinate system is defined so that the local Y axis makes an angle of 5
degrees with the ground. This coordinate system is referenced on the PLOAD4 continuation entry. Since
all of the elements in the model are exposed to the wind load, the simplest method is to use the thru
option on the PLOAD4 entry (fields 8 and 9).

The gravity load of 386.4 entered on the GRAV entry corresponds to a 1 G load in the English system.
Whenever you are working with an acceleration vector in terms of Gs, make sure that you convert the
load such that its units are consistent with your structure.

8-72 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 LOAD Example 2

$
$ FILENAME - TENT.DAT
$
:
LOAD=5
:
BEGIN BULK
$
CORD2R 1 0 0. 0. 0. 0. .08716 1.
1. .08716 1.
$
GRAV 1 0 386.4 0.0 0.0 -1.0
PLOAD4 2 1 .02 THRU 6
1 0. 1. 0.
LOAD 5 1. 1. 1 1. 2
$
PARAM POST 0
PARAM AUTOSPC YES
PARAM GRDPNT 0
$
PSHELL 1 4 .05
$
MAT1 1 10.E6 .3 2.59E-3
25000. 25000. 15000.
$
MAT1 4 2.E6 .3 2.59E-5
25000. 25000. 15000.
$
$ GRID, ELEMENT, AND CONSTRAINTS ARE NOT SHOWN
$
ENDDATA

Listing 8-13. Umbrella Tent with Wind and Gravity Loads

The gravity load is defined by the GRAV entry with a loading set ID of 1. The wind loading is defined
using the PLOAD4 entry with a loading set ID of 2. The LOAD entry (ID of 5) is used to combine the
gravity and wind loadings. The LOAD entry is selected by the LOAD = 5 command in the Case Control
Section. If the GRAV and the PLOAD4 entries were assigned the same ID, a fatal error would have
resulted.

LOAD Example 2

Assume that your model has one concentrated force of 15.2 lbs in the y direction applied to grid point
12, and one concentrated moment of 6.4 inch-lbs about the x-axis applied to grid point 127. It is
required to double the value of force and triple the value of moment for your next analysis.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 8-73

© 2024 Siemens
 8. Applied Loads

The LOAD Bulk Data entry may be written with an overall scale factor (S) of 1.0 and loadset scale factors
(Si) of 2.0 for force and 3.0 for moment. Thus,

In Case Control:

LOAD = 22

In Bulk Data:

1 2 3 4 5 6 7 8 9 10
LOAD 22 1.0 2.0 30 3.0 40
FORCE 30 12 15.2 0. 1. 0.
MOMENT 40 127 6.4 1. 0. 0.

Bolt Preload
Bolts (and certain types of threaded fasteners) are commonly tightened to levels producing very high
preload forces. Preloading bolts to about 75% of their proof strength is typical. The bolt preload
capability in Simcenter Nastran allows you to predict stresses in the bolts and the bolted medium that
arise from bolt preload forces alone or bolt preload forces and service loads.

Historically, bolt preload was modeled using either an equivalent thermal load approach or a multipoint
constraint (MPC) approach. Both methods are capable of providing accurate results. However, both
methods are labor intensive requiring multiple solutions, manual capture of data, and hand calculations.

The Simcenter Nastran approach is much more efficient because the entire run is automated and allows
for direct entry of the bolt preload forces. During the run, the model is solved twice. The first solution
calculates the strains in the bolts resulting from bolt preload forces. The second solution uses that strain
along with any other service loads as the total applied load.

In the first solution, the bolt is altered before applying the preload. For bolts modeled with beam or bar
elements, the axial stiffness is reduced to be negligible. For bolts modeled with solid elements, the bolt
mesh is actually cut. In both cases, the bolt is effectively split and the preloads are applied as equal and
opposite forces on the two sides. The resulting strain in the bolt from this solution is then used in the
second solution.

In the second solution, the preload strain load computed in the first solution is applied along with any
service loads. The preload strain load is applied differently depending on the element representation.
For the beam or bar element bolts, the axial stiffness is restored and the preload bolt strain is applied

8-74 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Bolt Preload

as a load on the element. For the solid element bolts, the grids on the two cut faces are held in their
relative state of axial interference using a stiff glue connection. This interference condition is the axial
bolt strain.

Note that when you are viewing results on solid element preloaded bolts in a post processor, the bolts
will be split and the displacements of the two cut faces will overlap. This axial displacement in the solid
element bolt does not reflect reality, but the strain and stress in the bolts is correct.

Bolts can be modeled using CBEAM and CBAR line elements or CHEXA, CPENTA, and CTETRA solid
elements.

Bolt preload is supported in SOLs 101, 103, 105, 107 through 112, 401 and 402. The description
provided here applies to SOLs 101, 103, 105, 107 through 112. For information regarding bolt preload
in SOLs 401 and 402, see the Multi-Step Nonlinear User's Guide.

Related Inputs for SOLs 101, 103, 105, 107 through 112

• The BOLT
bulk entry selects the following:

• If your bolts are modeled as CBEAM or CBAR elements, you will use ETYPE=1. With this format, you
select one of the elements representing the bolt.

• If your bolts are modeled as CHEXA, CPENTA, and CTETRA solid elements, you will use ETYPE=2.
With this format, you select the GRID entries that form a cross section through the bolt.

You can optionally leave both the CSID and IDIR fields on the BOLT entry blank and the software
will automatically determine the coordinate system and the bolt axis. The software requires
that the element faces which are associated with the grid points listed on the BOLT entry lie
approximately on a plane perpendicular to the intended bolt axis. This is important for the software
to automatically compute the bolt coordinate system and the bolt axis. If you leave both the CSID
and IDIR fields blank, and the cut you define is not close to being planar, a fatal error will occur.
In addition, if the cut is planar but the plane is not perpendicular to the intended bolt axis, the
software computed coordinate system and bolt axis will be skewed from the intended bolt axis, and
your preloads will not be accurate.

• The BOLTFRC bulk entry defines the bolt preload force for BOLT bulk entries. Note that only
TYPE=LOAD and TYPE=DISP are supported for the linear solutions. With TYPE=DISP, the displacement
you define on the BOLTFRC entry represents the resulting shortening of the bolt from the preload. The
software uses the bolt length to convert the preload displacement to an equivalent axial bolt strain.
The STRAIN preload option on the BOLTFRC bulk entry is not supported for the linear solutions and
will cause a fatal error if defined.

• The BOLTFOR bulk entry is a legacy input to define a preload force. The BOLTFRC bulk entry is the
recommended input to define a preload. For SOLs 101 and 401, if you are using the legacy BOLTFOR
entry to define preloads, and you have defined multiple BOLTFOR entries with the same ID, the
software will only apply the preloads for one of the BOLTFOR entries and will ignore the others. As a

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 8-75

© 2024 Siemens
 8. Applied Loads

result, if multiple BOLTFOR entries are desired, they should be defined with unique IDs and combined
into a single preload set with the BOLTLD bulk entry.

• The BOLTLD bulk entry

is optionally used to define combinations of BOLTFRC bulk entries and scale the corresponding preload
forces.

• The BOLTLD case control command is used to select a BOLTLD bulk entry, or BOLTFRC bulk entries to
include in either the global case or subcases.

• The BOLTFACT parameter is used

during the initial preload solution. During this solution, the deformed shape of the bolted medium
is obtained by solving the model with the axial stiffness of the bolts effectively eliminated from the
model. For CBEAM or CBAR element bolts, this is accomplished by reducing the bolt axial stiffness by
the BOLTFACT parameter (default value of 1 x 107). Additional information regarding this parameter is
provided below.

An input file can contain BOLTLD case control commands selecting BOLTLD and BOLTFRC bulk entries.
For such a case, if the set identification number (SID) referenced by a BOLTLD case control command
exists on BOLTLD or BOLTFRC bulk entries, the BOLTLD bulk entry is selected.

Bolt Preload Procedure

Modeling with Line Elements

CBEAM and CBAR elements are used to represent a bolt.

• A bolt can be modeled as a single CBEAM or CBAR, or as several.

• The material properties and cross section of the CBEAM or CBAR elements should be representative of
the corresponding bolt.

• The grids connected to CBEAM or CBAR elements representing bolts should be unconstrained.

BOLT bulk entries are used to define CBEAM or CBAR elements as bolts when ETYPE = 1.

• A single BOLT bulk entry should be created for each bolt incidence.

• If a bolt is modeled with multiple CBEAM or CBAR elements, only one of the elements needs to be
included in a BOLT bulk entry. If additional CBEAM and CBAR elements are included in the BOLT bulk
entry, the first element listed is used and all the other elements are ignored.

BOLTFOR or BOLTFRC bulk entries are used to assign bolt preload forces to bolts.

• At least one BOLTFOR or BOLTFRC bulk entry is required for each bolt preload force level.

8-76 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Bolt Preload Procedure

BOLTLD bulk entries are used to define combinations of BOLTFOR and BOLTFRC bulk entries and scale the
corresponding bolt preload forces. The linear solutions only support TYPE=LOAD or TYPE=DISP on the
BOLTFRC entry.

• The use of BOLTLD bulk entries is optional. BOLTLD bulk entries provide flexibility in creating
combinations of bolts and bolt preloads.

BOLTLD case control commands are used to select BOLTLD or BOLTFOR and BOLTFRC bulk entries to
include in either the global case or subcases.

• Use a BOLTLD case control command before the subcases to select a BOLTLD bulk entry or BOLTFOR
and BOLTFRC bulk entries to include in the global case.

• Use a BOLTLD case control command inside the subcases to select a BOLTLD bulk entry or BOLTFOR
and BOLTFRC bulk entries to include in the subcases.

The BOLTFACT parameter is the stiffness reduction factor.

• For most cases, the default value of 1 x 107 is appropriate.

• For cases where the bolts are very stiff relative to the bolted medium, the BOLTFACT parameter can be
increased to 1 x 1010 or so.

• For cases where constraints are sufficient to equilibrate the bolted medium if the bolts were not
present, the BOLTFACT parameter can be arbitrarily large.

• For other cases, caution should be exercised when specifying the BOLTFACT parameter.
Inappropriately large values can make the stiffness matrix nearly singular and potentially cause
numerical problems.

RBE2 spider elements are typically used to connect the end grids of CBEAM or CBAR elements that
represent bolts to the surrounding mesh.

CBEAM or CBAR bolts and contact conditions

When a bolt preload is defined on CBEAM or CBAR elements and contact is in the model, the software
runs an initial preload solution to compute the strain in the bolt and to compute a converged set of
contact status conditions before applying any service loads. The bolt preload behaviour depends on the
RESET contact parameter defined on the BCTPARM bulk entry.

• When RESET=1, for all consecutive solutions for the same subcase as the preload and for any
consecutive subcases, the software applies the computed bolt strain from the initial preload solution
and it uses the final contact status from the initial bolt preload solution as the starting contact status.

• When RESET=0 (default), for the consecutive solution for the same subcase as the preload, the
software applies the computed bolt strain from the initial preload solution and it uses the final contact
status from the initial bolt preload solution as the starting contact status (same as RESET=1). For

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 8-77

© 2024 Siemens
 8. Applied Loads

the solutions from any consecutive subcases, the software applies the computed bolt strain from the
initial preload solution, and it applies the final contact status from the end of the prior subcase.

• When both RESET=0 or RESET=1, if the subcase which computes the bolt preload includes only the
bolt preload and no service loads, the two pass solution described above still occurs. That is, the
first solution computes the strain in the bolt and the converged set of contact status conditions.
Since there are no service loads for this scenario, the second solution is almost a repeat of the first,
although the contact status results from the second solution may be slightly different than the contact
status results from the first. The reason is that the second solution begins with the final contact status
from the first, but also continues to iterate on this contact condition until the hard coded minimum
number of contact iterations is reached. You can optionally adjust the contact convergence criteria
with the NCHG parameter on the BCTPARM bulk entry if you would like the contact status between
the first and second solutions to be more similar.

Modeling with Solid Elements

The structure of an input file for modeling bolts with solid elements is identical to that for modeling
bolts with line elements except that:

• A mesh of CHEXA, CPENTA, and CTETRA elements are used to represent a bolt. Create the solid
element mesh such that at an intermediate position along the bolt axis the element edges and
element faces of the mesh form a cross section through the bolt. As a best practice, make the cross
section planar and normal to the axis of the bolt. Doing so will facilitate interpretation of the results.
The cross section of the mesh and the material properties associated with the mesh should also be
representative of the corresponding bolt.

• ETYPE = 2 should be specified on the BOLT bulk entries. When ETYPE = 2, the fields on the BOLT entry
are used to define the axis of the bolt and list the grid points that lie in a cross section of the bolt.

Output

Use a FORCE case control command to obtain element force output for the CBEAM and CBAR elements
that represent bolts.

Use GPFORCE, BGRESULTS (when SYSTEM(610)=0), or MPCFORCES (when SYSTEM(610)=10) case

control commands to obtain force output for bolts modeled with solid elements.

Subcases

Subcases can be used to examine different combinations of bolts, bolt preload forces, and service loads,
among others.

• Use the BOLTLD case control command to select a BOLTLD bulk entry or BOLTFOR and BOLTFRC bulk
entries to include in a subcase. All BOLTFOR and BOLTFRC bulk entries used in a subcase should have
the same SID.

8-78 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Bolt Preload Procedure

• For the special case where the sparse solver is used, the model does not contain contact, and the
keyword SCRATCH = YES, the BOLTLD case control commands in the subcases must be ordered. For
example, if a model contains three subcases with BOLTLD = 1 used twice and BOLTLD = 2 used once,
the BOLTLD case control commands should be ordered with:

• The first and second subcases containing BOLTLD = 1 and the third subcase containing BOLTLD = 2.

• The first subcase containing BOLTLD = 2 and the second and third subcases containing BOLTLD = 1.

Otherwise, the run will terminate without results. If the iterative solver is used or contact is used or
the keyword SCRATCH = NO, ordering the subcases in this way is not necessary, but is recommended.
Doing so minimizes the number of matrix decompositions required during the solution.

• Subcases without BOLTLD case control commands should be placed at either the beginning or end of
the subcases.

Superelements and Bolt Preloads

Models with bolt preload can contain superelements. However, the elements used to define the bolts
along with the bolt preload forces must be in the residual.

Contact and Bolt Preloads

When a contact definition exists, the contact conditions are included in both solutions. To decrease the
solution time, the second solution begins with the contact status from the end of the first solution. As
a result, the contact element creation steps and some of the initial contact iterations are avoided in the
second solution.

Enforced Displacements and Bolt Preloads

Enforced displacements defined with non-zero SPC or SPCD* bulk entries can be included with a bolt
preload. As described in the Bolt Preload introduction, when a bolt preload is defined, the software
automatically solves the model twice; the bolt preload solution step and the consecutive solution step.
The software behaviour is different during these solution steps for the following input scenarios.

* NOTE: When an SPCD is defined with a bolt preload, system cell 581 must be set to 1 to override an
error trap that exists in the software. Also note the general, Simcenter Nastran requirement that DOF
included on an SPCD must also be fixed on an SPC or SPC1 bulk entry, and selected by the SPC case
control.

• Contact is not defined. A bolt preload and non-zero SPC or SPCD* entries are defined.

When contact conditions are not included with the bolt preload and enforced displacements, for
example, glue conditions are defined at the bolted interface, the bolt preload solution step is
performed with all SPC and SPCD DOF fixed, including the DOF where non-zero SPC or SPCD
specifications are defined. In other words, enforced displacements are not applied during the bolt
preload solution step.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 8-79

© 2024 Siemens
 8. Applied Loads

The resulting preload is then carried forward for the consecutive solution step, which includes the
service loads and the enforced displacements.

Case control example without contact conditions:

...
$CASE CONTROL
SUBCASE 1
SPC=1 (Selects fixed*, and non-zero SPC entries)
LOAD=2 (Selects service loads including SPCD* entries)
BOLTLD=4 (Selects bolt preload)
...

• Contact, a bolt preload, and non-zero SPCD* entry are defined.

When the bolt preload, the SPC (fixed DOF), the service loads (including the SPCD), and contact are
all selected in a single subcase, the bolt preload solution step includes the enforced displacements
(non-zero SPCD specifications). In most instances, this is not desired.

Single subcase case control example with undesired results:

...
$CASE CONTROL
SUBCASE 1
SPC=1 (Selects fixed* DOF)
LOAD=2 (Selects service loads including SPCD* entries)
BOLTLD=4 (Selects bolt preload)
BCSET=3 (Selects contact conditions)
...

To apply the non-zero SPCD definition for the consecutive solution step only, two subcases are
required. The bolt preload, the contact set, and the SPC set are defined globally (above the subcases).
The first subcase can be left empty, but must be defined. The second subcase selects the service loads,
which includes the SPCD entry.

Two subcase case control example with desired results:

...
$CASE CONTROL
BOLTLD=4 (Selects bolt preload for all subcases)
BCSET=3 (Selects contact conditions for all subcases)
SPC=1 (Selects fixed* DOF only for all subcases)
SUBCASE 1
SUBCASE 2
LOAD=2 (Selects service loads, which includes the SPCD* entries)

...

8-80 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Bolt Preload Example — Line Element Model

• Contact, a bolt preload, and a non-zero SPC entry are defined.

The software always applies the non-zero SPC definition for both the bolt preload solution step,
and the consecutive solution, regardless of the number of subcases. Since this is likely undesired
behaviour, it is recommended that you use the two subcase option with the SPCD entry described
above.

Bolt Preload Example — Line Element Model

A structure contains two bolts. One bolt has a M10 x 1.5 thread and the other has a M20 x 2.5 thread.
Both bolts are modeled using CBAR elements. Four subcases are considered:

• Bolt preload forces at 70% of proof strength and the service loads excluded.

• Bolt preload forces at 70% of proof strength and the service loads included.

• Bolt preload forces at 80% of proof strength and the service loads excluded.

• Bolt preload forces at 80% of proof strength and the service loads included.

The following table contains bolt and proof strength data for use in the model. The equivalent radius is
calculated from the tensile stress area.

Bolt size Grade Tensile Equivalent Proof Preload force Preload force Preload force
stress radius (mm) strength at 70% of at 80% of at 100% of
area (MPa) proof proof proof
(mm2) strength (kN) strength (kN) strength (kN)
M10 x 5.8 58 4.30 380 15.4 17.6 22.0
1.5
M20 x 5.8 245 8.83 380 65.2 74.5 93.1
2.5

Three input files are provided that model this situation:

• The first input file does not include a BOLTLD bulk entry. In this case, the BOLTLD case control
commands select the BOLTFOR bulk entries having the same SID.

• The second input file includes BOLTLD bulk entries. In this case, BOLTLD case control commands select
the BOLTLD bulk entry having the same SID.

• The third input file demonstrates how both BOLTLD bulk entries and BOLTFOR bulk entries can
be selected by BOLTLD case control commands in the same input file. In this case, the BOLTLD
case control commands in two subcases select a BOLTLD bulk entry and the BOLTLD case control
commands in the other two subcases select BOLTFOR bulk entries.

You can find the complete bulk data for this example in install_directory\scnas\nast\tpl\boltld03.dat.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 8-81

© 2024 Siemens
 8. Applied Loads

Input file without BOLTLD bulk entries

SOL 101
........
........
$ SUBCASES
$
SUBCASE = 1
SUBTITLE = Bolt preload force at 70% of proof strength without
service loads
BOLTLD = 1
$
SUBCASE = 2
SUBTITLE = Bolt preload force at 70% of proof strength with service
loads
BOLTLD = 1
LOAD = 1
$
SUBCASE = 3
SUBTITLE = Bolt preload force at 80% of proof strength without
service loads
BOLTLD = 2
$
SUBCASE = 4
SUBTITLE = Bolt preload force at 80% of proof strength with service
loads
BOLTLD = 2
LOAD = 1
$
BEGIN BULK
........
........
CBAR 2237 2 8001 8000 1.0 0.0 0.0
CBAR 2238 3 8003 8002 1.0 0.0 0.0
$
$ BOLT ENTRIES
$
BOLT 1 1 2237
BOLT 2 1 2238
$
$ BOLT PRELOAD ENTRIES
$
BOLTFOR 1 15.4+3 1
BOLTFOR 1 65.2+3 2
BOLTFOR 2 17.6+3 1
BOLTFOR 2 74.5+3 2
........
........
ENDDATA

8-82 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Bolt Preload Example — Line Element Model

Input file with BOLTLD bulk entries

SOL 101
........
........
$ SUBCASES
$
SUBCASE = 1
SUBTITLE = Bolt preload force at 70% of proof strength without
service loads
BOLTLD = 1
$
SUBCASE = 2
SUBTITLE = Bolt preload force at 70% of proof strength with service
loads
BOLTLD = 1
LOAD = 1
$
SUBCASE = 3
SUBTITLE = Bolt preload force at 80% of proof strength without
service loads
BOLTLD = 2
$
SUBCASE = 4
SUBTITLE = Bolt preload force at 80% of proof strength with service
loads
BOLTLD = 2
LOAD = 1
$
BEGIN BULK
........
........
CBAR 2237 2 8001 8000 1.0 0.0 0.0
CBAR 2238 3 8003 8002 1.0 0.0 0.0
$
$ BOLT ENTRIES
$
BOLT 1 1 2237
BOLT 2 1 2238
$
$ BOLT PRELOAD ENTRIES
$
BOLTFOR 1 22.0+3 1
BOLTFOR 2 93.1+3 2
$
$ SCALE BOLT PRELOAD ENTRIES
$
BOLTLD 1 0.70 1.0 1 1.0 2
BOLTLD 2 0.80 1.0 1 1.0 2

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 8-83

© 2024 Siemens
 8. Applied Loads

........
........
ENDDATA

Input file with BOLTLD case control commands selecting both BOLTLD and BOLTFOR bulk entries

SOL 101
........
........
$ SUBCASES
$
SUBCASE = 1
SUBTITLE = Bolt preload force at 70% of proof strength without
service loads
BOLTLD = 1
$
SUBCASE = 2
SUBTITLE = Bolt preload force at 70% of proof strength with service
loads
BOLTLD = 1
LOAD = 1
$
SUBCASE = 3
SUBTITLE = Bolt preload force at 80% of proof strength without
service loads
BOLTLD = 2
$
SUBCASE = 4
SUBTITLE = Bolt preload force at 80% of proof strength with service
loads
BOLTLD = 2
LOAD = 1
$
BEGIN BULK
........
........
CBAR 2237 2 8001 8000 1.0 0.0 0.0
CBAR 2238 3 8003 8002 1.0 0.0 0.0
$
$ BOLT ENTRIES
$
BOLT 1 1 2237
BOLT 2 1 2238
$
$ BOLT PRELOAD ENTRIES
$
BOLTFOR 2 17.6+3 1
BOLTFOR 2 74.5+3 2
BOLTFOR 3 22.0+3 1

8-84 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Bar and Beam Bolt Preload Method

BOLTFOR 3 93.1+3 2
$
$ SCALE BOLT PRELOAD ENTRIES
$
BOLTLD 1 0.70 1.0 3
........
........
ENDDATA

Other sample input files are available at nastran_installation\scnas\nast\tpl. Files with bolt preload are
typically named boltld*.dat.

Bar and Beam Bolt Preload Method

The following figure depicts a bolted connection where L denotes the undeformed thickness of the
medium being bolted together. Suppose the bolt is modeled using a CBEAM or CBAR element that
is positioned along the centerline of the actual bolt and has an undeformed length L. Typically, the
element representing the bolt is connected to the mesh of the bolted medium using RBE2 spider
elements.

Figure 8-21.

Because the bolt head is in static equilibrium, a free body diagram of the bolt head gives

where Fb denotes the force the body of the bolt exerts on the bolt head and Fm denotes the force the
bolted medium exerts on the bolt head.

In the absence of service loads, these forces arise exclusively because of bolt preload. Therefore

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 8-85

© 2024 Siemens
 8. Applied Loads

where Fp denotes bolt preload force.

When the bolted joint is tightened, the loaded portion of the bolt elongates and the bolted medium
compresses. However, as the joint is tightened, the actual length of the bolt under load decreases.
Therefore, the undeformed length of the bolt L defined by the model geometry is larger than the actual
length of the bolt under load. The difference in length is termed the preload length and is denoted as Δp.

The method outlined below accounts for the difference in length by applying an appropriately sized
compressive force Fapplied to the ends of the element representing the bolt. An expression for this force
can be obtained by first noting that

Equation 8-34.

where Δb is the elongation of the bolt due to the bolt preload force, and Δm is the compression of the
bolted medium due to the bolt preload force.

If the elongation of the bolt and compression of the bolted medium are small in comparison to L, the
elongation of the bolt due to the bolt preload force is given by

Equation 8-35.

where A and E are the tensile stress area and modulus of the bolt, respectively.

To find Δm, the software automatically solves the model with:

• The bolt preload force Fp applied compressively to the ends of the bolt.

• The bolt axial stiffness reduced by the BOLTFACT parameter.

• The service loads excluded from the solution.

Because the bolt is extremely compliant in the axial direction during this solution, nearly all of the
preload force compresses the bolted medium. As a result, the nodal deflections at each end of the bolt
(Ui and Uj) due solely to bolt preload are obtained.

8-86 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Bar and Beam Bolt Preload Method

Because Δm is a positive quantity when the bolted medium is compressed and the positive sense for
nodal deflections is directed from node i to node j

Equation 8-36.

Substituting Equation 35 and Equation 36 into Equation 34 gives

Equation 8-37.

Recalling that the force producing this deflection is denoted by Fapplied, we have

Equation 8-38.

Substituting Equation 37 into Equation 38 yields

Equation 8-39.

The software calculates Fapplied and then solves the model a second time with:

• Fapplied applied compressively to each end of the bolt.

• The bolt axial stiffness not reduced.

• The service loads optionally included in the solution.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 8-87

© 2024 Siemens
 8. Applied Loads

Therefore, the second solution of the run predicts the stress state resulting from both the bolt preload
and, if applicable, the service loads.

Frequently, a model contains multiple bolts. These bolts might have a variety of sizes, and be fabricated
from different materials, and have differing preload forces, and connect mediums of various thickness
and material type. For such a case, all the bolts are included during the first solution with:

• The appropriate preload forces applied compressively to each bolt.

• The axial stiffness of each bolt reduced by either a factor specified using the BOLTFACT parameter or
the default value of 1 x 107.

• The service loads excluded from the solution.

The software then calculates Fapplied for each bolt from Equation 39 using Fp, A, E, L, Ui and Uj
corresponding to each bolt. During the second solution:

• Each Fapplied is applied compressively to the appropriate bolt.

• The axial stiffness of each bolt is not reduced.

• The service loads are optionally included in the solution.

Therefore, the second solution of the run predicts the stress state resulting from both the bolt preload
and, if applicable, the service loads. In so doing, the software accounts for interactions between the
stress fields arising from each bolt preload force and, if applicable, the service loads.

When a bolt is modeled with CHEXA, CPENTA, and CTETRA elements the formulation is different. The
software splits the bolt mesh by:

• Duplicating each grid point listed on the BOLT entry. The grid IDs for the duplicated grid points start at
the highest user-defined grid ID in the model plus one and continue sequentially higher. The duplicate
grid points are written to the .op2 file and are visible when post-processing results.

• Modifying the element connectivity on one side of the cross section to use the duplicate grid points.
The modified connectivity is written to the .op2 file and is visible when post-processing results.

The software adds a compliant spring between each pair of coincident grid points. The springs have
stiffness in the axial direction of the bolt. The spring stiffness is calculated from:

EA / (L x BOLTFACT x N)

8-88 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Solid Bolt Preload Method

where E is the modulus of the bolt material, A is the projected area of the bolt cross section, L is the
undeformed length of the bolt, and N is the number of grid points selected on the BOLT entry. The value
for BOLTFACT is determined from the BOLTFACT parameter. For additional information on the BOLTFACT
parameter, see the Simcenter Nastran Quick Reference Guide.

The projected area of the cross section is calculated using the grid points listed on the BOLT entry,
and the bolt axis definition on the BOLT entry (the CSID and IDIR fields). This area is used to apply the
equivalent bolt preload as two equal and opposite surface tractions in the bolt axial direction.

A solution is run with only these surface tractions applied. The deflection at the bolt cross section due to
only bolt preload is obtained. These deflections are then used to calculate the initial strain in the bolt.

The software solves the model a second time with:

• The initial strain applied.

• The springs removed.

• The service loads applied if applicable.

Solid Bolt Preload Method

When a bolt is modeled with CHEXA, CPENTA, and CTETRA elements the formulation, the software splits
the bolt mesh by:

• Duplicating each grid point listed on the BOLT entry. The grid IDs for the duplicated grid points start at
the highest user-defined grid ID in the model plus one and continue sequentially higher. The duplicate
grid points are written to the .op2 file and are visible when post-processing results.

• Modifying the element connectivity on one side of the cross section to use the duplicate grid points.
The modified connectivity is written to the .op2 file and is visible when post-processing results.

The following occurs after the software splits the bolt.

1. The projected area of the cross section is calculated using the grid points listed on the BOLT entry,
and the bolt axis definition on the BOLT entry (the CSID and IDIR fields). This area is used to apply
the equivalent bolt preload as two equal and opposite surface tractions in the bolt axial direction.
The preload portion is run with only these surface tractions applied and the deflection at the bolt
cross section due to the bolt preload is found.

2. In the second portion of the solution, the software then maintains the relative axial displacement
at the bolt cut determined from the preload portion with either a glue or an MPC condition, and
applies any service loads you have defined.

Bolt preload approach

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 8-89

© 2024 Siemens
 8. Applied Loads

• By default, the software uses a glue approach at the cut to hold the grid pairs in their relative
deformed state. For the preload portion, the software zeroes the translational glue stiffness in the bolt
axial direction but keeps the rotational stiffness. Because the glue condition couples the rotational and
translational DOFs, a non-zero axial stiffness remains. As a result, the glue approach allows bending
loads to transfer from one side of the bolt cut to the other during the preload portion.

For the service load portion, the software holds the grid pairs at the cut in their relative deformed
state with a large translational and rotational glue stiffness. As a result, bending effects are transfered
at the bolt cut during the second portion also.

With the glue approach, you can recover bolt forces using the glue results requested with the
BGRESULTS command.

• You can select the pre-NX Nastran 12 MPC based approach by defining SYSTEM(610)=10. Because no
axial MPC condition is present during the preload portion, there is no mechanism to transfer bending
loads at the bolt cut.

For the second portion, the software holds the grid pairs in their relative deformed state with
rotational and translational MPC conditions between the two sides of the bolt cut. As a result, bending
effects are transfered at the bolt cut in the second portion.

With the MPC approach, you recover bolt forces using the MPC force results requested with the
MPCFORCES command.

Note:
The glue approach is not supported by the element iterative solver. If you run with the element
iterative solver and you have defined the ETYPE=2 bolt, the software will use the MPC approach.

External Force Fields

You can include forces from an external force field in your Simcenter Nastran model. Although this
capability could be adapted to a wide variety of problems, the initial application is to structural problems
where electromagnetic fields are present. Simcenter Nastran can convert the electromagnetic field
surface loads computed by a third-party electromagnetic simulation product like MAXWELL, to Simcenter
Nastran structural loads. You can then examine the mechanical response of a structural component to
an electromagnetic field both alone and in combination with other structural loads. This capability is
useful for analyzing structural components in motors or other electromechanical devices.

External field surface loads are the components of force that act at discrete points on the free faces of
the external field simulation model. These discrete points are referred to as external force points. Surface
patches are either planar surfaces or cylindrical surfaces defined on free faces of the Simcenter Nastran
model that correspond to free faces of the external field simulation model. You define surface patches by
selecting sets of structural grids in the Simcenter Nastran model.

Simcenter Nastran uses a surface spline technique to convert the external field surface loads acting at
external force points that you select to mechanically equivalent structural loads acting at structural grids

8-90 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 EFPOINTS and EFFORCE Matrices

on the surface patch. The surface spline technique allows you to select external force points whose
locations do not exactly coincide with the locations of the structural grids that define the surface patch.
However, to obtain the most accurate surface load to structural load conversion, the projections of all
the selected external force points in the direction normal to the surface patch should fall within the
envelope of the surface patch.

The surface spline technique used by Simcenter Nastran is also used in aeroelastic analysis. For
more information regarding the surface spline technique, see “Interconnection of the Structure with
Aerodynamics” in the Simcenter Nastran Aeroelastic Analysis User’s Guide.

Simcenter Nastran can convert the surface loads from an external field solution to structural loads for
the linear static (SOL 101), normal modes (SOL 103), direct frequency response (SOL 108) solution
sequences, and for a rotor dynamic analysis with the direct complex eigenvalue solution sequence (SOL
107). For a linear static solution, the converted surface loads represent static loads that contribute
to the overall loading of the structural component. For a normal modes solution or a rotor dynamic
analysis with the direct complex eigenvalue solution, the converted surface loads represent static loads
that prestress the structural component. For a direct frequency response solution, the converted surface
loads represent scaling factors for the dynamic loads that excite the structural component.

EFPOINTS and EFFORCE Matrices

For Simcenter Nastran to convert surface loads from an external field solution to Simcenter Nastran
structural loads, you must use direct matrix input (DMI) to define an external force points (EFPOINTS)
matrix and a corresponding external field force (EFFORCE) matrix for each surface patch.

• EFPOINTS matrices contain the spatial coordinates of external force points that you select. To follow
the geometry conveniently, the coordinate basis for this data can be either rectangular or cylindrical.

• EFFORCE matrices contain external field surface loads that act at external force points defined in
the corresponding EFPOINTS matrices. The coordinate basis of these force components must be
rectangular.

For information on how to create EFPOINTS and EFFORCE matrices, see Creating an EFPOINTS Matrix
and Creating an EFFORCE Matrix.

Simcenter Nastran Input Requirements

A single Simcenter Nastran input file and solution is required for a linear static (SOL 101) or direct
frequency response (SOL 108) analysis. In the case control section of the input file, you must define a
separate subcase for each surface patch for which you want to convert the external field surface loads to
structural loads. Each of these subcases must include:

• A P2G case control command to select the applicable EFPOINTS and EFFORCE matrices from the bulk
section.

P2G = EFPOINTS, EFFORCE

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 8-91

© 2024 Siemens
 8. Applied Loads

• A SET case control command to select structural grids.

SET 3 = 25, 28, 30 THRU 42

• A PARTN case control command to define the set of selected structural grids as a surface patch.

PARTN 3

• An EFLOAD case control command to request that the external field solution surface loads are
converted to structural loads and to define the direction normal to the surface patch.

EFLOAD (NDIR = 1, SCID = 7) = YES

The final subcase in the input file can only include output requests and other structural loads. You cannot
include an EFLOAD case control command in the final subcase. Thus, for the case where external field
surface loads are converted to structural loads for only one surface patch, two subcases are required.

In the bulk section of the input file, you must include either DMI bulk entries defining the EFPOINTS
and EFFORCE matrices or INCLUDE bulk entries that reference external files that contain these DMI bulk
entries. For instructions on how to use DMI bulk entries to define EFPOINTS and EFFORCE matrices, see
Creating an EFPOINTS Matrix and Creating an EFFORCE Matrix.

Normal modes (SOL 103)

For a normal modes (SOL 103) analysis, external field surface loads represent a static loading that
prestresses the structure. To account for static prestress due to external field surface loads, two solves
are required. The first solve is a linear static (SOL 101) analysis where the external field surface loads are
converted to structural loads. The input file requirements for this solve are identical to those outlined
previously. During this solve, the presence of an EFLOAD case control command signals Simcenter
Nastran to automatically write a punch (.pch) file that contains the structural loads that were converted
from the external field surface loads.

The second solve is a normal modes (SOL 103) analysis. During this solution, the structural loads
converted from the external field surface loads during the linear static analysis are used to statically
prestress the structure. The input file is similar to that used to perform a normal modes analysis in which
differential stiffness is computed from static loads, except that you must include:

• EFLOAD = YES and no more in the first subcase, which is the static subcase. In the second subcase,
which is the normal modes subcase, you must include STATSUB = 1 and can optionally define output
requests and so on. The second subcase cannot contain an EFLOAD case control command.

SUBCASE 1
EFLOAD = YES
SUBCASE 2
STATSUB = 1
.....
.....

8-92 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Simcenter Nastran Input Requirements

• An INCLUDE bulk entry that references the punch file created during the linear static solve. Simcenter
Nastran then uses the structural loads in the punch file as static preload forces during the normal
modes solution.

INCLUDE ‘filename.pch’

Because the EFPOINTS and EFFORCE matrices are not used during the normal modes solution, the DMI
bulk entries that define them can optionally be omitted from the normal modes solution input file.

Rotor dynamic analysis (SOL 107)

An external force field can be included in a rotor dynamic analysis with the direct complex eigenvalue
solution sequence (SOL 107). The three solution steps are described as follows.

Run 1:

A linear static analysis (SOL 101) is executed with the standard external force inputs. The result of
this analysis is a punch file that contains the combined load vector representing the external field.
This simplified input example shows the external force field inputs required for this step.

The complete input file rotefa.dat is available in install_directory\scnas\nast\tpl\ to demonstrate

this step.

SOL 101
CEND
...
$* CASE CONTROL
...
Case control inputs for the external field
SET 1 = 5, 6, 7, 45, 46, 47
PARTN 1
p2g = ep2, ef2
efload (nDIR=3, SCID=9)=yes
....
BEGIN BULK

This coordinate system is referenced by the SCID case control.

CORD2C, 9, , 0.0, 0.0, 0.0, 0.0, 0.0, 1.0, +C1
+C1, 20.0, 0.0, 1.0
....
The EP2 matrix defines locations in the force field.
dmi,ep2,0,2,1,0,,3,2
dmi,ep2,1,1, 20.0, 180.0, 80.0
dmi,ep2,2,1, 20.0, 270.0, 80.0
....
The EF2 matrix defines the force components
at the locations defined in the EP2 matrix.
dmi,ef2,0,2,1,0,,3,2

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 8-93

© 2024 Siemens
 8. Applied Loads

dmi,ef2,1,1, -100.0, 0.0, 0.0

dmi,ef2,2,1, 0.0, -100.0, 0.0
....
ENDDATA

Run 2:

This SOL 107 run uses the force matrix stored in the punch file from run 1 to compute the
differential stiffness matrix due to the external field loads. Run 2 stores the differential stiffness
matrix into a new punch file, and then exits. It is an incomplete run since it does not compute the
complex eigenvalues. This computation will occur in run 3.

The input file for this run is defined with two subcases. Subcase 1 contains an EFLOAD=yes case
control command with no other data. Subcase 2 contains a STATSUB entry referring to the first
subcase. The bulk data section must include the punch file generated by run 1. The execution of
this abbreviated run is triggered by the presence of the EFLOAD case control command in the first
subcase.

The complete input file rotefb.dat is available in install_directory\scnas\nast\tpl\ to demonstrate

this step.

SOL 107
CEND
...
$* CASE CONTROL
...
The EFLOAD command is defined in subcase 1
SUBCASE 1
efload = yes
...
The STATSUB command is defined in subcase 2
SUBCASE 2
statsub = 1
...
BEGIN BULK
...
The punch file generated by run 1 is included here
include 'OUTDIR:rotefa.pch'
...
ENDDATA

Run 3:

This run is the complete rotor dynamic analysis, and includes the differential stiffness matrix written
to the punch file created in run 2. Your SOL 107 input file for this run will include any typical rotor
dynamic inputs, it will include EFLOAD=yes in the case control, and the contents of the run 2 punch
file in the bulk data section. The EFLOAD command will trigger the software to add the differential
stiffness matrix to the standard stiffness matrix before the rotor dynamic matrices are generated.

8-94 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Using the EFLOAD Case Control Command

The complete input file rotefc.dat is available in install_directory\scnas\nast\tpl\ to demonstrate

this step.

SOL 107
$
CEND
...
$* CASE CONTROL
The EFLOAD command will trigger the software to
include the differential stiffness matrix
...
efload = yes
...
spc = 1
cmethod = 1
rmethod = 2
...
OUTPUT
DISPLACEMENT(PRINT) = ALL
...
BEGIN BULK
...
Typical rotor dynamic analysis inputs
EIGC 1 CLAN 5
ROTORD 2 0.0 250.0 3 ROT 0.0 RPM
HZ +ROT0
$ ZSTEIN ORBEPS ROTPRT
+ROT0 NO 1.0E-5
3 +ROT1
$ RID1 RSET1 RSPEED1 RCORD1 W3-1 W4-1 RFORCE1
+ROT1 1 3 1.0 1 0. 0.
...
ROTORG 3 1 THRU 64
...
The punch file generated by run 2 is included here
include 'OUTDIR:rotefb.pch'
...
ENDDATA

Using the EFLOAD Case Control Command

The EFLOAD case control command has the following format:

EFLOAD (NDIR = n, SCID = m) = YES/NO

If you specify the YES describer, the external field surface loads are converted to structural loads.
Conversion does not occur if you specify the NO describer.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 8-95

© 2024 Siemens
 8. Applied Loads

The use of the describers NDIR = n and SCID = m depends on the shape and orientation of the surface
patch that is defined in the same subcase. You use the describer NDIR = n to specify the normal to the
surface patch. You use the describer SCID = m to specify that the surface patch is cylindrical.

• For a planar surface patch lying on either an xy-, yz-, or zx-plane, you should specify the describer
NDIR = n and omit the describer SCID = m. If the surface patch normal is the z-direction, the describer
NDIR = n can be omitted because n = 1 is the default.

• For a cylindrical surface patch or a planar surface patch lying in either an rθ- or rz-plane, you should
specify both the describer NDIR = n and the describer SCID = m. Regardless of the local coordinate
system referenced by describer SCID = m, Simcenter Nastran uses a cylindrical coordinate system
whose:

• Origin coincides with that of the basic coordinate system.

• Cylindrical axis coincides with the z-axis of the basic coordinate system.

• Tangential coordinate, θ, is measured from the x-axis of the basic coordinate system. (0 ≤ θ ≤ 360
degrees)

Table 8-44 lists the allowable values for n and the corresponding normal directions in rectangular and
cylindrical coordinate systems.

Table. Normal Direction for Surface Patches

n Rectangular CSYS Cylindrical CSYS

1 (default) z z
2 y θ
3 x r

Creating an EFPOINTS Matrix

An external force points (EFPOINTS) matrix specifies the spatial coordinates of external force points
corresponding to a surface patch. You can express the spatial coordinates in terms of either a
rectangular or cylindrical coordinate system.

• For a rectangular coordinate system, the spatial coordinates must be expressed in terms of the basic
coordinate system of the Simcenter Nastran model.

• For a cylindrical coordinate system, the spatial coordinates must be expressed in terms of a cylindrical
system whose:

• Origin coincides with that of the basic coordinate system of the Simcenter Nastran model.

• Cylindrical axis coincides with the z-axis of the basic coordinate system of the Simcenter Nastran
model.

8-96 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Creating an EFFORCE Matrix

• Tangential coordinate, θ, is measured from the x-axis of the basic coordinate system of the
Simcenter Nastran model. (0 ≤ θ ≤ 360 degrees)

The first row of the DMI bulk entry specifies parameters for the EFPOINTS matrix. For real, single
precision input of spatial coordinates, the format is as follows:

dmi,name,0,2,1,0,,3,N

where the EFPOINTS matrix can be named as you like and N is the number of external force points.
The remaining rows in the DMI bulk entry contain the spatial coordinates. One row is required for each
external force point. The format is as follows:

dmi,name,1,1,X1,Y1,Z1
.....
.....
dmi,name,n,1,Xi,Yi,Zi
.....
.....
dmi,name,N,1,XN,YN,ZN

where name must correspond with the first DMI entry, and Xi, Yi, and Zi are the spatial coordinates of
the ith external force point in the x-, y-, and z-direction, respectively.

For cylindrical coordinates, the format is identical except Xi, Yi, and Zi are replaced by Ri, θi, and Zi,
respectively. Ri, θi, and Zi are the spatial coordinates of the ith external force point in the r-, θ-, and
z-direction, respectively.

Creating an EFFORCE Matrix

An external field force (EFFORCE) matrix specifies the components of force at the external force points
of the corresponding EFPOINTS matrix. The force components must be expressed in terms of the
basic coordinate system of the Simcenter Nastran model. The first row of the DMI bulk entry specifies
parameters for the EFFORCE matrix. For real, single precision input of force components, the format is as
follows:

dmi,name,0,2,1,0,,3,N

where the EFFORCE matrix can also be named as you like and N is the number of external force points in
the corresponding EFPOINTS matrix. The remaining rows of the DMI bulk entry contain the components
of force. One row is required for each external force point. The format is as follows:

dmi,name,1,1,FX1,FY1,FZ1
.....
.....
dmi,name,n,1,FXi,FYi,FZi
.....

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 8-97

© 2024 Siemens
 8. Applied Loads

.....
dmi,name,N,1,FXN,FYN,FZN

where name must correspond with the first DMI entry, and FXi, FYi, and FZi are the components of
force that act at the ith external force point in the x-, y-, and z-direction, respectively.

Linear Static and Normal Modes Example

In this example, a linear static analysis is performed on a cylindrical tube that has one end fixed to
ground and the other end subjected to an external force field that produces external field surface loads.
Then, the converted external field surface loads from the linear static analysis are used to prestress the
cylinder during a normal modes analysis.

The cylinder is modeled with CQUAD4 elements. The bottom edge of the Simcenter Nastran model is
fixed to ground. The other end of the tube is subjected to an external force field. The external field
surface loads act on a ring of external force points that lie midway between the top two rings of
structural grids in the Simcenter Nastran model.

Structural grids used to define the surface patch

The relevant portions of the Simcenter Nastran input file are:

ASSIGN PUNCH = static_loads.pch,NEW,UNIT=7

$
SOL 101
$
CEND
$
$ CASE CONTROL
$
ECHO = NONE
SPC = 1
$
SUBCASE 1
$
$ The P2G case control command selects the EFPOINTS and EFFORCE matrices
from the bulk
$ section. The EFPOINTS matrix for this surface patch is named ‘ep1’.

8-98 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Linear Static and Normal Modes Example

The EFFORCE matrix

$ for this surface patch is named ‘ef1’.
$
P2G = ep1, ef1
$
$ The SET and PARTN case control commands define the surface patch
$
SET 1 = 1 thru 8, 44 thru 50, 58
PARTN 1
$
$ The EFLOAD case control command requests that external field surface
loads be converted
$ to structural loads. SCID=9 specifies that the surface patch is a
cylindrical surface.
$ NDIR=3 specifies that the normal to the surface patch is in the radial
direction.
$
$
EFLOAD (NDIR=3, SCID=9) = YES
$
SUBCASE 2
OUTPUT
DISPLACEMENT(PRINT) = ALL
STRESS(PRINT) = ALL
$
BEGIN BULK
$
$ The CORD2C bulk entry specifies a cylindrical local coordinate system
used to specify
$ the structural grids.
$
CORD2C, 9, , 0.0, 0.0, 0.0, 0.0, 0.0, 1.0, +C1
+C1, 20.0, 0.0, 1.0
$
$ PARAM CARDS
$
PARAM K6ROT 100.0
PARAM POSTEXT YES
PARAM UNITSYS MN-MM
$
$ GRID CARDS
$
GRID 1 9 20.07.016-15 80.0 0
.....
GRID 64 9 20.0 0.0 0.0 0
$
$ ELEMENT CARDS
$
CQUAD4 1 1 9 16 17 10

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 8-99

© 2024 Siemens
 8. Applied Loads

.....
CQUAD4 56 1 57 9 10 56
$
$ PROPERTY CARDS
$
PSHELL 1 1 1.8 1 1.0 1.8333333 0.0
$
$ MATERIAL CARDS
$
MAT1 1 2.0+5 .25 7.872-9
$
$ LOAD AND CONSTRAINT CARDS
$
SPC 1 51 123456 0.0
.....
SPC 1 64 123456 0.0
$
$ The EFPOINTS and EFFORCE matrices are constructed from the external
field solution.
$ The following DMI bulk entries define an EFPOINTS matrix named ‘ep1’.
$
dmi,ep1,0,2,1,0,,3,8
dmi,ep1,1,1, 20. , 0.0, 74.286
dmi,ep1,2,1, 20. , 45., 74.286
dmi,ep1,3,1, 20. , 90.0, 74.286
dmi,ep1,4,1, 20. , 135.0, 74.286
dmi,ep1,5,1, 20. , 180.0, 74.286
dmi,ep1,6,1, 20. , 225.0, 74.286
dmi,ep1,7,1, 20. , 270.0, 74.286
dmi,ep1,8,1, 20. , 315.0, 74.286
$
$ The following DMI bulk entries define an EFFORCE matrix named ‘ef1’.
Because external field
$ surface loads are always expressed relative to a rectangular
coordinate basis, the DMI bulk
$ entries define an external field surface load of FX = 100, FY = 0, and
FZ = 100 acting at
$ each external force point.
$
dmi,ef1,0,2,1,0,,3,8
dmi,ef1,1,1, 100.0, 0.0, 100.
dmi,ef1,2,1, 100.0, 0.0, 100.
dmi,ef1,3,1, 100.0, 0.0, 100.
dmi,ef1,4,1, 100.0, 0.0, 100.
dmi,ef1,5,1, 100.0, 0.0, 100.
dmi,ef1,6,1, 100.0, 0.0, 100.
dmi,ef1,7,1, 100.0, 0.0, 100.
dmi,ef1,8,1, 100.0, 0.0, 100.

8-100 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Linear Static and Normal Modes Example

$
ENDDATA

Note:
The use of an ASSIGN statement to specify the name of the punch file that is written during the
solve is optional.

A second run is used to perform a normal modes analysis on the cylinder. During this run, the converted
external field surface loads are used to prestress the cylinder. The relevant components of the input file
are:

SOL 103
$
CEND
$
$ CASE CONTROL
$
ECHO = NONE
SPC = 1
$
SUBCASE 1
$
EFLOAD = YES
$
SUBCASE 2
STATSUB = 1
METHOD = 99
OUTPUT
DISPLACEMENT(PRINT) = ALL
STRESS(PRINT) = ALL
$
BEGIN BULK
$
EIGRL, 99, , , 9
$
$ The CORD2C bulk entry specifies a cylindrical local coordinate system
used to specify
$ the structural grids.
$
CORD2C, 9, , 0.0, 0.0, 0.0, 0.0, 0.0, 1.0, +C1
+C1, 20.0, 0.0, 1.0
$
$ PARAM CARDS
$
PARAM K6ROT 100.0
PARAM POSTEXT YES
PARAM UNITSYS MN-MM

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 8-101

© 2024 Siemens
 8. Applied Loads

$
$ GRID CARDS
$
GRID 1 9 20.07.016-15 80.0 0
.....
GRID 64 9 20.0 0.0 0.0 0
$
$ ELEMENT CARDS
$
CQUAD4 1 1 9 16 17 10
.....
CQUAD4 56 1 57 9 10 56
$
$ PROPERTY CARDS
$
PSHELL 1 1 1.8 1 1.0 1.8333333 0.0
$
$ MATERIAL CARDS
$
MAT1 1 2.0+5 .25 7.872-9
$
$ LOAD AND CONSTRAINT CARDS
$
SPC 1 51 123456 0.0
.....
SPC 1 64 123456 0.0
$
$ The linear static solve wrote a punch file that contains the preload
forces at the structural
$ grids. The INCLUDE bulk entry references this file.
$
INCLUDE static_loads.pch
$
ENDDATA

8-102 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 9. Constraints
Overview of Constraints
A constraint is the enforcement of a prescribed displacement (i.e., component of translation or rotation)
on a grid point or points. There are two basic types of constraints in Simcenter Nastran: single point
constraints (SPCs) and multipoint constraints (MPCs).

• A single point constraint is a constraint applied to an individual grid point. Single point constraints can
enforce either zero displacement or nonzero displacement.

• A multipoint constraint is a mathematical constraint relationship between one grid point and another
grid point (or set of grid points).

The boundary conditions of a static structure (fixed, hinged, roller support, etc.) typically require
that various degrees of freedom be constrained to zero displacement. For example, consider a grid
point fixed in a rigid wall. All six displacement degrees of freedom—three translational directions and
three rotational directions—must be constrained to zero to mathematically describe the fixed boundary
condition.

Real world structures often don't have simple or ideal boundary conditions. Because a model's
constraints greatly influences its response to loading, you must try to constrain your model as accurately
as possible.

This chapter describes how you apply constraints. To understand how constraints are processed, you
need to be familiar with the Simcenter Nastran set notation and matrix operations.

Single-point Constraints
A single-point constraint (SPC) applies a fixed value to a translational or rotational component at an
individual grid point or to a scalar point. Common uses of single-point constraints are to specify
the boundary conditions of a structural model by fixing the appropriate degrees-of-freedom and to
eliminate unwanted degrees-of-freedom with zero stiffness. Multiple sets of single-point constraints can
be provided in the Bulk Data Section, with selections made at execution time by using the subcase
structure in the Case Control Section. This procedure is particularly useful in the solution of problems
having one or more planes of symmetry.

For more information about subcases, see Defining Subcases.

You can use single-point constraints to:

• To constrain a structure to ground.

• To apply symmetric or antisymmetric boundary conditions by restraining the degrees of freedom that
must have a zero value to satisfy symmetry or antisymmetry.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 9-1

© 2024 Siemens
 9. Constraints

• To remove degrees of freedom that are not used in the structural analysis (i.e., are not connected to
any structural elements or otherwise joined to the structure).

• To remove degrees of freedom that are very weakly coupled to the structure. This condition can occur,
for example, to the rotations about the normal of a slightly curved shell. In this case, a judgment must
be made whether to remove the degree of freedom using an SPC (in which case the structure may be
over-constrained), or to leave it in the problem (in which case the stiffness matrix is nearly singular).
A reasonable rule is to constrain the degree of freedom if its stiffness is less than 10-8 as large as the
stiffness in another direction at the same grid point.

The elements connected to a grid point may not provide resistance to motion in certain directions,
causing the stiffness matrix to be singular. Single-point constraints are used to remove these degrees-of-
freedom from the stiffness matrix.

For example, consider a planar structure composed of membrane and extensional elements. The
translations normal to the plane and all three rotational degrees-of-freedom must be constrained since
the corresponding stiffness matrix terms are all zero. If a grid point has a direction of zero stiffness, the
single-point constraint doesn't need to be exactly in that direction. It only needs to have a component
in that direction. This allows the use of single-point constraints for the removal of such singularities
regardless of the orientation of the global coordinate system. Although the displacements will depend
on the direction of the constraint, the internal forces will be unaffected.

Defining Single point Constraints in Simcenter Nastran

There are several different ways that you can define single point constraints:

• You can define single point constraints on SPC, SPC1 and SPCADD bulk data entries.

• You can define single point constraints on the GRID and GRDSET entries.

• You can use the PARAM,AUTOSPC,YES option to have the software automatically create SPCs to
remove degrees of freedom that are either unconnected or very weakly coupled.

You can obtain structural reaction forces called single point constraint forces (SPCFs) at grid points
constrained by SPCs. You can request printed output for single point forces in the Case Control section.
The software prints all nonzero forces, whether they originate from SPCi entries, the PS field on GRID
entries, or PARAM,AUTOSPC,YES.

When you apply a single-point constraint to remove a singularity, the restrained component of motion
to doesn't need to be aligned exactly with the singular direction of motion (however, it is highly
recommended). Consider the pair of co-linear pin-connected rods, shown in Figure 9-1, that permit
unrestrained motion at point G in any direction perpendicular to the axis of the rods.

9-2 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Using GRID Entries

Figure 9-1. Pin-Connected Rods

After SPCs are applied to the translation in the z-direction and to all three components of rotation,
unrestrained motion is still possible in the n-direction. An SPC in either the x- or the y-direction removes
the remaining singularity without disturbing the axial forces in the rods (in this special case). Note,
however, that the motion at point G has a zero component in the restrained direction, which may
produce incorrect results for the displacements. For this reason, you should select a local coordinate
system whose coordinate axes align with the singular components of motion.

Using GRID Entries

You can also define single point constraints using the PS option (field 8) on the GRID and GRDSET bulk
data entries. When you define constraints this way, the constraints are permanent and are constrained
for all subcases; you can't select them in the Case Control section. Their primary use, in addition to
applying permanent supports, is to remove unconnected or very weakly connected degrees of freedom.

The default value for enforced displacement on points that you constrain on GRID entries is zero. You
can override the default value at the subcase level with SPC entries.

See also

• “Specifying Permanent Single Point Constraints (Field 8)” in the Simcenter Nastran User's Guide

• “GRID” in the Simcenter Nastran Quick Reference Guide

• “GRDSET” in the Simcenter Nastran Quick Reference Guide

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 9-3

© 2024 Siemens
 9. Constraints

Grid Example 1: Cantilever Beam

In this example, we constrained Grid 1 in all six DOFs using field 8 (permanent single point constraint) of
the GRID Bulk Data entry:

1 2 3 4 5 6 7 8 9 10
GRID ID CP X1 X2 X3 CD PS SEID
GRID 1 0. 0. 0. 123456

Or, in free field format,

GRID,1,,0.,0.,0.,,123456

Note that with this option, you don't need to specify a constraint set in the Case Control section.

Using SPC and SPC1

You can define single point constraints on SPC, SPC1, and SPCADD bulk data entries.

• The SPC entry is the most general way of specifying single point constraints.

• The SPC1 entry is a less general entry that is more convenient when several grid points have the same
components constrained to a zero displacement.

• The SPCADD entry defines a union of single point constraint sets specified with SPC or SPC1 entries.

You should use the SPC and SPC1 Bulk Data entries if you need to apply different sets of boundary
conditions in different subcases. The constraints specified on the SPC and SPC1 entries belong to sets
identified by set identification numbers (SIDs) that must be selected in the Case Control Section to be
used. You can use the SPCADD entry in the Case Control section to merge individual SPC and SPC1 sets
into combined sets.

9-4 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 SPC Example 1

The format of the SPC entry is as follows:

1 2 3 4 5 6 7 8 9 10
SPC SID G1 C1 D1 G2 C2 D2

Field Contents

SID Identification number of the single-point constraint set.

Gi Grid or scalar point identification number.

Ci Component number.

Di Value of enforced displacement for all degrees of freedom designated by Gi and

Ci.

The SPC entry can also be used to specify a nonzero displacement constraint. The values specified in
fields 5 and 8 are enforced displacements for the components specified in fields 4 and 7, respectively.
When using an enforced constraint within multiple subcases, it is more efficient to use the SPCD entry.

See also

• “SPC” in the Simcenter Nastran Quick Reference Guide

• “SPC1” in the Simcenter Nastran Quick Reference Guide

• “SPCADD” in the Simcenter Nastran Quick Reference Guide

SPC Example 1

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 9-5

© 2024 Siemens
 9. Constraints

Grids 1, 2, 3, 4, 6, 7, 8, and 9 must be constrained in all six DOFs. Assume that a single point constraint
selection is defined in the Case Control Section (SPC = 42). Applying the same constraint on numerous
individual GRID Bulk Data entries is not particularly convenient-an SPC1 entry is a better choice:

1 2 3 4 5 6 7 8 9 10
SPC1 SID C G1 G2 G3 G4 G5 G6
G7 G8 G9 -etc.-

SPC1 42 123456 1 2 3 4 6 7
8 9

Or, in free field format,

SPC1,42,123456,1,2,3,4,6,7

,8,9

Note that if the “THRU” (alternate format) option were used (1 THRU 9), grid point 5 would have been
incorrectly fixed, and no displacement would occur anywhere in the model.

SCP Example 2: Fixed Hinged Beam

Here, an SPC entry is used to constrain both ends of the beam. Assume that a single point constraint set
selection is defined in the Case Control Section (SPC = 42). The values of enforced displacement (Di) are
left blank since zero displacement is the default.

9-6 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Automatically Applying Single-point Constraints

1 2 3 4 5 6 7 8 9 10
SPC SID G1 C1 D1 G2 C2 D2

SPC 42 1 123456 5 12346

Or, in free field format,

SPC,42,1,123456,,5,12346

Automatically Applying Single-point Constraints

Simcenter Nastran automatically identifies and constrains singularities in the stiffness matrix using a
parameter called AUTOSPC. When you add PARAM,AUTOSPC,YES to either the bulk data or case control
sections of your input file, Simcenter Nastran automatically removes degrees of freedom that are either
unconnected or very weakly coupled to the finite element model.

With AUTOSPC, you use the PARAM,EPZERO option to specify the minimum stiffness ratio the software
should use to determine a singularity. The stiffness ratio is the ratio of the stiffness in the weakest
direction to the stiffness in the strongest direction at the grid point (considering all possible directions
rather than just the coordinate directions). The software considers translational and rotational stiffness
separately.

See also

• “AUTOSPC” in the Simcenter Nastran Quick Reference Guide

• “EPZERO” in the Simcenter Nastran Quick Reference Guide

Grid Point Singularity Processor

When you use the AUTOSPC parameter, the software uses the Grid Point Singularity Processor (GPSP)
module to identify potential singularities. When you set the AUTOSPC parameter to YES, the GPSP
constrains the degrees of freedom associated with the obvious singularities. The GPSP identifies local
singularities in the stiffness matrix after the software processes all the SPC and MPC constraints in your
model.

The software begins identifying local grid point singularities by examining the 3 · 3 matrices that
represent the three translations and three rotations of each grid point. For each 3 · 3 matrix, the
software performs an eigensolution to obtain the principal values (eigenvalues) and the principal
directions (eigenvectors). The principal values are normalized to obtain a measure Ri as shown in Eq.
9-1:

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 9-7

© 2024 Siemens
 9. Constraints

Equation 9-1.

Ki = the diagonal stiffness term corresponding to the i-th degree of freedom

KMAX = the maximum principal stiffness

Any principal value for which |Ri| < EPZERO represents a singularity. The software examines the principal
direction corresponding to the singularity and identifies the component with the largest ratio as the
failed direction for the grid point. The software doesn't constrain any degrees of freedom that you
specify as belonging to the a-set.

See also

• “Constraint and Mechanism Problem Identification in SubDMAP SEKR” in the Simcenter Nastran
User's Guide

Understanding the Grid Point Singularity Table

When the GPSP examines the stiffness matrix for singularities at the grid point level, it lists any
singularities in the Grid Point Singularity Table (GPST). The software prints the GPST in the .f06 file.
The GPST lists all singular degrees-of-freedom, in the global coordinate system, and the ratio of
stiffness between the softest and stiffest degree-of-freedom for the grid point. It also lists any degrees-
of-freedom that the software automatically constrained.

You should always review the Grid Point Singularity Table carefully to understand which DOFs were
singular and why. Automatic constraint generation can mask actual modeling errors, such as your
forgetting to specify bending stiffness (via MID2 on the PSHELL entry).

The output includes a grid point singularity table with the following format:

• The TYPE specifies whether the failed degree of freedom is a component of a grid point (G) of a scalar
point (S).

• The component of motion that failed. Constraints are applied in the global coordinate direction
closest to the weakest direction as defined above.

• If possible, the failed degrees of freedom are placed in the sb-set if AUTOSPC is set to YES (the default
for Solution 101). If AUTOSPC is set to NO, the singularity is identified but not constrained and is not
placed in the sb-set.

9-8 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Understanding the Grid Point Singularity Table

• An asterisk (*) is printed at the end of the line and indicates that action was taken by Simcenter
Nastran.

To illustrate how failing to inspect the GPSP output can lead to a potential problem, consider the single
CBAR element example shown in Figure 9-2.

Figure 9-2. Single Element with Vertical Load

The CBAR element is fully constrained at end A and is loaded with a 100 N force in the Yb-direction at
end B. The input file for this model is shown in Listing 9-4. Both grid points are defined in the basic
coordinate system. The displacement coordinate system for grid point 1 is the basic system, but the
displacement coordinate system for grid point 2 is in the local coordinate system 1. It is typical to define
a local coordinate system when you need a displacement component in a specific direction.

A sample Grid Point Singularity Table is shown in Figure 9-3.

Figure 9-3. Grid Point Singularity Table

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 9-9

© 2024 Siemens
 9. Constraints

In this table, grid point 2, the free end of the rod, failed (i.e. had no stiffness connection) in five of its
six possible DOFs. DOF 2 (translation in the Y direction) of grid point 2 wasn't singular because it is
attached to the extensional stiffness of the rod. CROD elements support extension and torsion, but in
this example no value of J, the torsional constant, was provided in field 5 of the PROD element property
entry. As a result, the CROD element had no torsional stiffness. The resulting unconnected DOF was
detected and constrained by AUTOSPC.

$ FILENAME - BAR3A.DAT
ID LINEAR,BAR3A
SOL 101
TIME 5
CEND
TITLE = BAR WITH I1 AND I2
SPC = 1
LOAD = 1
FORCE = ALL
DISP = ALL
BEGIN BULK
PARAM POST 0
PARAM AUTOSPC YES
CORD2R 1 0 0.0 0.0 0.0 0.0 .7071068.7071068+
1. 0.0 0.0
$
GRID 1 0.0 0.0 0.0
GRID 2 10. 0.0 0.0 1
$
CBAR 1 1 1 2 1. 1. 0.0
$
SPC 1 1 123456 0.0
$
PBAR 1 1 1. .667 .167
FORCE 1 2 100. 0. 1. 0.
MAT1 1 1.+7 .3
ENDDATA

Figure 9-4. Single Element with a Vertical Load

For this example, the displacement should be exclusively in the Yb-direction. Since the displacement
is computed and output in the CD coordinate system, you expect that the displacement components
T2 (the Y1-direction) and T3 (the Z1-direction) for grid point 2 are the same for this model. The
displacement and force results shown in Figure 9-5 confirm this assumption.

9-10 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Enforced Displacements at Grid Points (SPCD, SPC)

Figure 9-5. Force and Displacement at Grid Point 2 with a Nonzero I2

However, suppose that when you construct the model, you know that the deflection is in plane 1 of the
element only, so you leave the I2 field blank. The resulting grid point singularity table, element forces,
and displacements are shown in Figure 9-6

Figure 9-6. Force and Displacement at Grid Point 2 with a Zero I2

As can be seen, the GPSP module identified a singularity that is caused by the missing bending stiffness
of the element. However, the constraint is placed on the Y1 and θy1 the degrees of freedom, not the
Y1 and θyb degrees of freedom. The force in the CBAR element is correct, but the displacements are
wrong. You made a common assumption that for a symmetric structure with in-plane loads, the out-of-
plane properties are not needed. However, the choice of the output coordinate system and parameter
AUTOSPC resulted in a modeling error. If you chose the output coordinate system for grid point 2 as the
basic system, the displacements would be correct. Also, for this model, if parameter AUTOSPC is set to
NO, the job will fail.

If this had been a production size model, the error would not have been so obvious and may well
have gone undetected. Always look at the Grid Point Singularity Table results to make sure that the
constrained degrees of freedom are acceptable.

Enforced Displacements at Grid Points (SPCD, SPC)

There are several methods available to you for specifying an enforced displacement at a component:

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 9-11

© 2024 Siemens
 9. Constraints

• Enter the value of the enforced displacement directly on an SPC entry.

• Enforce a displacement at a component by using the SPCD bulk entry. The SPCD entry is actually a
force, not a constraint, but it is used in conjunction with an SPC entry to enforce the displacement.
When you use an SPCD entry, internal forces are computed that are applied to the structure to
produce the desired enforced displacements.

• Apply a large number of frequency-dependent enforced motions on structural grid points by using the
SPCF bulk entry. For the enforced motions, you use displacement results from an OUTPUT2 (.op2) file
that was generated from a previous frequency response analysis.

The displacement coordinate systems of the grids in the referenced OP2 file and grids in your current
model must be identical, and the subcases in the referenced OP2 file and subcases in the current
solution must match.

For more information, see SPCF Procedure.

The format of the SPCD entry is as follows:

1 2 3 4 5 6 7 8 9 10
SPCD SID G1 C1 D1 G2 C2 D2

Field Contents

SID Identification number of a static load set.

Gi Grid or scalar point identification number.

Ci Component numbers.

Di Value of enforced displacement for at Gi and Ci.

For every solution except for the solutions listed below, grid points with an enforced displacement using
the SPCD entry must also appear on an SPC or an SPC1 bulk entry.

These solutions do not require the companion SPC or an SPC1 bulk entry:

∙ SOL 402

∙ SOL 414,101

∙ SOL 414,110

∙ SOL 414,129

9-12 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Enforced Displacements at Grid Points (SPCD, SPC)

Note that SOL 414,111 does require the SPC or SPC1 entry. If a static subcase is included for SOL
414,111, that subcase does not require the SPC or SPC1 entry.

The SPCD method of enforcing a nonzero constraint is more efficient than using an SPC entry alone
when you're using multiple subcases that specify different constraint conditions. Note also that when
you use an SPCD entry, the displacement values entered on the SPC entry are ignored. The software only
uses the SPCD values.

As an example of enforced displacement, consider the flat plate shown in Figure 9-7. The plate is
modeled with 2500 CQUAD4 elements.

Figure 9-7. Flat Plate with Various Corner Constraints

The purpose of this example is to compare the efficiency of using the SPC/SPCD combination to that
of using the SPC entry only. For the plate model, four runs are made, each having two subcases with
different constraint conditions. For the input files “const1.dat” and “const2.dat”, the locations of the
constraints are the same for both subcases, but the magnitudes of the constraints change. The file
“const1.dat” uses a different SPC for both subcases. The file “const2.dat” specifies the same SPC for both
subcases, but it has different SPCD entries.

The input files “const3.dat” and “const4.dat” are similar to the first two; however, the locations of the
constraints in the two subcases are different. Table 9-11 summarizes the constraints and CPU time for all
four analyses.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 9-13

© 2024 Siemens
 9. Constraints

Table. Comparison of the SPC Versus the SPC/SPCD Combination

Location of Enforced Displacement
const1.dat const2.dat const3.dat const4.dat
SPC Only SPC/SPCD SPC Only SPC/SPCD
Corner
Displacemen
t (in) Subcase 1 Subcase 2 Subcase 1 Subcase 2 Subcase 1 Subcase 2 Subcase 1 Subcase 2
Grid 1 0 0 0 0 - - - -
Grid 51 0 5 0 5 - 0 - 0
Grid 2551 5 5 5 5 0 5 0 5
Grid 2601 5 0 5 0 5 - 5 -
CPU Time* 7.3 (sec) 5.1 (sec) 7.3 (sec) 7.3 (sec)
– free to move

* CPU time on Windows 2000

The SPC/SPCD combination is more efficient for both modeling situations. It is most advantageous to
use this combination when the magnitudes but not the locations of the constraints change. The larger
improvement seen by the magnitude only change is reasonable because only one SPC needs to be
processed; the multiple boundary conditions are represented as multiple loads. In general, whenever
you have multiple subcases with different constraint conditions, the SPC/SPCD combination is the
preferred method for efficiency.

As mentioned previously, the SPCD is a force entry, not a constraint. Therefore, the SPCD is called out in
the Case Control Section with the LOAD command. The input file for “const2.dat” is shown in Listing 9-1.

$ FILENAME - CONST2.DAT
ID LINEAR,CONST2
SOL 101
TIME 200
CEND
TITLE = ENFORCED DISPLACEMENTS USING SPC1 ONLY
DISP(PLOT) = ALL
SPC = 1
SUBCASE 1
LOAD = 11
SUBCASE 2
LOAD = 12
BEGIN BULK
PARAM POST 0
$
SPC1 1 123456 1 51 2551 2601
$
SPCD 11 1 3 0.0
SPCD 11 51 3 0.0

9-14 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Multipoint Constraints

SPCD 11 2551 3 5.0

SPCD 11 2601 3 5.0
$
SPCD 12 1 3 0.0
SPCD 12 51 3 5.0
SPCD 12 2551 3 5.0
SPCD 12 2601 3 0.0
$
INCLUDE 'const.bulk'
$
ENDDATA

Listing 9-1. The SPC/SPCD Combination

As a final comment, if the SPCD entry is the only loading in the model, the software issues a warning
message stating that no loading is specified for your model. If you intend for the SPCD entry to be the
only loading condition, you can ignore the message.

Multipoint Constraints
In Simcenter Nastran, you can create linear constraint relationships between several degrees of freedom
known as multiple-point constraints or MPCs. Each multipoint constraint is described by a single
equation that specifies a linear relationship for two or more degrees-of-freedom.

In Simcenter Nastran, the term multiple-point constraint (MPC) describes an imposed linear relationship
between two or more degrees of freedom that are expressed in the form

Equation 9-2.

uj = any degree of freedom defined by a grid point or an S point

Rj = user-defined scale factor

See also

• Section 5.4 of The NASTRAN Theoretical Manual.

Uses of Multi-point Constraints

Multi-point constraints have many important practical applications and can be used to:

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 9-15

© 2024 Siemens
 9. Constraints

• Define the relative motion between two grid points as a degree of freedom.

• Define the average of several motions as a degree of freedom.

• Supply hinges or sliding joints between parts of a structure.

• Join dissimilar elements, for example, to join elements with rotational degrees of freedom to
elements that have only translational degrees of freedom (e.g., to join shell elements to solid
elements).

• Obtain the resultant of forces on a structure or on part of a structure.

• Distribute a force to several points in a structure. This is particularly useful if the force is an unknown
force, for example, the force required to compress a fluid.

• Join elements with noncoincident grid points, for example, to change mesh size within a structure.

• Replace extremely stiff structural members with rigid connections. This modeling technique should
only be used when necessary to improve the numerical conditioning of the stiffness matrix. MPCs can
be used for this application, but they are not recommended. The preferred method is to use an R-type
element because this element is simpler, and therefore less prone to error.

• Define a component of motion at a grid point that isn't aligned with the axes of the global coordinate
system, for example, to let you supply a constraint in such a direction with an additional SPC entry.

• Describe rigid elements and mechanisms such as levers, pulleys, and gear trains. In this application,
the degrees-of-freedom associated with the rigid element that are in excess of those needed to
describe rigid body motion are eliminated with multipoint constraint equations. Treating very stiff
members as rigid elements eliminates the ill-conditioning associated with their treatment as ordinary
elastic elements.

• Enforce zero motion in directions other than those corresponding with components of the global
coordinate system. In this case, the multipoint constraint will involve only the degrees-of-freedom at
a single grid point. The constraint equation relates the displacement in the direction of zero motion to
the displacement components in the global system at the grid point.

• Generate nonstandard structural elements and other special effects with scalar elements.

• Describe parts of a structure by local vibration modes (see Section 14.1 of The NASTRAN Theoretical
Manual) where the matrix of local eigenvectors represents a set of constraints relating physical
coordinates to modal coordinates.

Defining Multi-point Constraints in Simcenter Nastran

In Simcenter Nastran, you can define multi-point constraints on MPC and MPCADD bulk data entries.

9-16 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Defining Multi-point Constraints in Simcenter Nastran

• MPC is the basic entry for defining multipoint constraints. The first component specified on the entry
is the dependent degree-of-freedom, i.e., that degree-of-freedom that is removed from the equations
of motion. Dependent degrees-of-freedom may appear as independent terms in other equations of
the set; however, they may appear as dependent terms in only a single equation.

• The MPCADD entry defines a union of multipoint constraints.

You can output multipoint forces of constraint with the MPCFORCE Case Control command in Solution
Sequences 101 through 200.

In general, with these bulk data entries, you must provide the coefficients in the multi-point
constraint equations. However, you can also use several rigid elements to generate the MPC equations
automatically for some applications:

RROD A pin-ended rod which is rigid in extension.

RBAR A rigid bending element with six degrees-of-freedom at each end.

RTRPLT A rigid triangular plate with six degrees-of-freedom at each vertex.

RBE1 A rigid body connected to an arbitrary number of grid points. The

independent and dependent degrees-of-freedom can be arbitrarily
selected by the user.

RBE2 A rigid body connected to an arbitrary number of grid points. The

independent degrees-of-freedom are the six components of motion at
a single grid point. The dependent degrees-of-freedom are specified at
an arbitrary number of grid points.

RBE3 Defines the motion at a reference grid point as the weighted average
of the motions at a set of other grid points.

RSPLlNE Defines multipoint constraints for the interpolation of displacements at

grid points.

RSSCON Defines multipoint constraints to model clamped connections of shell-

to-solid elements.

The rigid elements will always meet equilibrium and continuity requirements, whereas this your
responsibility if you use the bulk data entries to define the MPC equations.

See also

• “MPC” in the Simcenter Nastran Quick Reference Guide

• “MPCADD” in the Simcenter Nastran Quick Reference Guide

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 9-17

© 2024 Siemens
 9. Constraints

Using MPC

The format of the MPC entry is as follows:

1 2 3 4 5 6 7 8 9 10
MPC SID G1 C1 A1 G2 C2 A2
G3 C3 A3 -etc.-

Field Contents

SID Set identification number.

Gj Identification number of grid or scalar point.

Cj Component number.

Aj Coefficient.

As an illustration of the use of the MPC entry, consider the following example.

Consider the model shown in Figure 9-8. The goal is to determine the relative motion between
component 1 of grid point 1 and component 1 of grid point 2 using an MPC entry and a scalar point.

Figure 9-8. Using an MPC to Compute Relative Displacement

Let the components of motion at the two grid points whose difference is desired be designated as and as
given by Eq. 9-3. The difference is represented by scalar point 3 defined on an SPOINT entry.

9-18 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Using MPC

Equation 9-3.

The equation of constraint can be in any of the forms shown in Eq. 9-4.

Equation 9-4.

Listing 9-2 shows the implementation of Eq. 9-4(a) to obtain the relative displacement. The first degree
of freedom on the MPC entry is the dependent degree of freedom-which in this case is scalar point 3.
It can just as well be made a component of motion at a grid point defined on a GRID entry. Grid points
1 and 2 are the independent degrees of freedom. Note that the MPC set ID (SID) is selected in the Case
Control Section.

The first listed degree of freedom on the MPC entry (i.e., x3 in this example) is designated as dependent
and is placed in the m-set. The software eliminates the degrees of freedom in the m-set early in the
solution process. If you want the relative motion to appear as an independent degree of freedom,
express the multipoint constraint in either the second or third forms of Eq. 9-4.

$ FILENAME - RELDIFF.DAT
ID LINEAR,RELDIFF
SOL 101
TIME 5
CEND
TITLE = RELATIVE DISPLACEMENT
SET 1 = 1,2,3
DISP = 1
MPC = 1
LOAD = 1
SPC = 1
BEGIN BULK
$
$ ONLY THE GRIDS, SPOINT AND MPC ENTRIES ARE SHOWN
$
SPOINT 3
MPC 1 3 0 1. 2 1 -1.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 9-19

© 2024 Siemens
 9. Constraints

1 1 1.
$
GRID 1 -10. 20. 0.0
GRID 2 10. 20. 0.0
GRID 4 -10. 10. 0.0
GRID 5 0.0 10. 0.0
GRID 6 10. 10. 0.0
GRID 7 0.0 0.0 0.0
ENDDATA
Listing 9-2. Relative Displacement Using an MPC Entry

Rules for Using Multi-point Constraints

There are three important rules for using MPCs (or R-type elements):

1. A member of the m-set can't be a member of any exclusive subset of the g-set, for example, it
cannot be a degree of freedom that is removed by single-point constraints.

2. A given degree of freedom cannot be designated as a member of the m-set (dependent) more than
once. Thus, it cannot appear as the first listed component on more than one MPC entry.

3. You must avoid redundant constraints. This redundancy occurs, for example, if two of the forms
in the following equation are supplied simultaneously on MPC entries. The technical difficulty that
occurs is that the matrix of constraint coefficients for the m-set is singular so that the dependent
degrees of freedom cannot be evaluated in terms of the independent ones. This problem also
occurs when R-type elements are interconnected in a closed loop.

Equation 9-5.

For more information, see Using MPC.

Multipoint constraints simultaneously supply relationships between components of motion and

relationships between the forces of constraint. This can be demonstrated by using the principle that
the forces acting on a rigid body can do no work. Consider, for example, the equation of constraint
among three degrees of freedom shown in Eq. 9-6.

9-20 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Rules for Using Multi-point Constraints

Equation 9-6.

The work done by the corresponding forces of constraint is given by Eq. 9-7.

Equation 9-7.

In order for both Eq. 9-6 and Eq. 9-7 to be valid, it is necessary that the following equation also be valid.

Equation 9-8.

Thus, the forces of constraint are proportional to the coefficients in the equations of constraint. As an
example, consider the use of a constraint to evaluate the average of four motions, i.e.,

Equation 9-9.

From Eq. 9-9, the forces of constraint are given by:

Equation 9-10.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 9-21

© 2024 Siemens
 9. Constraints

It is clear, from Eq. 9-10, that a force applied to u5 is distributed equally to u1, u2, u3, and u4.

In general, any constraint can be viewed either as a means to constrain a component of displacement
or as a means to distribute the net force acting on the constrained component of displacement (i.e., the
resultant of applied force and internal forces).

See also

• “Understanding Sets and Matrix Operations” in the Simcenter Nastran User's Guide

Rigid Body Supports

In the following discussion, a free body is defined as a structure that is capable of motion without
internal stress; i.e., it has one or more rigid body degrees-of-freedom. The stiffness matrix for a free
body is singular with the defect equal to the number of stress-free, or rigid body modes. A solid
three-dimensional body has up to six rigid body modes. Linkages and mechanisms can have a greater
number. In order to permit the analysis of mechanisms, no restriction is placed in the program on the
number of stress-free modes.

You can define free-body supports with a SUPORT or SUPORT1 entry. You must define free-body supports
in the global coordinate system. You must select the SUPORT1 bulk entry with the SUPORT1 Case Control
command in your input file.

In static analysis by the displacement method, the rigid body modes must be restrained in order to
remove the singularity of the stiffness matrix. The required constraints may be supplied with single point
constraints, multipoint constraints, or free body supports.

If you use free body supports, Simcenter Nastran calculates the rigid body characteristics and checks the
sufficiency of the supports. The software checks the supports by calculating the rigid body error ratio
and the strain energy as defined in the Rigid Body Matrix Generator operation.

This error ratio and the strain energy are automatically printed following the execution of the Rigid Body
Matrix Generator. The error ratio and the strain energy should be zero, but may be nonzero for any of
the following reasons:

• Round-off error accumulation.

• Insufficient free body supports have been provided.

• Redundant free body supports have been provided.

The redundancy of the supports may be caused by improper use of the free body supports themselves or
by the presence of single point or multipoint constraints that constrain the rigid body motions.

Static analysis with inertia relief is necessarily performed on a model that has at least one rigid body
motion. Such rigid body motion must be constrained by the use of free body supports. These supported

9-22 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Linear Contact Using Constraints

degrees-of-freedom define a reference system, and the elastic displacements are calculated relative to
the motion of the support points. The element stresses and forces will be independent of any valid set of
supports.

Rigid body vibration modes are calculated by a separate procedure provided that you supply a set of
free body supports. This is done to improve efficiency and, in some cases, reliability. The determinant
method, for example, has difficulty extracting zero frequency roots of high multiplicity, whereas the
alternate procedure of extracting rigid body modes is both efficient and reliable. If you don't specify free
body supports (or you specify an insufficient number of them), the (remaining) rigid body modes are
calculated by the method selected for the finite frequency modes, provided zero frequency is included in
the range of interest. If you don't provide free body supports, and if zero frequency isn't included in the
range of interest, the rigid body modes won't be calculated.

You must specify free body supports if the mode acceleration method of solution improvement is
used for dynamic problems having rigid body degrees-of-freedom. This solution improvement technique
involves a static solution, and although the dynamic solution can be made on a free body, the static
solution cannot be performed without removing the singularities in the stiffness matrix associated with
the rigid body motions.

See also

• “Static Solutions in SubDMAP SEKRRS” in the Simcenter Nastran User's Guide

• “Rigid-body Modes” in the Simcenter Nastran Basic Dynamic Analysis User's Guide

• “SUPORT Entry” in the Simcenter Nastran Basic Dynamic Analysis User's Guide

Linear Contact Using Constraints

Note:
Note: This section describes an older approach to defining linear contact conditions with the use
of constraints. When this procedure was created, the options for linear contact were limited and
required some creative modeling techniques. There are now improved procedures to define linear
contact conditions, the use of this procedure is no longer recommended. The documentation for
this procedure is included here for those interested in the technique, and to support legacy input
files with this type of definition. For a full discussion on defining the recommended linear contact
conditions, see the chapter “Contact for Solutions 101, 103, 111, 112” in this book.

Introduction

Traditionally, GAP elements were used to model contact-type problems and had required the use of
nonlinear solution sequences—specifically, SOL 106 or 129. In contrast to the gap element available
for nonlinear analysis (SOL 106 and SOL 129), the “linear” gap capabilities in SOL 101 did not require
estimation of gap stiffness. It used an iterative constraint type approach. The constraints were applied to
grid points or SPOINTs. The constraints ensured that:

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 9-23

© 2024 Siemens
 9. Constraints

1. The displacement (UR) cannot be negative. This is to ensure that there is no penetration. Therefore,
the chosen degree-of-freedom must be perpendicular to the contact surface and positive in the
opening direction.

2. The force of constraint (QR) cannot be negative. This is to ensure that there is no tension.

The constraints are satisfied by an iterative technique that is built into SOL 101. The iterative process
starts with a random vector. This random vector assumes certain grids to be in contact and other grids
to be in an open state. A solution is obtained when all the gap constraints are satisfied, i.e., there's
no penetration and no tension forces. If a limit cycle (return to a previous state) is detected during the
iterations, a new random start vector is tried.

This approach provides an alternate method to the use of GAP elements in SOL 106. Some experiments
have shown that the cost of analysis will be about the same. The advantage is that you do not have
to learn how to calculate the GAP stiffness nor how to control SOL 106. Multiple load conditions are
allowed, and each will be solved separately.

If the constraint is between a finite element model and a fixed boundary, then arrange to have one of
the degrees-of-freedom at the boundary grid points represent motion perpendicular to the boundary. A
positive displacement represents motion away from the boundary. If, on the other hand, the constraint
represents relative motion between two bodies, MPC equations are needed to define a relative motion
degree-of-freedom, which is then constrained to have a non-negative displacement.

Input

A SUPORT entry along with PARAM,CDITER are required. These two items plus some optional parameters
are described below.

SUPORT entry Selects constrained degrees-of-freedom. These points must be in the a-set
of the residual structure. This means they must not be dependent in an
MPC equation, constrained by SPC, partitioned by OMIT, or in an upstream
superelement.

PARAM,CDITER,x CDITER Constraints will be applied if CDITER is greater than zero. The value
x is the maximum number of iterations allowed. (Default = 0).

PARAM,CDPRT Controls the printing of constraint violations during the iterations. The
sparse matrix printer prints UR (negative displacements) and QR (negative
forces of constraint) for constrained degrees-of-freedom. (Default = `YES')

DMIG,CDSHUT Optional DMIG input of the vector defining the state of the constrained
degrees-of-freedom. A one indicates closed and a zero indicates open. If

9-24 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Output

DMIG,CDSHUT is not provided, all SUPORT degrees of freedom will assumed

to be closed initially.

PARAM,CDPCH Controls the PUNCH output of DMIG,CDSHUT records for the final state of
the constrained degrees-of-freedom. (Default = `NO'). This output will be
created in a separate .pch file.

Output

The output is similar to standard SOL 101 output. Forces for closed degrees-of-freedom will appear in
the SPCFORCE output. In addition, optional diagnostic information for the iterations can be requested
(PARAM,CDPRT). A final state vector may also be sent to a .pch file (PARAM,CDPCH).

The .f06 file should be examined to ensure that the iterations have converged, since the results of the
last iteration will be output. The last iteration should have zero changes. A typical converged output is
shown below. More detailed explanation are included in the actual examples.

Limitations

The following is a list of limitations for the linear gap elements.

• The only nonlinearity allowed is the constrained displacements.

• There is no sliding friction.

• Free bodies—using the SUPORT entry to define rigid body modes— cannot be used in the same model
with linear gap elements. The parameters INREL and CDITER are mutually exclusive. A fatal message is
issued if both parameters are present.

• No constraint changes are allowed between subcases.

• There is no guarantee that the solution will converge or that all systems will follow the same path.

Examples

Two examples are shown below to illustrate the use of the linear GAP element.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 9-25

© 2024 Siemens
 9. Constraints

EXAMPLE 1 (cd_0.dat)

This model consists of a cantilever beam with a stopper and a vertical load of 30 lbs applied at the
free-end. There is a clearance of 0.05 inch between the free-end and the stopper. The purpose of this
problem is to find out if 30 lbs is sufficient to cause the free-end to hit the stopper.

The first step in solving this problem is to define a variable S51 to monitor the opening between the
free-end (grid point 12) and the stopper (grid point 13). The following equation

Equation 9-11.

U12 = the displacement at grid point 12

U13 = the displacement at grid point 13
Gi = the initial gap opening
S51 = the relative distance between grids points 12 and 13 including the effect of the initial
opening

accomplishes this goal. Grid point 13 can be removed from the equation since the stopper does not
move. Eq. 9-11 can be simplified and recast as

Equation 9-12.

so that an MPC equation can be written. Two scalar points are introduced in this case: Scalar points
101 for G101 (Gi) and 51 for S51. Note that neither S51 nor G101 can be defined as dependent since they
already belong to the r-set and s-set, respectively. A copy of the input file is shown in Listing 9-3.

9-26 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 EXAMPLE 1 (cd_0.dat)

$
$ cd_0.dat
$
sol 101
cend
Title = Gap elements, Cantilever beam
mpc = 77
spc = 88
load = 300
disp = all
spcf = all
mpcf = all
$
begin bulk
$
grid 11 0. 0. 0.
grid 12 60. 0. 0.
cbar 15 1 11 12 0. 1. 0.
pbar 1 2 1. 1. 1. 1.
mat1 2 3.e7 0.3
force 300 12 -30. 0. 1. 0.
spoint 51 101
suport 51 0
spc 88 11 123456
spc 88 101 0 .05
mpc 77 12 2 -1.0 51 0 1.

Listing 9-3. Input file for Cantilever Beam with Stopper.

Eq. 9-12 is represented by MPC,77. Scalar point 51 must also be placed on the SUPORT entry.
In addition, an enforced displacement of 0.05 must also be imposed on scalar point 101.
PARAM,CDITER,10 requests that a maximum of 10 iterations be performed for this analysis. If more
than 10 iterations are required for this problem, the results at the end of the 10th iteration will be
printed. PARAM,CDPRT,YES is optional in this case since it is the default.

An abridged output is shown in the following equation:

Equation 9-13.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 9-27

© 2024 Siemens
 9. Constraints

For more information, see Rules for Using Multi-point Constraints.

The large non-zero values for epsilon and strain energy for the SUPORT degree of freedom can be
safely ignored as we are not performing inertia relief analysis. The solution converges in 1 iteration as
indicated by the following output.

^^^ END ITERATION 1, CHANGES 0

The gap is in contact as indicated by the following scalar point 51 output.

0
0 SHUT
POINT VALUE POINT VALUE POINT VALUE POINT VALUE
POINT VALUE
COLUMN 1
51 S 1.00000E+00

The displacement T2 at grid point 12 is –0.05, which implies that the cantilever beam hits the stopper.
The load required to close the gap is:

Equation 9-14.

which is what the spcforce output at grid point 11 shows. The rest of the applied load 9.167
(30-20.833) is reacted by the stopper as indicated by spcforce at grid point 51 and, similarly, the
mpcforce at grid point 12.

9-28 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 EXAMPLE 1 (cd_0.dat)

Figure 9-9. Abridged Output of Cantilever Beam with Stopper.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 9-29

© 2024 Siemens
 9. Constraints

EXAMPLE 2 (cd_1.dat)

The second example is a thick pad supported on a rigid base with a vertical point load applied at the
center. Due to symmetry, a half model is used by applying the appropriate boundary condition at the
center. The problem is to determine where the lift off occurs along the base (grid points 1 through
7.) An abridged input file for this example is shown in Listing 9-4. A planar problem (Figure 9-10a)
is considered in this case by constraining 3456 on the GRDSET entry. SPC1,200 applies the symmetric
boundary condition at the center line of the structure. The potential contact points are specified on
the seven SUPORT entries—one for each grid point at the bottom. In this case, since the contact is
between a structure and a rigid surface, and there is no initial gap, the use of the MPC is not required.
A maximum of up to 20 iterations (PARAM,CDITER,20) are allowed for this case. Furthermore, an initial
guess of where the contact may occur is provided (DMIG,SHUT.) In this case, grid points 1 and 7 are
assumed to be shut and grid points 2 through 6 are assumed to be opened initially. PARAM,CDPCH,YES
requests that the final open/shut configuration be created in a separate punch file. PARAM,CDPRT,YES
requests that the printing of the constraint violations during the iterations, which is the default.

$ cd_1.dat
$
SOL 101
CEND
TITLE = DEMONSTRATE CONSTRAINED DISPLACEMENT IN SOL 101, #1 CD_1
SPC = 200
LOAD = 300
DISPL = ALL
OLOAD = ALL
SPCFO = ALL
BEGIN BULK
$
$ SYMMETRIC BOUNDARY AT X=0, DOWNWARD LOAD AT TOP CENTER
$
SPC1 200 1 1 11 21 31 41
GRDSET 3456
FORCE 300 41 1. -1.
$
$ CONSTRAIN DISPLACEMENTS ON BOTTOM EDGE, OPTIONAL CDSHUT INPUT
$
$2345678 2345678
SUPORT 1 2
SUPORT 2 2
SUPORT 3 2
SUPORT 4 2
SUPORT 5 2
SUPORT 6 2
SUPORT 7 2
$

9-30 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 EXAMPLE 2 (cd_1.dat)

PARAM CDITER 20
PARAM CDPRT YES
PARAM CDPCH YES
$
DMIG CDSHUT 0 9 1 0 1
DMIG CDSHUT 1 0 1 2 1.
7 2 1.
$
.
ENDDATA

Listing 9-4. Input File for Pad on Rigid Foundation.

Figure 9-10. Undeformed Plot of the Thick Pad.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 9-31

© 2024 Siemens
 9. Constraints

Figure 9-11. Deformed Plot of the Thick Pad.

From Figure 9-11b, it is obvious that grid points 6, 7, and possibly 5 have lifted off the foundation due
to the downward vertical load. Looking at the output (Figure 9-12), one can see that grid points 1, 2,
3, and 4 are in contact as indicated under the “SHUT” output. Furthermore, the displacement output
illustrates that grid points 1, 2, 3, and 4 have zero displacement in the T2 direction—indicating that
these four points are in contact. The T2 displacements for grid points 5, 6, and 7 are positive—indicating
that these three points have lifted off the foundation. The applied vertical load is reacted at the four grid
points (1-4) as illustrated by the spcforce output.

9-32 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 EXAMPLE 2 (cd_1.dat)

Figure 9-12. Abridged Output of Thick Pad.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 9-33

© 2024 Siemens
 9. Constraints

9-34 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 10. Understanding Sets and Matrix
Operations
Overview of Sets and Matrix Operations
When you define a finite element model in Simcenter Nastran, the software writes equations to simulate
the necessary structural properties. These equations are rows and columns in the matrix equations
that describe your model's structural behavior. Certain data entries (i.e., SPCs, MPCs, ASETs, etc.) cause
matrix operations to be performed in the various stages of the solution process.

To organize the matrix operations, Simcenter Nastran assigns each degree of freedom (DOF) in your
model to a displacement set. Simcenter Nastran uses the sets to successively eliminate variables during
the solution process.

Most matrix operations used in a structural analysis involve partitioning, merging, and/or transforming
matrix arrays from one subset to another. For example, all components of motion of a given type form a
vector set that is distinguished from other vector sets. A given component of motion can belong to many
combined sets but can belong to only one exclusive set.

In Simcenter Nastran, there are two basic types of degree-of-freedom sets:

• Mutually exclusive sets

In Simcenter Nastran, mutually exclusive sets are the data sets that are removed by a constraint or
reduction operation during the analysis. If a DOF is a member of a mutually exclusive set, it can't be
a member of any other mutually exclusive set. For example, a degree-of-freedom may not be in the
m-set on two RBAR entries, or on a selected SPC entry and a SUPORT entry.

• Combined sets

Combined sets are formed through the union (combination) of two or more sets. Each degree-of-
freedom is also a member of one or more combined sets called “supersets.” Supersets are formed
when the software combines the mutually exclusive sets.

The contents of both of these types of sets is listed in the Simcenter Nastran Quick Reference Guide.

See also

• “Data Processing and Matrix Operations by Functional Module” in the Simcenter Nastran User's
Guide

• “Degree-of-Freedom Sets” in the Simcenter Nastran Quick Reference Guide

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 10-1

© 2024 Siemens
 10. Understanding Sets and Matrix Operations

Understanding the Global Displacement Set and General Solution

Process
A finite element model's degrees of freedom include all the components of displacement at the grid
points and scalar points. Each grid point has six degrees of freedom, while each scalar point has one.
The union of all the local displacement components at all the grid points and scalar points is called the
global displacement set or g-set. The g-set is the top level set in Simcenter Nastran and represents an
unconstrained set of structural equations.

The g-set displacement is represented as either ug or {ug}. Its significance is that the mathematical
expression for the static equilibrium of the finite element model can be expressed in matrix form as:

Equation 10-1.

where [Kgg] is the global stiffness matrix and {Pg} is the vector of explicit loads (e.g., loads generated
from FORCE entries) or implicit loads (e.g., loads generated from GRAV entries) applied to all of the grid
points and scalar points.

The stiffness matrix [Kgg] is formed by generating and assembling the stiffness matrices for all of
the elements. The load vector is formed by assembling all of the implicitly and explicitly defined
load components within a given loading condition. Loads defined within elements are automatically
transferred to the appropriate grid points in a consistent manner.

The number of terms in the {ug} and {Pg} vectors is equal to six times the number of grid points plus the
number of scalar points. [Kgg] is a real and symmetric matrix.

[Kgg] is usually singular; therefore, the equations of equilibrium cannot be solved in the form shown by
Eq. 10-1.

At the next step in the solution process, Simcenter Nastran partitions the g-set into two subsets: the
m-set and n-set.

The dependent DOFs of all multipoint constraint relations (MPCs, RBEs, etc.) define the m-set. During
the solution process, the contributions of these m-set DOFs are condensed into the set of independent
DOFs, the n-set. The n-set represents all the independent DOFs that remain after the dependent DOFs
are removed from the active set of equations. The m-set represents the DOFs that are linearly dependent
upon the n-set DOFs.

Using the n-set, SPCs are then applied to the independent equations to further partition the equations.

10-2 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Understanding Basic Partitioning Operations

• Degrees of freedom identified by PARAM,AUTOSPC and/or SPCi entries are assigned to the sb-set.

• Permanent constraints, those DOFs specified in field 8 of the GRID entry, are assigned to the sg-set.

• The union of the sb- and sg-sets compose the s-set.

After reducing the n-set by applying the s-set constraints, the f-set remains. The f-set represents the
“free” DOFs of the structure. If applied constraints are properly specified, then the f-set equations
represent a statically stable solution (i.e, a nonsingular stiffness matrix).

If you want the software to perform static condensation (Guyan reduction), the f-set is further reduced
by partitioning it into the o- and a-set. The o-set DOFs are those that are to be eliminated from the active
solution by static condensation. The a-set is the “analysis “ set and is often the partition at which the
solution is performed. By default, all DOFs in the f-set are assigned to the a-set. In other words, the o-set
is null, by default.

If you use the SUPORT entry, the software places the degrees-of-freedom defined on the SUPORT entry
are placed in the r-set or “reference” set. This set aids in the calculation of inertia forces, balances the
applied loads, and helps determine rigid body modes. When the r-set is partitioned from the t-set, the
l-set remains. This final set is termed the "leftover" set and is the lowest level of partitioning performed
in Simcenter Nastran static analysis. The l-set partition is the matrix on which the final solution is
performed.

See also

• “Static Condensation (Guyan Reduction)” in the Simcenter Nastran User's Guide

• “Inertia Relief in Linear Static Analysis” in the Simcenter Nastran User's Guide

Understanding Basic Partitioning Operations

Simcenter Nastran partitions the combined sets and removes one set from the analysis. Table 10-1, for
example, lists the basic set partitioning operations. The m-, s-, o-, and r-sets are the mutually exclusive
sets.

Table. Basic Set Partitioning Operations

g-set - m-set = n-set

n-set - s-set = f-set
f-set - o-set= a-set
a-set - r-set= l-set

Physically, the mutually exclusive set partitioning ensures that operations can't be performed on a DOF
that's no longer active. For example, if you apply an SPC to a DOF that is a dependent degree of freedom
on an RBAR, Simcenter Nastran issues User Fatal Message (UFM) 2101A. Using an SPC entry moves a

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 10-3

© 2024 Siemens
 10. Understanding Sets and Matrix Operations

DOF to the sb-set, but this cannot occur if the same DOF is already a member of the m-set. Both the
m-set and sb-set are mutually exclusive.

When a particular set has no DOFs associated with it, it is a null set. If a null set exists, Simcenter Nastran
applies a partition and moves the DOFs to the subsequent partition.

The final set of degrees of freedom remaining after the exclusive sets are eliminated is called the l-set
(degrees of freedom left over). The static solution is performed on the ul set. The reduced equations of
equilibrium are expressed in the matrix form

Equation 10-2.

The reduced stiffness matrix [Kll] is symmetric and nonsingular if all mechanisms and rigid body motion
are removed by the partitioning of the m-set, s-set, and r-set. Eq. 10-2 is solved for {ul} by Gaussian
elimination (decomposition and forward-backward substitution). All of the other subsets of vector—
namely ur, uo, us, and um—can be obtained subsequently by successive substitution into their defining
equations. By default, all f-set degrees of freedom are assigned to the a-set. If you don't have a rigid
body support (r-set) in your model, then the a-set is identical to your l-set.

Determining Set Membership

You can output the set membership of all of the degrees of freedom with the PARAM,USETPRT and
PARAM,USETSEL Bulk Data entries. Use this parameter with caution as it can produce a substantial
amount of output for a large model.

See also

• “USETPRT” in the Simcenter Nastran Quick Reference Guide

• “USETSEL” in the Simcenter Nastran Quick Reference Guide

Static Condensation (Guyan Reduction)

The terms static condensation and Guyan reduction represent the same matrix operations and are used
interchangeably. Guyan Reduction involves a reformulation of the stiffness and mass matrices based
on partitions of the stiffness matrix. If you start with the stiffness equation in terms of the set of the
unconstrained (free) structural coordinates, you have

10-4 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Static Condensation (Guyan Reduction)

Equation 10-3.

Partitioning the statically independent free degrees of freedom into two subsets of the f-set, you obtain

Equation 10-4.

ua = the a-set variables (the “analysis” set)

uo = the o-set variables, which are removed by static condensation (the “omitted” set)

Rewriting the static equation for uf in terms of o-set and a-set partitions, you obtain

Equation 10-5.

Expanding the bottom equation, you arrive at

Equation 10-6.

where:

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 10-5

© 2024 Siemens
 10. Understanding Sets and Matrix Operations

[Goa] = −[Koo]−1[Koa]
{Uoo} = [Koo]−1[Po]

Expanding the top part of Eq. 10-5 and substituting uo with Eq. 10-6, you arrive at the equation

Equation 10-7.

where:

[Kaa] =

{Pa} =

The solution to the f-set degrees of freedom is obtained in three phases. The first phase is to reduce
the f-set to the a-set using Guyan reduction. The second phase is to solve for {ua} in Eq. 10-7, which is
the analysis set. Once this is done, {uo} can then be solved by Eq. 10-6, which is the omitted set. The
conventional method involves solving Eq. 10-3 in a single step.

In static analysis, the results using static condensation are numerically exact. The partitioned solution
merely changes the order of the operations of the unpartitioned solution. However, Guyan reduction
isn't generally recommended for static analysis. This recommendation stems from the cost of the
Guyan reduction process since it creates reduced matrices [Kaa] and [Maa], which are typically smaller
but denser (i.e., matrix bandwidth is destroyed). Additional decomposition of the matrix [Koo] is also
required. In other words, even though the solution size is smaller, the solution time may take longer and
the disk space requirement may be larger due to a denser matrix. Furthermore, it requires additional
interaction on your part. However, some understanding of the concept of Guyan reduction in static
analysis is quite helpful if you ever decide to apply this feature in dynamic, cyclic symmetry, or
superelement analyses.

Guyan reduction has special applications in dynamic analysis. In dynamic analysis, the reduction is
approximate; the term {uoo} is ignored. The reduction is based solely on static transformation and is
exact provided that no loads are applied to the o-set degrees of freedom.

Using Static Condensation

In Simcenter Nastran, the ASETi and OMITi bulk data entries define the degrees-of-freedom in the
analysis and omitted sets, respectively. You can use either the ASET/ASET1 or the OMIT/OMIT1 entries to
specify only the a-set (with ASET/ASET1 entries) or o-set (with OMIT/OMIT1 entries) degrees of freedom.

10-6 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Format for ASET

The software automatically places the remaining f-set DOFs in the complementary set, as described
below. However, if you specify both the a-set and o-set DOFs, then the unspecified remaining f-set DOFs
are automatically placed in the o-set. If you specify the same DOF on both the a-set and o-set, the job
fails with UFM 2101A.

The a-set and the o-set are created in one of the following ways:

1. If OMITi entries are present, the o-set consists of degrees-of-freedom listed explicitly on OMITi
entries. The remaining f-set degrees-of-freedom are placed in the b-set. They are then also
members of the a-set.

2. If ASETi or QSETi entries are present, the a-set consists of all degrees-of-freedom listed on ASETi
entries and any entries listing its subsets, such as QSETi, SUPORTi, CSETi, and BSETi entries. The
remaining f-set degrees-of-freedom are placed in the o-set. Note, however, there must be at least
one explicit ASETi or QSETi entry type present for the o-set to exist, even if it is a redundant
specification.

3. If there are no ASETi or QSETi entries , but there are SUPORT, BSETi, or CSETi entries, the entire f-set
is placed in the a-set, and the o-set isn't created.

4. There must be at least one explicitly defined ASETi, QSETi, or OMITi entry for the o-set to exist, even
if the ASETi, QSETi, or OMITi entry is redundant.

Note:

In superelement analysis, exterior (“boundary”) degrees-of-freedom are automatically determined and

placed in the a-set of the superelement. Exterior degrees-of-freedom for superelements are therefore
equivalent to degrees-of-freedom listed on ASETi entries.

The a-set defines a combined set. Degrees-of-freedom listed on Bulk Data entries that define its subsets,
the q-, r-, c-, and b-set, may also be listed on ASETi Bulk Data entries, although this is a redundant
specification and isn't necessary. However, automatic restarts are more reliable if ASETi entries are
present for all degrees-of-freedom listed on SUPORT entries.

Since the reduction process is performed on an individual degree of freedom, it is possible to have some
of the degrees of freedom at a grid point in the a-set and other degrees of freedom at a grid point in one
of the other mutually exclusive sets. No additional user input is required.

Format for ASET

You can use ASET in conjunction with ASET1. The formats for the ASET/ASET1 are as follows:

1 2 3 4 5 6 7 8 9 10
ASET G1 C1 G2 C2 G3 C3 G4 C4

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 10-7

© 2024 Siemens
 10. Understanding Sets and Matrix Operations

Field Contents

Gi The i-th grid or scalar point identification number

Ci Component number(s) for the i-th grid or scalar point

Example

ASET 5 3 30 345 8 12345

In this example, component 3 of grid point 5, components 3, 4, and 5 of grid point 30, and components
1,2,3,4, and 5 of grid point 8 are assigned to the a-set. All other degrees of freedom that aren't
otherwise specified (e.g., on SPCi or MPC entries) are placed in the omitted set (o-set).

Format for ASET1

1 2 3 4 5 6 7 8 9 10
ASET1 C G1 G2 G3 G4 G5 G6 G7
G8 G9 G10 G11 -etc.-

Field Contents

C Component number(s).

Gi The i-th grid or scalar point identification number.

Example

ASET1 345 3 4 1 16 8 9 5
13 20 25

In this example, components 3,4, and 5 of grid points 1, 3, 4, 5, 8, 9, 13, 16, 20, and 25 are assigned to
the a-set. All other degrees of freedom that are not otherwise specified (e.g., on SPCi or MPC entries) are
placed in the omitted set (o-set).

Alternate Format for ASET1

1 2 3 4 5 6 7 8 9 10
ASET1 C ID1 “THRU” ID2

10-8 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Example

Field Contents

C Component number(s).

ID1 The first grid or scalar point identification number in the range.

ID2 The last grid or scalar point identification number in the range.

Example

ASET1 123456 10 THRU 909

In this example, components 1,2,3,4,5, and 6 of grid point 10 through 909 are assigned to the a-set.
As long as at least one grid point between 10 and 909 exists, all the grid points between grid points 10
and 909 don't need to exist . Simcenter Nastran places all other degrees of freedom that are outside of
grid points 10 through 909 and not otherwise specified (e.g., on SPCi or MPC entries) in the omitted set
(o-set).

Format for OMIT

OMIT can be used in conjunction with OMIT1. The formats for the OMIT/OMIT1 are as follows:

1 2 3 4 5 6 7 8 9 10
OMIT G1 C1 G2 C2 G3 C3 G4 C4

Field Contents

Gi The i-th grid or scalar point identification number.

Ci Component number(s) for the i-th grid or scalar point.

Example

OMIT 5 3 30 345 8 123456

For the above example, component 3 of grid point 5, components 3, 4, and 5 of grid point 30, and
components 1,2,3,4,5, and 6 of grid point 8 are assigned to the o-set. All other degrees of freedom that
are not otherwise specified (e.g., on SPCi or MPC entries) are placed in the analysis set (a-set).

Format for OMIT1

1 2 3 4 5 6 7 8 9 10
OMIT1 C G1 G2 G3 G4 G5 G6 G7
G8 G9 G10 G11 -etc.-

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 10-9

© 2024 Siemens
 10. Understanding Sets and Matrix Operations

Field Contents

C Component number(s).

Gi The i-th grid or scalar point identification number.

Example

OMIT1 345 3 4 1 16 8 9 5
13 20 25

For the above example, components 3, 4, and 5 of grid points 1, 3, 4, 5, 8, 9, 13, 16, 20, and 25 are
assigned to the o-set. All other degrees of freedom that aren't otherwise specified (e.g., on SPCi or MPC
entries) are placed in the analysis set (a-set).

Alternate Format for OMIT1

1 2 3 4 5 6 7 8 9 10
OMIT1 C ID1 “THRU” ID2

Field Contents

C Component number(s).

ID1 The first grid or scalar point identification number in the range.

ID2 The last grid or scalar point identification number in the range.

Example

OMIT1 123456 10 THRU 909

For this example, components 1,2,3,4,5, and 6 of grid points 10 through 909 are assigned to the o-set.
All grid points between grid points 10 and 909 need not exist as long as at least one of them exists.
All other degrees of freedom that are outside of grid points 10 through 909 and not otherwise specified
(e.g., on SPCi or MPC entries) are placed in the analysis set (a-set).

The choice of whether to use the ASET/ASET1 or OMIT/OMIT1 is really a matter of convenience.

Consider a simple example using Guyan reduction. The cantilever beam model in Figure 10-1 is used for
this purpose. The common Bulk Data entries for this model are contained in Listing 10-1.

10-10 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Example

Figure 10-1. Cantilever Beam Using Guyan Reduction

$ filename - bulkm.dat
$
CBAR 1 1 1 2 10
CBAR 2 1 2 3 10
CBAR 3 1 3 4 10
CBAR 4 1 4 5 10
CBAR 5 1 5 6 10
FORCE 10 6 100. 0. 0. 1.
$
GRID 1 0. 0. 0.
GRID 2 1. 0. 0.
GRID 3 2. 0. 0.
GRID 4 3. 0. 0.
GRID 5 4. 0. 0.
GRID 6 5. 0. 0.
GRID 10 0. 0. 10.
MAT1 1 7.1+10 .33 2700.
PBAR 1 1 2.654-3 5.869-7 5.869-7 9.9-7
SPC1 10 123456 1
$

Listing 10-1. Input File for Common Bulk Data Entries

The first run is a conventional run without any reduction (Listing 10-2). Since Guyan reduction isn't
used, then, by default, all degrees of freedom that aren't otherwise constrained are assigned to the
a-set. In other words, degrees of freedom 1 through 6 for grid points 2 through 6 are assigned to the
a-set. Components 1 through 6 of grid point 1 are assigned to the s-set due to the SPC Case Control
command and SPC1 Bulk Data entry.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 10-11

© 2024 Siemens
 10. Understanding Sets and Matrix Operations

$ filename - full.dat
id bar aset
sol 101
time 10
cend
$
spc = 10
disp = all
$
subcase 1
load = 10
$
begin bulk
$
include 'bulkm.dat'
enddata

Listing 10-2. Input File for Using Conventional Analysis

The second run uses Guyan reduction by specifying degrees of freedom 1 through 6 at grid points
2, 4, and 6 to the a-set by using the ASET1 entries (Listing 10-3). Due to this action, the remaining
unconstrained degrees of freedom (DOF 1 through 6 for grid points 3 and 5) are assigned to the o-set.
Components 1 through 6 of grid point 1 are assigned to the s-set due to the SPC Case Control command
and SPC1 Bulk Data entry.

$ filename - aset.dat
id bar aset
sol 101
time 10
cend
$
title =
subtitle =
spc = 10
disp = all
$
subcase 1
load = 10
$
begin bulk
$
aset1,123456,2,4,6
$

10-12 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Example

include 'bulkm.dat'
enddata

Listing 10-3. Input File Specifying A-Set DOFs

The third run is similar to the second run except the omitted degrees of freedom are specified rather
than the analysis degrees of freedom (Listing 10-4). Degrees of freedom 1 through 6 for grid points 3
and 5 are assigned to the o-set by using the OMIT1 entries. Due to this action, degrees of freedom 1
through 6 for grid points 2, 4, and 6 are automatically assigned to the a-set. Components 1 through 6 of
grid point 1 are assigned to the s-set due to the SPC Case Control command and SPC1 Bulk Data entry.

$ filename - omit.dat
id bar omit
sol 101
time 10
cend
$
title =
subtitle =
spc = 10
disp = all
$
subcase 1
load = 10
$
begin bulk
$
omit1,123456,3,5
$
include 'bulkm.dat'
enddata

Listing 10-4. Input File Specifying O-Set DOFs

Since Guyan reduction is numerically exact in static analysis, all three runs should yield the same results.
Figure 10-2 summarizes the results from the three runs-they are identical, as expected.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 10-13

© 2024 Siemens
 10. Understanding Sets and Matrix Operations

Figure 10-2. Results Comparison With and Without Guyan Reduction

Note that the only differences in the three input files are the ASET1/OMIT1entries. The common Bulk
Data entries are inserted in the run stream by using the INCLUDE statement.

Sets for Dynamic Reduction

One of the more important applications of partitioning is Guyan reduction. This technique is a means for
reducing the number of degrees-of-freedom used in dynamic analysis with minimum loss of accuracy.
The basis is that a lot fewer grid points are needed to describe the inertia of a structure than are needed
to describe its elasticity with comparable accuracy. The error in the approximation is small provided
that the set of displacements used for dynamic analysis is judiciously chosen. Its members should be
uniformly dispersed throughout the structure, and all large mass items should be connected to grid
points that are members of the analysis set.

Importantly, the matrix operations associated with this partitioning procedure tend to both create
nonzero terms and fill what were previously very sparse matrices. The partitioning option is most
effectively used if the members of the omitted set are either a very large fraction or a very small fraction
of the total set. In most of the applications, the omitted set is a large fraction of the total and the

10-14 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Sets for Aerodynamics

matrices used for analysis, while small, are usually full. If the analysis set isn't a small fraction of the
total, a solution using the large, but sparser, matrices may be more efficient.

In dynamic analysis, additional vector sets are obtained by a modal transformation derived from real
eigenvalue analysis of the a-set. These sets are as follows:

ξo = rigid body (zero frequency) modal degrees-of-freedom

ξf = finite frequency modal degrees-of-freedom
ξi = ξo + ξf, the set of all modal degrees-of-freedom

One vector set is defined that combines physical and modal degrees-of-freedom:

uh = ξi + ue, the set of all physical and modal degrees-of-freedom

In dynamic analysis, the a-set is further partitioned into the q-, r-, c-, and b-subsets.

• The q-set stores the coefficients for the generalized coordinates determined by dynamic reduction
and/or component mode calculations.

• The r-set is again used to determine rigid body modes.

• The c-set contains coordinates considered free to move while computing the generalized coordinate
functions.

• The b-set contains coordinates fixed during this process.

The d-set (“dynamic”) for direct formulations is formed from the union of the structural degrees-of-
freedom in the a-set and the extra points of the e-set. The d-set can be used to input unsymmetrical
terms into the equations of motion using DMIG and TF Bulk Data entries. In the modal formulations, the
h-set is composed of the union of the ci (modal) set and the e-set. Again, you can use the DMIG and TF
Bulk Data entries to introduce unsymmetrical matrix coefficients.

See also

• “Dynamic Reduction” in the Simcenter Nastran Basic Dynamic Analysis User's Guide

• “The Set Notation System Used in Dynamic Analysis” in the Simcenter Nastran Basic Dynamic Analysis
User's Guide

• “Direct Matrix Input” in the Simcenter Nastran User's Guide

Sets for Aerodynamics

Aerodynamic calculations are made in a Cartesian aerodynamic coordinate system. By convention, the
flow is in the positive x direction. The basic coordinate system was not chosen, since it would place an

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 10-15

© 2024 Siemens
 10. Understanding Sets and Matrix Operations

undesirable restriction upon the description of the structural model. Any Simcenter Nastran Cartesian
system may be specified, and flow will be defined in the direction of its positive x-axis. All element and
aerodynamic grid point data, computed initially in the basic coordinate system, will be converted to the
aerodynamic coordinate system. The global (displacement) coordinate system of the aerodynamic grid
points will have its T1 direction in the flow direction. T3 is normal to the element for boxes, and parallel
to the aerodynamic T3 in the case of bodies. Coordinate system data are generated for the aerodynamic
grid points.

The grid points are physically located at the centers of the boxes and body elements. Permanent
constraints are generated for the unused degrees-of-freedom. A second set of grid points, used only for
undeformed plotting, is located at the element corners. All six degrees-of-freedom associated with each
grid point in this second set are permanently constrained. Grid point numbers are generated based upon
the element identification number. For any panel, the external grid point numbers for the boxes start
with the panel identification number and increase consecutively.

Aerodynamic degrees-of-freedom, along with the extra points, are added after the structural matrices
and modes have been determined.

See also

• “Aerodynamic Data Input and Generation” in the Simcenter Nastran Aeroelastic Analysis User's Guide

Output Selection Via Set Specification

The membership of grid points and their components and scalar points in the various sets is stored
in the USET table. In Simcenter Nastran, sets are defined by their name. For example, if you want the
TABPRT module to print the M, R, N, SG, K, and SA sets, then specify:

PARAM,USETSTR1, M:R:N:SG

PARAM,USETSTR2, K:SA.

The data block naming convention in the solution sequences for matrices related to the sets is based on
the set names. For example, the system stiffness matrix [Kgg] has the name KGG. Some sets are used
that do not appear in the USET table. The h-set, for example, is the set used for modal formulations. Its
size is not known until after the USET table is formed, so it does not appear in the table. The j-set, a set
synonymous with the g-set, is used in superelement analysis. There are several other sets used that also
do not appear in the USET table.

Some names may be used for other purposes. You can input the name H, for example, to the MATGPR
module when you want to label rows or columns of matrices sequentially, rather than by their external
sequence numbers. This feature may be used to number vectors in load matrices or in other instances
not related to modal analysis, even though “H” is defined for modal analysis.

10-16 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 11. Direct Matrix Input
Overview of Direct Matrix Input
You can use the Bulk Data entry DMIG to input a stiffness (or mass) matrix which connects specified
degrees-of-freedom. The matrix so defined will be added to the stiffness (or mass) matrix computed
from finite element properties.

The DMIG entry includes provisions for unsymmetrical terms and complex values, both of which are
useful in dynamic analysis. These provisions should not be used in static or normal modes. Note that an
entry in the Case Control Section is required (K2GG = NAME for a stiffness matrix or M2GG= NAME for a
mass matrix).

The primary application of the DMIG Bulk Data entry is to enter stiffness and mass data for parts of the
structure which are obtained from another computer run. The format is cumbersome (two matrix terms
per continuation entry) and the matrix should be input to high precision. For stiffness matrices only, the
GENEL Bulk Data entry is an alternative for manually inputting data.

The finite element approach simulates the structural properties with mathematical equations written
in matrix format. Once you provide the grid point locations, element connectivities, cross-sectional
properties, material properties, applied loads, and boundary conditions, Simcenter Nastran then
automatically generates the appropriate structural matrices. The structural behavior is then obtained
by solving these equations.

If these structural matrices are available externally, you can input these matrices directly without
providing all the modeling information. Normally this is not a recommended procedure since it requires
additional effort. However, there are occasions where the availability of this feature is very useful and in
some cases is extremely crucial. Some possible applications are listed below:

• Suppose you are a subcontractor to a classified project. The substructure that you are analyzing is
attached to the main structure built by the primary contractor. The flexibility of this main structure
is crucial to the response of your component, but the geometry of the main structure is classified.
The main contractor, however, can provide you with the stiffness matrix of the classified structure. By
reading in this stiffness matrix and adding it to your Simcenter Nastran model, you can account for
the flexibility of the attached structure without compromising the security. The stiffness matrix is the
inverse of the flexibility matrix.

• Suppose you are investigating a series of design options on a component attached to an aircraft
bulkhead. Your component consists of 500 DOFs and the aircraft model consists of 100,000 DOFs.
The flexibility of the backup structure is somewhat important. You can certainly analyze your
component by including the full aircraft model (100,500 DOFs). On the other hand, if the flexibility
at the attachment points on the aircraft can be measured experimentally, then you can add the
experimental backup structure stiffness to your component without including the whole aircraft
model. The experimental backup structure stiffness matrix is the inverse of the measured flexibility
matrix. This way your model size remains at 500 DOFs, and you still have a good approximation of the
backup structure stiffness.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 11-1

© 2024 Siemens
 11. Direct Matrix Input

• The same concept can be applied to a component attached to a test fixture. The stiffness of the fixture
at the attachment locations can be read in as a stiffness matrix. Once again, the experimental test
fixture stiffness matrix at the attachment points is the inverse of the measured flexibility at these
points.

There are several ways that these matrices can be read in, such as DMIG, GENEL, and INPUTT4. Only
DMIG is covered here. The DMIG and the INPUTT4 options offer alternate methods for inputting large
matrices. Note that INPUTT4 provides more precision than the DMIG input; the DMIG yields more
precision than the GENEL on a short word machine.

See also

• “General Element Capability (GENEL)” in the Simcenter Nastran Element Library

• “INPUTT4” in the Simcenter Nastran DMAP Programmer's Guide

Using DMIG
You can use the DMIG feature to directly input a stiffness, mass, or loading matrix to the grid and/or
scalar points in static analysis. These matrices are referenced in terms of their external grid point IDs and
DOFs. The symbols for these g-type matrices in standard mathematical format are

All three matrices must be real. Furthermore,

must also be symmetric. These matrices are implemented on the g-set level. In other words, these terms
are added to the finite element model at the specified DOFs prior to the application of constraints. This
section focuses on the DMIG features that are used in static analysis. The DMIG has additional features
that can be used in dynamic analysis.

See also

• “DMIGs, Extra Points, and Transfer Functions” in the Simcenter Nastran Advanced Dynamic Analysis
User's Guide

11-2 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 DMIG Bulk Data User Interface

DMIG Bulk Data User Interface

In the Bulk Data section, the DMIG matrix is defined by a single DMIG header entry followed by a
series of DMIG data entries. Each of these DMIG data entries contains a column containing the nonzero
elements for the particular matrix that you are entering.

Header Entry Format

1 2 3 4 5 6 7 8 9 10
DMIG NAME “0" IFO TIN TOUT NCOL

Field Contents

NAME Name of the matrix. It consists of one to eight alphanumeric characters, the first
of which must be alphabetic. (Required)

IFO Form of matrix input: Integer, required

1 = square

9 or 2 = rectangle

6 = symmetric

TIN Type of matrix that is input: Integer, required

1 = real, single precision

2 = real, double precision

TOUT Type of matrix that is created: Integer

0 = set by your machine precision (default)

1 = real, single precision

2 = real, double precision

NCOL Number of columns in a rectangular matrix. Used only for IFO = 9. (Integer > 0).

Remarks

1. One header entry is required for each input matrix.

2. The matrix name used in field 2 must be unique.

3. Field 3 of the header entry must contain an integer 0.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 11-3

© 2024 Siemens
 11. Direct Matrix Input

4. The default value of 0 for TOUT is recommended.

5. If the input matrix is rectangular, using IFO = 9 is recommended instead of IFO = 2.

Column Data Entry Format

1 2 3 4 5 6 7 8 9 10
DMIG NAME GJ CJ G1 C1 A1
G2 C2 A2 -etc.-

Field Contents

NAME This is the same name that is used with its corresponding header entry as
described above.

GJ The external grid or scalar point identification number for the column.
(Integer > 0).

CJ The component number for the GJ grid point. (0 < Integer ≤ 6 if GJ is a grid
point; blank or zero if GJ is a scalar point).

Gi The external grid or scalar point identification number for the row. (Integer
> 0).

Ci The component number for the Gi grid point. (0 < Integer ≤ 6 if Gi is a grid
point; blank or zero if Gi is a scalar point).

Ai The value of the matrix term.

Remarks

1. Each non-null column (a column with at least one nonzero term) starts with a GJ, CJ pair. The
entries (Gi, Ci pairs) for each row of that column follow. The number of Gi, Ci pairs equals the
number of nonzero terms in that column. The terms may be input in arbitrary order. High precision
is required when using DMIG; therefore, in general, the use of large field input is recommended.
Each non-null column of your matrix makes up a DMIG column data entry. Therefore, the total
number of DMIG entries is equal to the number of non-null columns plus one. The extra entry is
the DMIG header entry.

2. For symmetric matrices (IFO = 6), you can input a given off-diagonal term either below or above
the diagonal. While the upper and lower triangle terms may be mixed, a fatal error results if the
corresponding symmetric element is input both below and above the diagonal. However, to keep
it simple, for a given matrix, you should decide to input the terms either completely below the
diagonal or above the diagonal, but not both. The symmetric option is generally used for the mass
and stiffness matrices.

11-4 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 DMIG Case Control User Interface

3. Rectangular matrices can be input using the option IFO = 9. This option is generally used to input
load matrices.

See also

• “DMIG” in the Simcenter Nastran Quick Reference Guide

DMIG Case Control User Interface

In order to include the direct input matrices you define with DMIG, the Case Control Section must
contain the appropriate K2GG, M2GG, B2GG, K42GG, or P2G command.

See also

• “K2GG” in the Simcenter Nastran Quick Reference Guide

• “M2GG” in the Simcenter Nastran Quick Reference Guide

• “B2GG” in the Simcenter Nastran Quick Reference Guide

• “K42GG” in the Simcenter Nastran Quick Reference Guide

• “P2G” in the Simcenter Nastran Quick Reference Guide

Examples

K2GG = mystif

The above Case Control command adds the terms of the matrix, which are defined by the DMIG entries
with a name of “mystif”, to the g-set stiffness matrix.

M2GG = yourmass

The above Case Control command adds the terms of the matrix, which are defined by the DMIG entries
with a name of “yourmass”, to the g-set mass matrix.

B2GG = mydamp

The above Case Control command adds the terms of the matrix, which are defined by the DMIG entries
with a name of “mydamp”, to the g-set viscous damping matrix.

K42GG = strdamp

The above Case Control command adds the terms of the matrix, which are defined by the DMIG entries
with a name of “strdamp”, to the g-set structural damping matrix.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 11-5

© 2024 Siemens
 11. Direct Matrix Input

P2G = hisload

The above Case Control command adds the terms of the matrix, which are defined by the DMIG entries
with a name of “hisload”, to the g-set load matrix.

DMIG Examples
The following examples of using DMIG features use the cantilever beam model shown in Figure 11-1.

Figure 11-1. Planer Cantilever Beam

Stiffness Matrix Using DMIG

First, we must analyze this problem using the conventional method with two loading conditions—a
vertical tip load and a gravity load. Listing 11-1 contains the input file for this run. Note that for
illustrative purposes, a planar problem (xz plane) is shown. However, this same concept is applicable to a
general three-dimensional structure.

$ FILENAME - DMIGFULL.DAT
ID BAR DMIG
SOL 101
TIME 10
DIAG 8
CEND
$
TITLE = PLANAR CANTILEVER BEAM

11-6 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Stiffness Matrix Using DMIG

SPC = 10
DISP = ALL
$
SUBCASE 1
LABEL = TIP LOAD AT END
LOAD = 10
$
SUBCASE 2
LABEL = GRAVITY LOAD
LOAD = 20
$
BEGIN BULK
$

CBAR 1 1 1 2
10
CBAR 2 1 2 3
10
CBAR 3 1 3 4
10
CBAR 4 1 4 5
10
CBAR 5 1 5 6
10
FORCE 10 6 100. 0. 0.
1.
GRAV 20 9.8 0. 0.
1.
GRID 1 0. 0. 0.
246
GRID 2 1. 0. 0.
246
GRID 3 2. 0. 0.
246
GRID 4 3. 0. 0.
246
GRID 5 4. 0. 0.
246
GRID 6 5. 0. 0.
246
GRID 10 0. 0. 10.
123456
MAT1 1 7.1+10 .33
2700.
PBAR 1 1 2.654-3
5.869-7
SPC1 10 123456
1

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 11-7

© 2024 Siemens
 11. Direct Matrix Input

$
ENDDATA

Listing 11-1. Input File for Conventional Analysis

The next run removes CBAR element number 5 from the input file and uses the DMIG feature to directly
read in the bar element stiffness matrix. The bar element stiffness matrix for a planar element and
ignoring the axial stiffness is as follows:

Since the matrix is symmetric, either the lower or upper triangular portion of the matrix must therefore
be provided. For E = 7.1 · 1010N/m2, I = 5.869 · 10−7m4, and l5 = 1m, the stiffness matrix for CBAR
element number 5 is as follows:

In this case, you chose UGSTIF as the name of the input stiffness for bar element number 5. Therefore, in
order to bring in this stiffness matrix and add it to the global stiffness matrix, the Case Control command
K2GG= UGSTIF is required. Note these stiffness terms are additions to the global stiffness matrix at that
location and are not a replacement of the stiffness terms at that location. In the Bulk Data Section, five
DMIG entries are required-one for the header entry and four for the data column entries—since there are
four non-null columns in the above matrix.

For the header entry, the same name UGSTIF must be used to match the name called out in the Case
Control Section. The third field is 0, which must be the value used for the header entry. The fourth field
(IFO) is set to 6 to denote a symmetric matrix input. The fifth field (TIN) is set to 1 to denote that the
matrix is provided as real, single precision.

Once again, the terms in the matrix are referenced in terms of their external grid IDs when using the
DMIG feature. Physically, each term in a particular column of the stiffness matrix (kij) represents the
induced reactive load in the i-th degree of freedom due to a unit displacement in the j-th direction with

11-8 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Stiffness Matrix Using DMIG

all other displacement degrees of freedom held to zero. Since the matrix is symmetric, only the lower
triangular portion of the matrix is read.

The first DMIG data column entry reads the first column of the above matrix. Field 2 of this DMIG entry
must have the same name UGSTIF as referenced by the Case Control K2GG = UGSTIF command. Fields
3 and 4 of this entry identify this column in terms of its external grid ID and corresponding degree of
freedom, respectively. In this case, it is grid point number 5, degree of freedom 3 (z-translation at grid
number 5).

Once this column is defined and by following the format description as described in the section for
column data entry format, you can then input the four terms in this column row by row. These four
terms are defined by sets of three fields. They are the external grid point ID number, corresponding
degree of freedom, and the actual matrix term, respectively. The first row of column one is defined by
external grid point ID 5, degree of freedom 3 (z-translation at grid point number 5) with a stiffness value
of 500039. The second row of column one is defined by external grid point ID 5, degree of freedom 5
(y-rotation at grid point number 5) with a stiffness value of -250019. The third row of column one is
defined by external grid point ID 6, degree of freedom 3 (z-translation at grid point number 6) with a
stiffness value of -500039. The fourth row of column one is defined by external grid point ID 6, degree of
freedom 5 (y-rotation at grid point number 6) with a stiffness value of -250019.

The next DMIG entry reads the second column of the above matrix starting with the diagonal term.
Fields 3 and 4 of this entry identify this column in terms of its external grid point ID and corresponding
degree of freedom, respectively. In this case, it is grid number 5, degree of freedom 5 (y-rotation at grid
point number 5). The rest of the procedure is similar to that of column one with the exception that only
three rows need to be read due to symmetry (rows two through four).

The next two DMIG entries read columns three and four of the stiffness matrix, respectively. Note that
due to symmetry, one less row needs to be read for each additional column.

The input file using DMIG is contained in Listing 11-2. The results as compared with the conventional
analysis are tabulated in the first two sections of Figure 11-2. The comparison is performed for Subcase
1 with a vertical load applied to the tip of the cantilever beam. In this case, the results are accurate to
within three to four digits.

As mentioned previously, a high degree of precision should be retained when inputting a matrix with
the DMIG method. In this case, since the small-field input format is used, the maximum number of
characters that can be used are eight (including sign and decimal place). The same job was then rerun
for the DMIG input with additional significant digits added using double-precision and large-field format
(Listing 11-3). In this case, the results are identical by comparing the results of the first and last sections
of Figure 11-2. Note that even the double-precision input on short-word machines does not contain all
of the significant digits of the machine.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 11-9

© 2024 Siemens
 11. Direct Matrix Input

$ FILENAME - DMIGSTFS.DAT
ID BAR DMIG
SOL 101
TIME 10
DIAG 8
CEND
$
TITLE = PLANER CANTILEVER BEAM
SUBTITLE = USE DMIG TO BRING IN STIFFNESS MATRIX OF ELEMENT # 5 USING
SMALL FIELD
K2GG = UGSTIF
SPC = 10
DISP = ALL
$
SUBCASE 1
LABEL = TIP LOAD AT END
LOAD = 10
$
BEGIN BULK
$
CBAR 1 1 1 2
10
CBAR 2 1 2 3
10
CBAR 3 1 3 4
10
CBAR 4 1 4 5
10
$ DMIG HEADER ENTRY
$
DMIG UGSTIF 0 6
1
$
$ DMIG DATA COLUMN ENTRIES
$
DMIG UGSTIF 5 3 5 3 500039. +000001
++0000015 5 -250019. 6 3 -500039. +000002
++0000026 5 -250019. +000003
$
DMIG UGSTIF 5 5 5 5 166680. +000004
++0000046 3 250019. 6 5 83340. +000005
$
DMIG UGSTIF 6 3 6 3 500039. +000006
++0000066 5 250019. +000007
$
DMIG UGSTIF 6 5 6 5 166680.

11-10 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Stiffness Matrix Using DMIG

$
FORCE 10 6 100. 0. 0. 1.
GRAV 20 9.8 0. 0. 1.
GRID 1 0. 0. 0.
246
GRID 2 1. 0. 0.
246
GRID 3 2. 0. 0.
246
GRID 4 3. 0. 0.
246
GRID 5 4. 0. 0.
246
GRID 6 5. 0. 0.
246
GRID 10 0. 0. 10.
123456
MAT1 1 7.1+10 .33
2700.
PBAR 1 1 2.654-3
5.869-7
SPC1 10 123456
1
$
ENDDATA

Listing 11-2. Input File for Reading a Stiffness Matrix Using DMIG with Small-Field Format

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 11-11

© 2024 Siemens
 11. Direct Matrix Input

$ FILENAME - DMIGSTFL.DAT
ID BAR DMIG
SOL 101
TIME 10
DIAG 8
CEND
TITLE = PLANER CANTILEVER BEAM
SUBTITLE = DMIG TO BRING IN STIFFNESS MATRIX FOR ELEMENT # 5 USING LARGE
FIELD
K2GG = UGSTIF
SPC = 10
DISP = ALL
$
SUBCASE 1
LABEL = TIP LOAD AT END
LOAD = 10
$
BEGIN BULK
$
CBAR 1 1 1 2
10
CBAR 2 1 2 3
10
CBAR 3 1 3 4
10
CBAR 4 1 4 5
10
$
$ DMIG HEADER ENTRY
$
DMIG UGSTIF 0 6
2
$
$ DMIG DATA COLUMN ENTRIES
$
DMIG* UGSTIF 5 3 *A
*A 5 3 5.000388 D+5 *B
*B 5 5 -2.500194 D+5 *C
*C 6 3 -5.000388 D+5 *D
*D 6 5 -2.500194 D+5
$
DMIG* UGSTIF 5 5 *A2
*A2 5 5 1.666796 D+5 *B2
*B2 6 3 2.500194 D+5 *C2
*C2 6 5 8.33398 D+4
$

11-12 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Stiffness Matrix Using DMIG

DMIG* UGSTIF 6 3 *A3

*A3 6 3 5.000388 D+5 *B3
*B3 6 5 2.500194 D+5
$
DMIG* UGSTIF 6 5 *A4
*A4 6 5 1.666796 D+5
$
FORCE 10 6 100. 0. 0. 1.
GRAV 20 9.8 0. 0. 1.
GRID 1 0. 0. 0. 246
GRID 2 1. 0. 0.
246
GRID 3 2. 0. 0.
246
GRID 4 3. 0. 0.
246
GRID 5 4. 0. 0.
246
GRID 6 5. 0. 0.
246
GRID 10 0. 0. 10.
123456
MAT1 1 7.1+10 .33
2700.
PBAR 1 1 2.654-3
5.869-7
SPC1 10 123456
1
ENDDATA

Listing 11-3. Input File for Reading a Stiffness Matrix with Large-Field Format

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 11-13

© 2024 Siemens
 11. Direct Matrix Input

Figure 11-2. Results Comparison Between the Conventional Method Versus DMIG

Mass Matrix Using DMIG

You can also include a mass matrix with DMIG in a similar manner as the stiffness matrix with the
following exceptions:

• You use the Case Control command M2GG = “name”, instead of K2GG= “name,” where “name” is the
name of the mass matrix referenced in field 2 of the DMIG Bulk Data entries.

• The matrix defined in the DMIG entries represents the mass matrix terms instead of the stiffness
matrix terms.

Viscous Damping Matrix Using DMIG

You can also include a viscous damping matrix with DMIG in a similar manner as the stiffness matrix with
the following exceptions:

11-14 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Structural Damping Matrix Using DMIG

• You use the Case Control command B2GG = “name”, instead of K2GG= “name,” where “name” is the
name of the viscous damping matrix referenced in field 2 of the DMIG Bulk Data entries.

• The matrix defined in the DMIG entries represents the viscous damping matrix terms instead of the
stiffness matrix terms.

Structural Damping Matrix Using DMIG

You can also include a structural damping matrix with DMIG in a similar manner as the stiffness matrix
with the following exceptions:

• You use the Case Control command K42GG = “name”, instead of K2GG= “name,” where “name” is the
name of the structural damping matrix referenced in field 2 of the DMIG Bulk Data entries.

• The matrix defined in the DMIG entries represents the structural damping matrix terms instead of the
stiffness matrix terms.

Load Matrix Using DMIG

The load vectors can also be read using the DMIG feature if desired, although this is not a commonly
used feature. It is similar to the format for reading in the stiffness or mass matrix, with the following
exceptions:

• You use the Case Control command P2G = “name”, instead of K2GG = “name” or M2GG = “name.”

• The matrix defined in the DMIG entries are columns of the load vectors instead of the stiffness or mass
matrix terms.

• The load matrix, in general, is rectangular and unsymmetrical (i.e., IFO = 9 on the header entry).

• The column number (GJ) you specify on the DMIG input must match the load sequence number. The
GJ field is the load sequence number, and the CJ field isn't used at all for a rectangular matrix. For
example, if there is a total of ten load conditions in your model and you want to add load vectors
to the third and seventh subcases using DMIG, then the NCOL field should be assigned as 10 on the
header entry. Two additional DMIG Data entries, representing the two load vectors added to the third
and seventh subcases, are needed. The GJ field for the first of these two data entries must have a
value of 3 to denote the third load sequence. The remaining data for this entry are the nonzero rows
of this load vector. Their input format is identical to that of the stiffness or mass matrix. The GJ field
for the second of these two data entries must have a value of 7 to denote the seventh load sequence
number. Again, the remaining data for this entry are the nonzero rows of this load vector.

Again the problem in the figure Planer Cantilever Beam is used to illustrate this feature. The input file
is shown in Listing 11-4. In this example, the DMIG feature is used to read in the load for the vertical
tip load in Subcase 10. In the Case Control Section, the P2G = UGLOAD command is used to activate the
DMIG entries with the name of UGLOAD. In the Bulk Data Section, the DMIG load header entry is similar
to that of the stiffness and mass header entry with the exception of the IFO and NCOL fields. In general,

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 11-15

© 2024 Siemens
 11. Direct Matrix Input

the matrix is rectangular; therefore, IFO is defined as 9 instead of 6. NCOL is the total number of load
cases for the problem.

Since there are two load cases, NCOL must be defined as 2. The number of DMIG load data entries is
equal to the number of external load columns added. In this case, you only need one DMIG load data
entry since only one load condition is added to the first subcase. The GJ field of 1 denotes that you are
adding the load to the first subcase (first load sequence). Note that GJ references the load sequence (1)
rather than the SUBCASE ID (10). The CJ field should be blank when using the DMIG load feature. The
rest of the fields are read in the same manner as the stiffness and mass matrices—only nonzero entries
need to be defined. Note that if you have reversed your subcase setup—that is, if you apply your gravity
load as the first subcase and bring in the tip load as the second subcase—then field three of the second
DMIG entry must then be modified to 2 instead of 1.

$ FILENAME - dmigload.dat
ID BAR DMIG
SOL 101
TIME 10
DIAG 8
CEND
$
TITLE = USE DMIG TO BRING IN LOAD VECTOR FOR SUBCASE 1
P2G = UGLOAD
SPC = 10
DISP = ALL
$
SUBCASE 10
LABEL = TIP LOAD AT END
$
SUBCASE 20
LABEL = GRAVITY LOAD
LOAD = 20
$
BEGIN BULK
$
CBAR 1 1 1 2 10
CBAR 2 1 2 3 10
CBAR 3 1 3 4 10
CBAR 4 1 4 5 10
CBAR 5 1 5 6 10
$
DMIG UGLOAD 0 9 1 2
DMIG UGLOAD 1 6 3 100.
$
GRAV 20 9.81 0. 0. 1.
GRID 1 0. 0. 0. 246
GRID 2 1. 0. 0. 246
GRID 3 2. 0. 0. 246

11-16 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Load Matrix Using DMIG

GRID 4 3. 0. 0. 246
GRID 5 4. 0. 0. 246
GRID 6 5. 0. 0. 246
GRID 10 0. 0. 10. 123456
MAT1 1 7.1+10 .33 2700.
PBAR 1 1 2.654-3 5.869-7
SPC1 10 123456 1
$
ENDDATA

Listing 11-4. Input File for Reading in a Load Vector Using DMIG

Figure 11-3. Results Comparison Between Conventional Load Input Versus DMIG

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 11-17

© 2024 Siemens
 11. Direct Matrix Input

Inputting large matrices by hand using the DMIG method can be time consuming, error prone,
and quite tedious. However, in most instances, the DMIG input may be automatically generated
by a DMAP module when using Simcenter Nastran or other external programs. You can also use
PARAM,EXTOUT,DMIGPCH to generate these DMIG entries.

See also

• “EXTOUT” in the Simcenter Nastran Quick Reference Guide

11-18 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 12. Element Data Recovery Resolved at Grid
Points
Overview of Element Data Recovery
Element stresses may be averaged at grid points and then used to calculate estimates of stress
discontinuities in the model mesh. Resolving element data at the grid points is supported only for
selected linear elastic element types (see Table 12-1 for supported element types). The calculation of
average stresses at grid points and the calculation of estimates of stress discontinuities use a common
mesh definition method.

You define the mesh in the OUTPUT(POST) section of the Case Control section using one of two
methods:

• SURFACE input for 2-D shell element meshes

• VOLUME input for 3-D solid element meshes

The averaging of grid point stresses (and strains) and stress discontinuities are calculated independently
for meshes composed of only 2-D shell elements and those composed of only 3-D solid elements.
You use the STRESS Case Control command to request the element stress and strain data used in the
calculation of grid point stresses and stress discontinuities. To substitute the calculation of strains for
stresses, use the STRAIN Case Control command in place of the STRESS command in the Case Control
Section.

You request grid point stresses for reports with the GPSTRESS Case Control command. The calculation of
stress discontinuities is supported for both elements and grid points associated with an element mesh by
the ELSDCON and GPSDCON commands, respectively.

The following sections explain in more detail the requesting and calculation of grid point stresses and
stress discontinuities, respectively.

Table. Elements Types Supported in Element Data Resolved at Grid Point Calculations (Hyperelastic
Elements Not Included)

Element Type 2-D Shell Stress 3-D Solid Stress Stress Discontinuity
CQUAD4 x x
CQUADR x x
CQUAD8 x x
CTRIA3 x x
CTRIAR x x
CTRIA6 x x

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 12-1

© 2024 Siemens
 12. Element Data Recovery Resolved at Grid Points

Element Type 2-D Shell Stress 3-D Solid Stress Stress Discontinuity
CHEXA x x
CPENTA x x
CPYRAM x x
CTETRA x x

See also

• “STRESS” in the Simcenter Nastran Quick Reference Guide

• “STRAIN” in the Simcenter Nastran Quick Reference Guide

• “ELSDCON” in the Simcenter Nastran Quick Reference Guide

• “GPSDCON” in the Simcenter Nastran Quick Reference Guide

Stress Recovery at Grid Points

You can have Simcenter Nastran compute stresses at grid points:

• On surfaces of two-dimensional plate elements, namely, the CQUAD4, CQUADR, CQUAD4, CTRIA3,
CTRIAR, and CTRIA6

• In volumes containing CHEXA, CPENTA, CPYRAM and CTETRA solid elements

By default, Simcenter Nastran outputs stresses at the center of each of these elements and also at the
vertices of CQUADR, CQUAD8, CTRIAR, and CTRIA6 shell elements and the CHEXA, CPENTA, CPYRAM and
CTETRA solid elements. However, the center output isn't always adequate. For example, when you're
working with the surfaces of CQUAD4 and/or CTRIA3 elements, you may also need realistic stress values
at the grid points. Also, in the case of surfaces of CQUADR, CQUAD8, CTRIAR, and CTRIA6 elements or of
volumes containing CHEXA, CPENTA, CPYRAM and CTETRA elements, the vertex stresses output for the
elements that connect to a common grid point are often not identical.

This section describes a rational method that yields accurate and unique stresses at each vertex grid
point in user-defined surfaces or volumes. Grid point stress recovery is only available in linear static
analysis, real eigenvalue analysis, and transient analysis. Grid point stress recovery isn't available for
lamina stresses.

You can also use Simcenter Nastran to estimate the probable error in these grid point stresses (see
“Error Estimates for Grid Point Stress Data”). Estimates are generated for each stress component. The
root mean square error of these estimates is then computed to provide a single measure of the error in
the computation of the stresses at a grid point. This data provides a gross indication of the adequacy of
a finite element model to represent the physical phenomena under investigation. The error estimator is
available in SOLution Sequence 101.

12-2 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Description of Method

Description of Method

A scheme to interpolate and/or extrapolate over a surface from a known set of stresses (the element
stresses) requires that these known stress components be transformed into a consistent coordinate
system prior to interpolation and extrapolation. This coordinate system is defined by the analyst and
it should be generally compatible with the surface of interest in order to obtain accurate grid point
stresses. Generally, such a coordinate system will be the natural coordinate system that the analyst
used in the generation of the model of the surface or volume in question. Calculated grid point stress
components are output in this coordinate system, called the output coordinate system of the surface or
volume.

Each stress component is treated independently of other stress components in the calculation of average
grid point stress components. The stress invariants at grid points (i.e., principal stresses, the von Mises
stress, and the mean pressure) are evaluated from the average values of the stress components at grid
points. The grid point stress components are output in the output coordinate system of the surface or
volume and the stress invariants are oriented relative to this output coordinate system.

Two methods are used to calculate grid point stress components for plate and shell elements:
topological and geometric. A single, simplified, topological method is used to calculate grid point stress
components in volumes containing solid elements. However, prior to a discussion of these methods, it
is necessary to define the four categories into which grid points are divided. The following graphic and
symbols demonstrate the locations of the interior (triangles), corner (squares), and edge (circle) grid
points.

1. Interior Grid Points

Grid points that are connected only to interior line segments; i.e., line segments that are coincident
with the edges of two or more elements.

2. Corner Grid Points

Grid points that are only connected to a single element.

3. Edge Grid Points

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 12-3

© 2024 Siemens
 12. Element Data Recovery Resolved at Grid Points

Grid points that are connected to both interior and exterior line segments. An exterior line segment
is coincident with the edge of only one element.

4. Exception Grid Points

Grid points at which discontinuities in stress can occur. Exception points will be described more
completely under “Treatment of Exception Points – 'BRANCHing'.”

Stresses at interior, corner, or edge grid points are considered to be continuous between directly
connected elements. Stresses at exception points may be discontinuous between connected elements
and, as will be seen, a different grid point stress may be output for each of the connected elements.
You must exercise good engineering judgment when you assess the validity of grid point stresses at
exception points.

Topological Method – Plate Elements and Solid Elements

The topological method for the evaluation of grid point stresses only recognizes the existence of interior,
edge, and corner points; i.e., exception points are not considered as a distinct category of grid point
(exception points arise as a result of geometrical rather than topological characteristics). The relations
used to compute average grid point stresses in surfaces of plate elements by the topological method
are described in Table 12-2. The relations used to compute the average grid point stresses in volumes
containing solid elements are described in the table Evaluation of Average Grid Point Stresses for
Plate and Shell Elements by the Geometric Method. Only the topological method is available for solid
elements.

Note:
The topological method requires the following STRESS recovery locations on the STRESS case
control command:
For the CQUAD4 and/or CTRIA3 elements, use STRESS(CENTER)=n.
For the CQUAD8 and/or CTRIA6 elements, use STRESS(CORNER)=n.
For the solid elements, use STRESS(CORNER)=n.

Table. Evaluation of Average Grid Point Stresses for Plate and Shell Elements by the Topological
Method

Type of Grid
Point Average Grid Point Stress Component Remarks

Interior If CQUAD4 and/or

CTRIA3 elements
connect to the same grid
point as CQUAD8 and/or
CTRIA6 elements, then
where: only the CQUAD8 and
or CTRIA6 elements are

12-4 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Topological Method – Plate Elements and Solid Elements

Type of Grid
Point Average Grid Point Stress Component Remarks

σI = average stress at interior grid point used in the calculation

of average grid point
Ne = number of directly connected elements
stresses.

Edge If an edge point is not

connected to an interior
point by line segments,
then NI = 0, the second
term in the relation σE
is set to zero, and the
where:
factor 2 in the first term
σE = average stress at edge grid point is replaced by 1.
NI = number of interior points connected by line Same as remark for
segments interior grid points.
σIi = average stress at interior grid points

Corner Corner points connected to CQUAD4 element with the Same as remark for
grid point diagonally opposite the corner point, an interior points
interior point, or an edge point.

where:
σc = average stress at corner grid point
σi = CQUAD4 center stress
σD = average stress at grid point diagonally opposite
corner point
(If point diagonally opposite corner point is not an
interior or edge point, then σc = σ1.

Corner Corner points connected to CTRIA3 elements and the Same as remark for
other two grid points, namely, E1 and E2, of the interior points
connected CTRIA3 are edge points:

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 12-5

© 2024 Siemens
 12. Element Data Recovery Resolved at Grid Points

Type of Grid
Point Average Grid Point Stress Component Remarks

where:
σc = average stress at corner grid point
σ1 = CTRIA3 center stressσE1, σE2 = stresses at edge
points E1 and E2
(if either E1 or E2 is not an edge point, then σc)
Corner points connected to one CQUAD4 or one CTRIA6
element:

where:
σr = stress at element vertex connected to corner point
Corner points connected to two or more CTRIA6
elements:

where:
Nc = number of CTRIA6 elements connected to corner
point.

12-6 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Geometric Method – Surface Elements

Table. Evaluation of Average Grid Point Stress for Solid Elements by the Topological Method

Type of Grid
Point Average Grid Point Stress Component Remarks

Interior and Mixtures of CHEXA, CPENTA,

Edge CPYRAM and CTETRA elements
may connect to an interior or
edge grid point.

where:
σI = average stress at interior or edge grid points
Ne = number of directly connected elements
σei = element vertex stresses

Corner The same relation is used

whether a CHEXA, CPENTA,
CPYRAM or CTETRA element is
where: connected to the grid point.
σr = stress at element vertex connected to corner
point

Geometric Method – Surface Elements

In the geometric method, grid point stresses for points which connect to CQUAD8 or CTRIA6 elements
use the average of the element corner stresses. If any CQUAD4 or CTRIA3 elements connect to these
points, their influence is not considered.

The geometric method for the determination of average grid point stresses which connect only to
CQUAD4 or CTRIA3 elements utilizes the numerical technique of least squares. The “best fit” value of the
stress Σ at a point (x, y) is functionally denoted as:

Equation 12-1.

where the coefficients A0, A1x, and A2 must be determined. The coordinates x and y are the locations of
points at which known stresses exist. In accordance with the least squares technique, it is required to
minimize the function:

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 12-7

© 2024 Siemens
 12. Element Data Recovery Resolved at Grid Points

Equation 12-2.

with respect to {A}. Here σ represents the known element stresses.

To perform the minimization, Eq. 12-1 is written in matrix form as:

Equation 12-3.

where:

Equation 12-4.

and Eq. 12-2 is differentiated with respect to the coefficients Ai.

Equation 12-5.

The coefficients {A} are then determined from Eq. 12-5 to be:

12-8 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Geometric Method – Surface Elements

Equation 12-6.

The average grid point stress at a given grid point can then be calculated by substituting Eq. 12-6 into
the matrix representation of Eq. 12-1 as follows:

Equation 12-7.

where:

Q = matrix of interpolation factors

Xo Yo = locations of grid points at which stress is required

Note that if ([T]T[T])−1 is singular, the matrix of interpolation factors is taken to be so that Eq. 12-7

becomes:

Equation 12-8.

where NI = number of known stress points associated with the grid point. In most cases, Eq. 12-8 is
equivalent to the topological interpolation method.

The strategy used in the evaluation of Eq. 12-7 for interior, edge, and corner grid points is presented in
Table 12-4.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 12-9

© 2024 Siemens
 12. Element Data Recovery Resolved at Grid Points

As previously mentioned, the coordinates x and y of Eq. 12-1 are the locations of points at which
known stresses exist. These locations are actually taken as the projected coordinates of the known stress
points on the average surface evaluated at the grid point at which the stress is to be computed. This
average surface is determined to be the surface whose normal is the average of the unit normals to the
elements connected to the grid point of interest. The local x-axis used in the evaluation of the geometric
coefficients of the matrix [T] is taken as the projection of the x-axis of the output coordinate system on
the average surface evaluated at the grid point of interest.

Table. Evaluation of Average Grid Point Stresses for Plate and Shell Elements by the Geometric
Method

Type of Grid
Point Number of Elements Involved in Stress Calculation
Interior All directly connected elements
Edge Two directly connected elements:
Use the two known element stresses plus the stresses at all interior points connected by
line segments of the directly connected elements.
Three or more directly connected elements:
Use the known element stresses of the directly connected elements.
Corner Same as topological method (see the table Elements Types Supported in Element
Data Resolved at Grid Point Calculations (Hyperelastic Elements Not Included)).

Treatment of Exception Points – “BRANCHing”

Exception points have been defined as grid points at which stress discontinuities may occur. Such
discontinuities may occur because a substantial difference in slope exists between adjacent elements
directly connected to the same line segment or if more than two elements are directly connected to the
same line segment. Examples of these two cases are illustrated below.

12-10 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Treatment of Exception Points – “BRANCHing”

Figure 12-1. Three or More Elements Connected to the Same Line Segment

Figure 12-2. Slope Difference between Adjacent Elements Connected to the Same Line

Note:
That these two situations are geometric rather than topological in nature and, therefore, are only
considered under the geometric interpolation option.

Since exception points are defined as grid points at which stress discontinuities may exist, it is
conceivable that a computed average grid point stress at an exception point may be different for each of
the directly connected elements. However, it is more common to find that a subset of the elements that
are directly connected to an exception point forms a subsurface on which stresses may be considered

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 12-11

© 2024 Siemens
 12. Element Data Recovery Resolved at Grid Points

to be continuous. A unique average grid point stress may then be computed for the elements of the
subsurface.

The calculation of average grid point stresses for a subsurface that contains two or more elements is
performed with the geometric interpolation method used for edge points. If a subsurface consists of
a single element, the calculation of an average grid point stress is performed in the same manner as
utilized in the calculation of average grid point stresses at corner points. It must be noted, however, that
σD, σE1, and σE2 may be interior, exterior, or corner points (see the table Evaluation of Average Grid
Point Stresses for Plate and Shell Elements by the Geometric Method).

User Information

The use of the grid point stress recovery option is demonstrated in this section through the
consideration of several example problems. These problems are discussed separately after the
presentation of the requirements that must be met in the Executive Control Section, the Case Control
Section, and the Bulk Data Section.

Executive Control Section

A typical Executive Control Section would be of the following form:

SOL 101
CEND

Case Control Section

The option to obtain grid point stress output is exercised through commands in the Case Control
Section. The general request for the recovery of grid point stresses is accomplished using commands
with the form:

GPSTRESS = ALL

or

SET j = k, l, m, . . .GPSTRESS = j

The k, l, m, . . . reference the identification numbers of the surfaces or volumes of elements that are
defined in the OUTPUT(POST) section of the Case Control Section. Grid point stress data will be output
only for those surfaces or volumes referenced through the GPSTRESS Case Control command.

Note:
If grid point stress data is desired, an ELSTRESS (or STRESS) request for all elements in the surfaces
or volumes of interest must be included in the Case Control Section. Such requests will cause the

12-12 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Grid Point Stress Output Description

output of element stress data as well as grid point stress data. If the output of element stress data
is not desired, the requests ELSTRESS(PLOT) = or STRESS(PLOT) = may be used.

In static analysis, one may place the GPSTRESS command above the subcase level or in individual
subcases. If the GPSTRESS command is above the subcase level, it will be utilized by all subcases and
subcoms unless overridden by a GPSTRESS command in a SUBCASE or SUBCOM.

The OUTPUT(POST) section of the Case Control Section contains specific requests for grid point stress
data for the surfaces or volumes of elements. Any number of element surfaces or volumes may be
defined, but only those surfaces or volumes that are referenced through the GPSTRESS command will
have grid point stress data generated and output.

Element surfaces and volumes are defined through a SURFACE and VOLUME commands in the
OUTPUT(POST) section of the Case Control Section.

Grid Point Stress Output Description

Listing 12-1 through Listing 12-3 are examples of grid point stress output produced for two-
dimensional plate elements requested on the SURFACE Case Control command. Listing 12-1 is the most
common form of output and is always the result when the TOPOLOGICAL method is used, and when
the GEOMETRIC method is used and no surface “BRANCHing” is detected. Note that when the “ELEMENT
ID” value is zero, it always indicates that all elements in the SURFACE connected to the grid point
contribute to the average grid point stress and lie in the “best” subsurface. Listing 12-2 and Listing 12-3
show examples of the output when the GEOMETRIC method is used and when “BRANCHing” occurs.
The negative value for the “ELEMENT ID” indicates the “best” average grid point stress and the element
identification number of an element associated with “best” subsurface. Similarly a positive element
identification numbers indicate another average grid point stress for a “less best” grid point stress and
the element identification number associated with it's subsurface. When TOLERANCE is set positive as
in Listing 12-2 only the “best” grid point stress is calculated when “BRANCHing” is detected and when
TOLERANCE is set negative all subsurface average grid point stress are calculated. See “SURFACE” in the
Simcenter Nastran Quick Reference Guide.

Listing 12-1. Grid Point Stress Output – Typical

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 12-13

© 2024 Siemens
 12. Element Data Recovery Resolved at Grid Points

Listing 12-2. Grid Point Stress Output with “BRANCHing” and TOLERANCE > 0

Listing 12-3. Grid Point Stress Output with “BRANCHing” and TOLERANCE < 0

General Remarks

Each element stress component (independently of all other stress components) is transformed into
the output coordinate system before interpolation. It is necessary that this output coordinate system
vary slowly over the surface for the interpolated stresses to be accurate. This approach is certainly
valid for stress interpolation over elements that are in a common plane and it should be adequate
for engineering analysis when the difference in slope between adjacent elements is small. However,

12-14 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Error Estimates for Grid Point Stress Data

when the elements do vary substantially from a smooth surface, the analyst must utilize engineering
judgment relative to the accuracy of the resultant stress data.

Average grid point stresses are only computed at element vertices. For elements with defined midpoints,
you can estimate the stress at these midpoints to be the average of the computed stresses at the grid
point at the ends of the edge in question.

Mesh Stress Discontinuities at Grid Points

Error Estimates for Grid Point Stress Data

This section describes the error estimation tools that can help you identify areas of your model that may
require refinement.

The essence of the previous averaging procedures used to obtain grid point stress data is to:

1. Convert these local element stress components into a common user-defined coordinate system.

2. Average the several values of each stress component to obtain a unique value of the stress
component that is to be associated with the grid point in question.

3. Compute the stress invariants at the grid points from the stress components at the grid points.

In the general case, the stress components are σx, σy, σz, τxy, τxz, and τyz.

For discussion purposes, the averaging process used to compute the stress components at the grid
points can be represented as:

Equation 12-9.

where:

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 12-15

© 2024 Siemens
 12. Element Data Recovery Resolved at Grid Points

σg = weighted mean value of the stress component computed at the grid point.
σei = value of the stress component in the ith element (i = 1,2,...,Ne) in the neighborhood of the
grid point. δgi is in the same coordinate system as δg.
Wi = weighting factor assigned to the ith element. The sum of the Ne values of Wi must equal
1. (This requirement assures that all computed statistics will be unbiased.) The attribute of
being unbiased implies that the variance is equal to the mean square error. Equal weighting
(i.e., Wi = 1/Ne) is assumed.

An estimate of the error in a particular component of stress at a grid point can then be computed by
assuming that the values of the corresponding stress components computed for the elements in the
neighborhood of the grid point are data points with uncorrelated random errors. It then follows that an
estimate of the probable error in the stress component δg at the grid point is:

Equation 12-10.

where δei = σei − σg. Thus, the probable error δg is the root mean square error in δei divided by .

It should be noted that the standard deviation or root mean square error is a reasonable measure of
precision in many practical cases, but it is easy to provide examples in which the standard deviation is a
poor measure of the concentration of the distribution about the mean. Eq. 12-9 is assumed to provide
an approximate error estimator for the grid point stress data.

Element and Grid Point Stress Discontinuity Output Requests

It should be clear that element and grid point stress discontinuity output can only be obtained if the
analyst has requested grid point stress output via the GPSTRESS command in the Case Control Section of
the Simcenter Nastran input file. The GPSTRESS command, of course, requires that the analyst define all
required SURFACEs and VOLUMEs in the OUTPUT(POST) portion of the Case Control Section.

To output Grid Point Stress DisCONtinuities:

12-16 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Discussion of Error Measures

GPSDCON = ALL

or

SET j = k, l, m, ....GPSDCON = j

To output ELement Stress DisCONtinuities:

ELSDCON = ALL

or

SET j = k, l, m, ....ELSDCON = j

The following remarks should be noted relative to the use of the GPSDCON and/or ELSDCON commands.

1. The GPSDCON and ELSDCON commands are honored only in SOLution Sequences 101, 114, 144,
and 200.

2. The GPSDCON and ELSDCON commands may be placed above the SUBCASE level or in individual
SUBCASEs and/or SUBCOMs. The use of these commands above the SUBCASE level causes stress
discontinuity data to be output for ALL SUBCASEs in the Simcenter Nastran input file. If the analyst
wishes to restrict stress discontinuity output to specific SUBCASEs, GPSDCON and/or ELSDCON
commands should appear only under these specific SUBCASEs.

3. The analyst is cautioned that these commands produce a substantial amount of output.

4. The GPSDCON and/or ELSDCON commands may be effectively used in restarts. For example, one
could include requests for GPSTRESS data during the initial run of a project and then obtain data on
stress discontinuities on a subsequent restart by including GPSDCON and/or ELSDCON commands
in the Case Control Section.

5. Stress discontinuity output will only be provided when grid point stresses have been previously
computed by the methods described in this article. Thus, stress continuity will not be computed
when, for example, both plate and solid elements are connected to a grid point that is involved in
stress discontinuity calculations.

Discussion of Error Measures

Large values of error estimators for CQUAD4 and CTRIA3 elements are not uncommon occurrences
in statistical error measures. For example, the mean and variance have direct analogs in engineering
mechanics. The mean is analogous to the centroid of a body and is thus a measure of where the mass
is centered. The variance is the second moment about the mean, and it tends to be small if the majority
of the mass is concentrated about the centroid. As the mass is dispersed further from the centroid, the
moment of inertia tends to increase. If the mass is concentrated at the centroid (as it might be in some
idealized lumped mass models), the moment of inertia becomes zero.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 12-17

© 2024 Siemens
 12. Element Data Recovery Resolved at Grid Points

Thus, the error estimates for CQUAD4 and CTRIA3 elements will tend toward larger values because the
data used to compute the estimators are dispersed relatively far from the mean. For those elements that
provide stress data at vertices, the estimators will tend toward smaller values because the stress data
tend to be concentrated about the mean in well-designed finite element models.

Generally, the goal is to design a finite element mesh for static analysis so that all important stress
gradients will be adequately represented. If the mesh is not sufficiently detailed, the stress data at
element vertices or element centroids will result in the relatively large values of the error estimators.

Very inaccurate values of these error estimators may be obtained at the edges of defined SURFACEs and
on the faces of defined VOLUMEs. Such edges and surfaces will often correspond with the boundaries of
superelements.

In summary, error estimators can, in some cases, be highly inaccurate. Nevertheless, error estimate data
is quite useful when properly interpreted by the analyst.

Grid Point Stresses and Mesh Stress Discontinuities

Grid point stresses may be requested for grid points connected by CTRIA3, CTRIA6, CTRIAR, CQUAD4,
CQUAD8, CQUADR, CHEXA, CPENTA, CPYRAM and CTETRA elements (with linear material properties
only). In linear static analysis only, a mesh stress discontinuity error analysis may then be performed
based on the computed element and grid point stresses.

Grid point stresses are not computed for laminar stresses.

The grid point stresses are calculated in the user-specified coordinate system. The following real
membrane stresses are output on request:

• Normal stresses in the x and y directions.

• Shear stresses on the x face in the y direction.

• Angle between the x axis and the major principal axis.

• Major and minor principal stresses.

• von Mises equivalent or maximum shear stress (see STRESS Case Control command).

The following real stresses are output on request for grid points connected to solid elements:

• The three normal stresses.

• The three shear stresses.

• Magnitude and direction of the three principal stresses.

12-18 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Grid Point Stresses and Mesh Stress Discontinuities

• The mean pressure.

• The von Mises equivalent or octahedral stress (see STRESS Case Control command).

Only real stresses are available at the grid points. The user has the following options:

• The capability to vary output requests by SUBCASE.

• Choice of elements in stress field.

• Choice of topological or geometric interpolation (surface fields only).

• Choice of output coordinate system.

• Choice of output stress fiber locations (surface fields only).

The grid point stresses are requested in the Case Control Section for printing and postprocessing using
the Case Control commands GPSTRESS (Grid Point Stress) and STRFIELD (Stress Field). The user is
required to request element stress output (STRESS = N) for all elements referenced by selected SURFACE
and VOLUME commands. The grid point stress requests a set of surface or volume fields which are
defined in the OUTPUT(POST) part of the Case Control Section on the SURFACE or VOLUME commands,
respectively. The postprocessing output requests are separated from case control and plot requests by
the OUTPUT(POST) command.

The general rules for Case Control syntax are the same as for plotting. The grid point stress output
is defined on the SURFACE and VOLUME commands. The SURFACE command defines the following:
output coordinate system, list of elements in the stress surface field, output stress fiber location for
stress evaluation and the choice of topological or geometric interpolation method used to determine the
grid point stresses from the elements. The VOLUME command defines the following: output coordinate
system, list of elements in the stress volume, and the choice of DIRECT and/or PRINCIPAL stresses.

In the linear static structured solution sequences only (SOLutions 101, 114, 144, and 200) mesh stress
discontinuities may be requested to estimate errors due to the coarseness of the finite element mesh.
The estimation of the error is computed for the element and grid point stresses and requested by
the ELSDCON and GPSDCON Case Control commands, respectively. These commands select the desired
surfaces and volumes which are defined in the OUTPUT(POST) section of the case control. This output
option also requires a corresponding GPSTRESS or STRFIELD request for the desired surfaces and volumes
and a STRESS case control request for the elements in those surfaces and volumes.

The following is computed and output for stress discontinuities on shell elements:

• Normal stress discontinuities in the x and y directions.

• Shear stress discontinuities on the xy direction.

• Major and minor principal stress discontinuities.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 12-19

© 2024 Siemens
 12. Element Data Recovery Resolved at Grid Points

• von Mises and maximum shear stress discontinuities.

• Error estimate.

The following is computed and output for stress discontinuities on solid elements:

• Normal stress discontinuities in the x, y, and z directions.

• Shear stress discontinuities on the xy, yz, and xz directions.

• Major and minor principal stress discontinuities.

• von Mises stress and mean pressure discontinuities.

• Error estimate.

12-20 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 13. Solution Sequences
Overview of Solution Sequences
Simcenter Nastran is comprised of a large number of building blocks called modules. A module is
a collection of FORTRAN subroutines designed to perform a specific task, such as processing model
geometry, assembling matrices, applying constraints, solving matrices, calculating output quantities,
conversing with the database, printing the solution, and so on. The modules are controlled by an
internal language called the Direct Matrix Abstraction Program (DMAP).

Each type of analysis available in Simcenter Nastran is called a solution sequence. Each solution
sequence is a prepackaged collection of hundreds or thousands of DMAP commands that you select with
a SOL Executive Control statement. Once you select a solution sequence, the software automatically
sends that sequence's particular set of DMAP commands to the modules that are needed to perform the
requested solution.

The major type is called a Solution Sequence. The solution sequences are listed in Table 13-1.

Note: You can also select a solution sequence by using the SOL keyword followed by the name of the
solution sequence. For example, SOL SESTATIC is equivalent to SOL 101. Using the solution numbers
is generally recommended rather than their names because the solution numbers are more easily
recognized.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 13-1

© 2024 Siemens
 13. Solution Sequences

Figure 13-1. Hierarchy in the Structural Solution Sequences

Table. Structured Solution Sequences

SOL Number SOL Name Description

101 SESTATIC Statics with options: Linear Heat Transfer, Alternate
Reduction, Inertia Relief
103 SEMODES Normal Modes
105 SEBUCKL Buckling
106 NLSTATIC Nonlinear Statics
See the Simcenter Nastran Basic Nonlinear Analysis
User's Guide
107 SEDCEIG Direct Complex Eigenvalues
108 SEDFREQ Direct Frequency Response

13-2 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Overview of Solution Sequences

SOL Number SOL Name Description

109 SEDTRAN Direct Transient Response
110 SEMCEIG Modal Complex Eigenvalues
111 SEMFREQ Modal Frequency Response
112 SEMTRAN Modal Transient Response
114 CYCSTATX Cyclic Statics with option: Alternate Reduction
115 CYCMODE Cyclic Normal Modes
116 CYCBUCKL Cyclic Buckling
118 CYCFREQ Cyclic Direct Frequency Response
129 NLTRAN Nonlinear Transient Response
See the Simcenter Nastran Basic Nonlinear Analysis
User's Guide
144 AESTAT Static Aeroelastic Response
See the Simcenter Nastran Aeroelastic Analysis User's
Guide
145 SEFLUTTR Aerodynamic Flutter
See the Simcenter Nastran Aeroelastic Analysis User's
Guide
146 SEAERO Aeroelastic Response
See the Simcenter Nastran Aeroelastic Analysis User's
Guide
153 NLSCSH Static Structural and/or Steady State Heat Transfer
Analysis
159 NLTCSH Transient Structural and/or Transient Heat Transfer
Analysis
187 DDAM DDAM analysis.
190 DBTRANS Database Transfer
200 DESOPT Design Optimization with additional option for only
Sensitivity Analysis
See the Simcenter Nastran Design Sensitivity and
Optimization User's Guide
401 NLSTEP Multistep, structural solution which supports a
combination of subcase types (static linear, static
nonlinear, nonlinear dynamic, preload, modal, Fourier,
cyclic symmetric, buckling) and supports large rotation
and large strain.
402 NLSTPKIN Multistep, structural solution which supports a
combination of subcase types (static linear, static

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 13-3

© 2024 Siemens
 13. Solution Sequences

SOL Number SOL Name Description

nonlinear, nonlinear dynamic, preload, modal, Fourier,
buckling) and supports large rotation kinematics and
large strain.
414,N ROTDYN Rotor dynamic analysis. N represents a valid solution
type. The available types are:

N SOL 414 solution type

101 Maneuvers (linear static)
103 Eigenvalues and
Superelement Reduction
110 Complex Modal Analysis
111 Harmonic Response
129 Transient Response

Obtaining DMAP Listings for Solution Sequences

You can obtain the DMAP listings for any of the solutions sequences by including a COMPILER LIST
statement in the Executive Control Section.

• A description of the data processing and matrix operations for the solution sequences is given in
“Summary of Solution Sequence Operations.”

• The complete details for each of the functional modules are given in Chapter 4 of the Simcenter
Nastran Programmer's Manual.

• Those DMAP modules which are of general use (matrix operations and utility modules) are described
along with examples of their use in “Data Processing and Matrix Operations by Functional
Module.”

Using Alters to Modify Solution Sequences

If you want to modify a DMAP sequence, you can use the alter feature described in Simcenter Nastran
DMAP Programmer's Guide.. For example, you can use an alter to:

• Schedule an EXIT prior to completion in order to check intermediate output

• Schedule the printing of a table or matrix for diagnostic purposes

• Delete or add a functional module to the DMAP sequence

To make alterations to a solution sequence, you should be familiar with the rules for DMAP
programming

13-4 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Understanding Scheduled and Unscheduled Exits

A library of DMAP alters is delivered with Simcenter Nastran. You can use these alters to request options
not available in the solution sequences. You can also use them to perform exceptional operations that
do not fit in with the normal flow of the solution sequences.

Understanding Scheduled and Unscheduled Exits

Scheduled exits can be requested at any point in a structured solution sequence format by means of
the SEMG, SEKR, SELG, SELR, SEMR, and SEALL Case Control commands. If none of these commands are
specified, then SEALL = ALL is the default.

Unscheduled exits are usually caused by errors on input entries or errors in the structural model
resulting from missing or inconsistent input data. When such errors are detected, an unscheduled exit is
performed accompanied with the output of the applicable user error messages. Following the correction
of the input data errors, a modified restart can be performed.

Unscheduled exits may also occur because of machine failure or insufficient time allowance. In these
cases, an unmodified restart is usually made. In some cases, where a portion of the problem has been
completed, including the output for the completed portion, a modified restart must be made following
an unscheduled exit due to insufficient time allowance. The situations are discussed under case control
requirements in the sections dealing with the individual solution sequences.

The initial execution of any problem must be made with complete Simcenter Nastran input data,
including all of the bulk data. However, all or part of the bulk data may be assembled from alternate
input sources.

Restart Procedures

Simcenter Nastran also has an automatic restart capability. The automatic restart is not only driven
by modifications to the Case Control and Bulk Data Sections but also by modifications in upstream
superelements.

In Simcenter Nastran, restarts are accomplished with the NASTRAN Data Definition Language feature,
a file that lists the names and attributes of data types saved for restart. The evidence that a Solution
Sequence has restart capability can be found by the existence of TYPE statements in its DMAP compile
listings. It uses the SubDMAP feature, which allow breaking a large collection of DMAP statements into
more manageable data structures similar to subroutines in other programming languages.

For a restart analysis, the Bulk Data Section consists only of delete “/” entries and any new entries
you want to add. The previous Bulk Data is read from the database. All other parts of the Simcenter
Nastran Input Data (including the File Management statements, the Executive Control statements, the
Case Control commands, the BEGIN BULK command and the ENDDATA entry) must be resubmitted
even though no changes are made in the case control and no new bulk data is added. When changing
solution sequences, the solution number (SOL) must be changed to the number or name of the new
solution sequence.

Any changes in the Case Control Section associated with Bulk Data entry selection or subcase definition,
or changes in the Bulk Data Section.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 13-5

© 2024 Siemens
 13. Solution Sequences

Restarts always start at the beginning of a specified DMAP or SOL sequence. In a restart analysis,
Simcenter Nastran queries the database in two phases.

Phase 1: The software marks all appropriate existing database data blocks and parameters as
existing for the current restart run.

Phase 2: The software checks the current input against the version from which the restart initiates
and deletes any data blocks with modified input from the Phase 1 determined database data blocks
and parameters. This phase is performed by the RESTART module(s) contained within the solution's
DMAP.

Restart only executes DMAP Modules for which some or all of the output does not exist on the current
version database. The S-type parameter is considered as an output data block for restart purposes,
hence if it is not listed in the NDDL, then the module that contains it will be re-executed. All DMAP
modules are executed until the first RESTART module after which output checking is performed. Forced
execution of modules after this point may be manually controlled by SYSTEM (109) flag.

If no such changes are made, the executive system performs an unmodified restart. If changes have
been made only in the output requests, the restart is considered unmodified. However, some modules
may have to be re-executed in order to prepare the output. Output requests that were satisfied, plots
made, etc., should be removed if they are not to be output again.

For modified restarts, a number of previously executed DMAP instructions may have to be re-executed,
depending on the nature of the modifications made by the user. The DMAP instructions that need to
be executed in a modified restart are automatically determined within the program by comparing all
changes made in Case Control commands and Bulk Data entries.

See also

• “Restarts” in the Simcenter Nastran User's Guide

Data Processing and Matrix Operations by Functional Module

The following sections will refer specifically to the structural analysis matrices such as stiffness,
damping, and mass. However, this information also applies to heat transfer analysis for conductance
and capacitance. There is no analogy for mass in heat transfer analysis.

See also

• Simcenter Nastran DMAP Programmer's Guide (for additional information regarding the use of DMAP
modules for matrix operations)

• Chapter 4 of the Simcenter Nastran Programmer's Manual (for detailed descriptions of the individual
functional modules)

13-6 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Geometry Processing in SubDMAP PHASEO

Geometry Processing in SubDMAP PHASEO

Simcenter Nastran begins the initial processing of your total model in PHASEO. In PHASEO, the software
takes data from the input files and stores it in data structures that feed more expensive processes in
later SubDMAPS. Flags are set to control branching for special features. All superelement inputs are
processed. Tables are indexed for faster processing in post-processors. Parameters are read from Case
Control and Bulk Data files by module PVT. The shell element normals at each grid point are computed in
module TASNP2, and output in table GPSNT.

Tables for internal program use are generated by the geometry processors from the grid point
entries, coordinate system definition entries, sequence entries, connection entries, static load entries,
temperature definition entries, and constraint entries. The computer time for all geometry processing is
usually only a small part of the time required for problem solution.

The basic geometry processing is performed by GP1 (Geometry Processor – Phase 1). The software
assembles a list of all grid points and scalar points in internal order for use in relating internal and
external identification numbers. Matrices for transformation from basic to all defined global coordinate
systems are computed, and all grid point locations are transformed to the basic coordinate system.

The element connection tables (ECT) are generated by GP2 (Geometry Processor – Phase 2) from the
connection entries. The external grid point numbers are converted to the internal indices defined in GP1.

The temperature data and static loads are processed by GP3 (Geometry Processor – Phase 3). The
temperature data is collected by sets and the external grid point numbers. Element numbers are
converted to the internal indices defined in GP1 and GP2. In the case of static analysis, the static loads
are also collected by sets and the external identification numbers are converted to internal indices.

The constraint data is processed by GP4 (Geometry Processor – Phase 4). The displacement set definition
table (USET) is generated from the constraint definition entries. USET contains one coded word for
each degree-of-freedom in the model to identify the set or sets of coordinates to which the degree-
of-freedom belongs (see “Understanding Sets and Matrix Operations” for the definition of degree-of-
freedom sets). The multipoint constraint matrix [Rmg] is formed from the MPC and rigid element Bulk
Data entries. In static analysis, the enforced displacement matrix [Ys] is formed for use in the generation
of equivalent static loads.

The BNDSPC module moves SPCs on boundary grid points to downstream superelements.

Model Checkout

A checkout procedure is provided for the development of new models. It includes the bookkeeping
and control blocks, and part of the generation block of the figure Structured Solution Sequence Flow
Chart. It is requested by the PARAM,CHECKOUT,YES option.

See also

• Superelement Analysis in the Simcenter Nastran User's Guide

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 13-7

© 2024 Siemens
 13. Solution Sequences

SEP1 Module

Generates the SEMAP table, which lists superelement membership of grid points and elements. Provides
estimates for computation costs and storage requirements. This module will identify any structural
partitioning errors.

There are two paths through the superelement capability. If the bulk data entries have a BEGIN SUPER
and/or BEGIN SUPER = [SEID] entries the data flow goes through the partitioned superelement capability
modules, SEP1X and SEP2X. If only a BEGIN BULK entry is present the grid point list capability modules
SEP1 and SEP2 are used. The SEP1X path also allows grid point list superelements, with the exception
of fluid structure analysis, SOL 200 (Design Optimization), and the RELEASE entry. Models using these
features must use the grid point list path.

See also

• “Superelement Analysis” in the Simcenter Nastran User's Guide

SEP2, SEP2CT Modules

Partitions out the data for any superelement selected by SE-type Case Control commands. These
modules identify load and rigid element partitioning errors, and errors in the Case Control Section.

Constraint Modules

The GP1 and GP4 modules process all selected superelements. They check for the presence of requested
coordinate systems; for illegal duplicate specifications of mutually exclusive sets, such as SPCi and ASETi
entries referencing the same point; and for correct specifications on SPCi, xSETi, and MPCi entries, and
rigid elements.

You may print the user set table with PARAM,USETPRT, when PARAM,CHECKOUT,YES is present. If
rigid elements or selected MPC entries exist, the constraint matrix [Rmg] (see “Multipoint Constraint
Operations in SubDMAPs SEKR and SEMR2 and SEMRB”) is given three tests:

1. A matrix of rigid body vectors [uogh] is calculated by the VECPLOT module. This matrix includes
zeroes for the rows corresponding to scalar points. The product

is calculated. Terms of [Emh] larger than PARAM,TINY are printed. If the constraint matrix contains
internal constraints, nonzero terms will exist in [Emh]. If an MPC equation includes scalar points,
nonzero terms may exist in the corresponding row of [Emh] although the equation may be in
equilibrium.

2. The product

13-8 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Matrix Assembly Operations in SubDMAP SEMG

is calculated. The product matrix is decomposed by the DCMP module. Diagonal terms of the
factor matrix that are larger than PARAM,MAXRATIO are printed. These terms indicate that the
printed row is linearly dependent on prior rows. Unless the constraint matrix is changed, it is likely
that the model will encounter fatal errors due to singularity in the MCE1 module, or will have a
poorly-conditioned constraint elimination process.

3. The product

is calculated, where [Rmm] is the partition of [Rmg]. This product is also processed by the DCMP
module. If a row appears here that also appears in the messages from the prior step, it confirms
that the constraint matrix is inherently ill-conditioned. If a row appears here that did not appear in
the prior step, it indicates that a change in the m-set exists that can also make the [Rmm] matrix
well conditioned.

The DBC module stores data for use in postprocessors. It runs at many points in the Solution Sequences.
In this SubDMAP it stores basic structural geometric data. The AEROE SubDMAP generates the geometry
of the aerodynamic model. The RESTART module controls the modules that will be run on restart, based
on changes in the input files. The run exits after these checks are performed.

Matrix Assembly Operations in SubDMAP SEMG

The stiffness, mass, damping, differential stiffness, and heat conduction matrices are formed during the
matrix assembly operations in subDMAP SEMG.

Stiffness is generated from the following sources:

• Structural elements defined on connection entries (e.g., CBAR and CROD).

• General elements defined on GENEL entries.

• Scalar springs defined on CELASi entries.

Mass is generated from the following sources:

• A 6 × 6 matrix of mass coefficients at a grid point defined on a CONM1 entry.

• A concentrated mass element defined on a CONM2 entry in terms of its mass and moments of inertia
about its center of gravity.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 13-9

© 2024 Siemens
 13. Solution Sequences

• Structural mass for all elements, except plate elements without membrane stiffness, using the mass
density on the material definition entry.

• Nonstructural mass for all elements specifying a value on the property entry.

• Scalar masses defined on CMASSi entries.

Damping is generated from the following sources:

• Viscous rod elements defined on CVISC entries.

• Scalar viscous dampers defined on CDAMPi entries.

• Element structural damping by multiplying the stiffness matrix of an individual structural element by a
damping factor obtained from the material properties (MATi) entry for the element.

These matrices are formed from the connection and geometric tables prepared by the geometry
processors and the material and element property tables prepared in the Phase 0. The matrix assembly
operations will usually represent a significant portion of the total solution time. The computer time for
matrix assembly operations is linear with the number and type of elements. Unit times for stiffness
matrix assembly are given in the Simcenter Nastran Installation and Operations Guide. Assembly times
for coupled mass matrices, differential stiffness matrices, and heat conduction matrices are about the
same as those for stiffness matrices. Assembly times for lumped mass matrices and damping matrices
are usually relatively small.

The matrix assembly process begins with the generation of various tables in TA1 (Table Assembler)
from the geometric data, connection data and property data. These tables are used in data recovery
operations as well as in the matrix assembly operations.

The matrix assembly operations are performed by EMG (Element Matrix Generator) and EMA (Element
Matrix Assembler) modules. All of the element matrices (KELM, KDICT, MELM, MDICT, BELM, and BDICT)
are generated by EMG. The stiffness matrix exclusive of general elements, [Kxgg], as well as the structural
damping matrix, [K4gg], the mass matrix, [Mgg], and the viscous damping matrix, [Bgg], are assembled in
separate executions of EMA.

The parameter COUPMASS is used to select either coupled mass or lumped mass generation in EMG. If
the parameter WTMASS is present, the mass matrices generated in EMG are multiplied term by term by
the value in field 3 of the PARAM entry. This operation allows you to specify mass data in units of weight.

The structural damping matrix [K4gg] is assembled from elements with structural damping. Structural
damping is specified in the GE field of the MATi entry.

13-10 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Multipoint Constraint Operations in SubDMAPs SEKR and SEMR2 and SEMRB

Equation 13-1.

where:

NE = number of elements
GEe = element structural damping coefficient from MATi entries
[Ke] = element stiffness matrix

The direct matrix inputs requested by the K2GG, K42GG, M2GG, and B2GG Case Control commands
define the stiffness matrix [K2gg], structural damping matrix [K4,2gg], mass matrix [M2gg], and damping
matrix [B2gg], and are assembled by the MTRXIN module. The outputs from EMA and MTRXIN are
combined by the ADD module.

The stiffness contributions from the general elements (GENEL Bulk Data entry) are assembled in SMA3
(Structural Matrix Assembler – Phase 3). The final stiffness matrix [Kgg] is formed by combining [Kxgg]
prepared by the EMA module with the contributions from general elements generated by SMA3.

Weight and balance information is calculated from the mass matrix [Mgg] by GPWG (Grid Point Weight
Generator). The execution of GPWG is controlled by the parameter GRDPNT. The details of the output are
described in “GRDPNT” in the Simcenter Nastran Quick Reference Guide.

Multipoint Constraint Operations in SubDMAPs SEKR and SEMR2 and SEMRB

The multipoint constraint operations apply constraints to the stiffness matrices using the linear
relationships among the displacements you specify on MPC and rigid element Bulk Data entries.

The multipoint constraint equations are initially expressed in the form,

Equation 13-2.

where you supply the coefficients. You also specify the degree-of-freedom that is made dependent by
each equation of constraint, so that the {ug} matrix may immediately be partitioned into two subsets,

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 13-11

© 2024 Siemens
 13. Solution Sequences

Equation 13-3.

where the set, umwhere you supply the coefficients, is the set of dependent degrees-of-freedom. The
matrix of constraint coefficients is similarly partitioned

Equation 13-4.

so that Eq. 13-2 becomes

Equation 13-5.

[Rm] is a nonsingular matrix. We can, therefore, form the multipoint constraint matrix,

Equation 13-6.

so that Eq. 13-4 may be stated as

Equation 13-7.

13-12 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Multipoint Constraint Operations in SubDMAPs SEKR and SEMR2 and SEMRB

Prior to the imposition of constraints, the structural problem may be written as

Equation 13-8.

or, partitioning in terms of the coordinate sets, un and um

Equation 13-9.

Bars over symbols are used to designate arrays that are replaced in the reduction process.

The addition of constraints to the structure requires that the forces of constraint be added to the
equilibrium equations. Thus, writing the equilibrium and constraint equations together in partitioned
form,

Equation 13-10.

where {qm} is the vector of constraint forces on {um}. Straightforward elimination of um and qm gives

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 13-13

© 2024 Siemens
 13. Solution Sequences

Equation 13-11.

or

Equation 13-12.

where

Equation 13-13.

and

Equation 13-14.

The initial partition of Kgg and the operations indicated by Eq. 13-6, Eq. 13-13 and Eq. 13-14 are
performed by appropriate modules of the program. The multipoint constraint matrix, Gm, is used in
structural matrix reduction (Eq. 13-13), load vector reduction, (Eq. 13-14) and data recovery (Eq. 13-7).

The multipoint constraint equations, Rmg, formed in the GP4 module, are partitioned by the MCE1
module (Multipoint Constraint Eliminator – Phase 1) as follows:

13-14 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Multipoint Constraint Operations in SubDMAPs SEKR and SEMR2 and SEMRB

Equation 13-15.

MCE1 also solves the equation

Equation 13-16.

for the transformation matrix [Gmn]. An alternate path is available for module MCE1 in SubDMAP
XMCE1, requested by PARAM,MCE1,NEW. It contains a partitioned solution for R−1mm · Rmn, where the
part of Rmm that is diagonal is partitioned from the coupled part. It was developed at a time when the
partitioned solution was faster than the single-pass solution used in MCE1. The single pass solution was
improved so that it is now faster than the partitioned solution. The partitioned solution is maintained
to provide an alternate solution when the single-pass solution is implausible, or does not complete in a
normal manner.

The MCE2 module (Multipoint Constraint Eliminator – Phase 2) partitions the stiffness matrix

Equation 13-17.

and performs matrix reduction

Equation 13-18.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 13-15

© 2024 Siemens
 13. Solution Sequences

The other structural matrices, [K4gg], [Bgg] and [Mgg], are transformed by formulas that are identical in
form to Eq. 13-13, in subDMAPs SEMR2 and SEMRB.

The MCE2 module is also used in subDMAP SEMR2 to obtain [Mnn].

The UPARTN module partitions [Kgg] to [Kmg] form which is equivalent to

Equation 13-19.

and is used to obtain multipoint forces of constraint. See “Data Recovery Operations in SubDMAP
SEDISP.”

Single Point Constraint Operations in SubDMAPs SEKR, SEMR2 and SEMRB

The single point constraint operations apply constraints to the structural matrices using the single point
constraints you specify on SPCi Bulk Data entries, the PS field on GRID entries, or constrained by the
AUTOSPC feature described in “Constraint and Mechanism Problem Identification in SubDMAP SEKR.”

Single point constraints are applied to the set, us, in the form

Equation 13-20.

where {YS} is a vector of enforced deformations, any or all of whose elements may be zero. The set, un, is
partitioned into us and uf (the free or unconstrained set)

Equation 13-21.

The stiffness matrix, Knn, is similarly partitioned

13-16 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Single Point Constraint Operations in SubDMAPs SEKR, SEMR2 and SEMRB

Equation 13-22.

The complete structural equations including the single point forces of constraint, qs, may be written in
partitioned matrix form as

Equation 13-23.

Straightforward elimination gives

Equation 13-24.

The forces of constraint are recovered by means of the middle row of Eq. 13-23, i.e.,

Equation 13-25.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 13-17

© 2024 Siemens
 13. Solution Sequences

The distinct partitions Kfs and Kss, are needed in subsequent calculations of SPC forces in static analysis.
For the other structural matrices (K4nn, Bnn, and Mnn) the (sf) and (ss) partitions are saved, for dynamic
SPC forces calculation.

The partially constrained stiffness matrix [Knn] is partitioned by UPARTN module as follows:

Equation 13-26.

Similar partitioning operations are performed on [Mnn] in subDMAP SEMR2 to obtain [Mff], [Msf] and
Mss. In subDMAP SEMRB, the structural and viscous damping matrices [K4sf], [K4ss], [B sf] and [Bss] are
obtained from the MATREDU module. All (sf) and (ss) matrices are required for the recovery of single
point constraint reaction forces described in “Data Recovery Operations in SubDMAP SEDISP.”

Constraint and Mechanism Problem Identification in SubDMAP SEKR

A singularity in a linear equation solution leads to a condition where a unique solution is not possible.
For example, if [K] {u} = {F}, and [K] is zero, then if {F} is not zero, no value of {u} will satisfy the
equation. However, if {F} is zero, any value of {u} will satisfy the equation. The equation can always
be satisfied by prescribing {u}, rather than {F}. This is equivalent to constraining u with a single-point
constraint.

There are two types of singularity:

1. The singularity can be identified by considering the stiffness terms of only one grid point. This is
called a grid point singularity.

2. The singularity requires consideration of stiffness terms of more than one grid point. This is called a
mechanism type of singularity.

Grid Point Singularity

Grid point singularities are automatically identified by the GPSP module. They may be automatically
constrained, under user control. First, multipoint constraints are eliminated, leading to the matrix [Knn]
(see “Multipoint Constraint Operations in SubDMAPs SEKR and SEMR2 and SEMRB”). Singularities
are identified by grid point and component number of the failed direction. The ratio of stiffness at the
degree-of-freedom to the largest stiffness at that grid point is listed. If automatic constraint is selected
(see “AUTOSPC” in the Simcenter Nastran Quick Reference Guide), the set membership of the singular
degrees-of-freedom is listed before the automatic constraint and after the automatic constraint. Both

13-18 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 User Options

the mutually exclusive set and the highest combined set identical to the mutually exclusive set are listed
for each identified singularity.

The output will include a Grid Point Singularity Table with the following format:

G R I D P O I N T S I N G U L A R I T Y T A B L E
POINT TYPE FAILED STIFFNESS OLD NEW
ID DIRECTION RATIO USET USET
3801 G 1 0.00E+00 L
S *
3801 G 2 0.00E+00 L
S *
3801 G 3 0.00E+00 L
S *
3801 G 4 0.00E+00 L
S *
3801 G 5 0.00E+00 L
S *
3801 G 6 0.00E+00 L
S *

In this table,

1. G (under the heading TYPE) stands for grid point (S for scalar point is also output).

2. FAILED DIRECTION is the degree-of-freedom component in the displacement coordinate system and
closest to the weakest direction.

3. The STIFFNESS RATIO is the ratio of the stiffness in the weakest direction to the stiffness in
the strongest direction at the grid point (considering all possible directions rather than just
the coordinate directions). Translational and rotational stiffnesses are considered separately. The
default value for the failure criterion is 10−8, which may be changed by the PARAM,EPZERO entry.

4. Note that, for the example shown, the failed degrees-of-freedom were originally in the ul set and
were placed in the us set. This may not always be the case, for reasons noted earlier, or because you
didn't request the AUTOSPC feature.

5. You should carefully inspect the Grid Point Singularity Table because a singularity is very often the
result of a modeling error, which should be corrected.

User Options

The identification of singularities is an automatic feature; however, you may exercise the following
options through PARAM Bulk Data entries described in “Parameters and Parameter Applicability Tables” in
the Simcenter Nastran Quick Reference Guide.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 13-19

© 2024 Siemens
 13. Solution Sequences

Parameter
Name Description
AUTOSPC Controls automatic constraint of a singularity.
EPPRT Specifies the maximum value of a singularity to print.
EPZERO Specifies the maximum value of a singularity.
PRGPST Controls printout of singularity table.
SPCGEN Writes SPC entries to the PUNCH file based on the singularity table.

The singular degrees-of-freedom are shifted to other sets by the following logic if PARAM,AUTOSPC,YES
is specified:

o-set Condition Set Before Constraint Set After Constraint

If o-set exists o s
c b
If no o-set c s
b s

Other mutually exclusive sets not listed above will not be constrained. In addition, singular degrees-of-
freedom on CYAX, CYJOIN, and SPCOFFi entries are identified but not constrained. The most usual cause
of grid point singularity is degrees-of-freedom not constrained by elements because they are not of
interest to the modeler. An example is out-of-plane rotations on a component made of plates lying in
a plane. The automatic constraint provides the correct action for this type of singularity. Another cause
is elements left out by oversight. Here, one typically observes that many or all degrees-of-freedom at a
grid point are singular. The automatic constraint may mask a modeling error in this instance. For this
reason, all singularity messages should be carefully inspected during the modeling checkout phase to
ensure that modeling errors are not being masked.

When using the K2PP, M2PP, B2PP, or TFL Case Control command, PARAM,AUTOSPC,NO is
recommended.

Mechanism-Type Singularities

A more complicated type of instability arises when two or more grid points participate in a singularity
in a linear equation solution. This is known as a mechanism type of instability. Again, for the simple
example shown in Figure 13-2, the equations of equilibrium can not be satisfied at all, or may have
nonunique solutions for special loading conditions. Note, however, that the structure of this example
will pass the grid point singularity checks.

13-20 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Mechanism-Type Singularities

Figure 13-2. A Mechanism Type of Instability

The static equilibrium equation for the example in Figure A Mechanism Type of Instability is

Equation 13-27.

If P1 = −P2, many solutions are possible, where (u1 − u2) = P1/k, but u1 may have any value. If P1 and P2
are arbitrary, the equation cannot be satisfied.

This condition is detected during the matrix decomposition phase of the linear equation solution process
by the DCMP module. The stiffness matrix is decomposed into a lower triangular matrix, [L], and a
diagonal matrix, [D]. As discussed in “User Interface” in the Simcenter Nastran Numerical Methods User's
Guide a diagonal term of [D] (“di”) will approach zero for every mode of instability of the structure.
(In this example there is one instability.) These instabilities are identified by dividing all di's into their
corresponding stiffness term, then listing all ratios larger than a large number. See “MAXRATIO” in the
Simcenter Nastran Quick Reference Guide for a description of several other parameters used to control
this operation.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 13-21

© 2024 Siemens
 13. Solution Sequences

Identify Grid Component for All Ratios > PARAM,MAXRATIO.

A corrective action for this example would be to constrain one of the grid points or to add springs
connecting this spring to other parts of the structure (not shown). Since, in general, there may be many
grid points active in a singularity, there is little reason for picking one over another for connection. The
usual cause of this type of instability is elements left out by oversight. While this may be your intention,
it cannot be detected by the program. For these reasons, there is no automatic method to constrain
mechanisms. The discussion of “MAXRATIO” in the Simcenter Nastran Quick Reference Guide describes
the various actions and user options that are available in the several contexts where singularities may be
detected. Diagnosis techniques and corrective actions are described in the Simcenter Nastran Numerical
Methods User's Guide.

Static Condensation in SubDMAPs SEKR and SEMR2

Structural partitioning is used to refer structural matrices to the boundary points in substructuring
operations and to perform a static condensation in dynamic analyses.

At user option the set of free coordinates, uf, may be partitioned into two sets, uo and ua, such that the
uo set is eliminated first. Thus

Equation 13-28.

The equilibrium equations after the elimination of constraints may be written in partitioned form as

13-22 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Static Condensation in SubDMAPs SEKR and SEMR2

Equation 13-29.

Rearrange the bottom half of Eq. 13-29:

Equation 13-30.

and solve for {uo}:

Equation 13-31.

Substitute for uo into the top half of Eq. 13-29:

Equation 13-32.

It is convenient to define the matrix

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 13-23

© 2024 Siemens
 13. Solution Sequences

Equation 13-33.

so the Eq. 13-32 becomes

Equation 13-34.

where advantage is taken of the symmetry of [Koo].

Following the practice of condensation established in preceding subsections,

Equation 13-35.

where

Equation 13-36.

Equation 13-37.

The [Go] matrix defined in Eq. 13-33 is obtained practically from the solution of

13-24 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Static Condensation in SubDMAPs SEKR and SEMR2

Equation 13-38.

where [Kao]T is treated as a set of load vectors. Each such vector produces a column of [Go].

Once {ua} is obtained the set of omitted coordinates, {uo}, is obtained as follows. Define the set as

the solution of

Equation 13-39.

Then, using Eq. 13-33 and Eq. 13-39 in Eq. 13-30,

Equation 13-40.

In subDMAP SEKR, the constrained stiffness matrix, [Kff], is partitioned by UPARTN as follows:

Equation 13-41.

DCMP and FBS are used to solve the equation

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 13-25

© 2024 Siemens
 13. Solution Sequences

Equation 13-42.

for the static transformation matrix, [Gtoa], and MPYAD performs the matrix reduction

Equation 13-43.

If PARAM,ALTRED,YES is specified then the [Kff] matrix is partitioned and reassembled to form [Kxff] which
has the a-set degrees-of-freedom sequenced last. [Kxff] is partially decomposed

Equation 13-44.

where [Lfo] is a trapezoidal matrix and [Kaa] is the boundary stiffness matrix.

The [Lfo] matrix is partitioned

Equation 13-45.

While [Lfo]T is of the same dimension as the [Gtoa] matrix, it is on the average half as dense, thereby
reducing the storage requirements.

In subDMAP SEMR2, UPARTN partitions the constrained mass matrix [Mff] as follows:

13-26 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Static Solutions in SubDMAP SEKRRS

Equation 13-46.

and successive executions of MPYAD and SMPYAD modules perform the matrix reduction

Equation 13-47.

In subDMAP SEMRB, the structural and viscous damping matrices [Bff] and [K4ff] are condensed by the
MATREDU module.

The partitioned solution for statics is an exact solution. It merely changes the order of operations of
the unpartitioned solution. The static reduction for dynamics is exact for o-set degrees-of-freedom that
are massless, but represents an approximate relumping of o-set masses to the a-set. You must select
enough a-set degrees-of-freedom to adequately define the eigenvectors of the system. The dynamic
reduction technique discussed in “Dynamic Reduction and Component Mode Synthesis in SubDMAP
SEMR3” provides a way of improving the approximation by computing the normal modes for the o-set
degrees-of-freedom.

Static Solutions in SubDMAP SEKRRS

There are two major branches through this subDMAP, standard static analysis and static analysis with
inertia relief effects.

Inertia Relief Not Selected

When inertia relief is not selected and there are no SUPORTi entries are present the a-set equations are
solved directly

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 13-27

© 2024 Siemens
 13. Solution Sequences

Equation 13-48.

Kaa is decomposed into its factor matrices Laa and Da. These factors are passed on to SubDMAP STATRS
for solution of ua.

When inertia relief is not selected but there are SUPORTi entries are present the a-set equations are
partitioned into the r-set, defined on the SUPORTi entries, and the l-set, what is “left over” after the r-set
is removed.

Equation 13-49.

In this context, the r-set is treated as an alternate form of single-point constraints. The partitions of K
involving r are discarded. ur is set to zero. ul is found from the first row of this equation. Loads in Pr go
directly to ground, and are output with the SPC forces, along with the product −Krl · ul.

Inertia Relief Selected

The basic assumption of inertia relief is that the structure is in a state of steady acceleration, and that
all transients due to dynamic effects are negligible. The solution is obtained in a moving reference
frame from which all displacements are measured. For inertia relief analysis inertial loads are computed
automatically that exactly balance the applied loads. A support system is applied to make the system
numerically stable. The inertial loads assume rigid body accelerations when computing the loads.

The derivations given here assume that the system mass matrix Mgg has been reduced to Maa and is
available, as described elsewhere. In the actual DMAP, equivalent methods that never generate Maa are
used instead, in the interests of efficiency. Maa can be a larger size data block, and costly to compute. It
can be shown that the results provided by the two derivations are identical. For example, the rigid body
mass matrix Mrr, defined below as

Equation 13-50.

13-28 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Auto-Support (PARAM,INREL,-2)

can just as easily be formed from the g-set equations

Equation 13-51.

Similar comments can be made about computing the inertial loads that balance the applied loads. The
derivatives are presented here in the form of a-set equations that are more readily understood, if less
efficient in practice, than the g-set equations. You can find equations for the more efficient method TANs
4002 and 4854 or in DMAP compilations. Current research indicates that this capability may shift to a-set
operations in future versions to obtain more automation.

There are at present two forms of inertia relief: automatic selection of a support system (“auto-support”);
and manual selection of the support system, using SUPORTi entries. Both react to PARAM,GRDPNT,
the parameter used to locate a reference point for grid point weight summaries. The more automatic
method is described first.

Auto-Support (PARAM,INREL,-2)

Vectors of rigid body motion Dgr for six unit motions of a reference point are computed from grid point
geometry in module VECPLOT. The Dar component is partitioned from this matrix. Dar is used to compute
the rigid body mass Mrr about the reference point.

Equation 13-52.

Equation 13-53.

Mrr is printed out with the matrix name of QRR.

The acceleration of the reference point ar due to the applied loads is computed,

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 13-29

© 2024 Siemens
 13. Solution Sequences

Equation 13-54.

The accelerations ar are printed with the matrix name of URA.

The resulting inertial loads Pai and total loads Pa are

Equation 13-55.

Equation 13-56.

A constraint equation is written that states that the average weighted motion of the system is zero,

Equation 13-57.

This equation is appended to the a-set to define the 1-set equations in their augmented form,

Equation 13-58.

13-30 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Manual Support (PARAM, INREL, -1 and SUPORTi)

Equation 13-59.

qr are the unknown forces required on the system to produce equilibrium. They should be computational
zeros, and are output with the SPC forces. The equation involving Kll is solved directly, with ua and qr
partitioned from ul.

Manual Support (PARAM, INREL, -1 and SUPORTi)

You prescribe a set of DOFs that constrain the structure in a statically-determinate manner on SUPORTi
entries. These entries define the r-set. There must be six or less DOFs listed, and they must constrain
all rigid body modes that are not constrained by SPCs or other modeling devices. The rigid body modes
are computed from the statics equation, where there are no loads applied, and urr is the identity matrix.
Given

Equation 13-60.

Dlr is found by solving the first row of this equation, then merged with Irr to form,Dar,

Equation 13-61.

The rigid body mass, reference point accelerations, and inertial loads are formed with the same
equations used for auto-support. The loads are partitioned into the l- and r-sets, and the partitioned
form of Kll is used for solution for displacements,

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 13-31

© 2024 Siemens
 13. Solution Sequences

Equation 13-62.

ua is merged from ul and a null partitioned for ur.

A Comparison of Auto- and Manual-Support Solution

If one analyzes the same model with both methods, you should find that stresses, constraint forces,
and other internal quantities are identical except for truncation errors, but the displacements may differ
appreciably. This is because they use a different reference frame. For the many supports, the reference
frame is set up so that the motion at the r-set DOFs are identically zero. One can think of a reference
frame drawn through these points, and all displacements measured relative to it. This reference frame
is moving through space at a constant acceleration rate, so the displacements are relative, rather than
inertial displacements.

For the auto-support option, one can visualize the reference frame as a set of axes connected to every
point with mass with a very complicated set of levers that allows the structure to “breathe”, as loads
are applied to it, but keep the weighted average motion to a zero value. If the model happens to have
a grid point at the reference point, you will observe very small motion there, computational zeros but
not the binary zeros of the manual support method. This is because this motion is computed for the
auto-support method, but assumed to be zero for the manual method.

The choice of an r-set as defined by SUPORTi DOFs requires meeting both a geometric requirement and
a stiffness requirement. The geometric requirement is that the set should form a statically determinate
tie to ground. This can be difficult to do with complicated geometries. The stiffness requirement is that
the DOFs should be stiff enough to provide a stable solution. A method that has been used by some
modelers is to pick points stiff enough that they could be used to lift the structure without damaging it.
Sometimes such points do not exist. The difficulties in picking a stable, reliable support system was one
of the main motivations for development of the more modern auto-support systems.

PARAM, GRDPNT, GID is optional for both methods when superelements are not present, but required by
the manual method when superelements are present. The auto-method does not support superelement
analysis. The default for GRDPNT is GID = -1, which is a flag to use the basic origin as the reference
point when computing rigid body modes. Use of a GID near the c.g. of the structure produces more
esthetically pleasing results. The rigid body accelerations produced by the solution are then measured
near the center of the structure, rather than at a point that may be outside of the structure.

Static and Dynamic Load Generation

Applied Static Loads in SubDMAPs SELG, SELR, and SELRRS

The static load vectors {Pg} are generated by SSG1 (Static Solution Generator – Phase 1) in subDMAP
SELG from your input of applied loads, temperature fields, and enforced deformations.

13-32 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Applied Static Loads in SubDMAPs SELG, SELR, and SELRRS

The direct matrix input requested by the P2G Case Control command define the load matrix [P2g] which
is assembled by the MTRXIN module and is added to the output from SSG1.

If PARAM,INREL,-1 is specified, then

1. The rigid body transformation matrix [Dgr] is generated with the VECPLOT module. Its columns
contain the motion of all degrees-of-freedom due to unit motion of the corresponding reference
point degree-of-freedom.

2. Loads due to unit rigid body accelerations are computed,

3. The loads above are appended to the applied loads {Pj},

The SELA module assembles {Pj} into {Pg}.

The constraints are applied to the static load vectors by SSG2 (Static Solution Generator – Phase 2) in
subDMAP SELR as follows:

Equation 13-63.

Equation 13-64.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 13-33

© 2024 Siemens
 13. Solution Sequences

Equation 13-65.

Equation 13-66.

Equation 13-67.

Equation 13-68.

If PARAM,ALTRED,YES is specified, the diagonal extracted from [Loo] → [Doo] and the reduced load matrix
is computed from

Equation 13-69.

where {uxo} is obtained on a forward pass from the equation

13-34 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Applied Static Loads in SubDMAPs SELG, SELR, and SELRRS

Equation 13-70.

for PARAM,ALTRED,NO, the SSG3 module (Static Solution Generation – Phase 3) solves the equation

Equation 13-71.

for {uoo}, the displacements of the omitted coordinates.

SSG3 also calculates the residual vector, δPo, and the residual vector error ratio, ϵo, for the omitted
coordinates

Equation 13-72.

Equation 13-73.

Except for round-off error, the error ratio ϵo should be zero. Large values of these error ratios usually
indicate singularities in the stiffness matrix. The residual load vector, RUOV, may be output by use of
PARAM,IRES,1.

The quantity

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 13-35

© 2024 Siemens
 13. Solution Sequences

Equation 13-74.

is calculated by the SSG3 module and printed under the heading “External Work.” This component of
strain energy includes effects of thermal loads, element deformations, and enforced displacements that
may be subtracted later in the solution process. For example, an enforced displacement that causes zero
strain will result in external work.

If a r-set is present, subDMAP SELRRS calculates determinate forces of reaction

Equation 13-75.

where:

[Dar] =

Applied Dynamic Loads

Dynamic loads are produced in many subDMAPs. They are generally of the form:

Equation 13-76.

The POi vectors are spatial functions that describe what DOFs are loaded and their scale factors. The
Fi functions are temporal functions that define how the loads change with time in transient analysis x
= t, or with excitation frequency in frequency response analysis x = f. All spatial loads except those on
extra points are calculated in the static analysis load generator module, SSG1, to obtain the Pg vector
described in the prior paragraphs.

13-36 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Applied Dynamic Loads

Enforced motions are produced similarly.

Equation 13-77.

The YOi vectors are again the spatial distribution, and the Fi(x) temporal functions. The YOi vectors are
produced by the GP4 module to produce the Yg vector. It is the same vector described in Applied Static
Loads, except that it is merged to g-set size with null terms from Ys.

The static load and enforced motion vectors are converted into dynamic loading tables in DPD module,
then passed on to the dynamic load generators, FRLG and TRLG. These modules produce the applied
loads and enforced motions at the p-set level, Pp(x ) and Yp(x ), as a function of excitation frequency or
time. For transient analysis velocity and acceleration vectors are also provided, Y (x) and Y (x).

A reduction matrix Rpx is made in the FDRMGEN module, where the subscript “x” implies the d-set, for
direct analysis, and h, for modal analysis. It combines the operations shown in Applied Static Loads
previously. In subDMAP CFORCE2, the following equation, which was first presented in Applied Static
Loads,

Equation 13-78.

is modified to add velocity and acceleration effects,

Equation 13-79.

In frequency response the velocity and acceleration effects can be computed from Ys by multiplying it by
i · ω once or twice, where ω is the excitation frequency.

Pf is reduced to Pd or Ph by the same techniques used in static analysis.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 13-37

© 2024 Siemens
 13. Solution Sequences

For modal analysis with enforced motion it is essential that PARAM,RESVEC,YES be used. It causes static
solutions to be formed from Kfs · Ys -type terms. When the structure is tied to ground in a statically-
determinate manner, these solutions are linear combinations of the rigid body modes that would exist
if the ties to ground were removed. If the ties to ground are redundant, a condition named multi-base
input, the resulting shapes will have significant stresses associated with them. They represent the effect
of low-frequency excitation on the structure when PARAM,RESVEC,YES is not present, for the statically
determinate case the element stresses will be correct but the displacement-type data will be missing the
effect of the static shapes. For the multi-base input case, both stresses and displacement-type output
will be incorrect.

Dynamic Reduction and Component Mode Synthesis in SubDMAP SEMR3

These capabilities use several subsets of the a-set not used by other solution techniques. The v-set
matrix is formed in subDMAP SEKR

Equation 13-80.

The v-set mass matrix is formed similarly in subDMAP SEMR3. The v-set degrees-of-freedom are free
to vibrate during dynamic reduction and component mode computation. They consist of the omitted
terms (o-set), the free boundary points used for rigid-body mode designation (r-set), and the other free
boundary points (c-set). The [Mv] matrix is formed in a similar manner.

If the METHOD command is specified for superelements, then the component modes are computed. The
auto-omit operations described in “Real Eigenvalue Analysis in SubDMAPs SEMR3 and MODERS” are
carried out resulting in the reduced eigensolution equation

Equation 13-81.

where Nz is the number of eigenvectors requested by the EIGR or EIGRL Bulk Data entries. The back-
transformations for the auto-omit-spc are performed, resulting in the expanded eigenvector matrix [φyz].
This matrix is used to purify the approximate eigenvectors found in dynamic reduction,

13-38 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Dynamic Reduction and Component Mode Synthesis in SubDMAP SEMR3

Equation 13-82.

The uncoupled eigensolution components are assembled from [φvz]

Equation 13-83.

Equation 13-84.

where [φbz] = [0]

[φoz] and [φaz] are stored for use with the uncoupled solution obtained using PARAM,FIXEDB,-1 as
described in “Data Recovery Operations in SubDMAP SEDISP.”

The residual vector capability is requested with PARAM,RESVEC,YES. Details on this capability are given
in “Matrix Assembly Operations in SubDMAP SEMG,” where real eigensolutions are discussed. The
RESVEC SubDMAP returns auxiliary functions uo2. They are appended to the eigenvectors, and are given
all of the operations given to eigenvectors,

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 13-39

© 2024 Siemens
 13. Solution Sequences

Equation 13-85.

The inertia relief mode shapes are calculated by statically applying the rigid-body inertial loads to the
structure.

Equation 13-86.

Setting [Hia] to zero and solving for [Hio] results in,

Equation 13-87.

The [Goq] matrix is formed,

Equation 13-88.

The [Goq] matrix is truncated or padded with null columns to provide Nq columns, where Nq is the q-set
size. [Goq] is expanded to a-size,

13-40 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Dynamic Reduction and Component Mode Synthesis in SubDMAP SEMR3

Equation 13-89.

The total transformation matrix is formed by adding dynamic and static components and is reassembled
whenever needed,

Equation 13-90.

The generalized stiffness coefficients are formed from the dynamic transformation,

Equation 13-91.

The physical and generalized degrees-of-freedom are not stiffness coupled. [Kqq] is merged to Na size,

Equation 13-92.

where superscript l refers to matrices formed from dynamic reduction. [Kqq] is stored in the database.
When the total stiffness boundary matrix is needed, it is formed from its partitions,

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 13-41

© 2024 Siemens
 13. Solution Sequences

Equation 13-93.

or

Equation 13-94.

The physical and generalized degrees-of-freedom are mass-coupled. The terms are formed from the
equations,

Equation 13-95.

Virtual mass effects, if any, are added to [Mlaa1] to form [Mlaa].

The total mass matrix is formed by:

Equation 13-96.

13-42 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Dynamic Reduction and Component Mode Synthesis in SubDMAP SEMR3

For the rigid formats [Maa] is formed directly from its partitions,

Equation 13-97.

The a-set matrices do not contain static effects in the t-set components, and dynamic effects in the q-set
components. Virtual mass effects, if any, are added to [Maa] to form [Mmaa].

The damping matrices [Bgg] and [K4gg] are reduced to the a-set by the MATREDU module which performs
the following operations:

1. Eliminate multipoint constraints

2. Eliminate single point constraints

3. Partition omitted degrees-of-freedom

4. Perform static condensation

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 13-43

© 2024 Siemens
 13. Solution Sequences

Formulation of Dynamic Equations in SubDMAP GMA

The tables prepared by the geometry processors are modified by DPD (Dynamic Pool Distributor) to
include the extra points (e-points) introduced for dynamic analysis.

[Kmaa], [Mmaa], [Baa], and [K4aa] are expanded to the d-set to form [K1dd], [M1dd], [B1dd], and [K4dd]. The
direct input matrices [K2pp], [M2pp], and [B2pp] requested by the K2PP, M2PP, and B2PP Case Control
commands are reduced to the d-set to form [K2dd], [M2dd], and [B2dd], respectively.

The general dynamic equation used in the direct methods is

Equation 13-98.

p is the derivative operator

Equation 13-99.

where:

{Ud} = the extra points

For direct frequency response and complex eigenvalue analysis the stiffness, damping, and mass
matrices are generated as follows:

13-44 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Formulation of Dynamic Equations in SubDMAP GMA

Equation 13-100.

Equation 13-101.

and

Equation 13-102.

For direct transient response the stiffness, damping and mass matrices are generated as follows:

Equation 13-103.

Equation 13-104.

and

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 13-45

© 2024 Siemens
 13. Solution Sequences

Equation 13-105.

where:

• [K1dd] is the reduced structural stiffness matrix plus the reduced direct input K2GG (symmetric).

• [K2dd] is the reduced direct input matrix K2PP plus the reduced transfer function input (symmetric or
unsymmetrical).

• [K4dd] is the reduced structural damping matrix obtained by multiplying the stiffness matrix [Ke] of
an individual structural element by an element damping factor ge and combining the results for all
structural elements (symmetric) plus the reduced direct input K42GG (symmetric).

• [B1dd] is the reduced viscous damping matrix plus the reduced direct input B2GG (symmetric).

• [B2dd] is the reduced direct input matrix B2PP plus the reduced transfer function input (symmetric or
unsymmetrical).

• [M1dd] is the reduced mass matrix plus the reduced direct input M2GG (symmetric).

• [M2dd] is the reduced direct input matrix M2PP plus the reduced transfer function input (symmetric or
unsymmetrical).

• g is the structural damping coefficient on the PARAM Bulk Data entry G. The frequencies ω3 and ω4 are
specified on the PARAM Bulk Data entries as W3 and W4, respectively.

For direct solutions [Kdd], [Mdd], and [Bdd] are examined to identify rows and columns which are null in
all three matrices. If the parameters ASING = 0 (the default) the singularities caused by the null rows and
columns are removed as follows:

1. For direct frequency and direct transient response, the stiffness matrix [Kxdd] is formed from [Kdd]
by placing unity on the diagonal for each null row and column.

2. For direct complex eigenvalue analysis null rows and columns are discarded from [Kdd], [Mdd], and
[Bdd] to form [Kddd], [Mddd], and [Bddd]. This is performed in subDMAP DCEIGRS. See “Complex
Eigenvalue Analysis in SubDMAP CEIGRS.”

If parameter ASING = -1 and null rows and/or columns exist, a fatal error will result.

In modal frequency response, and complex eigenvalue analysis, structural damping and viscous
damping are included in the dynamic matrices as follows:

13-46 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Formulation of Dynamic Equations in SubDMAP GMA

Equation 13-106.

and

Equation 13-107.

In modal transient response, the structural damping and viscous damping are included as follows:

Equation 13-108.

and

Equation 13-109.

The general dynamic equation used in the modal method is:

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 13-47

© 2024 Siemens
 13. Solution Sequences

Equation 13-110.

where p is the derivative operator

Equation 13-111.

ζi are the modal coordinates.

The transformation from {Uh} to {Ud} is

Equation 13-112.

where:

[φdh] =

The dynamic matrices expressed in modal coordinates are assembled in GKAM (General K Assembler-
Modal) as follows:

13-48 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Formulation of Dynamic Equations in SubDMAP GMA

Equation 13-113.

Equation 13-114.

and

Equation 13-115.

where, if PARAM,KDAMP = 1 (default)

If PARAM,KDAMP = –1 in complex eigenvalue analysis, modal frequency response, aerodynamic flutter,

and aeroelastic response the viscous modal damping will be incorporated into the complex stiffness
matrix as follows:

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 13-49

© 2024 Siemens
 13. Solution Sequences

Equation 13-116.

Values of g(fi) are specified on the TABDMP1 Bulk Data entry, selected by the SDAMPING Case Control
command. If KDAMP = 1 (default), the matrices [b] and [k] are formulated as in modal transient
response.

In SubDMAP CMPMODE damping of CMS (o-set) modes is calculated by similar techniques. This q-
set damping is expanded to g-set size, and placed in the Baa matrix for summing into downtown
superelements.

Data Recovery Operations in SubDMAP SEDISP

The dependent displacements and the single-point forces of constraint are calculated in SDR1 (Stress
Data Recovery – Phase 1). The inputs to SDR1 are columns of the solution vectors corresponding to
each eigenvalue, loading condition, or output time step. In the case of transient analysis, there are three
columns, corresponding to displacement, velocity and acceleration, for each output time step.

Load Modification

If PARAM,INREL,-1 or PARAM,INREL,-2 are specified, then any component of the load vector is first
multiplied by [Crx] to produce a set of loads that are a linear combination of applied and inertial loads
called {P1l}

Equation 13-117.

Static Analysis

In the case of static analysis and PARAM,ALTRED,NO, SDR1 recovers dependent displacements

13-50 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Static Analysis

Equation 13-118.

Equation 13-119.

where

Equation 13-120.

If PARAM,ALTRED,YES is specified, then the displacements {uo} are obtained from a backward pass from
the equation

Equation 13-121.

where {uyo} = {uxo} − [Lao]T{ua}

Enforced displacements are combined with uf

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 13-51

© 2024 Siemens
 13. Solution Sequences

Equation 13-122.

Equation 13-123.

where

Equation 13-124.

and recovers single point forces of constraint

Equation 13-125.

and multipoint forces of constraint

13-52 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Eigenvalue Analysis

Equation 13-126.

In superelement static analysis, if PARAM,FIXEDB,-1 is present, then ua is set to zero, and data recovery is
performed using the modified equation

Equation 13-127.

This is equivalent to obtaining a solution with all exterior (“boundary”) degrees-of-freedom fixed.
Displacements are due only to interior displacements due to interior loads, that is, the {uoo} component.

Eigenvalue Analysis

In the case of eigenvalue analysis SDR1 recovers dependent components of the eigenvectors

Equation 13-128.

Equation 13-129.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 13-53

© 2024 Siemens
 13. Solution Sequences

Equation 13-130.

Equation 13-131.

Equation 13-132.

and single-point forces of constraint

Equation 13-133.

If PARAM,DYNSPCF,NEW (default) then the single-point forces are computed as follows:

Equation 13-134.

In superelement analysis, if PARAM,FIXEDB,–1 is present, the uncoupled eigenvectors are recovered

from the modified equations

13-54 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Dynamic Analysis

Equation 13-135.

Equation 13-136.

with {φaz} and {φoz} from “Superelement Analysis.”

The multipoint forces of constraint qmq are recovered by

Equation 13-137.

Dynamic Analysis

In the case of dynamic analysis with the direct formulation the extra points are partitioned out of {ud}.

Equation 13-138.

SDR1 recovers the dependent components

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 13-55

© 2024 Siemens
 13. Solution Sequences

Equation 13-139.

Equation 13-140.

and recovers single point forces of constraint

Equation 13-141.

If PARAM,DYNSPCF,NEW (default) then the forces are computed as follows.

In transient response analysis, the formulation is

Equation 13-142.

where:

MxSf = MSf + M2ppSf

BxSf =

KxSf = KSf + K2ppSf

13-56 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Dynamic Analysis

In frequency response and complex eigenvalue analysis, the formulation is

Equation 13-143.

where:

MxSf = MSf + M2ppSf

BxSf = BxSf + B2ppSf
KxSf = (1 + ig)KSf + K2ppSf + iK4ggSf

The multipoint forces of constraint qmg are recovered

Equation 13-144.

where qmg is the reactions expanded to the g-set and Xm and Pm are the internal and external loads on
the m-set degrees-of-freedom, respectively. (See also Eq. 9 in Section 3.5.1 in The NASTRAN Theoretical
Manual).

In transient response analysis,

Equation 13-145.

where:

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 13-57

© 2024 Siemens
 13. Solution Sequences

Mxmg = Mmg + M2ppmg

Bxmg =

Kxmg = Kmg + K2ppmg

In frequency response and complex eigenvalue analysis,

Equation 13-146.

where:

Mxmg = Mmg + M2ppmg

Bxmg = Bmg + B2ppmg
Kxmg = (1 + igKmg + K2ppmg + iK4ggmg

For frequency response there is one complex vector for each loading condition. For complex eigenvalue
analysis there is one complex eigenvector for each complex root and there is no loading term {Ps}
associated with the determination of the single point constraints. For transient analysis there is a
displacement, velocity, and acceleration associated with each output time step.

In the case of dynamic analysis with the modal formulation, you may obtain improved element stresses
by requesting the mode acceleration data recovery technique.

The use of the mode acceleration technique requires the decomposition of the l-set stiffness matrix
[Kll]. This decomposition is not otherwise performed in dynamic analysis unless the model has rigid
body modes. Also, the technique requires a forward/backward substitution operation with a number
of righthand sides equal to the number of solution vectors. A subset of the solution vectors can be
obtained can be obtained by using the OTIME or OFREQUENCY entries in the Case Control Section. The
subset of solution vectors is prepared by the MODACC functional module.

In the case of dynamic analysis with the modal formulation, two methods of data recovery are
available. The default procedure is usually much more efficient for data recovery operations when the
modal formulation is used for dynamic analysis. In this procedure, the dependent components of the
eigenvectors used in the modal formulation are first recovered by SDR1 using the same sequence of
operations described for problems in real eigenvalue analysis (after eliminating unwanted modes). The
complete eigenvectors are used as the input to SDR2 in order to determine the requested forces and
stresses in terms of modal coordinates.

13-58 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Data Recovery Operations in SubDMAP SEDRCVR

In dynamic analysis, you may restrict the output request to the response of selected points in the
solution set. In this case, the time-consuming operations in the recovery of the dependent degrees-of-
freedom can be avoided. This solution set processing is performed by VDR (Vector Data Recovery).

The modal data generated by SDR1 and SDR2 are formed into a data matrix by DDRMM (Dynamic Data
Recovery Matrix Method). The output quantities are determined by multiplying the data matrix by the
modal solution vectors. The computer time for this matrix multiplication is proportional to the number
of output times, the number of modes used in the modal formulation, and the number of output
frequencies or output time steps. If both the number of output items and the number of modes are
small, the computer time required will not be significant when compared to the total problem solution
time. Although the DDRMM method of data recovery is more efficient, a complete {ug} is not formed.
For this reason the method does not permit the creation of deformed structure plots, grid point force
balance or the use of the mode acceleration technique.

The alternate procedure may be invoked using PARAM,DDRMM,-1. With this method the solution vectors
are first transformed from modal to physical coordinates

Equation 13-147.

This operation is performed by MPYAD in subDMAP MODACC. Having determined the vectors {ud} in
physical coordinates, the data recovery operations can be completed using the same sequence of
operations described for the direct formulation of dynamics problems.

Most of the operations in the recovery of the dependent components do not require significant amounts
of computer time. However, in the case of the multiply-add operation associated with the recovery of
the omitted coordinates ([Goa] {ua}, [Goa] {φa}) a significant amount of computer time will be required
if there is a large number of vectors and a large number of components in the a-set. This operation is
likely to be especially costly for transient problems because there is usually a large number of output
time steps and there are three vectors for each time step.

Data Recovery Operations in SubDMAP SEDRCVR

Simcenter Nastran calculates the element forces and stresses in SDR2 (Stress Data Recovery – Phase 2).
The matrix operations in this phase of data recovery are similar to those performed in EMG for stiffness
matrix generation. The computer time will be proportional to the number of elements for which output
is requested.

If static loads are applied to intermediate points on the CBAR element, you can request the output
of stresses and forces at a selected set of intermediate points. The additional calculations required for
output at intermediate points on CBAR elements is performed in functional module SDRX.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 13-59

© 2024 Siemens
 13. Solution Sequences

The solutions and data recovery operations for transient analysis are initially sorted by the time step
(SORT1). A transpose operation is performed by SDR3 (Stress Data Recovery – Phase 3) in order
to prepare the requested output sorted by external point identification number or external element
identification number (SORT2). In frequency response problems, you may request either SORT1 or
SORT2 output. The SORT2 output will usually be desirable if the solution is made for a large number of
frequencies. In any event, the SORT2 output must be prepared if XY-plots are requested.

Computer time for the preparation of SORT2 output will be directly related to the number of items
requested for output. If a large number of items are requested for output, the transpose operation will
require a significant amount of computer time.

You may request the output of a grid point force balance table for a selected set of points. This table
lists the forces acting at each selected point due to element forces, single point forces of constraints and
applied loads. The grid point force balance table is prepared by GPFDR (Grid Point Force Data Recovery).

You may also request the output of an element strain energy table and/or an element kinetic energy
table for a selected set of elements. The tables include the energy for each element, the percentage
of the total energy for each element, and the energy density for each element. The element energy is
computed from the equations

Equation 13-148.

Equation 13-149.

where the subscript e implies the data for one element. The energy density is computed by dividing
the element energy by the element volume. Some elements do not have enough data input to allow
calculation of element volume, such as CELASi elements. The density output for these elements is set to
zero. Simplified equations for other element volumes may be used. For example, offsets are ignored for
bar-like elements, and midside node locations are ignored for plate and solid elements. Element volumes
are calculated by the ELPRT module, and can be output by use of PARAM,EST,1. The strain energy table is
prepared by functional module GPFDR.

13-60 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Summary of Solution Sequence Operations

Summary of Solution Sequence Operations

This section describes the data processing and algorithms that are performed by special purpose
functional modules, sequences of functional modules (subDMAPs) and complete solution sequences.

See also

• Simcenter Nastran DMAP Programmer's Guide (for detailed descriptions of functional modules)

• Chapter 4 of the Simcenter Nastran Programmer's Manual

Static Solutions

There are several major paths for static solutions and some minor ones. For standard static analysis the
stiffness matrix is decomposed in SubDMAP SEKRRS, in preparation for a linear solution in SubDMAP
STATRS. There is also a form of static analysis called inertia relief analysis.

There are now several paths through inertia relief analysis, reflecting that it is still in a state of
development, with more changes likely in future systems. This revision of this manual will discuss inertia
relief in SOL 101 only. Each path has advantages and limitations. There is another method in SOL 144,
Static Aeroelasticity. It is documented elsewhere.

All methods of inertia relief analysis take the standard static analysis equation

Equation 13-150.

where Pa are the applied loads, and add a set of constraint equations Rra and automatically-computed
inertial loads Pi. The constraints are required because Kaa is for a free structure and cannot be solved
reliably without some form of constraint.

Equation 13-151.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 13-61

© 2024 Siemens
 13. Solution Sequences

Equation 13-152.

In theory, if Pai is a set of balanced loads (no resultant) the solution of ua is possible without the
need for constraint equations. In practice, Pai may have some unbalance due to numerical truncation
effects. Even a small resultant load will provide a displacement solution at infinity in theory, or with very
large numbers in practice, usually. The same truncation effects that unbalance the loads may provide
enough “ground springs” in the stiffness matrix to allow a solution, but it will be a poor quality solution.
Everything will first move 1.E5 units in the same direction. Stresses will be determined by subtracting
numbers of this size to find element displacement perhaps on the order of 1.E-3. Ironically, if all goes
well the forces in the support system are miniscule, but it must be present to control stability when
truncation errors are present.

Types of Inertia Relief

Inertia Relief analysis is a quasi-dynamic analysis. It is assumed that the structure is not tied to ground
and is in a state of steady acceleration caused by the applied loads. The displacements measured relative
to ground are constantly increasing and cannot be determined. Displacements are therefore determined
with respect to an accelerating reference frame that is attached to the structure. The applied loads are
balanced by inertial loads that are calculated automatically from the steady acceleration acting on the
mass properties of the model. Inertia relief analysis is requested with PARAM,INREL,[x], where [x] may
have the values -1 or -2.

The reference frame may be obtained by two different methods.

• For the manual support method (PARAM,INREL,-1) you select 6 or fewer DOFs using SUPORTi entries.
The reference frame is defined from these DOFs.

• For the auto support method (PARAM,INREL,-2) 6 constraint equations are written that constrain the
structure to zero motion, on average, while leaving it free to “breath” and deflect due to the applied
loads and inertial loads.

Both methods find it convenient to have “rigid body modes shapes” that represent a set of vectors that
represent all possible strain-free motion of the structure when not loaded. The technique used to obtain
rigid body mode shapes differs between the two methods.

Rigid Body Modes Using Manual Support (PARAM,INREL,-1 and SUPORTi)

You prescribe a set of DOFs that constrain the structure in a statically-determinate manner on SUPORTi
entries. These entries define the r-set. There must be six or less DOFs listed and they must constrain all
rigid body modes that are not constrained by SPCs or other modeling devices. The rigid body modes
are computed from the static analysis equation. There are no loads applied and Irr is the identity matrix.
Given

13-62 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Rigid Body Modes Using Auto-Support (PARAM,INREL,-2)

Equation 13-153.

Dlr is found by solving the first row of this equation, then merged with Irr to form Dar,

Equation 13-154.

Equation 13-155.

Rigid Body Modes Using Auto-Support (PARAM,INREL,-2)

Vectors of rigid body motion Dgr for six unit motions of a reference point are computed from grid point
geometry in module VECPLOT. The Dar component is partitioned from this matrix.

The Basic Equation of Motion for Inertia Relief

Inertia relief analysis starts with the a-set equation

Equation 13-156.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 13-63

© 2024 Siemens
 13. Solution Sequences

and adds a constraint equation

Equation 13-157.

The last equation implies a similar equation on a generalized constraint force variable vector, qr

Equation 13-158.

If Pai represents balanced loads, qr is zero. If there is a small unbalance due to numerical truncation
effects, qr is then small, but finite. There are two methods for defining Rar, as discussed below. Both have
the property of defining some part of ua to have zero motion, so that the other parts of ua are in effect
measured relative to the constrained DOFs.

On of the solution methods partitions Kaa into its r-set and l-set components. The first equation is
partitioned, and the a-set equations and the constraint equation may combine into one equation,

Equation 13-159.

The methods of computing Piare described below. Yr, the generalized displacement input, is set to zero.

Constraint Equations Using Manual Supports

In the constraint equation Rrr = Irr, the identity matrix. Rrl is null. The last row can then be solved for ur =
Yr = 0. The second equation can then be used to eliminate qr,

13-64 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Constraint Equations Using Automatic Supports

Equation 13-160.

Equation 13-161.

Because Rrl is null, the first equation then becomes

Equation 13-162.

This equation can be solved for ul reduced from ua directly. It will be shown later that Pr terms are
transferred to Pl terms in the inertial loads. ua is formed from a merge of ul and null terms for ur.

Constraint Equations Using Automatic Supports

A constraint equation is written that states that the average weighted motion of the system is zero,

Equation 13-163.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 13-65

© 2024 Siemens
 13. Solution Sequences

Equation 13-164.

Rra is generally a dense matrix, with non-zero terms at most DOFs with mass. It could be used to reduce
the Kaa matrix with equations similar to those used for multi-point constraints, but that would produce a
dense Kll matrix, much more expensive to decompose and solve than the Kll used in the manual support
method. A solution method is used which increases solution costs only slightly.

The r-set l-set partitioning is not needed. The constraint equation is appended to the a-set equations to
define the l-set equations in their augmented form,

Equation 13-165.

Equation 13-166.

qr are the unknown forces required on the system to produce equilibrium. They should be computational
zeros, and are output with the SPC forces. The equation involving the augmented Kll is solved directly
for ul · ua and qr are then partitioned from ul. In superelement analysis the constraints are applied in the
residual structure only.

Matrix Decomposition in SubDMAP SEKRRS

There are two major branches through this SubDMAP, standard static analysis and static analysis with
inertia relief effects.

Inertia Relief Not Selected

When inertia relief is not selected and there are no SUPORTi entries present the a-set is decomposed
directly in the DCMP module,

13-66 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Computation of Inertial Loads

Equation 13-167.

Laa is lower triangular, and Daa is diagonal. Both are stored in an unconventional (“packed”) format,
and output in a single data block named LLL. This name is used because the l-set and the a-set are
synonymous when no SUPORTi entries exist. The factor matrices Laa and Daa are passed on to SubDMAP
STATRS for solution of ua.

When inertia relief is not selected but there are SUPORTi entries present, the a-set equations are
partitioned into the r-set, defined on the SUPORTi entries, and the l-set, what is “left over” after the r-set
is removed.

Equation 13-168.

In this context, the r-set is treated as an alternate form of single-point constraints. The partitions of K
involving r are discarded. ur is set to zero. ul is found from the first row of this equation. Loads in Pr go
directly to ground, and are output with the SPC forces, along with the product −Krl · ul from the second
equation. Kll is decomposed with the same equations used for conventional static analysis.

Decomposition of Kll

For the manual support method Kll is defined above. Its factors, used to compute the rigid body modes
are also used later to compute the static solution. For the auto support method the factors of the
expanded form of Kll are passed on.

Computation of Inertial Loads

The basic assumption of inertia relief is that the structure is in a state of steady acceleration and
that all transients due to dynamic effects are negligible. The solution is obtained in an accelerating
reference frame from which all displacements are measured. Inertial loads are automatically computed
that exactly balance the applied loads. A support system is applied to make the system numerically
stable. The inertial loads assume rigid body accelerations when computing the loads.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 13-67

© 2024 Siemens
 13. Solution Sequences

The derivations given here assume that the system mass matrix Mgg has been reduced to Maa and is
available, as described elsewhere. In the actual DMAP, equivalent methods that never generate Maa are
used instead, in the interests of efficiency. Unit inertial loads are computed from Mgg, then reduced by
load reduction techniques, avoiding the need for Maa · Maa can be a large size data block, and costly to
compute. It can be shown that the results provided by the two derivations are identical. For example, the
rigid body mass matrix Mrr, defined below as

Equation 13-169.

can just as easily by formed from the g-set equations

Equation 13-170.

Dar is used to compute the rigid body mass Mrr about the reference point.

Equation 13-171.

Mrr is printed out with the matrix name of QRR.

The acceleration of the reference point ar due to the applied loads is computed,

Equation 13-172.

13-68 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 A Comparison of Auto- and Manual-Support Solutions

The acceleration ar are printed with the matrix name of URA.

The resulting inertial loads Pai and total loads Pal are

Equation 13-173.

Equation 13-174.

For both methods, the mass matrix is reduced from Mgg size to Mrr size with load vector operations that
can be shown to be identical to the Maa operations described here, except for the order of operations.
The load and mass transformations are based on rigid body modes produced from geometry.

A Comparison of Auto- and Manual-Support Solutions

When the same model is analyzed with both methods you should find that stresses, constraint forces,
and other internal quantities are identical except for truncation errors, but the displacements may differ
appreciably. This is because they use a different reference frame. For the manual supports, the reference
frame is set up so that the motion at the r-set DOFs are identically zero. Think of a reference frame
drawn through these points, and all displacements measured relative to it. This reference frame is
moving through space at a constant acceleration rate, so that displacements are relative rather than
inertial displacements. There are many sets of SUPORT i entries valid for one model. If you change
SUPORTi DOFs you may also expect to see changes in displacements, but not in internal loads or element
stresses.

For the auto-support option, one can visualize the reference frame as a set of axes connected to every
point with mass with a very complicated set of levers that allow the structure to “breathe” as loads are
applied to it, but keep the weighted average motion to a zero value. If the model happens to have a
grid point at the reference point, you will observe very small motion there, computational zeros but
not the binary zeros of the manual support method. This is because this motion is computed for the
auto-support method, but assumed to be zero for the manual method.

The choice of an r-set as defined by SUPORTi DOFs requires meeting both a geometric requirement and
a stiffness requirement. This geometric requirement is that the set should form a statically determinate
tie to ground. This can be difficult to do with complicated geometries. The stiffness requirement is that
the DOFs should be stiff enough to provide a stable solution. A method that has been used by some
modelers is to pick points stiff points do not exist. The difficulties in picking a stable, reliable support

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 13-69

© 2024 Siemens
 13. Solution Sequences

system was one of the main motivations for development of the more modern auto-support system.
SUPORTi entries are not allowed for the auto support method, and will cause a fatal error exit when
present.

PARAM,GRDPNT,GID is optional for both methods when superelements are not present. It is required
when superelements are present, and must be on a boundary point attached to all superelements.
This same point should be used for the SUPORTi DOFs for the manual support method. It defines the
reference point used for generating the geometry-based rigid body vectors. The default for GRDPNT is
GID=-1, which is a flag to use the basic origin as the reference point when computing rigid body modes.
Use of a GID near the c.g. of the structure produces more esthetically pleasing results. The rigid body
accelerations produced by the solution are then measured near the center of the structure, rather than
at a point that may be outside the structure.

Applications of Inertia Relief

Although this capability could be used in an environment where the reference frame is accelerating
steadily, such as a free-falling elevator, the more common usage is when the modeler attempts to apply
loads in equilibrium. He wishes the net acceleration of the structure to be zero. A ship floating in the
sea, with pressure loads balancing gravity loads, is an example. Developing balanced load sets can
be tedious on large models with complex geometry. The accelerations printed out in the URA matrix
can be used as a measure of load quality. If the accelerations are negligibly small, the load balance is
adequate. If the accelerations are large, some adjustment of the loads sets may be necessary. For good
inertia relief analysis a good mass distribution is needed. As the structural weight is usually only a small
proportion of the total weight, this means that a realistic payload mass distribution must be modeled
also.

Another use of inertia relief analysis becoming more prevalent with the advent of auto support is in
checkout of a new model. Although six and only six rigid body modes are provided, they are provided
in the form of a “flying waterbed”. To take an extreme example, suppose that a structure is made up
of two components bolted together, but the bolts were not modeled, inadvertently. This model has
twelve rigid body modes before the constraints are applied. Both components sit on the same waterbed.
This means that all singularities will be constrained, and the model will run to completion on the first
pass. If loads are applied to one component only, the constraints will cause the other component to
move too, in a strain-free manner. A plot of the motion of the model should reveal that there is no
displacement continuity between the two components. This should make the cause of the unexpected
behavior apparent. As the goal of this analysis is model checkout only the structure mass distribution is
usually adequate for stable solution. The payload masses may be added later.

Real Eigenvalue Analysis in SubDMAPs SEMR3 and MODERS

The basic equation solved in these SubDMAPs is

13-70 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Massless Mechanism Identification in SubDMAP MMFIND

Equation 13-175.

where K and M are real and symmetric, and M is positive semi-definite. This equation is defined for
the a-set variables when called from MODERS for overall structural modes of vibration, and for o-set
variables when called from SEMR3 during CMS (Component Mode Synthesis) analysis. Many steps are
taken in DMAP to better automate the removal of DOFs that are ill-defined, and to augment the mode
shapes with residual flexibility effects. The SubDMAPs that perform these operations are described
below.

Although a knowledge of DMAP is not necessary for dynamic analysis, such a study can increase the
understanding of the steps that are performed.

Massless Mechanism Identification in SubDMAP MMFIND

SubDMAP MMFIND identifies potential instabilities in the matrices used for real eigensolutions. The
K and M matrices presented to the READ module for real eigensolutions may contain massless
mechanisms (MMs). A massless mechanism has a constraint shape X such that

Equation 13-176.

Equation 13-177.

Equation 13-178.

If φ contains the eigenvectors of the system, then

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 13-71

© 2024 Siemens
 13. Solution Sequences

Equation 13-179.

Equation 13-180.

The constraint shapes have non-zero terms only where the eigenvectors have zero terms. The massless
mechanism shapes can be thought of as "deficient" eigenvectors describing parts of the system that
are unconnected (or very weakly connected) to other parts of the structure that have both stiffness
and mass that are active in the well-conditioned eigenvectors. ϵk and ϵm may be binary zeroes, or
computational zeroes due to truncation effects. An "eigenvalue" λ associated with massless mechanism
shapes is

Equation 13-181.

λ is indeterminate in the limit as the ϵi approach zero. It can be a very large number, a very small
number, and either negative or positive. It is likely to change a large amount when small changes are
made to the model. This is the definition of an unstable eigensolution. The assumption taken here is that
the cause of this condition is either stiffness and/or mass effects inadvertently left out of this part of the
model, with the result that the load paths are ill-defined. This load path is constrained, and the software
issues a warning message. That is, this mode of motion is later removed from the K and M matrices prior
to eigensolution by steps equivalent to single point constraint elimination.

The massless mechanisms are defined from the shifted stiffness matrix A = K + s · M, where s is a shift
that you may change from its default value with the parameter PARAM, SHIFT1, (default=1.234 Hz.).
The default is chosen to make detection of rigid body modes as MMs unlikely. A is decomposed into its
factors,

13-72 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Massless Mechanism Control in SubDMAP XREAD

Equation 13-182.

The ratios of the diagonal terms of A over D are stored in a vector Ry. Terms of D approach a value
of zero at the ends of massless mechanisms. Terms of Ry larger in magnitude than user parameter
MAXRATIO are reset to unity and are kept in vector R1. It can be shown that each of these terms is at the
end of a massless mechanism constraint shape. R1 is expanded into a set of loading vectors P, with one
unit term per column. P is used along with the factors to find the constraint shapes,

Equation 13-183.

The constraint shapes are normalized, and small numbers are discarded. Each column of u is then a
potential constraint mode. It could also be a rigid body mode. The rigid body modes are detected by
computing the diagonal terms of the generalized mass matrix for these shapes,

Equation 13-184.

Terms of Mg(i, i) greater than PARAM, MECHFIL, default value = 1.e-6 are an indication that the
corresponding column is a rigid body mode. Columns of this type in u are discarded to produce u1,
the matrix of constraint mode shapes. The location of the end of the constraint shapes is passed back
in R. The constraint mode shapes themselves are passed back in MECHU. R is used later to remove rows
and columns in K and M to constrain the massless mechanisms. The nonzero terms in MECHU are used
to indicate which DOFs will discard loads applied to them, because they will have zero coefficients in
the true eigenvectors of the system, leading to a null contribution to generalized force for all modes.
MECHLD contains P, and is intended for a feature not yet implemented.

Massless Mechanism Control in SubDMAP XREAD

At present, the auto-omit DMAP is duplicated at several places, followed by a call to XREAD. It is
therefore an implicit part of the XREAD operations now, and may be moved to XREAD in a later version
to reduce the duplication of code.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 13-73

© 2024 Siemens
 13. Solution Sequences

SubDMAP XREAD modifies the input matrices used for eigensolution, when needed, then restores the
modified solutions back to the original size of the matrices. Solutions for both structural vibration
and buckling analyses are provided. This discussion describes vibration analysis first, with a concluding
section on changes made for a buckling solution.

Auto-Omit Operation

All the eigensolution methods require that each column of [K] plus [M] have some nonnull terms. Any
degree-of-freedom that has only null terms in both matrices is regarded as undefined and causes a fatal
error. The HOU method of eigenvalue extraction also requires that the mass matrix be positive-definite.
The MHOU and the Lanczos methods allow null columns in the mass matrix but are generally faster
if the massless degrees-of-freedom are removed by the static reduction method described in “Static
Solutions in SubDMAP SEKRRS.” However, the static reduction tends to increase the number of active
columns in the a-set matrices. The SINV and Lanczos methods, by contrast, are generally slower if the
static reduction is performed on massless degrees-of-freedom, because these methods take advantage
of sparse matrices, whereas the other methods do not.

Three options are provided to deal with the various null column possibilities. If PARAM,ASING,-1 is
used, any degrees-of-freedom with null columns in both the stiffness and mass matrices (“undefined
degrees-of-freedom”) are identified and a fatal error exit is taken. This is an option to use to identify
modeling errors before performing an expensive eigenvalue solution.

If ASING = 0 (the default) the method used depends on the eigenvalue extraction method. In the
tridiagonal methods of eigenvalue extraction, degrees-of-freedom with null columns in the mass matrix
are identified and placed in the w-set.

Equation 13-185.

where [Mww] and [Mwx] are null by definition. Degrees-of-freedom in the r-set (SUPORT Bulk Data entry)
must have mass, or a fatal error will result. The stiffness matrix is similarly partitioned.

13-74 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Auto-Omit Operation

Equation 13-186.

with until diagonal terms placed on null columns of [Kww]. The static condensation matrices are formed

Equation 13-187.

and

Equation 13-188.

This method will prevent errors due to mass matrix singularly in the tridiagonal methods if the
singularity is caused by null columns in [M]. It will not detect singularities due to mechanisms (caused,
for example, by a point mass offset from a grid point). This type of singularity does not necessarily
prevent obtaining eigensolutions, but will cause poor numerical stability. The MHOU method does not
suffer from numerical instability due to singular or nearly singular mas matrices. They are therefore
more reliable than the HOU method. In well conditioned problems, however, it is unnecessary to ?
automatically select the appropriate method. The AHOU method is the most modern, reliable method.

The static reduction will reduce the cost of solution and minimize the number of uninteresting high
roots found at the value of computed infinity due to the singularity of the mass matrix. Note that the
static reduction used here does not introduce any approximation, as the “omitted” that is w-set terms
have null mass values. The introduction of unit diagonal terms on null columns of [Kww] does not
introduce approximations because the terms that multiply these diagonal terms (the corresponding rows
of [G??] are null. The same results can be achieved at higher cost by first eliminating null rows and
columns of [Kww] by the single point constraint operations of “Single Point Constraint Operations in
SubDMAPs SEKR, SEMR2 and SEMRB” followed by static reduction.

The solution for the system [Kxx − λ · Mxx] · φx = 0 is provided. Rigid body shapes Dxr are provided when
SUPORTi entries exist, causing an r-set. The factor of Kxx, named LLL, is used for computing approximate

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 13-75

© 2024 Siemens
 13. Solution Sequences

solutions for buckling analysis, to better set the range of eigenvalues to be searched. For the tridiagonal
methods, K and M have DOFs with zero mass removed by an auto-omit method in a prior subDMAP. For
other methods of analysis, Kaa and Maa are used directly for residual structure modes, and Kvv and Mvv
for CMS (Component Mode Synthesis) modes.

The modes of the system are computed in the READ eigensolver module. If the solution appears to be of
good quality the subDMAP returns with no other operations performed. If the modes can not be found,
or appear to be of low quality, the module sets a flag (Neigv parameter). This flag causes the MMFIND
subDMAP to be called. Given K and M, it finds a high-ratio vector R and a matrix of constraint mode
shapes MMECHA. These terms are defined in the MMFIND section above.

There are non-zero terms in R at the end of massless mechanisms. These mechanisms can be removed
with SPC-like operations using R as a partitioning vector,

Equation 13-189.

Mxx is partitioned similarly, to obtain M22. Partitions involving the "1" subscript are discarded. The
modified eigenproblem solved is then [K22 − λ · M22] · φ2 = 0. φ2 is merged to the size of the original K
matrix by inserting null rows for the "1-set" DOFs,

Equation 13-190.

The auto-omit reduction is backed out by the operations

Equation 13-191.

13-76 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Real Eigenvalue Analysis in the READ Module

Equation 13-192.

Real Eigenvalue Analysis in the READ Module

Real eigenvalues and their associated eigenvectors are determined in READ (Real Eigenvalue Analysis –
Displacement). In the case of normal mode analysis, READ extracts real eigenvalues from the equation

Equation 13-193.

There are seven basic eigensolution methods. The Inverse (SINV) Method obtains eigensolutions by
iterations based on the equation

Equation 13-194.

where λi is an estimate of the eigenvalue. This method is best suited to large problems with sparse
matrices, where only a few eigenvectors are desired.

The tridiagonal methods, Householder (HOU and MHOU), first mass-scale the equation

Equation 13-195.

where [L] is a lower-triangular matrix. The [K] and [M] matrices are pre- and post-multiplied by [L]−1 and
[L−1]T, respectively, to obtain the standard form

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 13-77

© 2024 Siemens
 13. Solution Sequences

Equation 13-196.

where [J] = [L−1][K][L−1]T and [I] is the identity matrix. Note that the mass matrix is in effect inverted,
so that it must be nonsingular. The [J] matrix is transformed to tridiagonal form for economical
computation of eigenvalues and eigenvectors. The eigenvectors in the tridiagonal basis are then back-
transformed to the initial basis. The Modified Householder (MHOU) Method uses the shifted matrix

Equation 13-197.

in place of [M] in the discussion above, thereby allowing the mass matrix to be singular if there are
compensating terms in the stiffness matrix. The MHOU method isbest suited to small problems, or
for large problems where many eigenvectors are required, after extensive static reduction or dynamic
reduction is performed.

The Lanczos method uses a block shifted algorithm. Sets of vectors obtained by a recursive form are
used to reduce the problem to a reduced block tridiagonal form. The eigensolutions are computed in
the reduced basis by a QL algorithm, then back-transformed to the original basis. This is currently the
most modern method, and should be considered for all large-size problems. Since the method takes
advantage of sparsity in the input matrices, it is most economical when used without static or dynamic
reduction. It is requested with either the EIGRL or EIGR Bulk Data entries.

The AHOU method combines the best features of all of the tridiagonal methods, and is the
recommended method at present. It first decomposes M. If the factors are stable, the HOU method
is used. If the factors may be unstable, the MHOU method is used instead.

Modal Solution Preparation in the MODERS subDMAP

MODERS is called when solving for the eigensolution of the residual structure, with a-set operations. It is
used for both structural mode and buckling analysis.

In buckling, the r-set caused by SUPORTi entries are treated as equivalent SPCs. The rows and columns
associated with r-set DOFs are discarded before eigensolution.

For structural analysis, the auto-omit operations take place for the tridiagonal methods. See the XREAD
Section for details.

13-78 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Residual Flexibility in the SubDMAP RESVEC

Table indexing is a method to allow direct access to data in large sequential files, to speed up interactive
analysis. This work is being done as part of the API project.

Residual Flexibility in the SubDMAP RESVEC

Residual flexibility augments the mode shapes with shape functions found from static analysis with
various types of loading functions. These auxiliary displacement shapes reduce errors due to modal
truncation, particularly near the area of load application. These shapes are modified or discarded to
make them linearly independent of the mode shapes. They are then made orthogonal to the mode
shapes with a second eigensolution in a reduced basis. This augmented system provides increased
accuracy for modal dynamic analysis.

The RESVEC subDMAP is called in several contexts. When used in the residual structure, it operates on
a-set matrices such as Kaa. When used in CMS, it operates on o-set matrices such as Koo. The naming
convention used in the subDMAP is to use the generic set name x. For example, Kxx implies Kaa when
called from the residual structure operations, and Koo during CMS. In this section we will use the a-set
nomenclature, to make the discussion less abstract. The notation "In CMS" below implies that we are
then talking about operations unique to CMS-only in that paragraph, that is, o-set operations. Other
operations for CMS operations are similar to those for the a-set.

As in other subDMAPs, fluid-structure analysis of modal data is broken into two parts, a part for
structural modes only, and a part for fluid modes only. These effects are coupled after RESVEC. A review
of the subDMAP compilation shows the complication resulting from this partitioning and the looping
through several operations, once for fluid and once for structure. The complexity is ignored in this
section by largely ignoring fluid analysis. The equations describe the structural path. The fluid path can
be inferred to be similar.

Finally, some of the steps described in this section are actually performed in other SubDMAPs. They are
described here for clarity, with a notation of the SubDMAP where the actual DMAP statements may be
found.

Parameters Unique To Residual Flexibility Operations

The comment section of the DMAP source code is reproduced here, with extensions such as the default
value of NDDL parameters. The recommended parameters to use are param, resvec, yes, and param,
resviner, yes. The first is required to select any type of residual vector calculation. The second adds load
cases for loads due to inertia effects from rigid body acceleration to the applied load cases. It has been
found to improve results for a small incremental cost.

$ resvec default='no'
$ The main user parameter. Selects all resvec operations, as qualified
$ by the other resv* parameters. Set to 'yes' for any resvec
$ operations.

Tuning parameters are used only for unusual models. The defaults were set after running many test
problems.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 13-79

© 2024 Siemens
 13. Solution Sequences

$ resvrat default = 1.e8

$ filter value for discarding trial vectors with little
$ independent content. Make it smaller to discard more vectors
$ resvpgf default = 1.e-6
$ filter on partitioning vector. Make it larger to dicard more vectors
$ resvofxs default = .001
$ [in the DMAP, but no longer in use]

Intermediate Output Control

$ resvse default='no'
$ The default 'NO' will skip the printing of the strain
$ energy of the static shapes (labeled EXTERNAL WORK).
$ Use RESVSE='YES' to print the strain energy.

Miscellaneous

$ resvsli default = 'yes'

$ The default 'YES' will test for linearly
$ independent shapes. Use RESVSLI='NO' to skip the tests.
$ resviner default = 'no'
$ The default (NO) is to NOT use inertial loads to obtain trial
vectors.
$ Use 'yes' to use inertial loads as a source for
$ trial vectors
$ resvso default = 'yes'
$ The default 'YES' will clean up the static vectors to
$ insure orthogonality to modes. If the stiffness
$ matrix is well conditioned this is unnecessary, use
$ RESVSO='NO' to skip the reorthogonalization.
$ resmeth default = 'ahou'
$ method used for eigensolution in reduced basis.
$ 'lan' is an alternate method, somewhat more robust but slower for this
$ class of reduced basis problem.

The research legacy of this capability is evident in the many user parameters for specific functions,
some of which appear only in TYPE statements. Most are used only when you suspect problems
in your analysis. Their defaults, and even names, are likely to change from version to version. The
res*-type parameters current with the version you are using can be determined by compiling the RESVEC
SubDMAP, and searching for, y, parameter names in TYPE statements. The default values of NDDL
parameters can be determined compiling the NDDL statements with an input file like this:

$ file nddl*.dat
compile nddl=nddl nolist noref deck $
cend
begin bulk
enddata

13-80 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Load Generation

The .pch file output format, provided when "deck" appears on the compile statement, is more readable
with an editor than the .f06 output file. An example of some lines of this output is:

PARAM RESVEC ='NO' TYPE=CHAR8 PATH=SEID LOCATION=MASTER

PARAM RESVSE ='NO' TYPE=CHAR8 PATH=SEID LOCATION=MASTER
PARAM RESVSLI ='YES' TYPE=CHAR8 PATH=SEID LOCATION=MASTER
PARAM RESVINER ='NO' TYPE=CHAR8 PATH=SEID LOCATION=MASTER
PARAM RESVSO ='YES' TYPE=CHAR8 PATH=SEID LOCATION=MASTER
PARAM RESVRAT =1.E+8 TYPE=RS PATH=SEID LOCATION=MASTER
PARAM RESVPGF =1.E-6 TYPE=RS PATH=SEID LOCATION=MASTER

Load Generation

Three types of loads are used to generate auxiliary shapes. Their selection is controlled by user
parameters and other bulk data input. The parameter name and its default are listed by each load
type. Adding new load columns has little incremental cost, so provisions are made for many types of
loading conditions, to avoid the need for re-calculation of residual flexibility effects on restarts. The types
of loads chosen are those most likely to contain effects that may be truncated by modal analysis.

• Point Loads (USET,U6 Bulk Data entry; PARAM, RESVEC,YES)

This is a provision for cases where the DOFs to be loaded in later runs are known a priori, but the
magnitudes of the loads are not known. This type of loading is described not by conventional load
bulk data entries, but by an entry used to define a special set, the u6 set. This option is best suited to
loads on a few DOFs. This set is not used for other purposes in the solution sequences, so it is used
in this context to define places where unit point loads will be applied later. Each u6 DOF causes one
additional loading condition consisting of a point load at DOF u6. These loads are labeled Pg1 in the
subDMAP.

• Applied Loads (Conventional dynamic loading entries; PARAM, RESVEC,YES)

All dynamic loads defined on DAREA Bulk Data entries are converted to equivalent FORCE entries
at the start of the program. All loads selected from a DLOAD Case Control command that apply to
grid and scalar points are processed for static loads. These load entries are processed by the static
load generator (SSG1 module) to produce the Pg load vectors. These are the spatial loads that will
be multiplied later by time-varying functions in transient analysis, or frequency-varying functions in
frequency response analysis.

• Inertia loads due to unit rigid body acceleration (param,RESVINER, YES; PARAM, RESVEC,YES). This
computation is described in the following section Appended, Reduced Loads.

Appended, Reduced Loads

Loads from the first two sources (point loads and applied loads) are appended into one load matrix,

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 13-81

© 2024 Siemens
 13. Solution Sequences

Equation 13-198.

The loads are partitioned and reduced,

Equation 13-199.

Equation 13-200.

where Rga symbolizes the several-step load reduction described for static loads earlier. Pa is carried
forward when processing residual structure modes. Po is carried forward when processing CMS modes.
They will be given the generic name Pa in the remaining discussion.

Inertia loads are due to unit rigid body acceleration (param,RESVINER, YES; PARAM, RESVEC,YES).

Rigid body displacements Dgr for 6 unit motions of a reference point are computed from geometry. The
a-set DOFs are partitioned out to form Dar. The inertia loads due to unit accelerations of these shapes are
computed,

Equation 13-201.

These loads are appended to all prior selected loads to make the final load matrix Pa1, consisting of
u6-set effects, applied load effects, and rigid body inertia load effects.

13-82 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Free Boundary Effects (CMS only)

Free Boundary Effects (CMS only)

When free boundary points are defined by CSETi entries, the mode shapes are computed for the v-set,
which includes the o-set and the c-set. It is necessary to reduce the v-set effects to o-set effects. The v-set
may include rigid body modes that the c-set constrains. The derivation and motivation for these steps is
described elsewhere. The resulting steps implemented in the subDMAP are:

• Discard any rigid body modes by counting the number of r-set points.

• Reduce the mode shapes from v-set to o-set size,

Equation 13-202.

where x symbolizes the union of the c- and r-sets, locally.

Po2 replaces Pa1 in the ensuing discussion.

Load Sweeping

Later steps will discard shape functions that are not linearly independent of the mode shapes and prior
shape functions. Some of these shapes that would be discarded can be identified by inspecting the loads
that produce them. These loads can then be discarded before their residual vectors are calculated, saving
the expense of calculating shapes that will eventually be discarded.

Loads in Pa1 are normalized to produce Pan. Some portion of the loads in Pan can be expressed as modal
loads, and therefore do not need to be carried further. The derivation of the sweeping equation used to
eliminate these loads starts by expressing the loads as a linear combination of the inertial modal loads

Equation 13-203.

where y is unknown. Multiplying both sides by φTa, and solving for y,

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 13-83

© 2024 Siemens
 13. Solution Sequences

Equation 13-204.

where Mi − 1 = [φTa · Maa · φa]−1, the inverse of the modal mass.

These equations are used to solve for the swept loads Pa2, based on the normalized loads Pan,

Equation 13-205.

If the modes adequately express the generalized load caused by a column of Pan, the corresponding
column of Pa2 will be numerical noise, consisting of numbers near computational zero. Your input
used to set this filter is PARAM,RESVPGF, with a default value of 1.E-6. Columns with little content are
discarded from Pa2, leading to the swept load vectors Paf.

Static Solution for Shape Functions

If the structure contains rigid body modes, it is not stable for static analysis. This problem is solved by
one of two methods.

Manual SUPORT (SubDMAP SEKRRS)

If you supply SUPORTi entries, the r-set DOFs it produces are used as "temporary" SPCs, applied to
allow solving for the static shapes, but removed before computing modes. Partitions involving "r" are
discarded.

Equation 13-206.

13-84 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Solution for Trial Vectors by Static Analysis

Equation 13-207.

Equation 13-208.

Equation 13-209.

Solution for Trial Vectors by Static Analysis

Inertia relief analysis is NOT performed, despite the presence of SUPORTi entries, usually used in the
context of inertia relief. In this context, SUPORTi data or its equivalent for the auto-support feature is
used for SPC-type operations. In CMS at present, Koo may not contain rigid body modes. There is no
provision to constrain them with SUPORT-type inputs. DOFs may be removed from the o-set to make it
stable. For example, in superelement analysis, an interior point active in the rigid body mode, which by
default is in the o-set, can be made an exterior point. This changes it to the a-set, which is held fixed
when computing CMS modes. If the rigid body modes still exist downstream, they can be constrained in
residual structure calculations.

The ua2 trial vectors, or "raw static shapes", as they are called in the SubDMAP comments, may not be
linearly independent of the mode shapes, and are certainly not orthogonal to the mode shapes or each
other. An orthogonalization process is now used to insure that the resulting shapes are orthogonal to
the mode shapes and each other. Shapes that have little information after this process are discarded.

Orthogonalization

The trial vectors ua2 may not be linearly independent of each other or the mode shapes. They are
processed to find a set of combined mode shapes and linear combinations of the trial shapes that are all
orthogonal to each other with respect to the mass matrix. This reduces the risk of ill-conditioning when
using these matrices for solution later.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 13-85

© 2024 Siemens
 13. Solution Sequences

The trial vectors are appended to φa to form the combined set φua1

Equation 13-210.

An optional pre-sweeping of trial vectors is done when requested by PARAM, RESVSO, YES. The
derivation of the sweeping equation is similar to that for sweeping loads above, with the equation
for x, the vector of linear combination coefficients, being

Equation 13-211.

The swept displacement shape vectors φua1 are then determined by eliminating x,

Equation 13-212.

If the optional sweeping is not selected φa1 is merely equal to φa2.

The ua1 vectors are mass-normalized into φua such that the diagonal terms of the product on the left side
are unity

Equation 13-213.

13-86 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Orthogonalization

Equation 13-214.

Mzz is generally fully coupled at this point. It is decomposed with an LTL factorization, Mzz = L · D · L,
where L is lower triangular and D is diagonal.

The ratio of D over the diagonal terms of Mzz is calculated in the vector R,

Equation 13-215.

Rows of R whose value exceeds PARAM, RESVRAT (default=1.e8) indicate columns of φua that have
little new content, and are discarded to produce Ua3 and Mzz3 · Mzz3 is diagonal in the region of the
eigenvectors, but coupled in the region of the auxiliary shapes. It is then re-named as Mzz.

The generalized stiffness corresponding to Mzz is calculated,

Equation 13-216.

The eigensolution of the z-set is obtained,

Equation 13-217.

The eigenvectors φz of this reduced basis are used as a transformation to orthogonalize Ua3,

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 13-87

© 2024 Siemens
 13. Solution Sequences

Equation 13-218.

The eigenvalue table of the combined system, λz, replace the prior eigenvalue table that contained
modes only. λz includes both structural modes and the effects of orthogonalized residual flexibility
vectors.

Model Partitioning in the MODEFSRS subDMAP

MODEFSRS is used in the residual structure to separate the fluid and structural components of the a-set.
The modes of these components are computed independent of one another to find φsa and φfa, the
structural and fluid modes, respectively. Once computed, residual flexibility effects are added with calls
to the RESVEC SubDMAP.

The eigenvectors with residual flexibility effects appended to them λa are then merged together to
produce one eigenvector matrix,

Equation 13-219.

For the cases where only fluid or only structural modes are requested, the identity matrix is substituted
for the eigenvector matrix that is not calculated.

Complex Eigenvalue Analysis in SubDMAP CEIGRS

Complex eigenvalues and the associated eigenvectors are calculated in CEAD (Complex Eigenvalue
Analysis – Displacement) using the inverse power method, the determinant method, the Hessenberg
method, or the Lanczos method, as you request on the EIGC Bulk Data entry. For direct complex
eigenvalue analysis and if PARAM,ASING = 0 (default) then null rows and columns are discarded from
[Kdd], [Mdd] , and [Bdd] to form [Kdxx], [Mdxx], and [Bdxx].

In the case of a direct formulation CEAD extracts the eigenvalues from the following equation.

13-88 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Frequency Response and Random Analysis in SubDMAP FREQRS

Equation 13-220.

{udx} is then augmented with null rows to form {ud}.

In the case of a modal formulation the following equation is used:

Equation 13-221.

CEAD also normalizes the eigenvectors according to one of the following user requests:

• Unit magnitude of a selected coordinate (“POINT”).

• Unit magnitude of the largest component (“MAX”).

as specified on the EIGC Bulk Data entry.

The complex Lanczos method is recommended for large problems with sparse matrices. The HESS
method is best suited to small dense matrices, typical of small models or those typical in a modal
formulation.

Frequency Response and Random Analysis in SubDMAP FREQRS

The frequency-dependent loads are generated in FRLG (Frequency Response Load Generation) from the
RLOAD1 and RLOAD2 Bulk Data entries. Constraints are applied to reduce the loads to the d-set, and in
the case of a modal formulation, the loads are transformed to the h-set.

The solution for the response is performed in FRRD1 or FRRD2 (Frequency Response Solution). In the
case of a direct formulation the following equation is solved.

Equation 13-222.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 13-89

© 2024 Siemens
 13. Solution Sequences

and in the case of a modal formulation, the following equation is used in:

Equation 13-223.

The arithmetic used in the solution may be real or complex and the solution procedure may be
symmetric or unsymmetrical. The choice of arithmetic and solution procedure is made by the program
depending on the form of the dynamic matrices.

Power spectral density functions and autocorrelation functions are computed in RANDOM (Random
Analysis) according to the information on RANDPS, RANDPEX, and RANDT1 Bulk Data entries. Random
analysis calculations are made for selected displacements, loads, single point constraint forces, element
stresses and element forces. The selection of items for random analysis is made in the XY-output request
part of the Case Control Section.

Transient Response in SubDMAPs SEDTRAN and SEMTRAN

The linear time-dependent loads are generated in TRLG (Transient Response Load Generation) from the
TLOAD1 and TLOAD2 Bulk Data entries. The nonlinear loads are generated from the NOLINi Bulk Data
entries. Constraints are applied to reduce the loads to the d-set, and in the case of a modal formulation,
the loads are transformed to the h-set.

The solution for the response is performed in TRD1 (Transient Response Solution). In the case of a direct
formulation the following equation is integrated over the time periods specified via the TSTEP Bulk Data
entry in subDMAP DTRANRS:

Equation 13-224.

in order to determine the displacement velocity and acceleration response.

In the case of a modal formulation, the following equation is used in subDMAP MTRANRS:

13-90 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Buckling Analysis in SubDMAP MODERS

Equation 13-225.

If there are no direct input matrices and no nonlinear loads, an analytical integration procedure is used
rather than numerical integration. The numerical integration procedure may use either symmetric or
unsymmetrical routines. The choice is made by the program depending on the form of the dynamic
matrices.

Buckling Analysis in SubDMAP MODERS

In linear buckling, a static load is applied to the structure and the following eigenvalue problem is solved
subsequently,

Equation 13-226.

where [Kaa] is the linear stiffness, [Kdaa] is the differential stiffness, λi are the eigenvalues and {φi} are the
eigenvectors (buckling modes). The critical buckling loads are

Equation 13-227.

where {Pa} is the vector of applied static loads including reaction forces.

The basic assumption in linear buckling is that the differential stiffness [Kdaa] is proportional to the
applied load {Pa}. The assumption implies a linear material law and small deformations. For buckling in
nonlinear problems see the Simcenter Nastran Basic Nonlinear Analysis User's Guide.

The linear stiffness [Kaa] is derived from the variation of the stresses, the differential stiffness [Kdaa] is
derived from the variation of the strains keeping the stresses constant.

SOL 105 is the recommended solution sequence for linear buckling in static analysis. The Case Control
Section must contain at least two subcases. A static loading condition is applied in the first subcase.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 13-91

© 2024 Siemens
 13. Solution Sequences

The load is defined with a LOAD, TEMP(LOAD), or DEFORM Case Control command. The load may also
be defined with enforced deformations on the SPC Bulk Data entries. In the second subcase, a METHOD
command must appear to solve the eigenvalue problem. The Lanczos method (EIGR Bulk Data entry) is
the recommended eigenvalue solver for linear buckling.

Linear buckling is available in the following solution sequence.

SOL 105 Linear buckling in statics with superelements.

The linear buckling capability is available for the following elements: CONROD, CROD, CTUBE, CBAR,
CBEAM, CBEND, CQUAD4, CTRIA3, CTRIA6, CSHEAR, CHEXA, CPENTA, CPYRAM, AND CTETRA.

The differential stiffness formulation for the CQUAD4 and CTRIA3 elements was changed in Version 68.
Numerous spurious modes appeared in linear buckling of thin shell structures using versions prior to
Version 68. The improved results of Version 68 can be produced with a version earlier than Version 68 by
overlapping two CQUAD4 or CTRIA3 elements, one with bending stiffness only and one with membrane
stiffness only. The pre-Version 68 method of calculating the differential stiffness may be activated in
Version 68 by setting the system cell NASTRAN SYSTEM(170)=1 in the NASTRAN statement. The Version
68 method is the default, NASTRAN SYSTEM(170)=0. The pre-Version 68 method remains in the code to
recover old results.

The following recommendations apply to linear buckling analysis.

• Use at least 5 elements per half sine wave to get reasonable results in the buckling load.

• For buckling of 3-D shell structures, use PARAM,K6ROT,100. to assign a stiffness to the sixth degree-of-
freedom. The default is PARAM,K6ROT,0.

The following restrictions apply to linear buckling analysis.

• Offsets should not be used in beam, plate or shell elements. The buckling loads for structures with
offsets are incorrect.

• Follower force effects are not included in the differential stiffness. The calculated buckling loads are
incorrect. The Bulk Data entries FORCE1, FORCE2, MOMENT1, MOMENT2, PLOAD, PLOAD2, PLOAD4,
RFORCE, TEMP, TEMPD, TEMPP1, TEMPP3, TEMPRB describe follower forces.

Superelement Analysis

The theoretical aspects of superelement analysis are discussed in the Simcenter Nastran Superelement
User's Guide. Superelements are mathematically equivalent to substructures. The major difference is in
a more convenient user interface and in the ability to process a much broader class of problems than is
practical for most other substructuring capabilities.

There are now two distinct paths through superelements in the solution sequences. There are also two
major types of superelements.

13-92 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Terminology

• When a BEGIN SUPER entry is present the program uses the more modern SEP1X module to make the
SEMAP table used to control partition of superelements.

• When there is a BEGIN BULK entry but no BEGIN SUPER entries a parallel path using the older SEP1
module is used instead.

While the newer module supports most features of both part and grid list superelements, it does not
have some of the advanced features supported in the older module. When starting a new project you
should decide at modeling initiation which path is most suitable because transitioning from one to
another can be painful, particularly when the model is in development. If one starts on the old path,
for example, and then adds a new feature that causes the program to automatically switch to the new
path, some of your modeling input may be invalidated. This may cause considerable, confusing, and
unnecessary work to transition to the new path.

The short answer is that the new path should be the best path for most projects, based on the appeal of
its advanced features and the likelihood that future improvements in Simcenter Nastran may require use
of the new path. The new path is forced by replacing the BEGIN BULK entry with a BEGIN SUPER entry in
your Main Bulk Data file. It should by the first entry in your bulk data. It allows grid list superelements,
but they need not be present.

The exception for the path chosen is when your project requires the use of features available in the old
path but not yet enabled in the new path. Some of these features may be used in the new paths by
special avoidance techniques, but these techniques can be cumbersome. It may be more convenient to
stay with the older methods until the newer methods catch up.

Terminology

• Newest Option: The partitioned bulk data file superelements (parts) are the form introduced in
Version 69. They are sometimes called substructures. A major feature of parts is ID independence.
That is, element IDs and grid points IDs may be duplicated across different parts, although they must
be unique within a part.

• Intermediate: The Main Bulk Data Superelements (superelements) introduced in Version 47 are
sometimes called grid list superelements (list s.e.s). When used in the new path, they have more
restrictive rules than in the old path.

• Oldest Path: List s.e.s only are used in the old path. There are no formal names at present to
differentiate the intermediate and old flavors. Some advanced features are available only in this path,
at present.

A subtle difference between the two paths is that the new path always uses a single level configuration
for list s.e.s if there is no SETREE-type entry. It does this by defining new boundary points in the residual
structure when some are needed but not supplied by you. The old path automatically converts to a
multilevel configuration when required, regardless of the presence of DTI, SETREE entries. It has no
ability to generate new boundary points. This type of change when switching from the old path to the
new, on purpose or by accident can be confusing.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 13-93

© 2024 Siemens
 13. Solution Sequences

Criteria for Selecting Part Versus List Superelements

Each of the two major forms has overwhelming advantages in certain contexts. Some of these
advantages are of a temporary nature, as the rest of the technology in supporting FE tools catch up to
the newest methods. However, other relative advantages are due to the inherent difference in concept
between the two major branches that will be relevant into the foreseeable future. It is likely that both
branches will be maintained and extended because of their unique capabilities. The following criteria
should be considered when making a choice:

• Converting an existing non-superelement model to a part configuration is a labor-intensive process at

present. No migration aides are provided. Converting it to a list superelement requires addition of only
SESET entries, a trivial task by comparison.

• When debugging a new model it is much easier to comprehend and more efficient when it is in
a superelement configuration. Errors in one area will not obscure errors in other areas, and the
model solved is smaller, leading to faster turnaround. It may be more economical to run it in a
non-superelement configuration for production runs after the model is stable and requires no more
changes. If it is a list superelement this means merely removing the SESET entries, after which it
is automatically an exactly equivalent non-superelement model if advanced features such as those
provided by the RELEASE and CSUPEXT entries are avoided. If it is a part superelement there can be
a large labor cost for conversation, as all duplicated IDs will need to be changed, duplicate points at
boundaries will need to be resolved, etc. A list s.e. model, to make virtue of necessity, enforces rules
such as unique IDs when updating the model that make re-conversion to a non-s.e. configuration an
almost automatic process.

• When starting a new modeling project where more than one person provides input, and most
particularly when more than one organization contributes to the model, the higher degree of data
control for part superelements is attractive. There is no need to legislate acceptable ID ranges for each
part. It is possible to lock the description of a part into its boundary matrices, in a data base marked
with a time stamp, so that there is little questions about who is responsible for the behavior of each
part.

• Many GUIs can be used to modify list superelements because the rules for ID independence are the
same for GUIs as they are for list s.e.s Part superelements can be modeled in GUIs only one at a time.
This makes boundary matching, in particular, cumbersome.

Features Not Supported in the New Path

• Fluid Structure Analysis: This capability is supported in the new path with the “old” path rules:

1. All fluid elements must be in a special list superelement, and/or in the residual structure

2. This special superelement may have fluid elements only; structural elements or grid points are
not allowed

3. One or several fluid cavities may be in the fluid superelement

13-94 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Features Not Supported in the Old Path

4. Fluid elements may be connected to structural elements in list superelements, or in the residual
structure.

5. Elements in part superelements may not be connected to fluid elements.

• RELEASE Entries: Allowed only in the old path. There are semi-heroic modeling techniques that may
be used to simulate this feature in the new path, but they are not documented.

• SExSET Entries: Most set entries for non-superelement models such as ASET have synonyms for use in
list s.e.s such as SEASET. The synonym is not accepted in PARTs, but the simpler format of ASET, for
example, is accepted.

Features Not Supported in the Old Path

In general, features associated with automatic boundary control are not supported. The QRG is not
explicit on this issue. A list of entries not supported in the old path is:

SETREE, SENQSET, SEBNDRY, SEBULK, SECONCT, SEEXCLD, SELABLE, SELOC, SEMPLN

The DTI, SETREE entry is used in place of SETREE. The action of the SEELT entry is somewhat different.
For the new path, the SEMAP will be changed to account for the action of SEELT entries. For the old
path, it is assumed that the element being moved is one that can be moved without changing the
SEMAP, such as an element attached only to boundary points. If the SEELT requires change of the SEMAP
the program processes conventionally until it encounters the s.e. with the “illegal” SEELT entry, then
gives a fatal error exit stating that the connectivity of the superelement is incorrect.

Check List for Transitioning an Old Model to the New Path

• Replace the BEGIN BULK entry with a BEGIN SUPER entry.

• Check the SEMAP table from an old run to determine if it is in a multilevel state. When it is, add
SETREE entries that define the tree in the old run, or convert it to a single level model if multilevel is
not essential. If you have many multilevel models to convert, there is a parameter that will request
that SETREE entries be made from the SEMAP, and placed in the .pch (punch) file.

• Discard all RELEASE and CSUPEXT entries. If you cannot find reasonable replacements for them you
may be forced to stay in the old path.

Decide if this is the right time to convert to a PART model, or whether the list s.e. capability in the new
path is more advantageous, based on the discussion above.

The Case Control commands for superelement processing selection all begin with the letters SE (SEMG,
SELG, SEKR, SEMR, SELR, SEALL, and SEEX). The SEFINAL command is used to control the order of
processing of the superelements. A subcase structure is required that defines constraints, loads, and
data recovery options for each superelement. Each subcase includes the Case Control command SUPER
in order to relate the subcase to a particular superelement and load condition. Examples of Case Control
(including plot requests) are given in the Simcenter Nastran Superelement User's Guide. A set of plotting

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 13-95

© 2024 Siemens
 13. Solution Sequences

commands is required for each superelement to be plotted, separated by the plotter commands SEPLOT
and SEUPPLOT.

Each superelement is given an identification number (SEID). You assign each grid point to a
superelement on a GRID or SESET Bulk Data entry. Identical image, mirror image, and external
superelements are defined by the CSUPER and optional SEQSEP Bulk Data entries. The SEELT Bulk
Data entry may be used to change the superelement membership of boundary elements. The RELEASE
Bulk Data entry may be used to disconnect selected degrees-of-freedom from exterior grid points of a
superelement.

The mesh generator option, called by MESH on the NASTRAN statement, supports the superelement
capability fully, in that all GRID entries generated for a field can be given the same SEID, and the SESET
Bulk Data entry can change these SEIDs at boundaries. The automatic grid point resequencing features
are fully implemented for superelement models.

The superelement partitioning module passes on resequencing data from SEQGP Bulk Data entries only
for grid points of a superelement which are not exterior points for any superelement. This class of grid
points can be determined from the SEMAP table printed by the program, but may change if the order of
processing is changed. The automatic resequencing option processes each superelement separately.

Component mode synthesis uses SEBSETi, SECSETi, SESUP and SEQSETi entries to define the boundary
condition when computing component modes. Uncoupled solutions for statics and modes can be
obtained by use of PARAM,FIXEDB,-1.

Superelement Bulk Data entries may be specified in nonsuperelement solution sequences, but those
entries pertaining to the superelement capability (e.g., the CSUPER entry) are ignored, as well as some
fields on other entries (e.g., Field 9 on the GRID entry). However, better automatic resequencing may be
possible if the SEID data is removed.

Superelement Processing in the Structured Solution Sequences

An overview of the superelement processing in SOLutions 101-200 is shown in Figure 13-3. Phase
0 performs the preparation and bookkeeping function. The input data is stored on the database and
scanned to prepare the SEMAP table. This table lists the superelement to which each Bulk Data entry
is to be applied. Phase 0 also partitions out the data that defines each superelement and performs
automatic restart processing. In Phase I this data is used to generate the boundary matrices to be
assembled when combining the superelements into the total structure. Phase II performs calculations
done only on the residual structure (the superelement processed last), such as transient response
calculations or eigensolutions. Phase III expands boundary solutions to the interior grid points of the
superelement, performs the data recovery operations for internal forces, stresses, and strain energy,
and outputs these quantities. Phase IV combines plot vectors from upstream superelements to provide
deformed structural plots for all or part of the structure.

13-96 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Superelement Processing in the Structured Solution Sequences

Figure 13-3. Structured Solution Sequence Flow Chart

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 13-97

© 2024 Siemens
 13. Solution Sequences

The input file processor operations are carried out in a conventional manner. The grid point resequencer
module resequences each superelement individually.

Some of the geometry processor operations of “Geometry Processing in SubDMAP PHASEO” are
carried out to prepare tables of all grid points in the entire model, useful for generation of the SEMAP
table. After the SEMAP is available, the SLIST table is assembled. This list controls the operations
to be performed in that particular run for generating boundary matrices, as selected with Case
Control commands. The GP4 module generates constraint data in the form of the USET table, as
described in “Geometry Processing in SubDMAP PHASEO.” A separate USET table is generated for each
superelement. It is similar to the USET table of the older rigid formats except that the exterior points are
automatically placed in the a-set.

Assembly and Reduction (Phase I) in SubDMAP Phase1DR

The following operations may be carried out, as you select with superelement-type Case Control
commands. By default, all superelements and the residual structure are given these operations serially.
If requests for the SEMA (or SEMR) operation are made on an upstream superelement and the residual
structure, all superelements in the tree between the selected superelement and the residual structure
will also be given the SEMA operation.

Superelement Matrix Generation (SEMG) in SubDMAP PhaselA

The element structural matrices (stiffness, structural damping, mass and viscous damping) are
generated and stored. The direct matrix input option (DMIG Bulk Data entry) can be used to add
stiffness, structural damping, mass or viscous damping terms to the system matrices in this operation.
These terms are selected by the Case Control commands K2GG, K42GG, M2GG, and B2GG. The
parameters CB1, CB2, CK1, CK2, CK41, CK42, CM1 and CM2 can be used to scale the structural matrices.

The SEMA module adds in stiffness terms from any upstream superelements. The SEEX Case Control
command may be used to eliminate stiffness contributions from selected upstream superelements. The
superelements that are upstream are defined by the current SEMAP. The number of rows and columns of
the upstream superelements must be consistent with the number of exterior grid points, as defined by
the SEMAP In order to complete execution of the SEMA module, the boundary matrices of all upstream
superelements must be present in the database, and consistent with the current SEMAP.

The SELA module adds loads from upstream superelements to the load vector for the current
superelement. The SEEX Case Control command may be used to eliminate load contributions from
upstream superelements. The requirements for conformity of load data in the database with the current
state of the SEMAP are similar to those for stiffness, except that only rows, not columns of the load
vectors are mapped using data in the SEMAP table. The requirement on columns of all load matrices is
that they be equal to the number of columns (number of load condition) in the residual structure. If the
number of columns in all superelements and the residual structure is not the same, a user fatal error
results. Static loads may be applied to any grid point, whether exterior or interior to a superelement.
Dynamic loads may be applied directly to residual structure points. Dynamic load entries may also
reference static load sets applied to the superelements via the LOADSET Case Control command and
LSEO Bulk Data entries.

13-98 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Superelement Load Generation (SELG) in SubDMAP PHASE1A

Superelement Load Generation (SELG) in SubDMAP PHASE1A

The load generation module SSG1 is executed in the static and dynamic response solutions. This
operation is not executed in the normal modes and complex eigenvalue solution sequences. As loads
may depend on structural matrices (i.e., thermal or inertial loads), an SELG operation must be preceded
by an SEMG operation. The resultants of these loads about the grid point listed on the PARAM,GRDPNT
Bulk Data entry are output.

Superelement Stiffness Reduction (SEKR) in SubDMAPs SEKR and SEKRRS

You may change SPCs and/or MPCs between subcases (“boundary condition changes”). The GPSP module
identities grid point singularities, and, optionally, provides SPCs for singular degrees-of-freedom. This is
followed by the constraint eliminations described in “Multipoint Constraint Operations in SubDMAPs
SEKR and SEMR2 and SEMRB,” “Single Point Constraint Operations in SubDMAPs SEKR, SEMR2
and SEMRB,” and “Constraint and Mechanism Problem Identification in SubDMAP SEKR.” If the
omitted degrees-of-freedom contain mechanism-type singularities, the grid and component number of
all downstream ends of the mechanism are output in the matrix named MECHS, and a user fatal error
stops execution. The approximate fixed-boundary natural frequencies of the superelement are printed in
the FAPPROX matrix. The boundary stiffness matrix [Kaa] is stored in the database. Mass and damping
matrices are not processed in this phase.

SUPORTi Bulk Data entries, which define the r-set, may reference only grid points in the residual
structure. The rigid body operations for the stiffness are performed as described in “Static Solutions
in SubDMAP SEKRRS.”

Superelement Load Reduction (SELR) in SubDMAP SELR

This operation processes loads in the static and dynamic response solution sequences. This operation
is not executed in the normal modes and complex eigenvalue solution sequences. The static load
reduction of “Static Condensation in SubDMAPs SEKR and SEMR2” is applied, stopping after
generation of the [Pa] matrix for superelements and [Pl] for the residual structure.

Superelement Mass Assembly and Reduction (SEMR) in SubDMAP SEMRM

In the dynamics solution sequences, the constraint operations of “Multipoint Constraint Operations in
SubDMAPs SEKR and SEMR2 and SEMRB,” “Single Point Constraint Operations in SubDMAPs SEKR,
SEMR2 and SEMRB,” and “Constraint and Mechanism Problem Identification in SubDMAP SEKR” are
applied to the mass and damping matrices. The rigid body operations of “Static Solutions in SubDMAP
SEKRRS” are applied to the mass of the residual structure only. Dynamic reduction and component mode
calculations are performed. This allows a mass-change-only type of restart by use of the SEMG and
SEMR options and avoids the costly reduction of the stiffness matrix in the SEKR operation if no stiffness
changes are made.

The optional Bulk Data entry PARAM,ERROR,0 affects operations in Phase I. If it is not used, the default
action is to terminate the run after the first user fatal error, as in the older rigid formats. If the parameter
is used, a user fatal error causes a branch to the end of Phase I for the superelement, and continuation
to the generation, assembly, and reduction operations for any remaining superelements. The five SE-

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 13-99

© 2024 Siemens
 13. Solution Sequences

type commands may be executed on one run, or may be done separately. The following sequence must
be observed if these commands are executed separately:

SEKR after SEMG

SELG after SEMG
SELR after SEKR, SELG, SEMR
SEMR after SEKR

Residual Structure-Only (Phase II)

The operations on the superelements described above are also performed on the residual structure.
This is followed by operations unique to the residual structure. Eigensolutions and direct and modal
forced response calculations take place at this point. Direct input matrices, when selected by the Case
Control commands M2PP, K2PP, and B2PP are constrained and added as in the “Formulation of Dynamic
Equations in SubDMAP GMA.”

Bulk Data entries used for static loads (i.e., FORCE, GRAV, LOAD, etc.) may be used in direct and modal
dynamic analysis. They are interchangeable with DAREA Bulk Data entries and may be referenced by
LSEQ Bulk Data entries.

Data Recovery Operations (Phase III) in SubDMAP SUPER3

The total solution vectors of the downstream superelement are partitioned to generate the boundary
solution vectors for the superelement being processed. The boundary solution is expanded to all interior
points of the superelement. The fixed boundary component is computed and added.

If SPC force output is requested, the resultants of the SPC forces about PARAM,GRDPNT or the basic
coordinate system origin are also output. The data recovery operations of “Static Solutions in SubDMAP
SEKRRS” are carried out. (Note that solution sequences involving complex eigenvalue or frequency
response analysis may produce complex displacement vectors. GPFDR module output (grid point force
balance, element strain energy) may be requested for complex vectors, but only the real component of
the vector is processed.) Deformed structure plots of only elements in the superelement are generated
here.

Upstream Plots (Phase IV) in SubDMAP SUPER3

The displacement vectors of a superelement and all of its upstream members are combined and plotted
at your request.

Superelement Processing

An overview of the superelement processing is shown in the figure Solution Sequence Flow Chart.
Phase 0 performs the preparation and bookkeeping function. The input data is stored on the database
and scanned to prepare the SEMAP table. This table lists the superelement to which each Bulk Data
entry is to be applied. Phase I partitions out the data that defines one superelement and processes this
data to generate the boundary matrices to be assembled when combining the superelements into the
total structure. Phase II performs calculations done only on the residual structure (the superelement

13-100 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Assembly and Reduction (Phase I)

processed last), such as transient response calculations or eigensolutions. Phase III expands boundary
solutions to the interior grid points of the superelement, performs the data recovery operations for
internal forces, stresses, and strain energy, and outputs these quantities. Phase IV combines plot vectors
from upstream superelements to provide deformed structural plots for all or part of the structure.

The input file processor operations are carried out in a conventional manner. The grid point resequencer
module resequences each superelement individually.

Some of the geometry processor operations of Geometry Processing in SubDMAP PHASEO are carried
out to prepare tables of all grid points in the entire model, useful for generation of the SEMAP table.
After the SEMAP is available, the SLIST table is assembled.

This list controls the operations to be performed in that particular run for generating boundary matrices
as selected with Case Control options.

Assembly and Reduction (Phase I)

The following operations may be carried out as you select on superelement-type Case Control
commands. By default, all superelements and the residual structure are given these operations serially.
If requests for the SEMA (or SEMR) operation are made on an upstream superelement and the residual
structure, all superelements in the tree between the selected superelement and the residual structure
will also be given the SEMA operation.

Superelement Matrix Generation (SEMG)

The element structural matrices (stiffness, structural damping, mass and viscous damping) are
generated and stored. The direct matrix input option (DMIG Bulk Data entry) can be used to add
stiffness, structural damping, mass or viscous damping terms to the system matrices in this operation.
These terms are selected by the Case Control commands K2GG, K42GG, M2GG, and B2GG. The
parameters CB1, CB2, CK1, CK2, CK41, CK42, CM1, and CM2 can be used to scale the structural
matrices.

Superelement Load Generation (SELG)

The load generation module SSGI is executed in the statics and dynamic response solutions. This
operation is not executed in the normal modes and complex eigenvalue solution sequences. As loads
may depend on structural matrices (i.e., thermal or inertial loads), an SELG operation must be preceded
by an SEMG operation. The resultants of these loads about the grid point listed on the PARAM,GRDPNT
Bulk Data entry are output.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 13-101

© 2024 Siemens
 13. Solution Sequences

Figure 13-4. Solution Sequence Flow Chart

Superelement Stiffness Matrix Assembly and Reduction (SEKR)

The SEMA module adds in stiffness terms from any upstream superelements. The SEEX Case Control
command may be used to eliminate stiffness contributions from selected upstream superelements. The

13-102 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Superelement Load Assembly and Reduction (SELR)

superelements that are upstream are defined by the current SEMAP. The number of rows and columns of
the upstream superelements must be consistent with the number of exterior grid points, as defined by
the SEMAP. In order to complete execution of the SEMA module, the boundary matrices of all upstream
superelements must be present in the database, and consistent with the current SEMAP. The GP4
module generates constraint data in the form of the USET table, as described in “Geometry Processing
in SubDMAP PHASEO.” A separate USET table is generated for each superelement. It is similar to the
USET table of the older rigid formats except that the exterior points are automatically placed in the a-set.

You may change SPCs and/or MPCs between subcases (“boundary condition changes”) in SOLs 101
and 103 only. The GPSP module identifies grid point singularities, and at user's option provides
SPCs for singular degrees-of-freedom. This is followed by the constraint eliminations described in
“Multipoint Constraint Operations in SubDMAPs SEKR and SEMR2 and SEMRB,” “Single Point
Constraint Operations in SubDMAPs SEKR, SEMR2 and SEMRB,” and “Constraint and Mechanism
Problem Identification in SubDMAP SEKR.” If the omitted degrees-of-freedom contain mechanism-type
singularities, the grid and component number of all downstream ends of the mechanism are output
in the matrix named MECHS, and a user fatal error stops execution. The approximate fixed-boundary
natural frequencies of the superelement are printed in the FAPPROX matrix. The boundary stiffness
matrix [Kaa] is stored in the database. Mass and damping matrices are not processed in this phase.

SUPORT Bulk Data entries, which define the r-set, may reference only grid points in the residual
structure. The rigid body operations for the stiffness are performed as described in “Static Solutions
in SubDMAP SEKRRS.”

Superelement Load Assembly and Reduction (SELR)

This operation processes loads in the static and dynamic response solution sequences. This operation
is not executed in the normal modes and complex eigenvalue solution sequences. The SELA module
adds loads from upstream superelements to the load vector for the current superelement. The SEEX
Case Control command may be used to eliminate load contributions from upstream superelements.
The requirements for conformity of load data in the database with the current state of the SEMAP are
similar to those of the SEKR operation above, except that only rows, not columns of the load vectors are
mapped using data in the SEMAP table. The requirement on columns of all load matrices is that they be
equal to the number of columns (number of load condition) in the residual structure. If the number of
columns in all superelements and the residual structure is not the same, a user fatal error results. Static
loads may be applied to any grid point, whether exterior or interior to a superelement. Dynamic loads
may be applied directly to residual structure points. Dynamic load entries may also reference static load
sets applied to the superelements via the LOADSET Case Control command and LSEQ Bulk Data entries.
The static load reduction of “Static Condensation in SubDMAPs SEKR and SEMR2” is applied, stopping
after generation of the [Pa] matrix for superelements and [Pl] for the residual structure.

Superelement Mass Assembly and Reduction (SEMR)

In the dynamics solution sequences, the constraint operations of “Multipoint Constraint Operations in
SubDMAPs SEKR and SEMR2 and SEMRB,” “Single Point Constraint Operations in SubDMAPs SEKR,
SEMR2 and SEMRB” and “Constraint and Mechanism Problem Identification in SubDMAP SEKR” are
applied to the mass and damping matrices. The rigid body operations of “Static Solutions in SubDMAP
SEKRRS” are applied to the mass of the residual structure only. Dynamic reduction and component mode
calculations of “Static and Dynamic Load Generation” are performed. This allows a mass-change-only

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 13-103

© 2024 Siemens
 13. Solution Sequences

type of restart by use of the SEMG and SEMR options and avoids the costly reduction of the stiffness
matrix in the SEKR operation if no stiffness changes are made.

The optional Bulk Data entry PARAM,ERROR,0 affects operations in Phase I. If it is not used, the default
action is to terminate the run after the first user fatal error, as in the older rigid formats. If the parameter
is used, a user fatal error causes a branch to the end of Phase I for the superelement, and continuation
to the generation, assembly, and reduction operations for any remaining superelements.

The five SE-type commands may be executed in one run, or may be done separately. The following
sequence must be observed if these commands are executed separately:

SEKR after SEMGSELG after SEMGSELR after SEKR, SELG, SEMRSEMR after SFKR

Residual Structure-Only (Phase II)

The operations on the superelements described above are also performed on the residual structure. This
is followed by operations unique to the residual structure. Eigensolutions and direct and modal forced
response calculations take place at this point. Direct input matrices, when selected by the Case Control
commands M2PP, K2PP, and B2PP are constrained and added as in “Formulation of Dynamic Equations
in SubDMAP GMA.”

Bulk Data entries used for static loads (i.e., FORCE, GRAV, LOAD, etc.) may be used in direct and modal
dynamic analysis. They are interchangeable with DAREA Bulk Data entries and may be referenced by
LSEQ Bulk Data entries.

Data Recovery Operations (Phase Ill)

The total solution vectors of the downstream superelement are partitioned to generate the boundary
solution vectors for the superelement being processed. The boundary solution is expanded to all interior
points of the superelement. The fixed boundary component is computed and added.

If SPC force output is requested, the resultants of the SPC forces about PARAM,GRDPNT or the basic
coordinate system origin are also output. The data recovery operations of “Data Recovery Operations
in SubDMAP SEDISP” are carried out. (Note that solution sequences involving complex eigenvalue or
frequency response analysis may produce complex displacement vectors. GPFDR module output (grid
point force balance, element strain energy) may be requested for complex vectors, but only the real
component of the vector is processed.) Deformed structure plots of only elements in the superelement
are generated here.

Upstream Plots (Phase IV)

The displacement vectors of a superelement and all of its upstream members are combined and plotted
at user's request.

13-104 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Superelement Alternate Statics Solution

Superelement Alternate Statics Solution

SOL 101 provides an alternate method of performing the superelement static condensation.
PARAM,ALTRED,YES must be entered in the Bulk Data Section. In some cases, the alternate solution
can result in significant reduction in computer cost for static analysis.

The stiffness and load reduction and displacement recovery are described in:

• “Static and Dynamic Load Generation”

• “Static Condensation in SubDMAPs SEKR and SEMR2”

• “Data Recovery Operations in SubDMAP SEDISP”

Static Superelement Analysis With lnertia Relief

Inertia relief analysis is available in SOL 101. PARAM,INREL,-1 or PARAM,INREL, -2 must be present for
either of these options. The following method transforms static loads based on model geometry rather
than the stiffness matrix as described in “Static Solutions in SubDMAP SEKRRS.”

The additional operations required to perform inertia relief are described in “Static and Dynamic Load
Generation” and “Static Solutions in SubDMAP SEKRRS.”

Assumptions and Limitations

1. The model must have six and only six rigid body modes for a continuous, three-dimensional
structure.

2. Masses on scalar points are ignored in calculating inertia loads.

3. The structural model, when given enforced displacements at the reference point, is assumed to
produce the same motion as a rigid body model. This will be true for any structural or rigid
element. It is possible, if unconventional, to use CELASi or general elements, MPC equations, or
DMIG coefficients that will not meet this assumption.

4. Identical and mirror image superelements are not processed correctly.

Nonlinear Heat Transfer Analysis

In Simcenter Nastran, you can perform:

• Steady state heat transfer analysis

• Transient heat transfer analysis

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 13-105

© 2024 Siemens
 13. Solution Sequences

Steady State Heat Transfer Analysis

The steady state heat balance equation is given by

Equation 13-228.

where:

[K] = a heat conduction matrix

[ ] = a radiation exchange matrix
{P} = a vector of applied heat flows that are constant
{N} = a vector of nonlinear heat flows that depend on temperature
{u} = a vector of grid point temperature
Tabs = the absolution temperature

The components of the applied heat flow vector, {P}, are associated either with surface heat transfer or
with heat generated inside the volume heat conduction elements. The vector of nonlinear heat flows,
{N}, results from boundary radiation, surface convection, and temperature dependent thermal loads.

The equilibrium equation is solved by a Newton iteration scheme, where the tangential stiffness matrix
is approximated by

Equation 13-229.

and the residual vector is

13-106 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Transient Heat Transfer Analysis

Equation 13-230.

Transient Heat Transfer Analysis

The general equation solved in transient analysis has the form

Equation 13-231.

To take phase change into consideration, the heat diffusion equation is converted into

Equation 13-232.

where:

[B] = heat capacity matrix

[K] = heat conduction matrix
[ ] = radiation matrix
{P} = vector of applied heat flows that are constant or functions of time
{N} = vector of nonlinear heat flows that depend on temperature
{H} = enthalpy vector

{ } = {dH / dt}

{u} = grid point temperatures

{ } = {du / dt}

Tabs = absolute temperature scale factor

The equilibrium equation is solved by Newmark's method with adaptive time stepping. Base on this
one-step integration scheme, the time derivative of the nodal temperatures at the (i + 1) th iteration of
the time step (n + 1) is expressed as

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 13-107

© 2024 Siemens
 13. Solution Sequences

Equation 13-233.

where

Equation 13-234.

and

Equation 13-235.

The parameter η is specified on the PARAM,NDAMP Bulk Data entry. When η = 0 (θ = 0.5), no numerical
damping is requested. In this case, Newmark's method is equivalent to the Crank-Nicolson method.

13-108 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 14. Modeling Guidelines
Overview of Modeling Guidelines
With finite element modeling, the quality of your results depends directly upon the quality of your
model. One of the more common errors that a beginning finite element analyst makes in modeling is to
simply simulate the geometry rather than to simulate both the geometry and the physical behavior of
the real structure. This section covers the following finite element modeling topics and guidelines:

• Choosing the right element

• Mesh density

• Mesh transitions

• Grid point stresses

• Consistent loading

• Symmetry

It is also good modeling practice to simulate and validate a new capability or a feature that you haven't
used before with a small prototype model before applying this feature to your production model. Model
verification techniques are covered in “Model Verification.”

Choosing the Right Element

Simcenter Nastran contains a large library of structural elements. Often, you could use several elements
to model the same structural effects. The criteria for the selecting an element may include its capabilities
(for example, whether it supports anisotropic material properties), its cost (in general, the more DOF an
element has, the more expensive it is), and/or its accuracy.

In many cases, the choice of the best element for a particular application may not be obvious. For
example, in the model of a space frame, you may choose to use CROD elements if end moments
are unimportant or to use CBAR elements if end moments are important. You may choose to use
CBEAM elements with warping if the members have open cross sections and torsional stresses are
estimated to be significant. You may even choose to represent the members with assemblies of plate
or solid elements. The choice of which type and number of elements to use depends primarily on your
assessment of the effects that are important to represent in your model and on the cost and accuracy
you are willing to accept.

Given this, you should have a fairly good idea of how the structure will behave prior to generating
your finite element model. In other words, understanding the load path is crucial in the selection of the
appropriate element. Additionally, a few hand calculations can usually provide a rough estimate of stress

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 14-1

© 2024 Siemens
 14. Modeling Guidelines

intensities. If you don't have a fairly good idea of how the structure will behave, you may be misled by
incorrect results due to errors or incorrect assumptions in your input data preparation.

General Guidelines for Element Selection

Always experiment with a small test model when using elements that you aren't familiar with. This
practice is cheaper than experimenting with a large production model, and it gives you a better
understanding of an element's capabilities and limitations prior to applying it to a large production
model.

Avoid using obsolete elements. Elements are classified as obsolete only if their utility is completely
superseded by newer elements and after the newer elements are in service long enough to be
reliable. In most cases, they are retained in Simcenter Nastran only to satisfy the requirements for
upward compatibility. Elements that are considered obsolete are no longer included in any updated
documentation.

Zero-dimensional Elements

When you use CELASi elements to represent concentrated springs between two components of
translation, the directions of the two components must be coaxial. Even small deviations in direction
can induce a significant moment to your model that does not exist in your physical structure. When you
use a CELASi element, the locations of the two end points should be coincident to avoid this type of
problem. If the two end points aren't coincident, consider using a CROD or CBUSH element instead.

One-dimensional Elements

If only an axial and/or torsion load is to be transmitted in an element, then the CROD is the easiest
element to use.

A CBAR is easier to use than a CBEAM element. The I1 and/or I2 values can be set to zero. Use the CBEAM
element instead of the CBAR element if any of the following features is important:

• The cross-sectional properties are tapered.

• The neutral axis and shear center do not coincide.

• The effect of cross-sectional warping on torsional stiffness is critical.

• The difference in the mass center of gravity and the shear center is significant.

The formulation for the CBEAM element is based on a flexibility approach; the element stiffness matrix is
generated by inverting the flexibility matrix. For this reason, I1 and I2 must not be zero for the CBEAM
element.

14-2 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Two-dimensional Elements

Two-dimensional Elements

In general, quadrilateral elements (CQUAD4 and CQUAD8) are preferred over the triangular elements
(CTRIA3 and CTRIA6). The CTRIA3 element is a constant strain element. It is excessively stiff, and
when used alone, it is generally less accurate than the CQUAD4 element, particularly for membrane
strain. Whenever feasible, you should use the CQUAD4 element instead of the CTRIA3 element. CTRIA3
should only be used when necessary for geometric or topological reasons, for example, mesh transition
between regions of quadrilateral elements with different meshes or near the polar axis of a spherical
shell.

Additoinally, you should avoid using CTRIA3s in locations where the membrane stresses are changing
rapidly, for example, in the web of an I-beam. Since CTRIA3 has constant membrane stresses, a large
number of them may be needed to obtain acceptable accuracy. It is better to use quadrilateral elements
or CTRIA6 elements, if possible.

Don't use plate or shell elements (CQUADi, CTRIAi) in stiffened shell structures with very thin panels that
can buckle. You should use shear panels (CSHEAR) in this case or in any situation where direct stresses
cannot be supported, such as in a very thin curved panel.

Avoid highly skewed elements (see Figure 14-1). The angle α should be as close to 90 degrees as
possible.

Figure 14-1. Highly Skewed Element

For the CTRIA3 element, the skew test is based upon the three vertex angles.

Aspect ratio is defined as l/ ω (length/width). Very high aspect ratio (see Figure 14-2) should also be
avoided, although it is no longer true that accuracy degrades rapidly with aspect ratios as it once did
with some of the obsolete elements.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 14-3

© 2024 Siemens
 14. Modeling Guidelines

Figure 14-2. Element with High Aspect Ratio

Warping is a measure of the amount the element deviates from being planar (see Figure 14-3). Element
warping should be minimized.

Figure 14-3. Highly Warped Element

For the CQUAD8 elements, if midside nodes are present, they should be located within the middle third
of the edge. If a midside node is located at one-fourth the distance of the edge as measured from either
corner node on the edge, the internal strain field becomes singular at the corners of the element. For
best results, the midside node should be located as close to the center of the edge as possible. If you
want to use midside nodes, you should include all of them. A CQUAD8 element with midside nodes
deleted is excessively stiff and inferior to a CQUAD4 element.

14-4 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Shell Normals

Figure 14-4. CQUAD8 with Missing Midside Node

For single curved structures (e.g., a cylinder), the CQUAD8, in general, yields better results than the
CQUAD4 element. For doubly curved structures (e.g., a spherical dome), the CQUAD4 element, in
general, performs better than the CQUAD8.

You should turn on the shell normal (PARAM,SNORM,X) option when using the CQUAD4, CTRIA3,
CQUADR, or CTRIAR element. See the next section on shell normals for further details.

The membrane properties for the CQUADR and CTRIAR elements are less sensitive to the element shape
than the CQUAD4 and CTRIA3. The same thickness should be used for all four corners (T1 = T2 = T3 = T4,
or use the T Field on the PSHELL entry). The CQUADR and CTRIAR elements can be used in conjunction
with each other, but not with other elements.

Shell Normals

By default, the direction of the normal rotation vector for flat plate elements is assumed to be
perpendicular to the plane of each element. If the model is curved, the shell bending and twist
moments must change direction at the element intersection. If transverse shear flexibility is present, the
deformations may be too large. (Because elements using low-order formulations ignore the edge effect,
this rarely causes any problems-a large value of the parameter K6ROT partially cures the problem.) With
the unique normal (SNORM) option, the rotational degrees of freedom at each corner of an element are
measured relative to the specified normal vector direction. Thus, all elements connected to a grid point
will use a consistent direction for defining shell bending and twisting moments.

In CQUAD4 and CTRIA3 elements, the stiffness matrices of the elements are modified to eliminate
the undesirable small stiffness in the rotational motions about the shell normal vector. In effect, the
transformation replaces the normal moments with in-plane forces. No changes were made to the basic
element stiffness matrix, and therefore flat plate models will not be affected. The objective of the new
transformation was to remove a potential weakness in curved shell models and allow the automatic
constraint process to remove the true singularity in the assembled stiffness matrix.

The CQUADR and CTRIAR elements are also improved for some types of shell problems. Because of
their extra degrees of freedom, these elements are more sensitive to the coupling between in-plane

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 14-5

© 2024 Siemens
 14. Modeling Guidelines

and out-of-plane motion due to curvature. The new formulation provides more consistency between
adjacent elements in a curved shell.

Shell normals are available for CQUAD4, CQUADR, CTRIA3, and CTRIAR elements. Normals are activated
if the actual angle between the local element normal and the unique grid point normal is less than 20°,
the default value for β (see Figure 14-5.) The default for β can be changed by setting PARAM,SNORM,β
to the desired real value up to 89 degrees. The unique grid point normal is the average of all local
shell element normals at a specific grid point. Generated grid point normals may be overwritten by
user-defined normals.

Figure 14-5. Unique Grid Point Normal

A shell normal defines a unique direction for the rotational degrees of freedom of all adjacent elements.
A shell normal vector is created by averaging the normal vectors of the attached elements. If the actual
angle is greater than the value defined on PARAM,SNORM, (default = 20.) the edge is assumed to be a
corner, and the old method is used. Shell normals improve the accuracy of the results in curved shells
where in-plane shear and twisting moments act together.

If the actual shell normal vector is known, such as in a cylinder or sphere, the automatically calculated
values may be superseded by the actual vectors using the Bulk Data entry

SNORM,GID,CID,V1,V2,V3

where:

• GID is a unique grid point,

• CID is the coordinate system for defining the shell normal vector,

14-6 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Remarks and Recommendations for Shell Normals

• V1, V2, and V3 are unscaled components of the vector in the coordinate system.

A second parameter, SNORMPRT, controls the print or punch of values of the internal shell normals.
The output format is the same as the SNORM Bulk Data so that the individual values may be used and
modified on a subsequent job.

Remarks and Recommendations for Shell Normals

1. The type of structure that exhibits the most change in results is a thick curved shell with large
in-plane shear forces and twisting moments.

2. Most other problems, such as flat plates and curved shells with pressure loads, show changes
in results of less than 1%. More degrees of freedom may be constrained using this improved
formulation. This formulation results in zero in-plane rotational stiffness values.

3. It has been observed from testing that the automatic normal vector calculations produced answers
nearly equal to and as accurate as those using the explicit SNORM vector input. In other words, the
results were insensitive to small differences in the direction of the vectors. The important fact is
that the shell normal vector requires the connected elements to use a consistent normal direction.

4. The CQUAD8 and CTRIA6 elements are not included in the shell normal processing. If they are
modeled correctly, they do not require shell normal processing. Connecting these elements to the
lower-order flat elements is not recommended.

5. Curved shell elements have no stiffness in the rotational degree of freedom when the new
formulation is used-consequently, mechanisms may be introduced. Mechanisms occur when
elements, RBEs, or MPCs are connected to the out-of-plane rotation of the shell grid points. Note
that PARAM,AUTOSPC,YES does not constrain these mechanisms.

6. In linear solution sequences, the values of “param,k6rot,0.” and “param,snorm,20.” are the default.

7. Transverse shear flexibility (MID3 on the PSHELL property entry) should be left on when normals
are used.

Three-dimensional Elements

Although the CHEXA and CPENTA elements are designed to behave reasonably well as thin shell
elements, in general, you shouldn't use them in this capacity. The high ratio of extensional stiffness
in the direction normal to the effective transverse shear stiffness can produce significant round-off errors.

As in the case of plate elements, if you include midside nodes for the solid elements, you should locate
them as close to the center of the edge as possible. Additionally, if midside nodes are desired, you
should include all of them.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 14-7

© 2024 Siemens
 14. Modeling Guidelines

R-type Elements

A high degree of precision must be maintained when specifying coefficients for MPCs to avoid
introduction of unintentional constraints to rigid body motions. You should use rigid elements (e.g.,
the RBE2, RBAR, etc.) whenever possible because their constraint coefficients are internally calculated to
a high precision. Furthermore, these R-type elements require much less user interaction.

Determining Mesh Density

The mesh density in a finite element model has important implications for both accuracy and cost.
Frequently, the minimum number of elements is set by topological considerations, for example, one
element per member in a space frame or one element per panel in a stiffened shell structure.
Historically, when problem size was more severely limited, it was not uncommon to lump two or more
frames or other similar elements in order to reduce the size of the model. With computers becoming
faster and cheaper, the current trend is to represent all major components individually in the finite
element model.

If the minimum topological requirements are easily satisfied, the question remains as to how fine to
subdivide the major components. The question is particularly relevant for elastic continua, such as slabs
and unreinforced shells. In general, the accuracy of your results increases with the mesh density. The
mesh density required can be a function of many factors. Among them are the stress gradients, the type
of loadings, the boundary conditions, the element types used, the element shapes, and the degree of
accuracy desired.

The grid point spacing should typically be the smallest in regions where you expect stress gradients to
be the steepest. Figure 14-6 shows a typical example of a stress concentration near a circular hole. The
model is a circular disk with an inner radius = a and an outer radius = b. A pressure load pi is applied
to the inner surface. Due to symmetry, only half of the disk is modeled. In the example, both the radial
stress and the circumferential stress decrease as a function of l/r2 from the center of the hole. The error
in the finite element analysis arises from differences between the real stress distribution and the stress
distribution within the finite elements.

Three different mesh densities are used in this example as shown in Figure 14-6.

• The first one is a coarse mesh model with the elements evenly distributed.

• The second model consists of the same number of elements; however, the mesh is biased toward the
center of the hole.

• The third model consists of a denser mesh with the elements evenly distributed.

These three models are then analyzed with three different element types-CQUAD4, CQUAD8, and
CQUAD4 with the corner stress option. The circumferential stress at the inner radius is always greater
than pi, which is the applied pressure load at the inner radius, and approaches this value as the outer
radius becomes larger. The theoretical circumferential stress σθ (see Timoshenko and Goodier, Theory of
Elasticity, McGraw-Hill Book Company, Third Ed., 1970) is given by the following equation:

14-8 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Determining Mesh Density

where:

a = inner radius
b = outer radius
r = radial distance as measured from the center of the disk
pi = pressure applied at the inner radius

The stresses are then plotted as a function of the radius in a nondimensional fashion-stress pi/ versus r/a.
The results are summarized in Table 14-2 and Figure 14-7.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 14-9

© 2024 Siemens
 14. Modeling Guidelines

Figure 14-6. Circular Disk with Different Meshes

14-10 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Determining Mesh Density

Figure 14-7. Stresses with Different Elements and Meshes

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 14-11

© 2024 Siemens
 14. Modeling Guidelines

Figure 14-8. Figure 9-7 Stresses with Different Elements and Meshes (continued)

14-12 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Determining Mesh Density

Figure 14-9. Figure 9-7 Stresses with Different Elements and Meshes (continued)

Table. Stresses Close to r = a for a Circular Disk

Case
No. Element Type DOF(L-Set) Description Stress Theoretical
1 CQUAD4 194 Coarse, Even Mesh 0.594
2 CQUAD4 194 Coarse, Biased Mesh 0.881
3 CQUAD4 868 Fine, Even Mesh 0.801
4 CQUAD8 550 Coarse, Even Mesh 0.929
5 CQUAD8 550 Coarse, Biased Mesh 1.041
6 CQUAD8 2538 Fine, Even Mesh 1.015
7 CQUAD4 with Corner 194 Coarse, Even Mesh 1.142
Option
8 CQUAD4 with Corner 194 Coarse, Biased Mesh 1.047
Option

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 14-13

© 2024 Siemens
 14. Modeling Guidelines

Case
No. Element Type DOF(L-Set) Description Stress Theoretical
9 CQUAD4 with Corner 868 Fine, Even Mesh 1.127
Option
10 THEORETICAL – – 1.000

For this particular case, since the stresses are proportional to 1/r2, you expect the highest stress to occur
at the inner radius. In order to take advantage of this piece of information, the obvious thing to do is to
create a finer mesh around the inner radius. Looking at the results for the first two cases in Table 14-2, it
is quite obvious that just by biasing the mesh, the results are 30% closer to the theoretical solution with
the same number of degrees of freedom.

A third case is analyzed with a finer but unbiased mesh. Note that for case number 3, even though it
has more degrees of freedom, the result is still not as good as that of case number 2. This poor result
is due to the fact that for the CQUAD4 element, the stresses, by default, are calculated at the center of
the element and are assumed to be constant throughout the element. This assumption is best illustrated
by the results shown in Figure 14-7(a). Looking at Figure 14-6, it is also obvious that the centers of
the inner row of elements are actually further away from the center of the circle for case number 3 as
compared to case number 2. The results for case number 3 can, of course, be improved drastically by
biasing the mesh.

You can request corner outputs (stress, strain, and force) for CQUAD4 in addition to the center
values. Corner results are extrapolated from the corner displacements and rotations by using a strain
rosette analogy with a cubic correction for bending. The same three models are then rerun with this
corner option—their results are summarized in cases 7 through 9. Note that the results can improve
substantially for the same number degrees of freedom.

You select corner output by using a corner output option with the STRESS, STRAIN, and FORCE Case
Control commands. When you select one of these options, output is computed at the center and four
corners for each CQUAD4 element, in a format similar to that of CQUAD8 and CQUADR elements.

There are four corner output options available:

• CORNER

• CUBIC

• SGAGE

• BILIN

These different options provide for different approaches to the stress calculations. The default option is
CORNER, which is equivalent to BILIN. BILIN has been shown to produce better results for a wider range
of problems.

14-14 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 h-elements

To carry it a step further, the same three models are then rerun with CQUAD8 (cases 4 through 6). In this
case, the results using CQUAD8 are better than those using the CQUAD4. This result is expected since
CQUAD8 contains more DOFs per element than CQUAD4. Looking at column three of Table 14-2, you
can see that due to the existence of midside nodes, the models using CQUAD8 contain several times the
number of DOFs as compared to CQUAD4 for the same number of elements. The results using CQUAD4
can, of course, be improved by increasing the mesh density to approach that of the CQUAD8 in terms of
number of DOFs.

It is important to realize that the stresses are compared at different locations for Cases 1 through 3
versus Cases 4 through 9. This difference occurs because the stresses are available only at the element
centers for Cases 1 through 3, but the stresses are available at the corners as well as the element centers
for Cases 4 through 9. When looking at your results using a stress contour plot, you should be aware of
where the stresses are being evaluated.

How fine a mesh you want depends on many factors. Among them is the cost you are willing to pay
versus the accuracy you are receiving. The cost increases with the number of DOFs. The definition
of cost has changed with time. In the past, cost is generally associated with computer time. With
both hardware and software becoming faster each day, cost is probably associated more with the time
required for you to debug and interpret your results. In general, the larger the model is, the more time
it takes you to debug and interpret your results. As for acceptable accuracy, proceeding from case 8 to
case 6, the error is reduced from 4.7% to 1.5%; however, the size of the problem is also increased from
194 to 2538 DOFs. In some cases, a 4.7% error may be acceptable. For example, in cases in which you
are certain of the loads to within only a 10% accuracy, a 4.7% error may be acceptable. In other cases, a
1.5% error may not be acceptable.

In general, if you can visualize the form of the solution beforehand, you can then bias the grid point
distribution. However, this type of information is not necessarily available in all cases. If a better
assessment of accuracy is required and resources are available (time and money), you can always
establish error bounds for a particular problem by constructing and analyzing multiple mesh spacings of
the same model and observe the convergences. This approach, however, may not be realistic due to the
time constraint.

You can use the stress discontinuity feature, as described in “Model Verification” in the Simcenter
Nastran User's Guide, to assess the quality of the mesh density for the conventional h-version elements.

See also

• “CQUAD4 and CTRIA3 Elements” in the Simcenter Nastran Element Library

h-elements

In traditional finite element analysis, as the number of elements increases, the accuracy of the solution
improves. The accuracy of the problem can be measured quantitatively with various entities, such as
strain energies, displacements, and stresses, as well as in various error estimation methods, such as
simple mathematical norm or root-mean-square methods. The goal is to perform an accurate prediction
on the behavior of your actual model by using these error analysis methods. You can modify a series of
finite element analyses either manually or automatically by reducing the size and increasing the number

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 14-15

© 2024 Siemens
 14. Modeling Guidelines

of elements, which is the usual h-adaptivity method. Each element is formulated mathematically with
a certain predetermined order of shape functions. This polynomial order does not change in the
h-adaptivity method. The elements associated with this type of capability are called the h-elements.

Creating Mesh Transitions

You can use mesh transitions, for example, to refine the mesh in a particular area, connect different
element types (for example, a CBAR element to a solid element), or provide transitions required to
model the geometry of the structure.

• Never place a mesh transition in an area of interest or in an area where there is a large variation in
stress.

• Mesh transitions should be located away from the areas of interest in a region.

Due to incompatibilities between finite element types, any transition between different element types
(even a transition from CQUAD4 to CTRIA3 elements) can result in local stress anomalies. Normally,
these stress anomalies are localized and dissipate quickly as you move away from the transition.
However, a problem arises when the transition occurs in an area of interest. In this case, the local stress
rises (or decreases) due to the effect of the transition; in other words, the results may be conservative
(or unconservative) in an area near a transition. However, if this localized stress variation occurs away
from areas of interest, the increase (or decrease) in stress caused by the transition should cause no
concern.

Transition from a Coarse Mesh to a Fine Mesh

Creating a transitions between corase and fine meshes can be challenging. One common method
of performing a transition is to use an intermediate belt of triangular elements as shown in Figure
14-10(a). It is also very tempting to selectively delete midside nodes of higher-order elements (e.g.,
CQUAD8) and use them as transition elements as shown in Figure 14-10(b). However, this method isn't
recommended because it severely distorts the stress distribution in the elements adjacent to the change
in mesh size.

Figure 14-10. Transition Methods

14-16 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Transition from a Coarse Mesh to a Fine Mesh

When using a higher-order element, such as a CQUAD8 or a CHEXA (nine to 20 nodes) for any purpose,
you should include all midside nodes. A CQUAD8 element with all midside nodes deleted is excessively
stiff and therefore is inferior to a CQUAD4. For the CHEXA element, you should use it with eight nodes
(no midside nodes), or use it with 20 nodes (all midside nodes). The midside nodes, if used, should be
located as close to the center of the edge as possible.

As an alternative mesh transition method, a spline element can be used. In Simcenter Nastran, the
RSPLINE element is an elastic interpolation element that connects N end points. Displacements for
intermediate grid points are interpolated from the equations of an elastic beam passing through these
grid points. The spline element assumes a linear interpolation for displacement and torsion along the
axis of the spline, a quadratic interpolation for rotations normal to the axis of the spline, and a cubic
interpolation for displacements normal to the axis of the spline.

Consider an example problem as shown in Figure 14-11. This problem involves a plate fixed at the left
end with a dimension of 14 in by 9 in, and modeled with 1 in by 1 in CQUAD4 elements. For clarity,
only grid points of interest are labeled. Five different loadings-as shown in Table 14-3 are applied to this
structure along the free edge (right end). For convenience, a RBE3 element is created along the free
edge by defining a dependent point (grid point 500) at the center. This grid point is connected to all the
edge grid points. The load can then be applied to grid point 500 and distributed to the rest of the edge
grid points.

Hypothetically, suppose that you're interested in the stresses along the line one inch from the fixed edge.
Since you're not interested in the stresses outside of this region, you can represent the rest of the region
with a coarser mesh. Two different transition methods will be used for this purpose-one with CTRIA3s
(Figure 14-12) and one with RSPLINEs (Figure 14-13). The transitions are performed along the line
five inches away from the fixed edge. An additional transition is performed at x = 11 inches in order
to maintain the same grid point density on the right edge for all three models. This condition assures
identical load applications for all three models. Normally, this additional transition is seldom done in a
real model.

Table. Loadings Used for the Mesh Transition Test

Loading Description
1. Membrane Loads

2. In-Plane Shear Loads (parabolic

distribution)

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 14-17

© 2024 Siemens
 14. Modeling Guidelines

Loading Description
3. In-Plane Bending Loads

4. Vertical Loads

5. Twisting Loads

14-18 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Transition from a Coarse Mesh to a Fine Mesh

Figure 14-11. Baseline Model

Figure 14-12. Transition with CTRIA3

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 14-19

© 2024 Siemens
 14. Modeling Guidelines

Figure 14-13. Transition with RSPLINE

The model with the mesh transition using CTRIA3s, as shown in Figure 14-12, is quite straightforward
and needs no further explanation. However, further discussion is warranted in the case of mesh
transition using the RSPLINE element.

The input format for the RSPLINE element is as follows:

1 2 3 4 5 6 7 8 9 10
RSPLINE EID D/L G1 G2 C2 G3 C3 G4
C4 G5 C5 G6 -etc.-

Field Contents

EID Element identification number.

Gi Grid point identification number.

Ci Components to be constrained.

See also

• “The RBE3 Element” in the Simcenter Nastran Element Library

14-20 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Transition from a Coarse Mesh to a Fine Mesh

Remarks

1. The first and last grid points must be independent.

2. Displacements are interpolated from the equations of an elastic beam passing through the grid
points.

A single RSPLINE is used to connect grid points 51 through 60 as shown in Figure 14-13. Grid points 51,
54, 57, and 60 are considered to be independent, while the other six points are dependent (grid points
52, 53, 55, 56, 58, and 59). This RSPLINE interpolates the motions at the intermediate points, indicated
by circles in Figure 14-13 (grid points 52, 53, 55, 56, 58, and 59), from the cubic equations of a beam
spline passing through the points indicated by squares (grid points 51, 54, 57, and 60). The input for the
RSPLINE element connecting grid points 51 through 60 is as follows:

1 2 3 4 5 6 7 8 9 10
RSPLINE 1000 51 52 12345 53 12345 54
55 12345 56 12345 57 58
12345 59 12345 60

The results using these two transition methods are compared with the baseline model with no
transition. These results are evaluated at two locations: at x = 1.0 in (location of interest) and at x
= 5.0 in (transition region). The five different loading conditions mentioned previously were used. The
Hencky-von Mises stresses using the grid point stress feature (“Reviewing Grid Point Stresses”) and the
CQUAD4 corner stress option are used for comparison. The results are shown in Figure 14-14 through
Figure 14-18. As you can see, the results compare well, using either method of transition, with the
baseline model at the area of interest (away from the transition region) for all five loading conditions.

At the transition region, depending on the loading and location, the transition using CTRIA3 elements
compares better with the baseline model in some cases, whereas the transition using RSPLINE elements
compares better with the baseline model in other cases. Note that there are only four stress values
available in the case of using CTRIA3s since there are only four grid points along the line joining grid
points 51 through 60 in Figure 14-12.

The CTRIA3 elements used in this model are all regularly shaped elements; in most real life problems,
it is very likely that the transition areas may consist of CTRIA3 elements that are more distorted. In this
case, the transition using the distorted CTRIA3 elements does not perform as well as the model in Figure
14-11 consisting of regularly shaped CTRIA3 elements.

In either case, there are noticeable differences in the stresses at the transition region, as expected.
However, at the area of interest and away from the transition region, the stresses are virtually the
same for both cases as compared to the baseline model. Once again, mesh transitions should always be
modeled away from the area of interest.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 14-21

© 2024 Siemens
 14. Modeling Guidelines

Figure 14-14. Resultant Stresses for Membrane Loads

14-22 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Transition from a Coarse Mesh to a Fine Mesh

Figure 14-15. Resultant Stresses for In-Plane Shear Loads

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 14-23

© 2024 Siemens
 14. Modeling Guidelines

Figure 14-16. Resultant Stresses for In-Plane Bending Loads

14-24 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Transition from a Coarse Mesh to a Fine Mesh

Figure 14-17. Resultant Stresses for Vertical Loads

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 14-25

© 2024 Siemens
 14. Modeling Guidelines

Figure 14-18. Resultant Stresses for Twisting Loads

Mesh Transition Between Dissimilar Element Types

Attaching a plate or bar element to a solid element is a case of transition between dissimilar element
types. This process is more involved than it appears at first glance. Solid elements have stiffness only
in the translational DOFs at the attachment grid points; they have no stiffness for rotational DOFs. A

14-26 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Mesh Transition Between Dissimilar Element Types

simple visualization is to think of the attachment of a solid element to a grid point as a “ball-and-socket”
joint, that is, translational forces may be transmitted, but no moment may be transmitted through the
connection.

This incompatibility of the element stiffness matrices represents a modeling problem whenever plate or
bar elements are attached to solid elements. Both plate and bar elements have stiffness for rotational
DOFs (although the plate element may not have a stiffness for the normal rotation). Therefore, special
modeling must be performed whenever a plate or bar is connected to a solid element. Otherwise, the
connection becomes a hinge (for plate elements) or a pinned connection (for bar elements).

Several methods are available to handle the transition between these elements. These methods range
from adding extra elements (for example, adding an additional plate or bar that continues into the solid
element) to using special (R-type) elements for the transition.

One method of handling this transition is to use RBE3 elements. The RBE3 is an interpolation element,
which is ideally suited for this application. By using RBE3s, the rotations of the attached grid points is
simply dependent to the translations of the adjacent grid points.

Examples of using RBE3 elements to connect a solid element to a plate element and bar element are
shown in Figure 14-19. The RBE3 elements attach the rotational DOFs to the translational DOFs on the
solid element.

Figure 14-19. Typical Transition Between Dissimilar Elements

For the plate to solid connection, two RBE3 elements suffice:

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 14-27

© 2024 Siemens
 14. Modeling Guidelines

1 2 3 4 5 6 7 8 9 10
$RBE3 EID REFGRID REFC WT1 C1 G1,1 G1,2
RBE3 901 1 456 1.0 123 2 3
4

RBE3 902 2 456 1.0 123 1 3

4

For the bar to solid connection, one RBE3 element can make the connection:

RBE3 903 12 456 1.0 123 11 13

22

These RBE3 elements transmit the loads to the independent DOFs. If RBE2 elements are used, then the
connection is “rigid.”

Remember, when handling these connections, the solid elements have no stiffness for rotational DOFs,
whereas the real structure does. This means that a special modeling effort is needed when any element
with bending stiffness is connected to a solid element.

When using the RBE3 element, care must be taken to ensure that the independent DOFs are sufficient to
transfer any applied loads. For the bar-to-solid connection in the figure Resultant Stresses for Twisting
Loads, if only two independent grid points are used, the element is “unstable,” that is, since only the
translational DOFs are used as independent DOFs in the sample, the element is unstable for rotation
about the axis connecting the two points. Therefore, three non-colinear grid points are used. A simple
way to remember this is to ask, “If I constrain the DOFs that I list as independent on the RBE3, can I
prevent any possible rigid body motion?” If the answer to this question is “yes,” then the RBE3 element is
capable of transferring any applied loads. In this way, you can avoid possible problems in processing the
RBE3 elements.

See also

• “The RBE3 Element” in the Simcenter Nastran Element Library

Shell-to-solid Transition Element (RSSCON)

The RSSCON method is another way to model shell-to-solid transitions (see Figure 14-20). This
capability conveniently eliminates the need to define RBE3s or MPCs (multipoint constraints) to constrain
a shell element's translational and rotational degrees of freedom to a solid element's translational
degrees of freedom. When using the RSSCON capability, the shell element mesh must line up with the
solid element mesh so that there is an exact element-to-element correspondence.

14-28 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Shell-to-solid Transition Element (RSSCON)

Figure 14-20. Clamped Connection of a Shell to a Solid

Through the RSSCON Bulk Data entry, the shell elements CQUAD4, CQUAD8, CTRIA3, CTRIA6, CQUADR,
and CTRIAR can be connected to the solid elements CHEXA, CPENTA, CPYRAM, and CTETRA. Elements
with midside nodes are also supported.

RSSCON generates a multipoint constraint, which puts the shell degrees of freedom in the dependent
set (m-set). The three translational degrees of freedom and the two rotational degrees of freedom of
the shell edge are connected to the three translational degrees of freedom of the upper and lower solid
edge. Poisson's ratio effects and temperature loads are modeled correctly. The generated multipoint
constraints produce six zero-energy modes for rigid-body motion.

The RSSCON Bulk Data entry defines the connection of a shell element to a solid element. Within
Simcenter Nastran, however, there are two options for making this connection using the RSSCON Bulk
Data entry, as shown in Figure 14-21:

1. The ELEM option, which allows you to specify the element ID of the shell and the element ID of the
solid to which the shell is to be connected.

2. The GRID option, which allows you to specify the grid point ID of the shell and the upper and lower
grid point IDs of the solid. Two triplets of grid point IDs may be specified on one RSSCON Bulk Data
entry.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 14-29

© 2024 Siemens
 14. Modeling Guidelines

Figure 14-21. ELEM and GRID Option on the RSSCON Bulk Data Entry

The best modeling practice is illustrated in Figure 14-22. The height of the connected solid element
should be chosen equal to the thickness of the shell. If the height of the connected solid element is
much larger than the thickness of the shell element, then the connection modeled with RSSCON will
be stiffer than the continuum model. For example, in a mesh where shell grid points are identical or
coincide with solid grid points, the RSSCON Bulk Data entry may model a connection that is too stiff
(see Figure 14-23[a]). Note also that in the model of Figure 14-23[b], the GRID option must be used to
connect a shell element to more than one solid element.

The RSSCON connector element can accommodate a variety of mesh topologies, as shown in Figure
14-24.

14-30 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Shell-to-solid Transition Element (RSSCON)

Figure 14-22. Best Modeling Practice for RSSCON

Figure 14-23. RSSCON in Meshes Where Solid and Shell Elements Share Grid Points

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 14-31

© 2024 Siemens
 14. Modeling Guidelines

Figure 14-24. Modeling Options with RSSCON

The only modeling requirement is that the shell grid point must lie on the line connecting the lower and
upper solid grid points. By default, the RSSCON allows the shell grid point to have an offset from this line
of up to 5% of the distance between the two solid grid points. If a shell grid point does not lie on the line
joining the two solid grid points, but within this 5% tolerance, the shell grid point will be moved to lie on
the line. On the other hand, if a shell grid point lies outside this 5% tolerance, the run terminates with a
fatal message. You can modify the default 5% tolerance by adding “PARAM,TOLRSC, ϵ ” to the Bulk Data
Section, where ϵ is the desired tolerance percentage. The default value for ϵ is 0.05. You can print the
old and new location of the moved shell grid points with the parameter “PARAM,SEP1XOVR,16".

14-32 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 RSSCON Modeling Recommendations

When you use the GRID option, Simcenter Nastran does not verify that the grid points are valid shell or
solid grid points. When the ELEM option is used, if a solid-shell connection has horizontal edges that
are curved in the shell plane, Simcenter Nastran assumes that the geometry of the shell element is
compatible with that of the solid element. In the case of vertical edges, the shell element grid point
must lie on the line between the upper and lower solid element grid point. A default offset tolerance of
5% (user modifiable via "param,tolrsc,ϵ") of the distance between the solid grid points is allowed.

By default, the moved shell grid points are not printed.

Figure 14-25. RSSCON Modeling Practice Parameters

RSSCON Modeling Recommendations

The RSSCON connector produces excellent results if you follow good modeling practices. The geometry
of the RSSCON connector is checked, and fatal messages are issued for invalid connections. Additional
recommendations are given below.

1. Midside-noded elements should only be connected to other midside-noded elements. For example,
a CQUAD8 element connected to a CHEXA20 element is acceptable; however, a CQUAD4 element
connected to a PENTA15 element is not allowed. Elements with missing midside grid points are not
allowed.

2. For midside-noded elements, avoid using the RSSCON where shells connect to triangular solid
faces.

3. Do not mix coordinate systems when using midside nodes. In other words, when using midside
nodes, do not define a local coordinate system containing all the shell grid points and the basic
coordinate system to define the solid grid points. Otherwise, incorrect answers will be generated.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 14-33

© 2024 Siemens
 14. Modeling Guidelines

4. Only one edge of a single shell element may connect to a solid surface.

5. Do not connect more than one shell element to one solid element when using the ELEM option.

6. Do not attempt solid-shell connections to severely warped (greater than 20%) midside-noded solid
elements.

7. The plane of the shell element should not be parallel to the plane of the connecting face of the
solid element.

8. The connecting edge of the shell element must lie in the face of the solid element. In addition, the
vertex grid points of the shell edge connecting to a quadrilateral solid face must line up with its
opposite edges; for triangular solid faces, the connections may be made between any of the edges.
The tolerance in the geometry may be modified; see PARAM,TOLRSC,ϵ described above.

9. When using the ELEM option in the case of partitioned superelement Bulk Data, you must ensure
that the solid element, RSSCON connector, and shell element which define a shell-solid connection
are all contained within the same superelement.

Example - Cantilever Clamped to a Wall

A cantilever (4.1 in long, 0.1 in thick, 0.5 in wide) is clamped to a wall and loaded at the tip. This
example demonstrates how accurate the local stress distribution is when a RSSCON entry is used to
model the clamped condition. Four different finite element models are used, and their results are
compared with classical beam theory. The deflection and bending stress from beam theory are 0.1103 in
and 9,840 psi, respectively, at the clamped end of the cantilever.

1. The first model is a plate model consisting of all CQUAD4s. The stress contours of this model are
shown in Figure 14-26.

2. The second model is a plate model consisting of all CHEXAs. The stress contours of this model are
shown in Figure 14-28.

3. The third model is a plate-solid model using MPCs to transition between the plate and solid model.
The input file and stress contours of this model are shown in Listing 14-1 and Figure 14-29,
respectively.

4. The fourth model is a plate-solid model using RSSCONs to transition between the plate and solid
model. The input file and stress contours of this model are shown in Listing 14-2 and the figure
Uniaxial Load for a 3 x 3 Mesh, respectively.

An exploded view of the transition used in the third and fourth models are shown in Figure 14-26.
Both the deflections and stress contours of the four models are in good agreement with the theoretical
results, as shown in Table 14-10.

14-34 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Example - Cantilever Clamped to a Wall

Figure 14-26. Stress Distribution for CQUAD4 Model

Table. Tip Displacements and Maximum Stresses at the Clamped End

Cantilever Clamped to Wall Displacement at Tip Stress at Clamped End (psi)

Beam Theory (Rigid Wall) 0.1103 9840.
All CQUAD4s 0.1103 9720.
All CHEXAs 0.1087 9720.
Plate-to-Solid Transition Using MPCs 0.1103 9720.
Plate-to-Solid Transition Using 0.1103 9720.
RSSCONs

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 14-35

© 2024 Siemens
 14. Modeling Guidelines

$
$ mpc1.dat
$
TIME 100
SOL 101
CEND
TITLE= cant cantilever clamped with HEXA
$
ECHO=BOTH
$SEALL=ALL
DISPLACEMENT =ALL
stress(corner) = all
OLOAD =ALL
spc= 100
mpc= 200
$
SUBCASE 10
LOAD = 100
BEGIN BULK
$
param,post,-1
$
grid,1,, 0., 0., 0.
grid,2,,0.1, 0.,0.
grid,3,,0.1,0.5,0.
grid,4,,0.,0.5,0.
grid,5,,0.,0.,0.1
grid,6,,0.1,0.,0.1
grid,7,,0.1,0.5,0.1
grid,8,,0.,0.5,0.1
grid,9,,0.1,0.,0.05
grid,10,,0.1,0.5,0.05
grid,11,,1.1,0.,0.05
grid,12,,1.1,0.5,0.05
grid,13,,2.1,0.,0.05
grid,14,,2.1,0.5,0.05
grid,15,,3.1,0.,0.05
grid,16,,3.1,0.5,0.05
grid,17,,4.1,0.,0.05
grid,18,,4.1,0.5,0.05
$
spc1, 100, 123, 1, 4, 5, 8
$
MPC, 200, 9,5, 1., 6, 1, -10.,,+MP01
+MP01, , 2,1, 10.
MPC, 200,10,5, 1., 7, 1, -10.,,+MP02

14-36 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Example - Cantilever Clamped to a Wall

+MP02, , 3,1, 10.

MPC, 200, 9,4, 1., 2, 2, -10.,,+MP03
+MP03, , 6,2, 10.
MPC, 200,10,4, 1., 3, 2, -10.,,+MP04
+MP04, , 7,2, 10.
MPC, 200, 9,1, 1., 6, 1, -0.5,,+MP05
+MP05, , 2,1, -0.5
MPC, 200, 9,2, 1., 6, 2, -0.5,,+MP06
+MP06, , 2,2, -0.5
MPC, 200, 9,3, 1., 6, 3, -0.5,,+MP07
+MP07, , 2,3, -0.5
MPC, 200,10,1, 1., 7, 1, -0.5,,+MP08
+MP08, , 3,1, -0.5
MPC, 200,10,2, 1., 7, 2, -0.5,,+MP09
+MP09, , 3,2, -0.5
MPC, 200,10,3, 1., 7, 3, -0.5,,+MP10
+MP10, , 3,3, -0.5
$
chexa, 200, 210, 1, 2, 3, 4, 5, 6, +HX1
+HX1, 7, 8
CQUAD4, 300, 310, 9, 11, 12, 10
CQUAD4, 301, 310,11, 13, 14, 12
CQUAD4, 302, 310,13, 15, 16, 14
CQUAD4, 303, 310,15, 17, 18, 16
$
PSOLID 210 100
PSHELL 310 100 .1 100 1. 100
MAT1 100 1.+7 0.
$
FORCE 100 17 1. 0. 0.0 -1.
FORCE 100 18 1. 0. 0.0 -1.
$
ENDDATA

Listing 14-1. Input File for Solid-Plate Model Using MPCs for Transition

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 14-37

© 2024 Siemens
 14. Modeling Guidelines

$ rsscon1.dat
$
TIME 100
SOL 101
CEND
TITLE = CANTILEVER CLAMPED WITH HEXA
$
ECHO=BOTH
DISPLACEMENT = ALL
stress(corner) = all
OLOAD = ALL
SPC = 100
$
SUBCASE 10
LOAD = 100
$
BEGIN BULK
$
param,post,-1
$
GRID,1,,0.0,0.0,0.0
GRID,2,,0.1,0.0,0.0
GRID,3,,0.1,0.5,0.0
GRID,4,,0.0,0.5,0.0
GRID,5,,0.0,0.0,0.1
GRID,6,,0.1,0.0,0.1
GRID,7,,0.1,0.5,0.1
GRID,8,,0.0,0.5,0.1
GRID,9,,0.1,0.0,0.05
GRID,10,,0.1,0.5,0.05
GRID,11,,1.1,0.0,0.05
GRID,12,,1.1,0.5,0.05
GRID,13,,2.1,0.0,0.05
GRID,14,,2.1,0.5,0.05
GRID,15,,3.1,0.0,0.05
GRID,16,,3.1,0.5,0.05
GRID,17,,4.1,0.0,0.05
GRID,18,,4.1,0.5,0.05
$
SPC1, 100, 123, 1, 4, 5, 8
$
RSSCON, 200, elem, 300, 200
$
CHEXA, 200, 210, 2, 3, 7, 6, 1, 4, +HX1
+HX1, 8, 5
CQUAD4, 300, 310, 9, 11, 12, 10

14-38 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Example - Cantilever Clamped to a Wall

CQUAD4, 301, 310,11, 13, 14, 12

CQUAD4, 302, 310,13, 15, 16, 14
CQUAD4, 303, 310,15, 17, 18, 16
$
PSOLID 210 100
$
PSHELL 310 100 .1 100 1. 100
MAT1 100 1.+7 0.
$
FORCE 100 17 1. 0. 0.0 -1.
FORCE 100 18 1. 0. 0.0 -1.
$
ENDDATA

Listing 14-2. Input File for Solid-Plate Model Using RSSCONS for Transition

Figure 14-27. Stress Distribution for CQUAD4 Model

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 14-39

© 2024 Siemens
 14. Modeling Guidelines

Figure 14-28. Stress Distribution for CHEXA Model

14-40 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Example - Cantilever Clamped to a Wall

Figure 14-29. Stress Distribution for Solid-Plate ModelUsing MPCs for Transition

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 14-41

© 2024 Siemens
 14. Modeling Guidelines

Figure 14-30. Stress Distribution for Solid-Plate ModelUsing RSSCONs for Transition

Reviewing Grid Point Stresses

When you are working with a structure that is modeled with either plate or solid elements, you will
generally want to review the element component stresses. If this is the case, extra care must be taken
on your part to ensure that the component stresses, which you are evaluating, are output in some
consistent manner.

Consider a simple problem as shown in Figure 14-31. This plate, which is 0.1 inch thick, is subjected to
a uniaxial tensile load of 1 lb/in, and the boundary condition is as shown in Figure 14-31. This model
is represented with a 3 x 3 mesh of CQUAD4 elements. If the elements are connected in a manner
as shown in Figure 14-32, then all the element σx values are in the direction of the applied loads. In
this case, σx = 10 psi for all nine elements. On the other hand, if you were to change, for instance,
the connectivity order for element 5 from grid points 6-7-11-10 to 7-11-10-6, the component stresses
and would be swapped. Note that this is the only difference betweenFigure 14-32 and Figure 14-33.
However, if you are not aware of it and you are making a contour plot of σx, the results will be quite
different for the two cases.

14-42 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Reviewing Grid Point Stresses

Figure 14-31. Uniaxial Load for a 3 x 3 Mesh

Figure 14-32. Element Component Stresses

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 14-43

© 2024 Siemens
 14. Modeling Guidelines

Figure 14-33. Element Component Stresses

To avoid this type of surprise, it may be easier and more meaningful to look at the invariant stress
quantities, such as Hencky-von Mises stresses, rather than the component stresses. The grid point stress
(GPSTRESS) option in Simcenter Nastran offers you another alternative. The grid point stress option
calculates the stresses at the grid points from the adjoining plate and solid elements in a coordinate
system defined by you.

This option enables you to request the output of stresses at grid points in surfaces of two-dimensional
plate elements—namely, the CQUAD4, CQUADR, CQUAD8, CTRIA3, CTRIAR, and CTRIA6 elements—and
in volumes containing CHEXA, CPENTA, CPYRAM, and CTETRA solid elements.

See also

• “Element Data Recovery Resolved at Grid Points” in the Simcenter Nastran User's Guide

Defining Consistent Loading

In finite element analysis, you often apply loads to the elements rather than to the grid points. Examples
of such loadings include the edge load on a CQUAD4 element or a pressure load on one of the faces of a
solid element.

When you convert these element loads to grid point loads, a common tendency is to spread them
equally at the connected grid points. Let us look at an example of a typical plate element with uniform
in-plane loading at one of the edges as shown in Figure 14-34. The total load (P) is to be uniformly
distributed along the edge of the element with an intensity of P/l, where l is the length of the edge. This
same load is to be applied to six different elements-CQUAD4, CTRIA3, CQUAD8, CTRIA6, CQUADR, and
CTRIAR. If you lump these loads equally to the grid points, the load at each node is P/2 for the CQUAD4,
CTRIA3, CQUADR and CTRIAR, and P/3 for the CQUAD8 and CTRIA6. Since the geometry and boundary
condition are identical for the six models, and the loads are “seemingly equivalent,” you will expect
the results (displacements, stresses, etc.) to be identical. However, in general, this is not the case. The

14-44 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Defining Consistent Loading

results for the cases using CQUAD8, CTRIA6, CQUADR, and CTRIAR are different from the results using
CQUAD4 and CTRIA3. The reason for this is because these “seemingly equivalent” loads are actually not
quite equivalent. They are considered as lumped loads. In order to obtain the correct load distribution,
these loads need to be converted to consistent loads.

The equivalent loads at the grid points computed from the element loads are known as consistent loads,
and they are calculated by applying the principle of virtual work. The same shape function that is used in
deriving the element stiffness is used for arriving at this load-and hence the word consistent load. They
are a function of the element types and the applied loads.

Depending on the element, these lumped loads in general are not equal to the consistent loads. The
consistent loads for each of these elements for a uniform in-plane edge load of P/l are shown next to
each element. In this case, the lumped loads and consistent loads are the same for the CQUAD4 and
CTRIA3. They are not the same for the CQUAD8, CTRIA6, CQUADR and CTRIAR. As you can see, they
are substantially different from the lumped loads that were discussed in the previous paragraph. It is
interesting to note that the loads are distributed as 1/6, 4/6, and 1/6 along the edge for the CQUAD8
and CTRIA6, which is quite different from the lumped load approach. For the CQUADR and CTRIAR, an
additional moment Pl/12 is needed to arrive at the consistent loads. When the corresponding consistent
loads shown in Figure 14-34 are applied to each of the respective elements, the force distributions are
then equivalent for all the elements.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 14-45

© 2024 Siemens
 14. Modeling Guidelines

Figure 14-34. Consistent Loads Due to Uniform In-Plane Element Edge Load

Consider another example with solid elements. A load P is to be applied evenly as an outward pressure
load (P/A) to a surface of each of the four solid elements shown in Figure 14-35. Again, if you lump
these loads equally to the grid points, the lumped load at each node is P/4 for the CHEXA (with eight
nodes) and CPENTA (with six nodes), and P/8 for the CHEXA (with 20 nodes) and CPENTA (with 15
nodes). The consistent loads are shown in Figure 14-35. The consistent and lumped loads are the same
for the CHEXA (with eight nodes) and the CPENTA (with six nodes) for this case. The consistent loads
for the CHEXA (with 20 nodes) and CPENTA (with 15 nodes) are shown in Figure 14-35(c) and Figure
14-35(d), respectively. The load distribution along each edge is –.0833P, .3333P and –.08333P, or a –
1/4/–1 ratio, which is quite counter intuitive, especially for the sign change. Fortunately, you do not have
to calculate this consistent load for the pressure load. This consistent load is generated automatically
inside Simcenter Nastran by using the PLOAD4 entry.

14-46 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Symmetry

Figure 14-35. Consistent Loads Due to Uniform Loads on a Solid Element Face

As you can see, consistent loads are functions of the element types and applied loads.

Symmetry
When you think of a symmetrical structure, you most likely think of a structure that has one or more
planes of reflective symmetry. Although there are other kinds of symmetry available in Simcenter
Nastran, reflective symmetry is the only type that is discussed in this section. Simcenter Nastran also
provides a series of special solution sequences that automates some of these other types of analyses.
They are known as the cyclic symmetry features. If a structure is symmetric, then the size of your
finite element model can be reduced, which, in turn, reduces the time and cost of your analysis. For
each plane of symmetry that you have in your model, the model size can be reduced by a factor of
approximately two. Figure 14-36(a) and Figure 14-36(b) illustrate structures that contain one and two

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 14-47

© 2024 Siemens
 14. Modeling Guidelines

planes of symmetry, respectively. In the first case, only half the model needs to be represented. In the
second case, only a quarter of the model needs to be represented.

Figure 14-36. Structures with Planes of Symmetry

If the loads applied to the structure are symmetric relative to the plane of symmetry, then the full model
can be replaced with half the model by applying a symmetric boundary condition. On the other hand,
if the loads are antisymmetric, the same simplification can be achieved by applying the antisymmetric
boundary condition.

A symmetric boundary condition implies that the displacements normal to the plane of symmetry and
rotations about the axes in the plane of symmetry are zero at the plane of symmetry. An antisymmetric
boundary condition implies that the displacements in the plane of symmetry and rotations normal to the
plane of symmetry are zero at the plane of symmetry.

Any general loading condition can be broken into a combination of symmetric and antisymmetric loads
relative to the plane of symmetry. You can work through an example of applying symmetry using the
model in Figure 14-37(a). This load can be broken into two separate loads as shown in Figure 14-37(b)
and Figure 14-37(c).

14-48 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Symmetry

Figure 14-37. Decomposing a General Loading into Symmetric and Antisymmetric Loadings

Note that Figure 14-37(a) is a case of a general loading applied to a structure with one plane of
symmetry. Figure 14-37(b) and Figure 14-37(c) represent a symmetric loading and an antisymmetric
loading, respectively, applied to the same structure. By using symmetry, only one-half of the structure,
as shown in Figure 14-37(b), needs to be modeled by applying the symmetric boundary condition at the
plane of symmetry (see Figure 14-38).

Figure 14-38. Symmetric Model

Similarly, only one-half of the structure shown in Figure 14-37(c) needs to be modeled by applying the
antisymmetric boundary condition as shown in Figure 14-39.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 14-49

© 2024 Siemens
 14. Modeling Guidelines

Figure 14-39. Antisymmetric Model

User Interface

No special user interaction is required in the Executive Control Section or Bulk Data Section. Depending
on the number of planes of symmetry, approximately one-half or less of the structure needs to be
modeled. For the above frame model, only one-half of the model needs to be analyzed.

This feature is basically activated by Case Control commands. The above frame example is used to
illustrate the required Case Control commands. Listing 14-3 contains the input file for this problem.

14-50 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 User Interface

$
$ FILENAME - symbar.dat
$
ID SYM ANTI
TIME 5
SOL 101
CEND
$
TITLE = SYMMETRIC AND ANTISYMMETRIC
SUBCASE 1
LABEL = SYMMETRIC CONSTRAINTS - Y LOAD
SPC = 1
LOAD = 2
$
SUBCASE 2
LABEL = ANTISYMMETRIC CONSTRAINTS - Y LOAD
SPC = 2
LOAD = 2
$
SUBCOM 3
LABEL = LEFT SIDE OF MODEL - Y LOAD
SUBSEQ 1.0, 1.0
DISP=ALL
$
SUBCOM 4
LABEL = RIGHT SIDE OF MODEL - Y LOAD
SUBSEQ 1.0, -1.0
DISP=ALL
$
BEGIN BULK
CBAR 1 100 1 2 -1.0 0.0 0.0
CBAR 2 100 2 3 0.0 1.0 0.0
FORCE 2 2 2500. 0.0 -1. 0.0
GRID 1 0.0 0.0 0.0 123456
GRID 2 0.0 10.0 0.0
GRID 3 5.0 10.0 0.0
MAT1 1 3.+7 0.3
PBAR 100 1 5.0 5.0 5.0 10.
SPC1 1 156 3
SPC1 2 234 3
ENDDATA

Listing 14-3. Input File for Using Symmetric and Antisymmetric Boundary Conditions

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 14-51

© 2024 Siemens
 14. Modeling Guidelines

The first subcase calls out the symmetric boundary condition and load as shown in the figure Symmetric
Model (b). The second subcase calls out the antisymmetric boundary condition and load as shown in
the figure Antisymmetric Model (b).

The third subcase is a “results combination” subcase. No boundary or load conditions need to be applied
for this loading condition (see “Modeling Guidelines”). This subcase produces results for the portion of
the structure that you have modeled (the left-hand side of the frame in this case) by performing a linear
combination of the first two subcases. This combination is achieved by adding 100% of the results for
the first subcase to 100% of the second subcase using the SUBSEQ command. Note that in this case of
results combination, the SUBCOM instead of the SUBCASE command is used.

The fourth subcase is also a “results combination” subcase. It produces results for the portion of the
structure that you did not model (the right-hand side of the frame in this case). These results are
achieved by subtracting 100% of the results of the second subcase from 100% of the results from the
first subcase. As in Subcase 3, the SUBCOM command instead of the SUBCASE command is used in this
case. Once again, no load or boundary condition is needed for this subcase; this is an optional subcase.
It is not required if you do not want to obtain results for the other half of the structure that you did not
model.

The displacements for SUBCOM 3 and SUBCOM 4 are shown in Figure 14-40 along with the results
obtained from a full model run. The results for SUBCOM 3 (left-hand side of the frame) correlate with
the full model results. The results for SUBCOM 4 (right-hand side of the frame) correlate with the
full model results except for the sign changes for components x, θy, and θz. The sign changes occur
because the results of the right half (reflected half) are produced in terms of its left half using the
left-hand coordinate system. Note that these are the same degrees of freedom that are constrained for
the symmetric boundary condition.

14-52 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 User Interface

Figure 14-40. Results Comparison Using Symmetry Versus a Full Model

In the past, when computers were not as fast as they are now, the use of symmetry to reduce the finite
element model size was more popular than it is today. The advantage of using symmetry is obviously
the reduction in model size. The disadvantage is that it requires more effort on your part and it is more
prone to errors since you have to provide equivalent symmetric/antisymmetric boundaries and loading
conditions to your model. In Simcenter Nastran, the use of symmetry is built into some of the solution
sequences-they are known as the cyclic symmetry solution sequences. Similar approaches can be used
with superelement analysis.

The SYM/SYMCOM Case Control command combination can be used instead of the SUBCASE/SUBCOM
combination. If you use this combination, the only changes you have to make is to replace the SUBCASE
and SUBCOM Case Control commands with the SYM and SYMCOM Case Control commands, respectively.
When the SUBCASE/SUBCOM combination is used, output is available in both the SUBCASE and SUBCOM
subcases. On the other hand, when the SYM/SYMCOM combination is used, output requests are only
available in the SYMCOM subcases. There is no real advantage to using the SYM/SYMCOM command as
compared to the SUBCASE/SUBCOM command.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 14-53

© 2024 Siemens
 14. Modeling Guidelines

14-54 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 15. Model Verification
Overview of Model Verification
This chapter is dedicated to various tools that are available for verifying and improving the quality of
your models. They are broken into five different categories:

• Preprocessor checks.

• Strain energy output.

• Diagnostic tools.

• Stress error estimators.

• Postprocessor checks.

Some of these categories overlap each other. Most of these checks provide you with a tremendous
amount of information, but yet their usages are quite straightforward and require very little effort on
your part.

Preprocessor Checks
Preprocessors are used extensively for the generation of finite element models. They also contain
features that can help you to improve the quality of your model and detect errors in your model prior to
analysis. These features can result in substantial savings in terms of time and money. Depending on the
graphics package that you use, some of these features may or may not be available to you.

Creating a Shrink Plot of the Elements

The shrink option allows you to shrink your elements by a specified percentage. This feature is an
excellent tool for identifying missing elements. As an example, let us look at a plate model with 12
CQUAD4 elements surrounded by 31 CBAR elements as shown in Figure 15-1. (A) shows the regular
plot, while (B) shows the plot with the shrink option turned on. The CBAR elements aren't visible since
they lie on top of the edges of the CQUAD4 elements.

If you turn on the shrink option, as shown in Figure 15-1(b), you can easily see that you're missing
a CBAR element at the top and a CQUAD4 element close to the center. You may argue that had
the element labels been turned on, the missing elements would probably have been noticed. This is
certainly a valid argument in this case since the model is a simple one. However, for complex models,
the picture would have become too crowded to be useful had all the labels been turned on. This same
shrink feature can also be applied to solid elements in a similar manner.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 15-1

© 2024 Siemens
 15. Model Verification

Figure 15-1. Shrink Plot

Hidden Line/Shaded Plots

For a complicated model, looking at a wireframe model by itself can be quite confusing since you lose
the depth of field. However, hidden line and shaded plots can serve as the perfect complement to the
wireframe plots. The box model in Figure 15-2 is a perfect example. This model is made of CHEXA
elements and has a hole in the front. By looking at the wireframe plot on the top left-hand corner of
Figure 15-2, you may not notice that there is a hole in the front. However, the hidden line (top right),
shaded (bottom left), and shaded with shrink option (bottom right) plots all reveal the hole in front of
the box. Furthermore, when using these features, you can see the structure being plotted from back to
front, which gives you a better perspective of what the structure looks like. As the structure becomes
more complicated, you will appreciate these features even more.

15-2 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Free Edge/Face

Figure 15-2. Wireframe, Hiddenline, and Shaded Plots

Free Edge/Face

A free edge is an edge that is connected by one single two-dimensional element (e.g., CQUAD4). The
existence of free edges does not necessarily indicate a modeling problem. Figure 15-3 represents a
crude wing model. The bold lines around the outside are examples of legitimate free edges. On the
other hand, whether the internal vertical bold line is a legitimate free edge or not depends on your
design intent. This free edge indicates that elements 12 and 14, 11, and 13 are not connected to each
other. If this is not your intention, then this free edge indicates a potential modeling error.

This can happen, for example, if you create this wing with four separate surfaces as shown in Figure
15-4. Most preprocessors typically create geometric surfaces and then generate the finite element mesh
for each of these surfaces. Since elements 12 and 14 are created from two separate geometric surfaces,
their connecting grid points have distinct IDs. After these elements are generated, if you want them to
share the same edge, then you must perform some sort of equivalencing operation. The exact operation
depends on the graphics package itself. This free edge indicates that you have either forgotten to
perform this equivalencing operation or the equivalencing tolerance level is not tight enough for the
program to perform this task.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 15-3

© 2024 Siemens
 15. Model Verification

Figure 15-3. Wing Free Edge Check

Figure 15-4. Wing Geometric Surfaces

The free face concept is similar to the free edge concept except that it applies to three-dimensional
elements instead of two-dimensional elements. A free face is a face that is occupied by one single
three-dimensional element (e.g., CHEXA). The example in Figure 15-5 contains two volumes. After
these two volumes are meshed with solid elements (e.g., CHEXAs), a free face check is performed. The
outside surfaces of these two volumes indicate that they are free faces that are legitimate. In addition,
it indicates that the adjoining surface (hashed area) is also a free surface. This indicates that the two
volumes are not connected, which may or may not be your intention. If it is not your intention, there is
probably a modeling error.

15-4 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Zipper Effect

Figure 15-5. Possible Modeling Error as Indicated by Free Face

Zipper Effect

When adjoining surfaces are connected at very few points, the interface has a tendency to open up
when it is loaded. This looks similar to a zipper and hence is classified as the “zipper effect.” Use the
surfaces in the figure Wing Geometric Surfaces and create new meshes. Surfaces 1, 2, 3, and 4 contain
meshes of 4X2, 4X2, 3X2, and 3X2, respectively. As you can see in Figure 15-6 the interface between
surfaces 2 and 3 is only connected at two grid points—grid points A and B. The loads can only be
transferred between these two portions of the structure through these two locations. In other words,
the load path may be quite different from what the actual structure does.

Figure 15-6. Model with Zipper Effect

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 15-5

© 2024 Siemens
 15. Model Verification

The “zipper effect” can occur in both plate and solid element models. Therefore, when generating
meshes for a structure with multiple surfaces and/or volumes, you should keep this in mind so that the
interfaces are connected properly. You should always perform a quick check by zooming in at all the
interface locations.

CBAR/CBEAM Orientation and Offset Check

One common error that many users make is the incorrect orientation of the CBAR/CBEAM inertia
properties. If the graphics package that you are using plots these orientation vectors, you should take
advantage of this feature to ensure that you have oriented these elements properly. You should also
check the CBAR/CBEAM element offset plots, if they are available.

See also

• “CBAR Element” in the Simcenter Nastran Element Library

• “CBEAM Element” in the Simcenter Nastran Element Library

Duplicate Grid Points/Elements

The “duplicate grid point check” feature highlights grid points that occupy the same location. Having
grid points occupy the same space may or may not be your intention. An example of unintentional grid
points occupying the same location can be a result of not performing an equivalencing operation after
meshing more than one curve, surface, or volume. An application of intentional duplicate grid points
can be a model used to simulate a bolted joint. This effect can be achieved by connecting a stiff spring
between these two points that occupy the same location but are connected to different portions of the
structure.

The “duplicate element check” feature is similar to the “duplicate grid check” except the check is done
on the elements rather than the grid points. These errors are often due to unintentionally meshing the
same line, surface, or volume more than once.

Properties/Material Plots

You can also assign different colors to elements based on their property IDs (e.g., PSHELL IDs). If you
have assigned an incorrect property, graphically this becomes quite obvious. Differentiating by color can
also be used to highlight material properties (e.g., MATi IDs).

Using Consistent Normals

When you create a model, you should always generate the elements in a consistent manner. In some
cases, if you don't generate your model in a consistent manner, you might inadvertently apply the loads
in the wrong directions.

An example of such a case is when the applied load is a pressure load The plate model, consisting of
four CQUAD4s as shown in Figure 15-7, is used to illustrate this point. As mentioned in Chapter 4,

15-6 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Using Consistent Normals

each CQUAD4 has a positive normal direction associated with it. This direction is defined by the way
you connect the element using the right-hand rule. The symbols used to denote the directions of the
normals are shown in Figure 15-8.

Figure 15-7. CQUAD4 Model

Now apply a 100 psi pressure load to these four elements. The direction of the pressure load is pointing
into the paper. Listing 15-1 contains the relevant partial input file for this job. In this case, the grid
points for each element are defined in a consistent manner. They are connected in a clockwise direction.
Therefore, the positive normals are all pointing into the paper as indicated by Figure 15-8(a).

ID RUN1 PLOAD4
SOL 101
CEND
TITLE = CONSISTENT ELEMENT NORMAL DIRECTIONS
LOAD = 100
.
.
BEGIN BULK
$
CQUAD4,1,10,1,4,5,2
CQUAD4,2,10,2,5,6,3
CQUAD4,3,10,4,7,8,5
CQUAD4,4,10,5,8,9,6
$
PLOAD4,100,1,100.0,,,,THRU,2
PLOAD4,100,3,100.0
PLOAD4,100,4,100.0
.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 15-7

© 2024 Siemens
 15. Model Verification

.
ENDDATA

Listing 15-1. Consistent CQUAD4 Connectivity

Figure 15-8. CQUAD4s with Pressure Loads

On the other hand, perhaps for some reason, you connect element 3 in a counterclockwise direction
(Figure 15-8(b)) instead of a clockwise direction like the other three elements. If this is the case, you
then need to change the sign of the pressure load for element number 3 in order to have identical
loads for both cases (see Listing 15-2). Determining the positive direction of the pressure load for the
CQUAD4 element can be achieved by checking the connectivity order using the right-hand rule. If the
sign for the pressure load applied to element number 3 is not switched, then you have three elements
with the pressure load pointing into the paper and the fourth element with the pressure load pointing
out of the paper.

ID RUN1 PLOAD4
SOL 101
CEND
TITLE = INCONSISTENT ELEMENT NORMAL DIRECTIONS
LOAD = 100
.
.
BEGIN BULK
$
CQUAD4,1,10,1,4,5,2

15-8 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Strain Energy Output

CQUAD4,2,10,2,5,6,3
CQUAD4,3,10,4,5,8,7
CQUAD4,4,10,5,8,9,6
$
PLOAD4,100,1,100.0,,,,THRU,2
PLOAD4,100,3,-100.0
PLOAD4,100,4,100.0
.
.
ENDDATA

Listing 15-2. Inconsistent CQUAD4 Connectivity

This type of inconsistency can occur if the whole model is generated with a preprocessor using multiple
surfaces and the consistency of the connectivity direction is not retained. This problem is also likely
to occur if you edit some of the elements manually and forget to account for the sign change in the
pressure load. Many graphics packages offer this element normal check feature. This type of check
requires very little effort, and in many cases the graphics processor may also allow you to reverse the
direction of the normals.

The CQUAD4 element outputs are in the element coordinate system. Different connectivity orders affect
the way the results are printed. If you are not aware of this fact, you may interpret the results improperly
(see the example in “Stress Error Estimators”).

Strain Energy Output

Frequently you are faced with the decision of modifying your structure in order to meet stringent
design criteria. In most instances, your objective is to maximize the benefits with minimal changes. The
element strain energy output is an excellent tool for identifying areas of modification that will reap the
most benefits for design changes.

The element strain energy is basically the elastic energy stored in the structural element. As an example,
if you hold onto one end of the spring and push slowly on the other end starting from rest, the load
deflection curve looks something like Figure 15-9 for small deflections.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 15-9

© 2024 Siemens
 15. Model Verification

Figure 15-9. Load Deflection Curve

This energy is defined as

Since F = k < d,

Equation 15-1.

By working through a couple of examples, you can see how the element strain energy output is used to
identify areas to be modified in order to reduce deflections.

The problem of interest is shown in Figure 15-10. It consists of two springs in a series with a tip load
applied at grid point 2. The stiffnesses of the springs and the load applied are as shown in Figure 15-10.
The goal is to reduce the tip deflection at grid point 2. Without performing any calculation, it is quite
obvious that stiffening K2 is more efficient than stiffening K1. The next step is to calculate the strain
energy and see if it also guides you in the same direction.

15-10 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Strain Energy Output

Figure 15-10. Spring Sample Problem

The deflections at grid points 1 and 2 can be calculated as follows:

Equation 15-2.

From Eq. 15-1 and Eq. 15-2,

As you can see, U2 is an order of magnitude larger than U1. Therefore, it concurs with our intuition that
stiffening K2 is more effective than stiffening K1 for reducing the deflection at the tip.

The second example is a classic cantilever beam with a vertical tip load applied at the end as shown in
Figure 15-11. This finite element model is made up of five bar elements of equal length with square
cross-sectional properties (0.05m x 0.05m) as shown in Table 15-1.

Table. Cross-Sectional Properties of a Cantilever Beam

Elem. No. L(m) W(m) D(m) A(m2) I1(m4) I2(m4) J(m4)

1 1.0 0.05 0.05 2.5E-3 5.208E-7 5.208E-7 8.789E-7
2 1.0 0.05 0.05 2.5E-3 5.208E-7 5.208E-7 8.789E-7
3 1.0 0.05 0.05 2.5E-3 5.208E-7 5.208E-7 8.789E-7
4 1.0 0.05 0.05 2.5E-3 5.208E-7 5.208E-7 8.789E-7
5 1.0 0.05 0.05 2.5E-3 5.208E-7 5.208E-7 8.789E-7

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 15-11

© 2024 Siemens
 15. Model Verification

Figure 15-11. Cantilever Beam

As is commonly known, the largest deflection occurs at the tip of the cantilever beam. The vertical
deflection at grid point 6 is equal to 1.126833 · 10−3 meters. The goal is to minimize this deflection
at grid point 6. To make the problem more interesting, hypothetically impose the following constraint
such that all that is available is one structural member that is 0.06m · 0.06m and 1m long. In other
words, you can only replace one of the existing five CBAR elements with this new CBAR element. Which
one should you replace to minimize the deflection at grid point 6? Since grid point 6 has the largest
deflection, combined with our experience with the spring problem, you may be tempted to replace CBAR
element number 5 with this new element. If you do this, the vertical displacement at grid point 6 is then
reduced from 1.126833 · 10−3 meters down to 1.122165 · 10−3 meters, a reduction of only 0.414%.
Now run this problem using Simcenter Nastran and see what the element strain energy output suggests.
Listing 15-3 contains a listing of the Simcenter Nastran input file used for this purpose. Note that the
only additional request required is the following command in the Case Control Section:

ESE = ALL

or

SET a = x,y,......ESE = a

or

ESE(PLOT) = ALL

For large models, the element strain energy request can potentially generate a large amount of printout.
If this is not desired, you can use the “plot” option shown above to create a postprocessing file
containing element strain energy data without generating printed output (see Section ). The rest of
the input file is standard Simcenter Nastran input.

15-12 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Strain Energy Output

$
$ FILENAME - secant.dat
$
ID CANT ESE
SOL 101
TIME 10
CEND
$
TITLE = BASELINE MODEL
SUBTITLE =
$
SPC = 10
LOAD = 10
$
SUBCASE 1
SET 1 = 6
DISP = 1
ESE = ALL
$
BEGIN BULK
$
CBAR 1 10 1 2 10
CBAR 2 100 2 3 10
CBAR 3 100 3 4 10
CBAR 4 100 4 5 10
CBAR 5 20 5 6 10
FORCE 10 6 1. 0. 0. 1.
GRID 1 0. 0. 0.
GRID 2 1. 0. 0.
GRID 3 2. 0. 0.
GRID 4 3. 0. 0.
GRID 5 4. 0. 0.
GRID 6 5. 0. 0.
GRID 10 0. 0. 10.
MAT1 1 7.1+10 .33 2700.
PBAR 10 1 2.5-3 5.208-7 5.208-7 8.789-7
PBAR 20 1 2.5-3 5.208-7 5.208-7 8.789-7
$
$ REPLACE PBAR,10,...
$ WITH THE FOLLOWING FOR .06 X .06 CROSS SECTION
$
$PBAR,10,1,3.6-3,1.08-6,1.08-6,1.823-6
$
PBAR 100 1 2.5-3 5.208-7 5.208-7 8.789-7
SPC1 10 123456 1

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 15-13

© 2024 Siemens
 15. Model Verification

$
ENDDATA

Listing 15-3. Input File for the Strain Energy of a Cantilever Beam

The Simcenter Nastran element strain energy output for this run is summarized in Figure 15-12. As
it turns out, element number 1 has the highest element strain energy among the five elements. In
fact, it contains 48.8% of the total strain energy of the whole structure. On the other hand, element
number 5 contains the lowest element strain energy among the five elements—only 0.8%. In other
words, element number 5 is the least effective element to modify if you wish to increase the stiffness of
the structure for the same amount of weight increase.

This job is now rerun by replacing element number 1 with the new 0.06 × 0.06 element. The deflection
at grid point 6 is now reduced from 1.126833 × 10−3 meters down to 8.421097 × 10−4 meters, or a
reduction of 25.27% as compared to a mere 0.414% reduction if you had replaced element number 5
instead.

Figure 15-12. Element Strain Energy Output for a Cantilever Beam Model

As you can see, the element strain energy output is an extremely useful tool in helping you to identify
the most efficient locations for modification. The same concept used for the previous two simple
examples can be applied to complex models in the same manner. Again, the only additional input
needed is the “ESE” request in the Case Control Section. The output format is identical to that of the
table Contributions Used for the Grid Point Force Balance. Furthermore, most modern plot packages
also support element strain energy output. For plate and solid elements, these strain energy plots look
similar to stress contour plots.

A car model (courtesy of Lapcad Engineering), containing CQUAD4 and CTRIA3 elements, is used to
illustrate the element strain energy contour plot. The Executive Control and Case Control Sections are
shown in Listing 15-4. As you can see, the amount of user input required is quite small. The Bulk Data
for this model is inserted into the input file using the INCLUDE statement. A 1g loading is applied to each
of the three orthogonal axes. The element strain energy contour plot for a 1g loading in the z-direction
is shown on a deformed plot in Figure 15-13.

15-14 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Strain Energy Output

$ FILENAME - secar.dat
$
$ MODEL COURTESTY OF LAPCAD ENGINEERING
$
ID LSUG, CAR
SOL 101
TIME 60
CEND
TITLE = CAR MODEL WITHOUT SUSPENSION SPRINGS
ECHO = UNSORT
SPC = 10
DISP(PLOT) = all
ESE(PLOT) = all
$
SUBCASE 10
LABEL = 1G LOAD IN X-DIRECTION
LOAD = 1
$
SUBCASE 20
LABEL = 1G LOAD IN Y-DIRECTION
LOAD = 2
$
SUBCASE 30
LABEL = 1G LOAD IN Z-DIRECTION
LOAD = 3
$
BEGIN BULK
$
INCLUDE 'carbulk.dat'
$
ENDDATA

Listing 15-4. Input File for the Strain Energy of a Car Model

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 15-15

© 2024 Siemens
 15. Model Verification

Figure 15-13. Element Strain Energy Contour Plot of an Automobile Model

(Geometry model courtesy of LAP CAD Engineering)

Diagnostic Tools
Numerous diagnostic tools are available in Simcenter Nastran for debugging and understanding the
solution process. Some of them are standard output; others are available with requests in the Executive
Control Section, Case Control Section, and/or Bulk Data Section.

This section covers many of these tools. Whenever possible, the problem in Listing 15-5 is used to
demonstrate these features. The problem in Listing 15-5 is similar to the cantilever beam shown in the
figure Possible Modeling Error as Indicated by Free Face (listing Input File for the Strain Energy of a
Cantilever Beam) with the exception of an additional load case (gravity load) and output request. The
modifications and additions are indicated by the shaded regions.

$ FILENAME - diag.dat
SOL 101
TIME 10
CEND
$
TITLE = DIAGNOSTIC TOOLS
$
SPC = 10
DISP = ALL

15-16 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Element Summary Output

SPCF = ALL
OLOAD = ALL
$
SUBCASE 1
LABEL = VERTICAL TIP LOAD
LOAD = 10
$
SUBCASE 2
LABEL = GRAVITY LOADS
LOAD = 20
$
BEGIN BULK
$
$PARAM,GRDPNT,0
$
CBAR 1 10 1 2 10
CBAR 2 10 2 3 10
CBAR 3 10 3 4 10
CBAR 4 10 4 5 10
CBAR 5 10 5 6 10
FORCE 10 6 100. 0. 0. 1.
GRAV 20 9.8 0. 0. 1.
GRID 1 0. 0. 0.
GRID 2 1. 0. 0.
GRID 3 2. 0. 0.
GRID 4 3. 0. 0.
GRID 5 4. 0. 0.
GRID 6 5. 0. 0.
GRID 10 0. 0. 10.
MAT1 1 7.1+10 .33 2700.
PBAR 10 1 2.5-3 5.208-7 5.208-7 8.789-7
SPC1 10 123456 1
$
ENDDATA
Listing 15-5. Input File for Illustrating Diagnostic Checks

Element Summary Output

You can use the ELSUM case control command to request that an element summary be output. The
element summary includes element-id, material-id, length (or thickness), area, volume, structural mass,
non-structural mass, total mass, and weight (weight mass * total mass). You can specify that the
element summary be grouped according to either element type or property type. You can also specify
whether the software should:

• Include only mass totals for the element type or property type grouping.

• Include non-structural mass in the mass totals for the element type or property type grouping.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 15-17

© 2024 Siemens
 15. Model Verification

Prior to the ELSUM case control command, the EST parameter was used to obtain length, area, and
volume output for each element in the model.

The following example uses the ELSUM case control command to request summary information for the
elements included in SET 1.

0
0 C A S E C O N T R O L E
C H O
COMMAND
COUNT
1 $
2 SET 1 = 401 THRU 403
3 DISP(PRINT,PUNCH)=ALL
4 ELST(PRINT,PUNCH)=ALL
5 ELFO(PRINT,PUNCH)=ALL
6 SPCF(PRINT,PUNCH)=ALL
7 ESE(PRINT,PUNCH)=ALL
8 SPC=50
9 $
10 SUBCASE 100
11 ELSUM=1
12 LOAD=100
13 ESE(PRINT)=ALL
14 $
15 BEGIN BULK

The corresponding element summary is as follows:

Figure 15-14.

See also

• ELSUM in the Simcenter Nastran Quick Reference Guide

15-18 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Element Geometry Checks

Element Geometry Checks

Thorough evaluations of the various elements occurs prior to and during element matrix generation
operations. There are two categories of these element geometry checks.

• The system element checks determine if the element geometry is adequate for finite element matrix
generation. You cannot control their thresholds. The system controlled checks will always produce a
fatal error if a condition is found which prevents the analysis to proceed.

• The user controlled element checks are controlled with the GEOMCHECK executive statement.
These checks report on the element quality even when the system checks consider all elements
adequate for matrix generation. They can be turned off, or alternately, you can request that they
generate a fatal error when an element fails your defined criteria. By default, they do not produce
fatal errors, and occur using default values even when the GEOMCHECK statement is not included.

Taking a beam element for example, the system checks will report a fatal geometry condition if the two
end points are in the same location resulting in a zero length element. The user controlled checks on
the other hand check the beam for an excessive ratio of the length with offset vectors to the length
of the beam without them. In the first case, the system check is fatal because the element cannot be
formulated from the given geometry. In the second case, the geometry check is informational and the
simulation analysis proceeds when the GEOMCHECK defaults are used. It is your responsibility to inspect
these informational messages and determine if the results are acceptable.

System Element Checks

The system element checks described below are part of the GEOMCHECK functionality. You cannot
control their thresholds. These checks are performed as follows:
● For all solutions except SOLs 401, 402, and 414, the system geometry checks are performed unless
GEOMCHECK=NONE. If any tests exceed the tolerance, the solution behavior is the same as for the
optional tests and is governed by the MSGTYPE setting.
● For SOL 401, the system geometry checks are always performed. If any tests exceed the tolerance,
the solution will issue a fatal error. You can select the pre-Simcenter Nastran 2021.1 system check by
setting NASTRAN SYSTEM(760)=1, which does not include the Jacobian check at the CTETRA element
corners.
● For SOLs 402 and 414, the GEOMCHECK functionality is not supported.

Shell Element System Check

Jacobian check. For the CQUAD4, CQUADR, CQUAD8, and CTRIA6 shell
elements, the determinant of the Jacobian at the Gauss points
Occurs for all shell elements: CQUAD4,
is calculated.
CQUADR, CQUAD8, and CTRIA6.
A failure occurs if any DET J ≤ 0.0.
In addition, the average DET J is calculated on the CQUADR
and CQUAD4 elements and multiplied by 1E-10.
A failure occurs if any DET J < (DET J)AVG x 1E-10.
3-D Solid Element System Checks

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 15-19

© 2024 Siemens
 15. Model Verification

Jacobian check. a. The determinant of the Jacobian is calculated at the

Gauss points for the CHEXA, CPENTA, CPYRAM, and CTETRA
Occurs for all 3-D solid elements:
elements.
CHEXA, CPENTA, CPYRAM, and
CTETRA. SOL 401 also calculates the determinant of the Jacobian at
the corners for the CTETRA element. You can turn the CTETRA
corner check off by setting NASTRAN SYSTEM(760)=1 (pre-
Simcenter Nastran 2021.1 behavior).
b. The sign of DET J at the first calculated Gauss or corner
point is determined. If it is negative, then the value of
DET J for all points is multiplied by -1. This step is not a
requirement for the Jacobian check, but is done for elements
having reversed connectivity. 3D-Solid elements which have
a consistent Jacobian at the Gauss points but with reversed
connectivity will have a negative DET J at all Gauss points.
To allow for this type of connectivity, the software simply
reverses the sign of DET J by multiplying by -1.
c. The sign of all DET J is checked for consistency (all positive
or all negative). A failure occurs if the sign of any DET J is
inconsistent.
d. Out of all the elements failing the Jacobian check, the
element which has the most negative DET J is reported.
Volume Check. A failure occurs if Volume ≤ 0.0.
Occurs for the CTETRA and CPYRAM
elements.
Axisymmetric Element System Checks
Jacobian check. The determinant of the Jacobian at the Gauss points is
calculated. A failure occurs if any DET J ≤ 0.0.
Occurs for the CTRAX3, CTRAX6,
CQUADX4 and CQUADX8 elements.
Consistent Y check when elements are When axisymmetric elements are modeled in the XZ plane,
modeled in XZ plane. the grids used must all lie in the same Y-plane within the
tolerance +/- 1.E-6, or a failure occurs.
Consistent Z check when elements are
modeled in XY plane. When axisymmetric elements are modeled in the XY plane,
the grids used must all lie in the same Z-plane within the
Occurs for the CTRAX3, CTRAX6,
tolerance +/- 1.E-6, or a failure occurs.
CQUADX4 and CQUADX8 elements.
X ≥ 0.0 check when elements are When axisymmetric elements are modeled in the XZ plane,
modeled in XZ plane. the X location for all grids must be ≥ 0.0 or a failure occurs.
Y ≥ 0.0 check when elements are When axisymmetric elements are modeled in the XY plane,
modeled in XY plane. the Y location for all grids must be ≥ 0.0 or a failure occurs.
Occurs for the CTRAX3, CTRAX6,
CQUADX4 and CQUADX8 elements.

15-20 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 User Controlled Element Checks

Plane Stress and Plane Strain Element System Checks

Jacobian check. The determinant of the Jacobian at the Gauss points is
calculated. A failure occurs if any DET J ≤ 0.0.
Occurs for the plane stress elements
CPLSTS3, CPLSTS4, CPLSTS6, CPLSTS8
and the plane strain elements
CPLSTN3, CPLSTN4, CPLSTN6,
CPLSTN8.
Consistent Y check when elements are When plane stress or plane strain elements are modeled in
modeled in XZ plane. the XZ plane, the grids used must all lie in the same Y-plane
within the tolerance +/- 1.E-6, or a failure occurs.
Consistent Z check when elements are
modeled in XY plane. When plane stress or plane strain elements are modeled in
the XY plane, the grids used must all lie in the same Z-plane
Occurs for the plane stress elements
within the tolerance +/- 1.E-6, or a failure occurs.
CPLSTS3, CPLSTS4, CPLSTS6, CPLSTS8
and the plane strain elements
CPLSTN3, CPLSTN4, CPLSTN6,
CPLSTN8.

User Controlled Element Checks

The GEOMCHECK executive statement is used to control the optional, user controlled element checks.
By default, these checks do not produce fatal errors and occur using default values even when the
GEOMCHECK statement is not included. One exception is that the aspect ratio checks (test_keyword
= TET_AR, HEX_AR, PYR_AR, PEN_AR) for 3D solid elements will produce a fatal message by default
(MSGTYPE=FATAL).

GEOMCHECK controls are available for the shell elements CQUAD4, CQUAD8, CQUADR, CTRIA3, CTRIA6,
CTRIAR, the solid elements CHEXA, CPENTA, CPYRAM, CTETRA, the axisymmetric elements CQUADX4,
CQUADX8, CTRAX3, CTRAX6, the plane stress elements CPLSTS3, CPLSTS4, CPLSTS6, CPLSTS8, the plane
strain elements CPLSTN3, CPLSTN4, CPLSTN6, CPLSTN8, the CBAR element, and the CBEAM element.

The GEOMCHECK executive control statement format is as follows:

Also see GEOMCHECK in the Simcenter Nastran Quick Reference Guide.

The following table describes the user controlled element checks in detail.

Shell Element GEOMCHECK Descriptions

Quad Skew for CQUAD4, CQUAD8, and CQUADR elements.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 15-21

© 2024 Siemens
 15. Model Verification

Test_keywords: Q4_SKEW, Q8_SKEW

The smallest angle θ between vectors joining side mid-points PQ and RS.
Default: Fails if θ < 30.0 degrees

Quad Taper for CQUAD4, CQUAD8, and CQUADR elements.

Test_keywords: Q4_TAPER, Q8_TAPER

The element is transformed to a mean plane such that grids A, B, C, and D all lie in one plane. Taper is
given by
(Amax - Q) ) / Q (non-dimensional)
where Amax is the largest of the four corner triangular areas ABD, BCA, CDB, DAC,
and Q = 0.5 * total quad area.
Default: Fails if Q4_TAPER or Q8_TAPER > 0.5
Note: When a quad element is in the shape of a triangular as shown below, the taper value reaches
1.0.

Quad Warp for CQUAD4 and CQUADR elements.

Test_keyword: Q4_WARP

15-22 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 User Controlled Element Checks

Quad warp is defined as the ratio of HH to the sum of diagonal lengths.

W = HH / (DAC + DBD), where HH is the dot product AB•PK, PK = (AC×BD) / |AC×BD| and DAC and DBD
are distances from A to C and from B to D. In more detail, the normalized PK is an out-of-plane unit
vector. HH is the amount vector AB is projected onto vector PK. If an element is perfectly flat, HH will
be zero (no projection) and there is no warp. Dividing HH by the sum of diagonal lengths DAC and DBD
makes warp independent of the element size.
Default: Fails if W > 0.05

Tria Skew for CTRIA3 and CTRIAR elements.

Test_keyword: T3_SKEW

The smallest included (interior) angle θ from all three corners.

Default: Fails if θ < 10.0 degrees

Interior Angle Checks for CQUAD4, CQUAD8, CQUADR, CTRIA3, CTRIA6, and CTRIAR elements.
Test_keywords: Q4_IAMAX, Q4_IAMIN, Q8_IAMIN, Q8_IAMAX, T3_IAMAX, TA6_IAMN, TA6_IAMX

The extreme interior angles are designated by αmin & αmax.

Defaults:
For CQUAD4, CQUAD8, and CQUADR: Fails if αmax > 150.0 degrees or αmin < 30.0 degrees
For CTRIA3 and CTRIAR: Fails if αmax > 160.0 degrees
For CTRIA6: Fails if αmax > 160.0 degrees or αmin < 10.0 degrees

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 15-23

© 2024 Siemens
 15. Model Verification

Note: αmin for CTRIA3 and CTRIAR is checked by T3_SKEW.

Aspect Ratio for CQUAD4, CQUAD8, and CTRIA6 elements.

Test_keywords: QAD_AR, Q8_AR, TA6_AR

Aspect Ratio = longest element edge / shortest element edge

Default: Fails if Aspect Ratio > 100.0

Edge-Point-Length-Ratio for CQUAD8 and CTRIA6 elements.

Test_keywords: Q8_EPLR, TA6_EPLR

Default: Fails if Min(AB/BC, BC/AB) < 0.5

Note: A, B, C are grid locations on edge ABC.

3D-Solid GEOMCHECK Descriptions

Aspect Ratio for CTETRA, CHEXA, CPENTA, CPYRAM, CHEXCZ, and CPENTCZ elements.
Test_keywords: TET_AR, HEX_AR, PEN_AR, PYR_AR
Default: Fails if AR > 100.0 for all 3-D Solid elements

• For HEX_AR and PEN_AR, it is defined by

AR = longest element edge / shortest element edge

• For TET_AR, it is defined by

AR = longest element edge / shortest height

15-24 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 User Controlled Element Checks

The shortest height is the minimum of H1, H2, H3, and H4

where Hi = (6* Element Volume)/Area(i),

and (6 * Element Volume) is a parallelepiped volume calculated by

Area(i) is calculated for each of the four element faces and is the magnitude of the cross product
of two of the edge vectors on a face. The Area(i) calculation result is equivalent to the area of a
parallelogram.

• For PYR_AR, it is defined by

AR = longest element edge / shortest element edge or pyramid height

The shortest element edge or height is the smallest of one of the eight element edges or the
element height.

The element height is calculated by H = (3* Element Volume)/A

where A is the area of the quad face.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 15-25

© 2024 Siemens
 15. Model Verification

The element volume is calculated by dividing the pyramid element into two tetrahedral elements.
The tetrahedral volumes are calculated as described under the TET_AR description above, each
divided by 6, then added together to obtain the pyramid total volume.

Note: Unlike the other solid element aspect ratio checks, the PYR_AR check calculates the element
volume explicitly. A failure occurs if Volume = 0.0. A CPYRAM element volume check also occurs in
the system element checks.

Face-Planarity for CHEXA, CPENTA, CPYRAM, CHEXCZ, and CPENTCZ elements.

Test_keywords: HEX_WARP, PEN_WARP, PYR_WARP

In the example above, the hex element face ABCD is heavily warped. The face-planarity criterion is
defined as the cosine of the angle θ made by opposite corner face vectors V1 and V2.
Default: Fails if COS(θ) < 0.707

Edge-Point-Included-Angle for CTETRA, CHEXA, CPENTA, CPYRAM, CHEXCZ, and CPENTCZ elements
with midside grids.
Test_keywords: TET_EPIA, HEX_EPIA, PEN_EPIA, PYR_EPIA

Default: Fails if θ1 + θ2 > 30.0 degrees

Note: A, B, C are grid locations on edge ABC.

Edge-Point-Length-Ratio for CTETRA, CHEXA, CPENTA, CPYRAM, CHEXCZ, and CPENTCZ elements with
midside grids.
Test_keywords: TET_EPLR, HEX_EPLR, PEN_EPLR, PYR_EPLR

15-26 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 User Controlled Element Checks

Default: Fails if Min(AB/BC, BC/AB) < 0.5

Note: A, B, C are grid locations on edge ABC.

Ratio of longest edge to shortest edge in stacking direction for CHEXA and CPENTA elements that
reference PCOMPS entries only.
Test_keywords: HEX_TP, PEN_TP

• For HEX_TP

Assume edges AE, BF, CG, and DH coincide with the stack direction

CG is the longest

DH is the shortest

Default: Fails if CG/DH > 1.2

• For PEN_TP

Assume edges AD, BE, and CF coincide with the stack direction

BE is the longest

CF is the shortest

Default: Fails if BE/CF > 1.2

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 15-27

© 2024 Siemens
 15. Model Verification

Ratio of the difference between thickness in stacking direction as defined by grids and as defined by
ply thickness specification to the thickness in the stacking direction as defined by grids for CHEXA and
CPENTA elements that reference PCOMPS entries only.
Test_keywords: HEX_TK, PEN_TK

• For HEX_TK

Assume AB coincides with the stack direction

A and B are located at the center of the bottom and top faces of the element, respectively

AB is thickness as defined by grids.

AB’ is thickness as defined by ply thickness specification.

Default: Fails if |AB – AB’|/AB > 0.05

• For PEN_TK

Assume AB coincides with the stack direction

A and B are located at the center of the bottom and top faces of the element, respectively

AB is thickness as defined by grids.

AB’ is thickness as defined by ply thickness specification.

Default: Fails if |AB – AB’|/AB > 0.05

15-28 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 User Controlled Element Checks

Axisymmetric, Plane Stress, and Plane Strain GEOMCHECK Descriptions

Skew for the axisymmetric elements CQUADX4 and CQUADX8, the plane stress elements CPLSTS4
and CPLSTS8, the plane strain elements CPLSTN4 and CPLSTN8, the chocking elements CCHOCK4 and
CCHOCK8.
Test_keyword: QDX_SKEW

The smallest angle θ between vectors joining side mid-points PQ and RS.
Default: Fails if θ < 30.0 degrees

Taper for the axisymmetric elements CQUADX4 and CQUADX8, the plane stress elements CPLSTS4
and CPLSTS8, the plane strain elements CPLSTN4 and CPLSTN8, the chocking elements CCHOCK4 and
CCHOCK8.
Test_keyword: QDX_TAPR

The element is transformed to a mean plane such that grids A, B, C, and D all lie in one plane. Taper is
defined as
Q4_TAPER = (Amax - Q) ) / Q (non-dimensional)
where Amax is the largest of the four corner triangular areas ABD, BCA, CDB, DAC

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 15-29

© 2024 Siemens
 15. Model Verification

and Q = .5 * total quad area.

Default: Fails if Q4_TAPER > 0.5
Note: When a quad element is in the shape of a triangular as shown below, the taper value reaches
1.0.

Aspect Ratio for the axisymmetric elements CTRAX3, CTRAX6, CQUADX4, CQUADX8, the plane stress
elements CPLSTS3, CPLSTS4, CPLSTS6, CPLSTS8, and the plane strain elements CPLSTN3, CPLSTN4,
CPLSTN6, CPLSTN8, the chocking elements CCHOCK3, CCHOCK4, CCHOCK6, and CCHOCK8.
Test_keyword: TRX_AR, QDX_AR

Aspect Ratio = longest element edge / shortest element edge

Default: Fails if Aspect Ratio > 100.0

Edge-Point-Length-Ratio for the axisymmetric elements CTRAX6 and CQUADX8, the plane stress
elements CPLSTS6 and CPLSTS8, the plane strain elements CPLSTN6 and CPLSTN8, the chocking
elements CCHOCK6 and CCHOCK8.
Test_keywords: TRX_EPLR, QDX_EPLR

Default: Fails if Min(AB/BC, BC/AB) < 0.5

Note: A, B, C are grid locations on edge ABC.

15-30 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Sample Output

Interior Angle Checks for axisymmetric elements CTRAX3, CTRAX6, CQUADX4, CQUADX8, the plane
stress elements CPLSTS3, CPLSTS4, CPLSTS6, CPLSTS8, the plane strain elements CPLSTN3, CPLSTN4,
CPLSTN6, CPLSTN8, the chocking elements CCHOCK3, CCHOCK4, CCHOCK6, and CCHOCK8.
Test_keywords: QDX_IAMX, QDX_IAMN, TRX_IAMN, TRX_IAMX

The extreme interior angles are designated by αmin & αmax.

Defaults:
For the axisymmetric elements CQUADX4 and CQUADX8, the plane stress elements CPLSTS4
and CPLSTS8, the plane strain elements CPLSTN4 and CPLSTN8, the chocking elements
CCHOCK4 and CCHOCK8:
Fails if αmax > 150.0 degrees or αmin < 30.0 degrees
For the axisymmetric elements CTRAX3 and CTRAX6, the plane stress elements CPLSTS3 and
CPLSTS6, the plane strain elements CPLSTN3 and CPLSTN6, the chocking elements CCHOCK3
and CCHOCK6:
Fails if αmax > 160.0 degrees or αmin < 10.0 degrees

CBAR and CBEAM GEOMCHECK Descriptions

BAR and BEAM OFFSET Ratio for CBAR and CBEAM elements with offsets defined.
Test_keywords: BAR_OFF, BEAM_OFF

BAR and BEAM OFFSET Ratio = ABS(Offset Length - Original Length) / Original Length
Default: Fails if Ratio > 0.15

Sample Output

The following output from a small example provides a sample of the message formats that are produced
for geometry tests that can be controlled using the GEOMCHECK statement. For this output, most of the
default test tolerances were modified so that the effect could be observed in the tolerance information
lines of the messages that were produced.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 15-31

© 2024 Siemens
 15. Model Verification

Note that a small table is generated at the end of all messages that summarizes the number of tests
that actually exceeded the tolerance value for each element type as well as a list of the elements
that produced the worst violations. The summary table itself can be produced by using the keyword
SUMMARY on the GEOMCHECK statement.

15-32 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Weight Center of Gravity and Moment of Inertia Check

Weight Center of Gravity and Moment of Inertia Check

You can use the Grid Point Weight Generator (GPWG) to check whether the weight of your model is
correct. To turn on the Grid Point Weight Generator, include the parameter GRDPNT in the Bulk Data or
Case Control Section of your input file.

The data entry format is

PARAM,GRDPNT,i

where “i” is an integer value defining a reference point. The value of “i” can be any grid point in the
model, or if it is set to zero, the reference point is the origin of the basic coordinate system.

The output from the GPWG includes a rigid body mass matrix, various coordinate transformations, and
the location of the center of mass. The output from the GPWG is generally more than what you need.
The mass and center of gravity (CG) location is typically all that is used. A partial output for the problem
in the listing Input File for Illustrating Diagnostic Checks is shown in Figure 15-15.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 15-33

© 2024 Siemens
 15. Model Verification

Figure 15-15. Partial Output from the Grid Point Weight Generator

Simcenter Nastran does not keep track of your units. Therefore, you must input all of your properties
using a consistent set of units. For example, if you are using meters (m) for defining locations on
your grid entries, then your properties, such as areas (A), should be in terms of m2. Simcenter Nastran
expects your mass input (MATi, CONMi, etc.) to be in terms of mass units. However, if you would rather
input your mass in terms of weight units, then you must add the following entry to your Bulk Data
Section:

PARAM,WTMASS,x

In the parameter statement above, x is the appropriate conversion factor based on the units that you are
using with a default value of 1.0. In other words, the weight density is related to the mass density by the
relationship

where:

ρw = weight density
ρm = mass density
g = gravitational acceleration constant

In the example below, the mass came from the density entered on the MAT1 entry associated with the
CBAR elements. In this case, the density is in terms of mass units. Since the grid point locations are
defined in terms of meters, the density should then be in units of kg/m3. Using a consistent set of units,
the Young's modulus (E) and density (ρ) are 7.1E10 N/m2 and 2700 kg/m3 as indicated by Fields 3 and 6,
respectively, on the following MAT1 entry.

$ MASS DENSITY (kg/m3)

$
MAT1 1 7.1+10 .33 2700.

On the other hand, if you want to input your density in weight units, then you should replace the above
MAT1 entry with the following two entries.

15-34 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Mechanisms and Singularities

$ WEIGHT DENSITY (N/m3)

$ ACCELERATION (m/sec2)
$
MAT1 1 7.1+10 .33 2.65+4
PARAM WTMASS 0.1019

This density and all the other masses (excluding M2PP or M2GG input) are multiplied by this scale factor.
In this case, (0.1019 · 2.65E4) = 2700. The results (e.g., displacements, stresses, etc.) are the same
using either of the above mass or weight units.

The only difference is in the grid point weight generator output. The GPWG output is in the same units of
your density or point masses. In other words, if your input is in mass units, then the GPWG output is also
in mass units. If your input is in weight units, then the GPWG output is also in weight units.

If you are using English units, then the following two sets of entries yield the same results. Of course,
other units in your model must also be consistent, e.g., the grid point locations and cross-sectional
properties.

$ MASS DENSITY (lb-sec2/in4)

$
MAT1 1 1.0 +7 .33 2.51-4

$ WEIGHT DENSITY (lb/in3)

$ ACCELERATION (in/sec2)
$
MAT1 1 1.0 +7 .33 0.097
PARAM WTMASS 0.002588

The output from GPWG is for informational purposes only and is not used in any subsequent steps in the
solution process.

See also

• “Grid Point Weight Generator” in the Simcenter Nastran Basic Dynamic Analysis User's Guide

Mechanisms and Singularities

When performing the solution to a system of linear equations, singularities lead to conditions in which a
unique solution is not possible. Simcenter Nastran considers two types of singularities:

• Grid point singularity that is identified by considering the stiffness terms of only one grid point.

• A mechanism type of singularity that requires the consideration of the stiffness terms of more than
one grid point.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 15-35

© 2024 Siemens
 15. Model Verification

Singularities cause ill-conditioned matrices that can be detected in several phases of the Simcenter
Nastran execution.

After matrix assembly, the grid point singularities are detected. At each grid point, a 3X3 partition of the
stiffness matrix for each of the three translational and three rotational DOFs is solved as an eigenvalue
problem to determine the principal stiffnesses. Each stiffness term is compared to the principal stiffness
using the formula

where Kii is the term in the i-th row and i-th column of the matrix and Kmax is the principal stiffness. If ε
is less than the value of PARAM,EPZERO, the global direction nearest i is considered singular. The default
value for EPZERO is 10–8. A list of potential singularities is printed in the grid point singularity table (see
“Strain Energy Output” for further details).

PARAM,AUTOSPC. If PARAM,AUTOSPC,YES is specified (this is the default in the Structured Solution

Sequences, except SOLs 106 and 129), the potential singularities are automatically constrained if
possible.

During decomposition, mechanisms can be detected based on the maximum ratio of the matrix diagonal
to the factor diagonal

where Kii is the i-th diagonal term of the original stiffness matrix and Dii is the i-th diagonal term of the
factor diagonal matrix. For a symmetric matrix K it can be represented as

where:

[L] = lower triangular factor

[D] = factor diagonal matrix

PARAM,MAXRATIO. Refer to the Simcenter Nastran Numerical Methods User's Guide for further
details on the subject of system of linear equations. All terms whose ratio exceed the value of
PARAM,MAXRATIO are printed. The default for MAXRATIO is 107. UIM 4158 prints the statistics for the
decomposition that include the number of negative terms on the factor diagonal and the maximum

15-36 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Mechanisms and Singularities

ratio of matrix diagonal to factor diagonal at a specified row number and its corresponding grid point.
User Warning Message (UWM) 4698 prints the degrees of freedom that have a factor diagonal ratio
greater than the MAXRATIO value or have negative terms on the factor diagonal. Both of these messages
are issued by the DECOMP module and are shown below:

*** USER INFORMATION MESSAGE 4158---STATISTICS FOR SYMMETRIC DECOMPOSITION OF DATA BLOCK
KLL FOLLOW
NUMBER OF NEGATIVE TERMS ON FACTOR DIAGONAL = 1
MAXIMUM RATIO OF MATRIX DIAGONAL TO FACTOR DIAGONAL = 7.2E+15 AT ROW NUMBER
16
*** USER WARNING MESSAGE 4698. STATISTICS FOR DECOMPOSITION OF MATRIX KLL
.
THE FOLLOWING DEGREES OF FREEDOM HAVE FACTOR DIAGONAL RATIOS GREATER THAN 1.00000E+05 OR
HAVE
NEGATIVE TERMS ON THE FACTOR DIAGONAL.
GRID POINT ID DEGREE OF FREEDOM MATRIX/FACTOR DIAGONAL RATIO
MATRIX DIAGONAL
6714 T1 -7.19297E+15
6.02908E+07
^^^ DMAP FATAL MESSAGE 9050 (SEKRRS) - RUN TERMINATED DUE TO EXCESSIVE PIVOT RATIOS
IN MATRIX KLL. USER PARAMETER BAILOUT MAY BE USED TO CONTINUE THE RUN.

If the MAXRATIO value is exceeded, your job terminates with DMAP Fatal Message 9050. You can
override this fatal message by inserting “PARAM,BAILOUT,–1" in your input file. You should, however, be
aware that a large value of the MATRIX/FACTOR DIAGONAL may be an indication of a potential modeling
problem. Taking the log10 of MAXRATIO indicates how many significant digits may have been lost during
the decomposition. The MAXRATIO may change slightly if a different sequencer is used. This change is
due to the fact that the order of operation may change with a different sequencer that yields a different
numerical roundoff.

After decomposition, a singularity may lead to an incorrect solution. In static analysis, Simcenter Nastran
solves

to obtain u (displacements). Using these displacements, Simcenter Nastran then calculates a “residual”
loading vector as follows:

This residual vector should theoretically be null but may not be null due to numeric roundoff. To obtain a
normalized value of the residual loading, an error measure ε is calculated by

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 15-37

© 2024 Siemens
 15. Model Verification

δP can be printed by including PARAM,IRES,1 in your Bulk Data Section. The value ε (epsilon) is printed
in User Information Message (UIM) 5293 as shown below. Epsilon values that are greater than 0.001 are
flagged for a possible loss of accuracy due to numeric conditioning as shown below:

*** USER INFORMATION MESSAGE 5293 FOR DATA BLOCK KLL

LOAD SEQ. NO. EPSILON EXTERNAL WORK EPSILONS LARGER THAN .001 ARE FLAGGED
WITH ASTERISKS
1 -4.5961966E-14 5.6341639E+00
2 -9.2748066E-15 9.4523830E+00

One ε is generated for each loading condition. An acceptable value of ε depends on the model
complexity and the machine that it runs on. An epsilon value of |ε| in the neighborhood of less than 10-9
is generally considered acceptable.

Some general causes for singularity can include:

• Degrees of freedom without stiffness because of missing elements.

• A 2-dimensional plate problem with the normal rotation unconstrained.

• A solid model with rotational DOFs at the corners unconstrained.

• Incorrect modeling of offset beams.

• Incorrect multipoint constraints.

• Mechanisms and free bodies, such as sloped plates, beam to plate connections, beam to solid
connections, and plate-to-solid connections.

• Low stiffness in rotation.

• A stiff element adjacent to a very flexible element.

Applied Loads Check

There are two checks in Simcenter Nastran to let you verify that loads are being applied correctly to the
structure. The first one is the summation of the total loads applied to the structure about the reference
point x as specified on the PARAM,GRDPNT,x entry. If PARAM,GRDPNT,x is not specified, this summation
is performed about the origin of the basic coordinate system. This quantity is known as the OLOAD
RESULTANT output and will output by default. It can be deactivated by adding PARAM,PRTRESLT,NO in
the bulk data. It consists of seven lines of output per load case. The OLOAD RESULTANT output for

15-38 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Applied Loads Check

the problem in the listing Input File for Illustrating Diagnostic Checks is shown in the figure Partial
Output from the Grid Point Weight Generator. Note that loads from scalar points are not included in
the OLOAD RESULTANT output.

Figure 15-16. OLOAD Resultant

T1, T2, T3, R1, R2, R3 are the basic coordinate directions of the summed loads in the table. The row
headings Fx, Fy, Fz, Mx, My, Mz are relative to the applied loads, also in basic coordinates.

Look at Subcase 1 from the problem in the listing Input File for Illustrating Diagnostic Checks which
has a vertical tip load applied to grid point 6 in the +z direction. Since there are no loads applied to the
structure in the x- or y-direction, the loads in those two directions are equal to zero. The total z-load is
equal to 100 Newtons contributed solely by the applied load at grid point 6. The total moment about
the origin of the basic coordinate system is −5 · (100 N) = −500 NM about the y-axis. This R2 moment
is contributed by the Z load as indicated by the -500.0 in row FZ, column R2. There is no moment about
the other two axes. The next seven lines in the OLOAD RESULTANT output is due to the gravity load in
subcase 2.

So far, only the resultant applied load was discussed. You can also request more detailed information
regarding the exact location on the structure where loads are applied. This can be done with the OLOAD
request in the Case Control Section. The format is as follows:

OLOAD = All

or

SET 10 = 1,2,3,6
OLOAD = 10

In the case of OLOAD = All, a printout of all the applied loads is requested. As shown in Figure 15-17,
there is one load applied to grid point 6 in the +z direction for Subcase 1. However, for the gravity load

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 15-39

© 2024 Siemens
 15. Model Verification

in Subcase 2, each grid point that has mass associated with it has a 1g load applied to it. Each CBAR
element contributes {(2700)(1.0)(2.5E-3)(9.8)/2} = 33.075 N to each end of its connecting grid points.
Therefore, a 33.075 N load is applied to the end points and a 66.15 N load is applied to each of the
intermediate points since there are two CBAR elements connected to each of these intermediate grid
points. As you can see, for a large model, OLOAD output for gravity loads can potentially generate many
lines of output.

Figure 15-17. OLOAD Output

Verifying Reaction Loads

There are two checks available in Simcenter Nastran to let you check whether the reaction loads are
correct. The first one is the summation of the total reaction loads about the reference point specified on
the PARAM,GRDPNTx entry. If PARAM,GRDPNT,x is not specified, this summation is performed about the
origin of the basic coordinate system. This operation is known as the SPCFORCE RESULTANT output and
this output is obtained if the SPCFORCE output is requested. It consists of seven lines of output per load
case. The SPCFORCE RESULTANT output for the above problem (in the listing Input File for Illustrating
Diagnostic Checks) is shown below:

15-40 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Verifying Reaction Loads

Figure 15-18. SPCFORCE RESULTANT

Since the SPCFORCE RESULTANT and OLOAD RESULTANT are summed about the same reference point,
you would expect the total applied loads and total reaction loads to be equal and opposite of each other.
As indicated by the figure OLOAD Resultant and Figure 15-18, this is indeed the case.

Similar to the applied loads, you can also request more detailed information regarding where exactly
on the structure is reacting to these applied loads (grid point locations and directions). These are also
known as boundary loads. This operation can be done with the SPCFORCE request in the Case Control
Section. The format is as follows:

SPCFORCE = All

or

SET 10 = 15, 16, 18, 25

SPCFORCE = 10

Figure 15-19. Single-Point Constraint Forces

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 15-41

© 2024 Siemens
 15. Model Verification

In this case, since there is only one reaction point (fixed at grid point 1) in the model, the SPCFORCE
output only contains one line of output per subcase (see Figure 15-19). Some guidelines to consider
when examining the reaction loads include:

1. Are they equal to the applied loads? Is the structure in equilibrium?

2. Are the boundary loads coming out in the directions and magnitudes that you are expecting? As
an example, if one of your reaction loads is to simulate a shear bolt and you are getting a tension
reaction load at that location, the chances are that you have modeled the boundary condition
incorrectly.

3. Is the load distribution over the structure what you expect? If not, you may want to go back and
check your modeling of your reaction points to make sure that they are what you intended.

4. Grid points with reaction force equal to 0.0 on all six components are suppressed from the printed
output.

1g Load

A 1g gravitation load applied separately in each of the three orthogonal directions is suggested for
model checkout. This check is accomplished with three separate subcases. This is a basic check to
determine grid points on the structure that are either loosely connected or have very small stiffnesses.
A displacement output indicates that these points are moving excessively as compared to other parts
of the model, especially if you are using a postprocessor. To use this feature, you must ensure that
your structure contains masses-either by direct masses (e.g., CONM2) or indirect masses (e.g., from the
density on the MATi entries), or nonstructural mass (e.g., a nonzero value on the NSM field on the PBAR
entry).

Output of Maximum, Minimum Grid Point-based Responses

Documenting the validation of a mathematical simulation model can be a very tedious and time
consuming process. A portion of this process usually requires the generation of reports to completely
describe the results of the performed analysis. Although a substantial amount of the reporting
requirements can be satisfied using graphical results plots, tabular data that summarizes the information
depicted in the graphics are often required. Also, during the validation process itself, many tools are
employed that assist users in evaluating the results produced by the simulation. Maximum/minimum
surveys of output quantities are one such tool provided by the MAXMIN Case Control command.

Historically, Simcenter Nastran users have been able to obtain simple surveys of grid point related results
output such as displacements, applied loads, and forces of single point constraint. The output consists of
a summary that identifies the grid point at which the maximum and minimum values occurred in each of
the six output directions. This information is provided for each subcase.

The max/min information appears as typical grid point related output. You can request a number of the
highest and lowest values for a particular component. All components of the grid related quantities are
output together with the one component that is being surveyed. For example, if 10 maximum and 10
minimum values of T3 displacements are requested, the output will also contain the T1, T2, R1, R2,

15-42 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Example: Using the MAXMIN Command

and R3 displacement values associated with the grid points identified in the T3 component survey. The
output heading clearly identifies the component being surveyed. Output can be surveyed for SUBCASEs
and SUBCOMs. You can also specify sets of points to be surveyed and coordinate system to be used for
the survey output. Only SORT1 output is available.

If output coordinate systems are specified for any of the grid points being surveyed, the user has the
option of specifying one of several coordinate system views of the max/min output. The data can be
displayed in the basic system, a local system, or even the global system. Any grid point that is output in
the max/min survey will have its output presented in both the survey coordinate frame of reference and
its original frame of reference if they are different. In this way, the user can see both the relative position
of the grid point component in the max/min output as well as the original output quantities without
having to research other portions of the displacement output produced by the analysis.

See also

• “MAXMIN” in the Simcenter Nastran Quick Reference Guide

Example: Using the MAXMIN Command

The following sample output demonstrates some features available with the MAXMIN Case Control
command.

Inspection of the Case Control Echo reveals that the MAXMIN command at line 7 requests print and
punch of max/min survey results for all available output quantities generated (via keywords PRINT,
PUNCH and ALL). It also specifies that survey results be limited to component directions T1 and T2 for
grid points defined in case control set 100. Because the command has been placed above the subcase
level, it applies to all subcases and subcoms.

The MAXMIN command at line 11 requests a survey of the forces of single point constraint output
(keyword SPCF) for component direction T1 only (keyword T1) in the basic coordinate system frame of
reference (keyword CID selects BASIC). It also requests that the highest and lowest 15 values be output
considering ALL grid points. Because it has been placed within subcase 1000, it only pertains to that
subcase.

The MAXMIN command at line 14 requests print and punch of max/min survey results for single point
forces of constraint. A vector magnitude survey is requested (keyword VMAG) for the four highest
and lowest values. The results for these types of surveys appear in the columns of output normally
containing component directions T1 and R1. Note that the specification of component directions to be
surveyed (keyword COMP) has no effect in this case. All grid points will be included in the survey.

C A S E C O N T R O L E C H O
COMMAND
COUNT
1 TITLE = EXAMPLE OUTPUT FROM THE NEW MAXMIN CASE CONTROL COMMAND
2 SUBTITLE = ONLY GRID POINT RELATED OUTPUT QUANTITIES CAN BE
PROCESSED

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 15-43

© 2024 Siemens
 15. Model Verification

3 LABEL = COMPONENT DOF SPECIFICATIONS ARE POSSIBLE

4 SPC=100
5 SET 100 = 2,51,59
6 DISP=100
7 MAXMIN (PUNCH,PRINT,ALL ,COMP=T1/T2) = 100 $
8 SPCF=ALL
9 OLOAD=ALL
10 SUBCASE 1000
11 MAXMIN (SPCF,CID=BASIC,COMP=T1,BOTH=15) = ALL
12 LOAD=1000
13 SUBCASE 2000
14 MAXMIN (PUNCH,PRINT,SPCF,VMAG=4,COMP=T1/T3) = ALL
15 LOAD=1000
16 SUBCOM 3000
17 SUBSEQ = 0.5,0.5
18 BEGIN BULK

Examination of the following actual MAXMIN survey output results reveals that the three heading
lines contain important information that describe the options used to produce the results. The first
line identifies the component direction that was searched, the type of output being searched, and
the SUBCASE/SUBCOM identification number. The second line summarizes the options specified on the
MAXMIN Case Control command used to perform the search. The third line contains column headings
for the data values that follow in the output listing. Note that the field, TYPE, identifies whether it is a
grid point (G) or scalar point (S) and the field, POINT ID, identifies the grid point or scalar ID number.
In addition, the coordinate system reference frame for the output values is identified. If data values for
a point were transformed to a reference frame different from the one originally defined in the output,
then the data values for both frames of reference are output. This is shown in the output for the first grid
point (ID = 2) in the sample output. Here it is seen that values for the forces of single point constraint are
provided for the point in the basic and CID 100 systems. This occurs because the output for the point has
been specified in CID=100, but the max/min survey was requested in the basic coordinate system. See
the MAXMIN Case Control command for further details.

15-44 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Example: Using the MAXMIN Command

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 15-45

© 2024 Siemens
 15. Model Verification

15-46 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Performing an Unconstrained Equilibrium Check

Performing an Unconstrained Equilibrium Check

If a structure is truly unconstrained, then movement at one point of this structure should cause the
whole structure to move as a rigid body. The following procedure can be used to perform this check.

1. Remove all the constraints.

2. Apply a unit enforced displacement in the x-direction at one selected grid point while constraining
the other five components at this selected grid point to zero. This grid point should be close to the
center of gravity (C.G.) of the structure, although this is not a requirement. The x-displacements

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 15-47

© 2024 Siemens
 15. Model Verification

at all the other grid points should also be equal to unity if the structure is truly unconstrained. If
the displacement at a certain grid point is not equal to unity in the x-direction, it is very likely that
it overstrained. The likely causes for this can include incorrect modeling of rigid elements, offset
beams, non-colinear CELASi, etc. See “Constraints” for details regarding enforced displacement.

3. Repeat the same procedure for the other two orthogonal directions. You can perform similar checks
for rotations; however, it is more difficult to interpret the results.

At the conclusion of this check, you should remember to put your original constraints back into your
model.

Automatic Identification of Unintentional Grounding

In addition to the unconstrained equilibrium check mentioned previously, you may also request a more
robust and more automatic grounding check of the stiffness matrix. This grounding check is also referred
to as rigid body check because the stiffness matrix is multiplied by the rigid body transformation matrix.
The grounding check identifies unintentional constraints and ill-conditioning in the stiffness matrix.

The grounding check is requested with the GROUNDCHECK Case Control command.

For example, the following command

GROUNDCHECK=YES

will request a grounding check of the g-set. The following command

GROUNDCHECK(SET=ALL)=YES

will request a grounding check of all DOF sets.

The check may be performed at any or all stages of the stiffness reduction based on the specification of
the SET keyword.

SET Keyword DOF Set Description

G g-set before single point, multipoint constraints, and rigid elements are
applied
N n-set after multipoint constraints and rigid elements are applied

15-48 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Automatic Identification of Unintentional Grounding

SET Keyword DOF Set Description

N+AUTO n-set with same as the n-set with the rows/columns in the stiffness
AUTOSPC matrix corresponding to degrees-of-freedom constrained by the
PARAM,AUTOSPC operation zeroed out
F f-set after single point, multipoint constraints, and rigid elements are
applied
A a-set after static condensation

The THRESH keyword specifies the maximum strain energy that passes the grounding check.
DATAREC=YES requests data recovery of grounding forces. The RTHRESH=r keyword prints the grounding
forces larger than r percent of the largest grounding force if DATAREC is set to YES.

User Information Message 7570 is issued by GROUNDCHECK for each DOF set requested. The strain
energy is computed in each direction of the rigid body motion and by default, if the strain energy
exceeds the tolerance, then “FAIL” is printed out for that directory. The tolerance is set by dividing the
largest stiffness term by 1.E10. The THRESH keyword can be used to further control the amount of
output (see Appendix A). Possible reasons for failure are also printed after the strain energies. Here is an
example for the g-set that indicates the stiffness matrix passes the grounding check in all six rigid body
directions:

RESULTS OF RIGID BODY CHECKS OF MATRIX KGG (G-SET) FOLLOW:

PRINT RESULTS IN ALL SIX DIRECTIONS AGAINST THE LIMIT OF
2.725275E-04
DIRECTION STRAIN ENERGY PASS/FAIL
--------- ------------- ---------
1 1.484295E-09 PASS
2 2.182787E-10 PASS
3 1.637090E-11 PASS
4 1.619810E-10 PASS
5 2.727802E-10 PASS
6 1.054841E-07 PASS
SOME POSSIBLE REASONS MAY LEAD TO THE FAILURE:
1. CELASI ELEMENTS CONNECTING TO ONLY ONE GRID POINT;
2. CELASI ELEMENTS CONNECTING TO NON-COINCIDENT POINTS;
3. CELASI ELEMENTS CONNECTING TO NON-COLINEAR DOF;
4. IMPROPERLY DEFINED DMIG MATRICES;

Here is an example for the f-set that indicates the stiffness matrix fails the grounding check in all six rigid
body directions:

*** USER INFORMATION MESSAGE 7570 (GPWG1D)

RESULTS OF RIGID BODY CHECKS OF MATRIX KFF (F-SET) FOLLOW:
PRINT RESULTS IN ALL SIX DIRECTIONS AGAINST THE LIMIT OF
1.464858E-04
DIRECTION STRAIN ENERGY PASS/FAIL

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 15-49

© 2024 Siemens
 15. Model Verification

--------- ------------- ---------

1 2.564102E+05 FAIL
2 7.326008E+05 FAIL
3 2.237437E+03 FAIL
4 6.057062E+02 FAIL
5 4.015165E+03 FAIL
6 8.747863E+04 FAIL
SOME POSSIBLE REASONS MAY LEAD TO THE FAILURE:
1. CONSTRAINTS WHICH PREVENT RIGID-BODY MOTION.

If the DATAREC keyword is specified and the rigid body check fails then data recovery will be performed
to compute and print the grounding forces to make it easy to locate the source of the failure. By default,
only those grounding forces larger than 10 percent of the largest grounding force will be printed (see
RTHRESH keyword). Here is an example of grounding forces created by moving the model in rigid body
direction 1 associated with the f-set failure shown above:

DIRECTION 1
G R O U N D C H E C K F O R C E S ( F - S E T )
POINT ID. TYPE T1 T2 T3
R1 R2 R3
3 G 2.564102E+05 -2.884615E+05 .0 .0
.0 .0
4 G 2.564102E+05 2.884615E+05 .0 .0
.0 .0

Thermal Equilibrium Check

The following procedure can also be used to check for unconstrained thermal expansion if the structure
is going to be subjected to thermal loads.

1. Remove the actual boundary conditions and apply a set of statically determinate constraints.
Typically this procedure is done by constraining all six DOFs at a single grid point. If a single grid
point is used, make sure this single grid point contains six degrees of freedom. A single grid point
of a model consisting of all solid elements, for example, cannot satisfy this requirement since each
grid point of a solid element contains only three degrees of freedom.

2. Change all the thermal coefficients of expansion to a single value.

3. Apply a uniform ΔT to the structure.

If the model is “clean,” then the structure should be strain free; in other words, there should be no
reaction loads, element forces, or stresses. If this is not the case, then you may want to investigate
around the vicinity where the element forces or stresses are nonzero. Incorrect modeling of rigid
elements or offsets is a common cause of these types of errors. Once you are satisfied with your model,
remember to change the boundary condition, thermal coefficients of expansion, and ΔT back to their
original values.

15-50 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Grid Point Force Balance

Grid Point Force Balance

Frequently, several elements are connected to a single grid point. When this happens, you may want to
know how the loads are distributed among these elements at the common grid point. In other words,
you want to understand the load path. Is the load path correct? These questions can be answered by
using the GPFORCE (grid point force) option in Simcenter Nastran.

Grid point force balance is computed only from linear stiffness elements, the sum of applied loads and
thermal loads, and SPC forces. Effects not accounted for include those from mass elements in dynamic
analysis (inertia loads), rigid elements and MPCs, general elements, DMIG entries, and boundary loads
from upstream superelements. These effects may lead to an apparent lack of equilibrium at the grid
point level. Table 15-5 summarizes those effects that are considered and those effects that are ignored
in the calculation of grid point forces in the global coordinate system.

Table. Contributions Used for the Grid Point Force Balance

Contributions Included Contributions Ignored

Applied Loads Boundary Loads from Upstream Superelements
SPC Forces GENEL Forces
Element Elastic Forces DMIG and DMI Forces
Thermal Loads

The model in Figure 15-20 illustrates this feature. This model consists of nine CQUAD4 and six CBAR
elements and is subjected to vertical loads applied at the free end. The input file is shown in Listing
15-6.

EXAMPLE

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 15-51

© 2024 Siemens
 15. Model Verification

Figure 15-20. Clamped Plate with Stiffeners

$ FILENAME - q4bargpf.dat
$
SOL 101
TIME 5
CEND
TITLE = ILLUSTRATE USE OF GPFORCE
SUBTITLE = INCORRECT PROPERTIES
DISP = ALL
GPFORCE = ALL

15-52 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Grid Point Force Balance

STRESS(CORNER) = ALL
SPC = 1
LOAD = 1
PARAM AUTOSPC YES
PARAM POST 0
$
$ THIS SECTION CONTAINS BULK DATA
$
GRID 1 0.0 0.0 0.0
GRID 2 10. 0.0 0.0
GRID 3 20. 0.0 0.0
GRID 4 30. 0.0 0.0
GRID 5 0.0 10. 0.0
GRID 6 10. 10. 0.0
GRID 7 20. 10. 0.0
GRID 8 30. 10. 0.0
GRID 9 0.0 20. 0.0
GRID 10 10. 20. 0.0
GRID 11 20. 20. 0.0
GRID 12 30. 20. 0.0
GRID 13 0.0 30. 0.0
GRID 14 10. 30. 0.0
GRID 15 20. 30. 0.0
GRID 16 30. 30. 0.0
$
CBAR 101 100 5 6 0.0 0.0
1.
+C101 0. 0. .55 0. 0.
.55
CBAR 102 100 6 7 0.0 0.0
1.
+C102 0. 0. .55 0. 0.
.55
CBAR 103 100 7 8 0.0 0.0
+C103 0. 0. .55 0. 0.
.55
CBAR 104 200 9 10 0.0 0.0
+C104 0. 0. .55 0. 0.
.55
CBAR 105 200 10 11 0.0 0.0
+C105 0. 0. .55 0. 0.
.55
CBAR 106 200 11 12 0.0 0.0
+C106 0. 0. .55 0. 0.
.55
$

Listing 15-6. Input File for the Demonstration of Grid Point Force Balance (Continued)

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 15-53

© 2024 Siemens
 15. Model Verification

CQUAD4 1 1 1 2 6 5
CQUAD4 2 1 2 3 7 6
CQUAD4 3 1 3 4 8 7
CQUAD4 4 1 5 6 10 9
CQUAD4 5 1 6 7 11 10
CQUAD4 6 1 7 8 12 11
CQUAD4 7 1 9 10 14 13
CQUAD4 8 1 10 11 15 14
CQUAD4 9 1 11 12 16 15
$
$ THIS SECTION CONTAINS THE LOADS and CONSTRAINTS
$
$
FORCE 1 4 0 50. 0. 0. -1.
FORCE 1 8 0 100. 0. 0. -1.
FORCE 1 12 0 100. 0. 0. -1.
FORCE 1 16 0 50. 0. 0. -1.
$
SPC1 1 123456 1 5 9 13
$
$ THIS SECTION CONTAINS THE PROPERTY AND MATERIAL BULK DATA ENTRIES
$
PSHELL 1 1 .1 1
$
PBAR 100 1 .50 .0417 .01042 .0286
+P100 .5 .25 -.5 .25 -.5 -.25 .5 -.25
PBAR 200 1 .01042 .0286
+P200 .5 .25 -.5 .25 -.5 -.25 .5 -.25
$
$ CORRECT PBAR ENTRY WITH A AND I1
$
$PBAR 200 1 .50 .0417 .01042 .0286
$
$
MAT1 1 1.+7 .3
ENDDATA

Listing 15-6. Input File for the Demonstration of Grid Point Force Balance

If everything is done correctly, expect the stress contour plot to look something like Figure 15-21(a).
However, say that you made a modeling mistake somewhere, and the stress contour plot is coming out
as shown in Figure 15-21(b). You know that something is wrong because the stress contour plot is
not symmetrical. Since the structure, boundary conditions, and applied loads are all symmetrical, you
also expect the stress contour plot to be symmetrical. By inspecting the grid point force balance output
(Figure 15-22), you notice that the stiffeners (CBAR 104, 105, and 106) are not picking up any axial,
vertical, or out-of-plane bending loads at grid points 9, 10, 11, and 12.

15-54 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Grid Point Force Balance

A review of the input file explains the reason for this result. The area (A) and the out-of-plane bending
moment of inertia (I1) were "accidentally" left out for CBAR elements 104, 105, and 106 (PBAR 200).
Hence, all the axial and out-of-plane bending loads at grid points 9, 10, 11, and 12 were picked up
completely by the surrounding CQUAD4 elements. By correcting the PBAR element properties, note that
CBAR elements 104, 105, and 106 now pick up the correct loads as shown in Figure 15-23.

Figure 15-21. Stress Contour Plots of a Cantilever Plate

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 15-55

© 2024 Siemens
 15. Model Verification

Figure 15-22. Grid Point Force Balance Output for a Model with Incorrect CBAR Properties

Figure 15-23. Grid Point Force Balance Output for a Model with Correct CBAR Properties

15-56 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Stress Error Estimators

Both Figure 15-22 and Figure 15-23 are reduced versions of the full output. Only the grid point force
balance output as related to grid points 1, 9, 10, 11, 12, and 16 are shown in these two figures for
clarity. Notice that for grid points 1 and 9, in addition to the elastic element forces, the SPC forces are
also printed. Similarly, for grid points 12 and 16, the applied loads are also printed. As you can see, the
grid point force balance feature can provide you with a tremendous insight regarding the load path of
your structure.

Stress Error Estimators

Reviewing Grid Point Stresses is devoted to a discussion of the averaging procedures utilized to provide
meaningful stresses at the grid points of finite element models in Simcenter Nastran. However, when
you look at a stress distribution, frequently you may not be sure of whether further model refinement is
necessary. This section focuses upon the description of error estimators that help you identify regions of
your model that may require refinement.

The essence of the averaging procedures used to obtain grid point stress data are to:

1. Convert these local element stress components into a common coordinate system defined by you.

2. Average several values of each stress component to obtain a unique value of the stress component
that is to be associated with the grid point in question.

3. Compute the stress invariants at the grid points from the stress components at the grid points.

In the general case, the stress components are σx, σy, σz, τxy, τxz, and τyz.

For discussion purposes, the averaging process used to compute the stress components at the grid
points can be represented in the form

Equation 15-3.

where:

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 15-57

© 2024 Siemens
 15. Model Verification

σg = the weighted mean value of the stress component computed at the grid point.
σei = the value of the stress component in the i-th element i = 1, 2, ..., Ne connected to the grid
point. σei is in the same coordinate system as σg.
Wi = the weighting factor assigned to the i-th element. The sum of the values of must equal 1. This
requirement assures that all computed statistics are unbiased. The attribute of being unbiased
implies that the variance is equal to the mean square error. Equal weighting, i.e., Wi = 1/Ne, is
assumed in Simcenter Nastran.

An estimate of the error in a particular component of stress at a grid point can then be computed by
assuming that the values of the corresponding stress components computed by Simcenter Nastran for
the elements in the neighborhood of the grid point are data points with uncorrelated random errors. It
then follows that an estimate of the probable error in the stress component δg at the grid point is

Equation 15-4.

where δei = σei − σg. Thus, the probable error δg is the root mean square error in δei divided by .

It should be noted that the root mean square error is a reasonable measure of precision in many practical
cases, but it is easy to provide examples in which this value is a poor measure of the concentration
of the distribution about the mean. This subject is discussed further in the section "Discussion of Error
Measures." Eq. 15-4 is assumed to provide an approximate error estimator for the grid point stress data
output by Simcenter Nastran.

This stress error assessment is often referred to as stress discontinuity, and it is evaluated in two different
ways in Simcenter Nastran: as grid point and element stress discontinuities.

Grid Point Stress Discontinuities

To illustrate the computations described in the previous section, consider the following portion of a
finite element model consisting of CQUAD4 elements. For clarity, only a portion of the model is plotted.
These elements lie on a surface (defined as surface 91 in this case).

15-58 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Grid Point Stress Discontinuities

Figure 15-24. Partial Model for the Stress Discontinuity Calculations

The following data is extracted from the Simcenter Nastran output. Table 15-7 and Table 15-8 contain
partial listings of the element stresses and grid point stresses for the above model. The grid point
stresses are obtained following the procedures described in Reviewing Grid Point Stresses.

Table. Element Stresses for CQUAD4

Stresses in the Element Coordinate System
Element ID NORMAL-X NORMAL-Y SHEAR-XY
19 –4.006485E+03 –8.949428E-09 –1.045506E+02
20 –3.820136E+03 2.255547E-09 –1.045506E+02
.

.
59 –1.335495E+03 8.658390E-09 –2.287827E+02
60 –1.273379E+03 –1.818989E-09 –2.287827E+02

Table. Grid Point Stresses for Surface 91

Stresses in the Element System

Grid ID NORMAL-X NORMAL-Y SHEAR-XY
20 –5.218E+03 –6.730E–09 –4.243E+01
21 –4.969E+03 1.692E–09 –4.243E+01
.
.
.
60 –2.733E+03 –3.274E–10 –1.667E+02
61 –2.609E+03 –3.638E–11 –1.667E+02

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 15-59

© 2024 Siemens
 15. Model Verification

Stresses in the Element System

Grid ID NORMAL-X NORMAL-Y SHEAR-XY
62 –2.485E+03 1.273E–10 –1.667E+02
63 –2.360E+03 –5.457E–11 –1.667E+02
.
.
.
102 –6.104E–05 6.803E-09 –2.909E+02
103 –9.155E–05 –1.437E–09 –2.909E+02

The probable error for stresses at both grid points 20 and 61 can be computed using the following
equation:

Equation 15-5.

For more information, see Stress Error Estimators.

The probable error δg is expected to provide some measure of the error in the magnitude of the grid
point stress values. You must first compute the values δei = σei − σg for each stress component and
then compute the probable error, δg. This calculation is done for both grid points 20 and 61. The
symbols “GID” and “EID”, which represent grid point ID and element ID, respectively, are used extensively
throughout this section. Consider an example calculation using the NORMAL-X stress component.

NORMAL-X Stress

GID = 20
EID σei σg δei
19 –4.006485E+03 –5.218E+03 1.212E+03
20 –3.820136E+03 –5.218E+03 1.398E+03

Using the error estimate equation presented in Stress Error Estimators,

15-60 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 NORMAL-X Stress

GID = 61
EID σei σg δei
19 –4.006485E+03 –2.609E+03 -1.397E+03
20 –3.820136E+03 –2.609E+03 -1.211E+03
59 –1.335495E+03 –2.609E+03 1.274E+03
60 –1.273379E+03 –2.609E+03 1.336E+03

Again, using the error estimate equation presented in Stress Error Estimators,

The other stress components (σy and τxy) can then be obtained in a similar manner.

This same finite element model is then remodeled using different element types—CQUAD8, CHEXA(8)
as well as CQUAD4 elements with the corner output option. The values of δg for each of the three
stress components are then calculated at two locations: x = 28.5 and y = 0.0, as well as x = 28.5 and
y = 1.5. Note that these two locations correspond to GID 20 and GID 61, respectively, for the above
CQUAD4 model. This process is then repeated for each of the above element types. The results (δg and
ERROR ESTIMATE) from Simcenter Nastran are summarized in Table 15-11. Note that these values may
be slightly different than the hand-calculated values since the computer provides more precision than
the hand calculation.

The probable error δg provides an estimated error for each of the three stress components at each of
the two grid point locations considered. It is probably more useful and more desirable to combine these
three estimated errors into a single representative error measure at each grid point. An approximate root
mean square value of the three estimated errors for each of the three stress components offers such a
representative error measure. That is,

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 15-61

© 2024 Siemens
 15. Model Verification

Equation 15-6.

where Nc, which is the number of stress components, is equal to three for plate elements and six for
solid elements.

Table. The Values of δ g and ERROR ESTIMATE for Different Element Types(X = 28.5, Y = 0.0)

X = 28.5 Y = 1.5
ERROR
Element Type NORMAL-X (δg) NORMAL-Y (δg) SHEAR-XY (δg) ESTIMATE
CQUAD4 9.247E+02 4.628E–09 4.392E+01 5.345E+02
CQUAD8 0.000E+00 1.204E–09 0.000E+00 9.840E–10
CHEXA (8) 8.839E+01 2.841E+01 0.000E+00 5.729E+01
CQUAD4 (CORNER) 9.653E+01 2.896E+01 3.548E+02 2.129E+02

Table. The Values of δ g and ERROR ESTIMATE for Different Element Types (X = 28.5, Y = 1.5)

X = 28.5 Y = 1.5
ERROR
Element Type NORMAL-X (δg) NORMAL-Y (δg) SHEAR-XY (δg) ESTIMATE
CQUAD4 6.531E+02 3.196E-09 3.106E+01 3.775E+02
CQUAD8 0.000E+00 3.634E-09 0.000E+00 2.098E-09
CHEXA (8) 3.125E+01 1.497E+01 3.125E+01 2.148E+01
CQUAD4 (CORNER) 7.298E+01 2.153E+02 2.532E+02 1.965E+02

The following remarks should be noted in regard to Table 15-11 and Table 15-12.

1. By default, the stress components for the CQUAD4 elements are only output at the element
centroids, while the stress components for the CQUAD8, and CHEXA(8) elements are output at the
element centroids as well as the element vertices.

2. If vertex stresses are also desired for the CQUAD4 elements, then use the STRESS(CORNER) = x
option in the Case Control Section. This output corresponds to the last rows of Table 15-11 and
Table 15-12.

15-62 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Element Stress Discontinuities

3. The calculation of δg for the CQUAD8 and CHEXA(8) elements involves grid point and element
vertex stresses that are defined at the same points.

4. The calculation of δg for the CQUAD4 elements involve grid point stresses and element stresses
provided at a different geometric location if the default center option is requested. They are
evaluated at the same location when using the corner option.

5. The ERROR ESTIMATE for the CHEXA(8) also includes the contribution of σz, σxz, and σyz. These
three values are not listed in Table 15-11 and Table 15-12. The procedure is identical to that of the
plate element with the exception that it includes six instead of three stress components.

Table 15-11 and Table 15-12 indicate that the smaller values of δg and the ERROR ESTIMATE are
obtained from the models composed of elements for which vertex stresses are computed by Simcenter
Nastran. A detailed discussion of the error estimator used by Simcenter Nastran is presented later in the
section “Discussion of Error Measures.”

In addition to values of δg for each stress component and the ERROR ESTIMATE of Eq. 15-6 Simcenter
Nastran output also contains values of δg for each of the stress invariants. These quantities are output
for each grid point in the defined SURFACE (plate elements) or VOLUME (solid elements).

Element Stress Discontinuities

Other statistical error estimates may be generated by associating the estimates with the elements rather
than with the grid points. For example, root mean square errors for each stress component of an
element may be computed from the values of δei that are computed for each of the Ng connected vertex
grid points where δei = (σe − σgi). This computation is done for each stress component for all of the
elements of interest.

Next is an example calculating the NORMAL-X stress components for elements 20 and 60 in the figure
Partial Model for the Stress Discontinuity Calculations.

NORMAL-X Stress

EID=20
GID σe σgi δei
20 –3.820136E+03 –5.218E+03 1.398E+03
21 –3.820136E+03 –4.969E+03 1.149E+03
62 –3.820136E+03 –2.485E+03 –1.335E+03
61 –3.820136E+03 –2.609E+03 –1.211E+03

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 15-63

© 2024 Siemens
 15. Model Verification

Equation 15-7.

EID = 60
GID σe σgi δei
61 -1.273379E+03 -2.609E+03 1.336E+03
62 -1.273379E+03 -2.485E+03 1.212E+03
103 -1.273379E+03 -9.155E-05 -1.273E+03
102 -1.273379E+03 -6.104E-05 -1.273E+03

The other two components (σy and τxy) can be calculated in a similar manner.

The above data provides an estimated error for each of the three stress components for each of the
two elements considered. As discussed in the previous section, it is desirable to combine these three
estimated errors into a single representative error estimate for each element. The root mean square
value of the three error estimates for each of the three stress components offers such a representative
error measure. In other words,

Equation 15-8.

15-64 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Discussion of Error Measures

(The results (δe and ERROR ESTIMATE) from Simcenter Nastran for this model, using CQUAD4, CQUAD8,
and CQUAD4 with the corner option, are summarized in Table 15-15 and Table 15-16.

Table. The Values of δ e and ERROR ESTIMATE for Different Plate Element Types

Location at EID = 20
ERROR
Element Type NORMAL-X (δe) NORMAL-Y (δe) SHEAR-XY (δe) ESTIMATE
CQUAD4 1.277E+03 4.757E–09 6.212E+01 7.382E+02
CQUAD8 0.000E+00 4.014E–09 0.000E+00 2.318E-09
CQUAD4 (CORNER) 1.407E+02 2.990E+02 4.932E+02 3.428E+02

Table. The Values of δ e and ERROR ESTIMATE for Different Plate Element Types

Location at EID = 60
ERROR
Element Type NORMAL-X (δe) NORMAL-Y (δe) SHEAR-XY (δe) ESTIMATE
CQUAD4 1.274E+03 4.520E–09 6.212E+01 7.365E+02
CQUAD8 5.038E-10 3.362E–09 0.000E+00 2.020E-09
CQUAD4 (CORNER) 1.003E+02 2.969E+02 4.915E+02 3.366E+02

The CHEXA(8) element is not included in the above tables; however, the calculations are similar with the
exception that six rather than three components of stress need to be considered. In addition to values
of δe for each stress component and the ERROR ESTIMATE of Eq. 15-8, Simcenter Nastran output also
contains values of for each of the stress invariants. For plate elements, all of these quantities are output
at the neutral plane and at fiber distances Z1 and Z2 for each element in the defined SURFACE.

Discussion of Error Measures

An examination of the tables from NORMAL-X Stress through The Values of δ e and ERROR ESTIMATE
for Different Plate Element Types clearly shows that the error estimates for CQUAD4 elements without
vertex stress data are substantially larger than the corresponding error estimates for those elements that
provide vertex stress data as a default. In view of the fact that the finite model under discussion was
constructed with a rather high mesh density to produce results that agreed very closely with theory, you
may perhaps be surprised by the large values of the error estimates for those elements that only provide
stress data at the element centroid.

These seemingly large values of error estimators for CQUAD4 elements are not uncommon occurrences
in statistical error measures. For example, the mean and variance have direct analogies in engineering
mechanics. The mean is analogous to the centroid of a body and is thus a measure of where the mass
is centered. The variance is the second moment about the mean, and it tends to be small if the majority
of the mass is concentrated about the centroid. As the mass is dispersed further from the centroid, the
moment of inertia tends to increase. If the mass is concentrated at the centroid (as it might be in some
idealized lumped mass models), the moment of inertia becomes zero.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 15-65

© 2024 Siemens
 15. Model Verification

Thus, the error estimates for CQUAD4 elements tend toward larger values because the data used to
compute the estimators is dispersed relatively far from the mean. For those elements that provide
stress data at vertices, the estimators tend toward smaller values because the stress data tends to be
concentrated about the mean in well-designed finite element models.

Generally, you should design a finite element mesh for static analysis so that all important stress
gradients are adequately represented. If the mesh is not of sufficient detail, the stress data whether
available at element vertices or at element centroids results in the relatively large values of the error
estimators discussed here.

You are cautioned that very inaccurate values of these error estimators may occur at the edges of
defined SURFACEs and on the faces of defined VOLUMEs. In summary, the error estimators under
discussion can, in some cases, be highly inaccurate. Nevertheless, these data are quite useful when
interpreted properly.

User Interface

The element and grid point stress discontinuity output can only be obtained if the grid point stress
output is requested via the STRFIELD Case Control Command. The STRFIELD Case Control command is
also used for graphical postprocessing of grid point stress, element stress discontinuity, and grid point
stress discontinuity. The STRFIELD command, however, does not provide printed output for the grid
point stress. The GPSTRESS Case Control command is needed if printed output for the grid point stress
is desired. Similar to the GPSTRESS command, the STRFIELD command also requires that you define
all applicable SURFACEs and VOLUMEs in the OUTPUT(POST) portion of the Case Control Section. The
element stresses must also be requested for those elements that lie on the SURFACEs and/or VOLUMEs.
For a more detailed description of GPSTRESS, refer to Reviewing Grid Point Stresses.

Two other Case Control commands are needed–GPSDCON for grid point stress discontinuity and
ELSDCON for element stress discontinuity. Note that you can also request grid point stress discontinuity
without requesting element stress discontinuity, and vice versa, by omitting the appropriate Case
Control command (ELSDON or GPSDCON).

The following remarks should be noted relative to the use of the GPSDCON and/or ELSDCON commands.

1. The GPSDCON and ELSDCON commands are supported only in Solution Sequence101.

2. The GPSDCON and ELSDCON commands may be placed above the SUBCASE level or in individual
SUBCASEs and/or SUBCOMs. The use of these commands above the SUBCASE level causes the
stress discontinuity data to be output for all SUBCASEs in the Simcenter Nastran input file. If
you wish to restrict stress discontinuity output to specific SUBCASEs, GPSDCON and/or ELSDCON
commands should appear only under these specific SUBCASEs.

3. You are cautioned that these commands can produce a substantial amount of data. Therefore, you
should be judicious in their use.

4. Stress continuity data is not provided when both plate and solid elements are connected to a grid
point that is involved in stress discontinuity calculations.

15-66 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 User Interface

The following is a listing of the input file used to generate the results for the CQUAD4 element
model shown in NORMAL-X Stress. The example problem below contains the full list of Case Control
commands required to obtain Grid Point/Element Stress Discontinuity output.

$
$ FILENAME - q4sdcon.dat
$
ID STRESS DISCON
TIME 25
SOL 101
CEND
TITLE = CANTILEVER BEAM WITH PLATES
$
SPC=1
LOAD=1000
$
$
$ ELEMENT OUTPUT REQUIRED
$
SET 22 = 19,20,21,59,60,61
$
ELSTRESS = 22
$
$ FOLLOWING REFERENCES SURFACE 91
$
SET 21 = 91
GPSTRESS=21
STRFIELD=21
$
$ FOLLOWING ARE FOR STRESS DISCONTINUITIES
$
GPSDCON=21 $ FOR GRID POINT DISCONTINUITIES
ELSDCON=21 $ FOR ELEMENT DISCONTINUITIES
$
$ FOLLOWING REQUIRED FOR GRID POINT STRESS
$
OUTPUT(POST)
SET 95 = 19,20,21,59,60,61
$
SURFACE 91 SET 95 NORMAL Z SYSTEM ELEMENT
$
BEGIN BULK
$
$ BRING IN THE REST OF THE BULK DATA FILE
$
INCLUDE 'q4sdcon.blk'
$

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 15-67

© 2024 Siemens
 15. Model Verification

$
ENDDATA
Listing 15-7. Input File for the Error Estimators for CQUAD4

Postprocessor Checks
Once you have completed your Simcenter Nastran job, you must interpret the results. For a large model,
it is impractical to go through every page of your output to look at the results. Graphical postprocessors
can help you more easily examine and interpret large amounts of output data. This section describes
some of the basic output quantities that you may want to consider when using a postprocessor.

Examining Displacements

When you post process your model, you should create a displacement plot for each of your loading
conditions. If there are abrupt changes in the displacements at a certain region, you may want to
zoom in to inspect that local region and see if they are justifiable. One possible cause can be improper
modeling at that location or certain elements not being connected.

Examining Stress Contours

Stress contour plots are calculated using a stress averaging technique of neighboring elements. Once
these average stresses are calculated, the postprocessor assigns different colors to these stresses based
on the stress ranges. These stresses can then be plotted as stress contour plots with various colors
designating different stress ranges. This feature is probably the most frequently used postprocessing
option. When used properly, it can help you quickly identify high stress regions and provide you with
insight regarding your model.

When you create stress contour plots of component stresses, you should ensure that the component
that you are plotting is in a consistent direction for all the elements that are included in your plot. The
default component direction may be a function of element type. Some postprocessing packages may
also transform them into a common coordinate system. The following example problem in Figure 15-25
illustrates the importance of consistent direction. The stresses in a CQUAD4 element are in terms of the
element coordinate system.

15-68 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Examining Stress Contours

Figure 15-25. CQUAD4 Model

The complete input file used for this problem is sown in Listing 15-8.

$
$ FILENAME - (q4consc.dat)
$
SOL 101
TIME 5
CEND
TITLE = CONSISTENT CQUAD4 CONNECTIVITIES
SPC =1
LOAD =1
STRESS = ALL
BEGIN BULK
$
$
PARAM,POST,0
$
$ THIS SECTION CONTAINS GRID POINT LOCATIONS
$
GRID 1 0.0 0.0 0.0
GRID 2 10. 0.0 0.0
GRID 3 20. 0.0 0.0
GRID 4 0.0 10. 0.0
GRID 5 10. 10. 0.0
GRID 6 20. 10. 0.0
$
$ THIS SECTION CONTAINS ELEMENT CONNECTIVITIES
$
CQUAD4 1 10 1 2 5 4

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 15-69

© 2024 Siemens
 15. Model Verification

CQUAD4 2 10 2 3 6 5
$
$CQUAD4 2 10 3 6 5 2
$
$ LOADS
$
FORCE 1 3 100. 1.
FORCE 1 6 100. 1.
$
$ CONSTRAINTS
$
SPC 1 1 12345 0.0
SPC 1 4 1345 0.0
$
$ SECTION AND MATERIAL PROPERTIES
$
PSHELL 10 10 .1 10
MAT1 10 1.+7 .32
$
ENDDATA

Listing 15-8. Input File for CQUAD4 Stress Contour Plotswith Consistent Connectivity

In this example, the connectivities for both CQUAD4 elements are described in a consistent manner as
shown in Figure 15-26. Due to this consistency, the corresponding component stresses (e.g., ) point in
the same direction for both elements. Note that for this model is equal to zero for both elements.

Figure 15-26. Consistent Connectivity

15-70 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Examining Stress Contours

If you perform a stress contour on this model using σx, the contour plot for σx across the two elements
can be calculated as follows:

The contour plot for σvon across the two elements can be calculated as follows:

On the other hand, if the two elements are not connected in a consistent matter as shown in Figure
15-27, then the corresponding component stresses (e.g. σx, ) do not necessarily point in the same
directions for both elements. As shown in Figure 15-27, the σx component stresses point in the
directions of the x- and y-axis for elements 1 and 2, respectively.

Figure 15-27. Inconsistent Connectivity

Once again, if you perform a stress contour on this model using , the value used in the contour plot for
across the two elements in this case is as follows:

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 15-71

© 2024 Siemens
 15. Model Verification

As you can see, the σx value used in the contour plot for this case is basically meaningless and could lead
you to an erroneous conclusion.

The von Mises stresses used for the contour plot can be calculated as follows:

As you can see, σvon remains the same since it is an invariant quantity and is independent of the order of
connectivity. Note that the input file used for Figure 15-27 is identical to Listing 15-8 with the exception
of the shaded CQUAD4 entry. In other words, replace the following CQUAD4 entry:

1 2 3 4 5 6 7 8 9 10
CQUAD4 2 10 2 3 6 5

with the following:

1 2 3 4 5 6 7 8 9 10
CQUAD4 2 10 3 6 5 2

and the two input files are then identical.

As you have probably observed by now, using contour plots of component stresses can lead to
misleading results unless they are transformed into a common coordinate system. Therefore, in most
instances, it is generally easier and more meaningful to look at the invariant stress quantities, such as
von Mises stresses, instead of the component stresses. However, there are situations where you may
want to look at component stresses. An example is a cylindrical model whereby you are interested in
looking at the hoop stress. In this case, you should be aware of the consequences if your model is not
created in a consistent manner. Alternate invariant stress quantities are the principal stresses.

See also

• “CQUAD4 and CTRIA3 Elements” in the Simcenter Nastran Element Library

Creating Grid Point Stress and Stress Discontinuity Plots

The grid point stress feature enables you to obtain stress output at the grid points instead of the
elements. If the contour plot using grid point stress is substantially different from the contour plot using
element stress, it is probably an indication that the model needs to be refined or that you may have
inconsistent element orientations as described in the previous section.

Stress discontinuity is a measure of the probable error of the stresses across the grid points or elements.
This type of plot assists you in the identification of regions of your model that may require refinement.
This feature is discussed in detail in “Stress Error Estimators.”

15-72 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Viewing Element Strain Energy Plots

See also

• “Reviewing Grid Point Stresses”

Viewing Element Strain Energy Plots

The element strain energy plot is an excellent tool for identifying areas that are most effective in
influencing your design. For a large model, this output request can generate a large amount of printed
output. The plot option can be used to generate plots without generating printed output. This feature is
discussed in detail in “Strain Energy Output.”

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 15-73

© 2024 Siemens
 15. Model Verification

15-74 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 16. Restarts
Overview of Restarts
A restart is a logical way of continuing from a previous run without having to start from the beginning.
The savings can be in terms of both time and money. In the case of a static analysis, the most expensive
and time consuming part of the run is the decomposition of the stiffness matrix, especially with large
models. Now suppose after completing the original run, you want to obtain additional output (e.g.,
stresses, displacements, etc.) or add more load cases. You can always start from the beginning and
redo the whole problem, or you can perform a restart at a fraction of the time and cost. In the case
of additional output requests and additional load conditions, the decomposition of the stiffness matrix,
which was performed in the previous run, is not redone if you use the restart feature.

However, a restart may not always be the best option compared to rerunning the job. Determining
whether to save the database for restarting or rerun the analysis is based on:

• The amount of available disk space

• Computer speed

• The types of restarts

• The size of your model

For conventional static analysis runs (i.e, those not involving superelements), if a restart involves
model changes, the savings is probably minimal. However, in the case of additional output requests
or additional load cases, the savings can be substantial as indicated by the table Bulk Data Section.

In addition to needing to rerun a job, you may also want to use the restart capabilities of Simcenter
Nastran if you need to access data from an old file and the input file is no longer available. If you still
have the database archive for that solve, the following sample run (Listing 16-1) shows how you can
use restart to either print or punch a copy of the Bulk Data from the database that you can subsequently
use to create a new input file.

$ FILE - bulklist.dat
$
restart
assign master='archive1.MASTER'
sol 101
time 10
cend
title = run to get listing of bulk data
echo = sort
$

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 16-1

© 2024 Siemens
 16. Restarts

$ replace echo=sort with echo=punch if a

$ punched output is desired instead of a listing
$
param,serst,semi
begin bulk
$
enddata

Listing 16-1. Input File for Printing/Punching a Listing of Bulk Data File from the Database

See also

• Restart Procedures in the Simcenter Nastran User's Guide

Types of Restarts
Two types of restarts are available in Simcenter Nastran:

• Automatic

• Manual

With automatic restarts, Simcenter Nastran automatically determines which operations need to be
performed during the current run based on the changes that are made in your input file. Automatic
restarts are easy to use, and they can potentially save you a substantial amount of time—both CPU and
elapsed time—by not having to resolve a problem from the beginning.

With manual restarts, in addition to providing the necessary input file changes, you are also responsible
for telling Simcenter Nastran which major operations need to be performed. Manual restarts are more
difficult to use and are more prone to user error.

Restarts can use the same solution sequence or a different solution sequence. Restarting from a statics
run and requesting stress output is an example of a restart using the same solution sequence. Restarting
from a statics run into a normal modes run is an example of a restart across solution sequences.

Within automatic restart, there are two different options:

• A read-write restart

• A read-only restart

Each of these is described below.

16-2 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Understanding Restart Terminology

Understanding Restart Terminology

When you are performing restart analyses in Simcenter Nastran, your analyses are categorized into cold
start and restart runs.

Cold Start Runs

Your initial run is known as a “cold start” run and is identical to a regular Simcenter Nastran job, except
that you must save your database. In addition to your normal output files (e.g., the .F06 file), Simcenter
Nastran creates four database files as a result of this run. The naming convention for the filenames is
machine dependent-you should refer to Simcenter Nastran Installation and Operations Guide for the
exact syntax for your machine. For a Simcenter Nastran input file called “stat1.dat”, the following sample
submittal command can be used:

nastran stat1 scr=no

In this case, nastran is the name of the shell script for executing Simcenter Nastran. The “scr=no” option
in the above submittal command is not required if you have not modified the default value of the
submittal command as provided on the delivery media. By default, the following two database files are
created as a result of the above command:

• stat1.DBALL

• stat1.MASTER

Restart Runs

Once you have performed the cold start run and saved the database, subsequent runs are referred to as
“restart” runs.

Read-write Restarts
The following sections describe the changes you'll need to make to your input file to restart the analysis
for a read-write restart.

NASTRAN Statement Section

This section is normally the same as your cold start run. You should not change the BUFFSIZE in a restart
run.

File Management Section (FMS)

This section tells Simcenter Nastran that you are performing a restart run. In the FMS Section, only
two statements are needed: the “restart” statement to tell Simcenter Nastran that this current run is a

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 16-3

© 2024 Siemens
 16. Restarts

restart run and the “assign” statement that tells Simcenter Nastran the name of the database you want
to attach.

The RESTART statement is required in any restart run. The general format for the RESTART statement is:

RESTART VERSION=a,b

where, “a” is the version from which you restart (the default value for “a” is LAST), and “b” indicates
whether version “a” will be kept (KEEP) or deleted (NOKEEP) at the end of the run. The default value for
“b” is NOKEEP. Due to the default values, the following two restart statements are identical:

RESTART VERSION=LAST,NOKEEP

RESTART

Whenever you perform, the software automatically creates a new version number in the database. For
each restart, the current version number is incremented by one, regardless of whether the job ran
successfully or not. However, there are two exceptions to this rule (which are discussed later on in the
chapter).

Restart–Method 1

You also need to tell Simcenter Nastran the database that you want to attach to your current run. There
are two ways of accomplishing this goal. One way is to use the ASSIGN statement. For example, if you
are restarting from the database created by stat1.dat, the following FMS statements can be used in your
current run

RESTART
assign master= 'stat1.MASTER'
$
$ the following statement is optional
$
assign dball='stat1.DBALL'

Depending on the machine, the filenames may be case sensitive. They should be entered exactly as they
were created in the cold start run and enclosed with a single quotes as shown above. Note that the
use of the second ASSIGN statement is optional since the “master” DBset can find the associated DBALL
database.

Restart–Method 2

An alternate way to attach a restart database is to use the DBS keyword on the submittal command
instead of the ASSIGN statement. Assuming the current run is called stat2.dat, then the equivalent
submittal command is as follows:

nastran stat2 dbs=stat1

16-4 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Executive Control Section

The ASSIGN statement is not needed in this case; however, the RESTART statement is needed using
either option. While the first method requires more input than the second method, it also provides you
with information about the database used for the restart run although this information is also provided
in the Execution Summary Table (.F04 for most machines).

Executive Control Section

In a restart analysis, the executive control section is identical to one in a cold start analysis.

Case Control Section

When you're performing a restart analysis, the software's automatic restart logic compares the
modifications you made to the Case Control and Bulk Data Sections to determine which operations
need to be processed upon restart. Therefore, you must be very careful in the changes that you make
in your restart run. Adhering to the following rules will avoid unnecessary reprocessing of previously
completed operations:

1. You must include in your restart run all solution-type related Case Control commands that are
unchanged as compared to the cold start run. In other words, do not make unnecessary load, SPC,
or MPC command changes or remove them from the Case Control Section unless these are actual
changes. This process is clarified later on with the example problems.

2. Output requests can be modified. A typical example can be a request of the element stress output
that was not requested in the cold start run.

Bulk Data Section

The software stores a copy of the Bulk Data for each version. Your restart run must not contain any Bulk
Data entries that were included with the previous version that you are restarting from since they are
already saved in the database for that particular version. The Bulk Data section in the current restart
run should only contain the addition of new entries, the modification of old entries, or the deletion
of old entries from the database. The following is a series of three runs illustrating an application of a
restart. The first run (Listing 16-2) is a sample statics run with one loading condition, and it requests
displacement output.

$
$ filename - stat1.dat
$
ID ROD,TEST
SOL 101
TIME 20
DIAG 8
CEND
TITLE= COLD START RUN
$

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 16-5

© 2024 Siemens
 16. Restarts

SPC = 1
$
SUBCASE 1
LOAD = 1
DISP = ALL
$
BEGIN BULK
$
GRID,1,,0.,0.,0.
GRID,2,,5.,0.,0.
GRID,3,,10.,0.,0.
CROD,1,1,1,2
CROD,2,1,2,3
PROD,1,1,.2
MAT1,1,1.+7,,.32
FORCE,1,2,,1000.,1.,0.,0.
SPC1,1,123456,1
PARAM,AUTOSPC,YES
$
ENDDATA

Listing 16-2. Input File for a Cold Start Run

The second run (Listing 16-3) is a restart run with an additional load condition. Displacement and
stress outputs are desired for this new loading condition. Furthermore, stress output for all the elements
is desired for Subcase 1. However, since you already printed the displacement for Subcase 1 in your
original run, you may elect not to reprint this information by removing the displacement request from
Subcase 1. There are no model changes as compared to the original run. Note that the Bulk Data entries
that were present in the original run must not be included in the current run. The only entry in the
Bulk Data Section is the new FORCE entry for Subcase 2, which was not included in the original run.
Since this load is a new loading condition, the FORCE entry must have an unique ID (2 in this case) as
compared to the existing FORCE entry ID (1 in this case) that is stored in the database.

This type of restart run constitutes an efficient restart since the re-decomposition of the stiffness matrix
is not performed; for a large problem, this can result in substantial CPU savings.

$ filename - stat2.dat
$
restart
assign master='stat1.MASTER'
ID ROD,TEST
SOL 101
TIME 20
DIAG 8
CEND
TITLE= RESTART RUN

16-6 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Bulk Data Section

$
SPC = 1
$
SUBCASE 1
LABEL = ADDITIONAL STRESS OUTPUT REQUEST
LOAD = 1
STRESS = ALL
$
SUBCASE 2
LABEL = ADDITIONAL LOAD CASE
LOAD = 2
DISP = ALL
STRESS = ALL
$
BEGIN BULK
$
$
$ NOTE THAT THE FORCE ID IS UNIQUE
$ WITH RESPECT TO EXISTING LOAD ID IN
$ THE DATABASE.
$
FORCE,2,2,,-500.,1.,0.,0.
$
ENDDATA

Listing 16-3. Input File for a Restart Run with an Additional Load Case

The third run (Listing 16-4) is a restart run with a modification in the material property. The same two
loading conditions are used as run number two. A new material entry is added in this run and the old
material entry (the eighth sorted Bulk Data from the stat2.dat run) is deleted. The general format for
deleting a range of entries from the database is

/,x,y

where x and y are the range of the first and last sorted Bulk Data count to be deleted from the database,
respectively. You can have as many of these entries as you need. If you are only deleting one entry (y=x),
then either one of the following two formats can be used:

/,x,x
/,x

The sorted Bulk Data is output by default. If the ECHO Case Control command is specified in the cold
start run, it should be set to output the sorted Bulk Data (e.g., ECHO=SORT).

There may be times when you may be making substantial changes to the model. In this case, it may be
easier to delete the whole Bulk Data listing from the database and insert a complete new Bulk Data file
as shown below:

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 16-7

© 2024 Siemens
 16. Restarts

/,1,z
$ include a complete new bulk data listing
$

In this case, z stands for an integer larger than the total number of Bulk Data entries (e.g., 999999).
Alternatively, you could delete the database files and run the problem as a cold start run.

$ filename - stat3.dat
restart
assign master='stat1.MASTER'
ID ROD,TEST
SOL 101
TIME 20
DIAG 8
CEND
TITLE= RESTART RUN
$
SPC = 1
$
SUBCASE 1
LABEL = DISPLACEMENT AND STRESS OUTPUT REQUEST
LOAD = 1
DISP = ALL
STRESS = ALL
$
SUBCASE 2
LABEL = ADDITIONAL LOAD CASE
LOAD = 2
DISP = ALL
STRESS = ALL
$
BEGIN BULK
$
/,8
mat1,1,3.+7,,.32
$
ENDDATA

Listing 16-4. Input File for a Restart Run with Modified Material Property Entry

In this example, the old MAT1 entry is the eighth sorted Bulk Data (from run stat2.dat) as shown by the
following partial output:

16-8 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Bulk Data Section

S O R T E D B U L K D A T A E C H O

CARD
COUNT . 1 .. 2 .. 3 .. 4 .. 5 ..
6 .. 7

1- CROD 1 1 1
2
2- CROD 2 1 2
3
3- FORCE 1 2 1000. 1.
0.
4- FORCE 2 2 -500. 1.
0.
5- GRID 1 0. 0.
0.
6- GRID 2 5. 0.
0.
7- GRID 3 10. 0. 0.
8- MAT1 1 1.+7 .32
9- PARAM AUTOSPC
YES
10- PROD 1 1
.2
11- SPC1 1 123456
1
ENDDATA
TOTAL COUNT= 12

Listing 16-5. Sorted Bulk Data Listing from a Cold Start Run

This type of restart (run number 3) requires the reformulation of the stiffness matrix since the material
was changed. For a nonsuperelement run, this type of restart does not save you much in terms of
runtime, if any. This run (run number 3) is included merely to show you how to delete information from
the database. However, for superelement type runs, this type of restart can still save you time since you
may only be resolving a portion of the structure.

The model used in the above series of runs is quite simple. In most cases, such simple models do
not warrant the use of a restart since the savings are insignificant for such a small model. This simple
example is chosen to illustrate the mechanics of how simple the restart feature is without being
distracted by the details of a complex model.

The same restart procedure can now be applied to a larger model. The model used is “BCLL16". “BCLL16"
is one of the standard test problems that is run for each new version of Simcenter Nastran. It is basically
a 16 × 16 × 16 cube of CHEXA elements overlaid with a CQUAD4 elements on the face of each
CHEXA element to simulate a relatively dense stiffness matrix. The model consists of 29,400 degrees of
freedom. The following table illustrates the benefit of using restarts for large models.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 16-9

© 2024 Siemens
 16. Restarts

Table. BCLL16 Performance Summary

Run Number Description CPU Time (Sec)

1 Static analysis with one load case and the following output 372.1
request:
SET n = a,b,c,d
DISP = n
STRESS (PLOT) = All
2 After the completion of run number 1, you now decide that you 364.5
want additional printed stress output for a few selective elements.
One option is to rerun the same job from the beginning with the
following modified output request in the Case Control Section:
SET m = x,y,z
STRESS = m
3 The second option is to restart from run number 1 with the 5.0
following modified output request in the Case Control Section:
SET m = x,y,z
STRESS = m

Note:
Note the tremendous savings in the CPU time (364.5 versus 5 seconds) between the run numbers
2 and 3 with both runs accomplishing the same goal.

The cold start and restart input files used for run numbers 1 and 3 are shown in Listing 16-6 and Listing
16-7.

See “Mini-Database Restarts for efficient method of restarts when only data recovery is desired.”

$ FILENAME - BCLL16A.DAT
INIT DBALL,LOGICAL=(DBALL(200000))
INIT SCRATCH,LOGICAL=(SCRATCH(200000))
ID NXN, BCLL16A
TIME 300
SOL 101
CEND
TITLE = STATIC ANALYSIS OF A CELLULAR CUBE BCLL16A
SUBTITLE = 16 X 16 X 16 CUBE
ECHO=NONE
STRESS(PLOT)=ALL
SET 100 = 30080,40088,70880,80888

16-10 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Bulk Data Section

DISP = 100
LOAD=100
SPC=10
BEGIN BULK
$ BRING IN THE REST OF THE BULK DATA FILE
$
include 'bcllblk.dat'
ENDDATA

Listing 16-6. Input File for a BCELL16 Cold Start Run

$ FILENAME - RBCLL16A.DAT
$
RESTART
ASSIGN MASTER='bcll16a.MASTER'
$
$ THE ABOVE STATEMENT ASSUMES THAT YOUR
$ COLDSTART INPUT FILENAME AS
$
$ bcll16a.dat
$
ID NXN, BCLL16A
TIME 100
SOL 101
CEND
TITLE = STATIC ALALYSIS OF A CELLULAR CUBE BCLL16A
SUBTITLE = 16 X 16 X 16 CUBE
ECHO=NOSORT
SET 100 = 30080,40088,70880,80888
set 200 = 1001,2001,3001,4001
$DISP = 100
stress = 200
LOAD=100
SPC=10
BEGIN BULK
$
$ SINCE THERE IS NO MODEL CHANGES, NO BULK
$ DATA ENTRIES ARE NEEDED
$
ENDDATA

Listing 16-7. Input File for a BCELL16 Restart Run

Note the user interface is quite straightforward considering the tremendous benefits in terms of CPU
reduction.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 16-11

© 2024 Siemens
 16. Restarts

Read-only Restarts
Read-only restart is very similar to read-write restart, except that original database is accessed in a
read-only mode. No new data is added to the existing database. Instead, all new data is written to
the new database, leaving the original database intact. You can delete the new database or save it for
another restart. The new database created during the restart run retains an internal link to the original
database.

To use read-only restart for the same example problem as in the listing Input File for a Restart Run with
an Additional Load Case, all you need to do is modify the FMS statements from:

Restart

Assign master='stat1.MASTER'

to:

Restart logical=run2
Assign run2='stat1.MASTER'

In this case, run2 is an arbitrary logical name.

Mini-database Restarts
You can save on the database only those data blocks that are required for data recovery. The size of
the database, called a mini-database, is on average about 25% the size of a full database. For pure data
recovery run (no SPC or loading changes), this type of restart is very efficient.

To use this feature, use the following submittal command in the cold start run:

nastran jobname scr=mini

In the restart run, no special input is required because Simcenter Nastran automatically detects a
mini-database and will process any output request from the table below.

The mini-database restart processes the following Case Control commands and user parameters:

DISPLACEMENT STRAIN
ELSDCON STRFIELD

ESE STRESS
SUBCOM
FORCE
SUBSEQ
GPFORCE
SURFACE
GPSDCON
SYMCOM

16-12 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Determining the Version for a Restart

GPSTRESS SYMSEQ
MPCFORCES THERMAL
PARAM,POST VOLUME
REPCASE
SPCFORCES

For cyclic symmetry analysis, you can't change the HOUTPUT and NOUTPUT in the restart run.

Determining the Version for a Restart

Not all versions in the database are restartable. For each run, Simcenter Nastran prints a message near
the top of the .F04 file indicating the version number of the current job. In general, if a job completes
without any fatal messages, and the database is saved, then that particular version is restartable. It is a
good idea to keep a brief log of all the restartable versions since this is probably the most reliable source
of information regarding whether a version is restartable. If a restart job fails, e.g., due to Bulk Data
error, then this newly created version is not restartable, and it is indicated as such by the following error
message at the bottom of the .F06 file.

*** USER WARNING MESSAGE 784 (XCLNUP)

VERSION = yyy PROJECT = “zzz” OF THIS DATA BASE IS NOT VALID FOR
RESTART
PURPOSES.
USER ACTION:
SUBSEQUENT RESTARTS SHOULD REFERENCE VERSION = xxx OR A PRIOR VALID
VERSION

In this case, you must restart from a previous valid version. The “xxx” and “yyy” above denote
version numbers. The “zzz” denotes a project description provided by you. This project description is
alphanumeric and can contain up to 40 characters with the default being blank. The project description
is rarely used and is an optional statement.

If for some reason the records for the old runs are no longer available, then the DBDIR FMS statement
can be used to query the database contents to find out which versions are stored in the database. The
following simple setup is all that is required for this purpose:

ASSIGN MASTER='ddddd.MASTER'
DBDIR VERSION=*,PROJECT=*
ENDJOB

Neither the Executive, Case Control, nor Bulk Data Section is required in this case. Furthermore, a new
version is not created because the ENDJOB statement is present. Near the top of the .F06 output, a
PROJECT VERSION DIRECTORY TABLE is printed listing all the versions in the database. A “**” next to a
version number indicates that this particular version was deleted from the database. This deletion may
be due to the “NOKEEP” option or the use of the DBCLEAN statement when performing a restart run.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 16-13

© 2024 Siemens
 16. Restarts

A version number with a “**” next to it is not restartable. A version number without a “**” next to
it is restartable if the run that created the version did not terminate with UFM 784. This DBDIR FMS
statement can also be used to check the database directory for the existence of data blocks (e.g., UG). If
the job fails very early in the run (e.g., error in the FMS section), then a new version may not be created.

Always back up your database on a regular basis. If the system aborts the run, then your database
may get corrupted. Another good practice, which ensures that only good models are retained in the
database, is to perform the following:

1. Use RESTART VERSION=a,KEEP

where “a” is a version number.

2. If a version contains errors, or is no longer of interest, then you can use the FMS statement
DBCLEAN to remove obsolete or incorrect versions of the model from the database. Using DBCLEAN
allows the Executive System to reuse some of this space for new versions.

16-14 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 17. Database Concepts
Overview of Database Concepts
All Simcenter Nastran runs require a database. A database allows for the storage and retrieval of data:
data blocks, parameters, scratch files, and DMAP object and source files.

During a run, data may be written onto only one database. This is called the primary database which
is attached automatically or by the INIT MASTER statement. Additional databases may be attached for
“read-only” purposes. This includes a “delivery” database and one or more “located” databases.

• The delivery database contains solution sequences and is automatically attached by the SOL Executive
statement or the ACQUIRE FMS statement.

• “Located” databases are attached by the DBLOCATE statement.

Before presenting the details of the database structure, the definitions for some of the basic terms that
are used throughout this chapter need to be presented.

DBset Database set. DBset consists of a Simcenter Nastran logical name, such as
MASTER, DBALL, etc., which refers to an entity within the Simcenter Nastran
database. It consists of a collection of physical files.

Database The collection of all DBsets assigned to a run.

Data block A matrix or table (e.g., KAA, LAMA) that is stored in the database.

Logical name The local internal name (log-name) used by Simcenter Nastran for a file or
DBset.

Word For 32- and 64-bit machines, each word is equivalent to four and eight bytes,
respectively.

BUFFSIZE The length of a Simcenter Nastran buffer in terms of words (32- or 64-bit
words). It controls the physical record size for data storage/transfer that
is contained in many Simcenter Nastran logical units. The default and
maximum allowable BUFFSIZE is machine dependent. The default value is
recommended except for large problems. It can be modified by using the
following NASTRAN statement:

NASTRAN BUFFSIZE = xxxxx

Block A block is often referred to as a Simcenter Nastran GINO block. Each GINO
block contains one Simcenter Nastran buffer. The size of each block is equal
to BUFFSIZE-1 words. (GINO is an acronym for general input/output.)

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 17-1

© 2024 Siemens
 17. Database Concepts

{} A brace indicates that the quantity within this bracket is mandatory. The
underlined item within { } is the default value.

[] A square bracket indicates that the quantity within this bracket is optional.

_____ The underlined value is the default value for a particular option.

Understanding DBsets
A database is divided into several DBsets which are initialized by the INIT file management statement.
There are two types of DBsets:

• Permanent DBsets, which you can save at the end of the run and reuse in a restart run

• Scratch DBsets, which are automatically deleted at the end of the run

A DBset may be composed of up to twenty concatenated physical files or DBset members. Each member
is assigned to a physical file by the ASSIGN statement.

For a Simcenter Nastran input file named “stat1.dat”, the following sample submittal command can be
used:

nastran stat1 scr=no

In the command above, nastran is the name of the shell script used to execute Simcenter Nastran. The
following two permanent physical database files are created as a result of the above command.

• stat1.MASTER

• stat1.DBALL

Note that the temporary scratch files are allocated and used by Simcenter Nastran during the solve and
are deleted at the end. Unless otherwise stated, the input filename is assumed to be “stat1.dat” in this
chapter.

By default, there are four DBsets—two permanent and two scratch—which are predefined and
automatically assigned by the program. The DBset names of the predefined permanent DBsets are:
MASTER and DBALL. The DBset-names of the predefined scratch DBsets are: SCRATCH and OBJSCR. The
DBsets are briefly described below:

• The MASTER DBset is the directory of the database. It contains the names of all DBsets, DBset
members and their physical file names, and a directory of projects, versions, data blocks, parameters,
DMAP source and object files. It also contains the NDDL scheme used to describe the database. The
default maximum size for MASTER is 5,000 blocks.

17-2 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Understanding DBsets

• The DBALL DBset contains all the DMAP data blocks which may be saved permanently for reuse in a
subsequent run. In the solution sequences in which restarts are not possible, this DBset is empty.

• The SCRATCH DBset is a temporary DBset for all scratch data blocks and files. It has two partitions: one
for DMAP data blocks and one for DMAP module internal scratch files.

• The SCR300 DBset is used as the temporary workspace for the modules. This space is released at the
end of the module execution to be reused by subsequent modules. The default size is 250,000 blocks

• The OBJSCR DBset is a temporary DBset for DMAP compilation.

• If you want to create your own DMAP source and object files, you need to allocate the USROBJ
and USRSOU DBsets. They contain DMAP source and object files that may be saved permanently for
execution in a subsequent run. See the COMPILE Executive Control statement.

If an INIT and ASSIGN statement aren't specified for one or more of the predefined DBsets, then the
corresponding statement(s) are provided by default.

ASSIGN MASTER='dbs-name.MASTER'
INIT MASTER(RAM) LOGICAL=(MASTER(5000))
ASSIGN DBALL='dbs-name.DBALL'
INIT DBALL LOGICAL=(DBALL(25000))
ASSIGN USROBJ='dbs-name.USROBJ'
INIT USROBJ LOGICAL=(USROBJ(5000))
ASSIGN USRSOU='dbs-name.USRSOU'
INIT USRSOU LOGICAL=(USRSOU(5000))
ASSIGN SCRATCH='temp-name.SCRATCH'
ASSIGN SCR300='temp-name.SCR300'
INIT SCRATCH(MEM) LOGICAL=(SCRATCH(175000)),
SCR300=(SCR300(175000))
ASSIGN OBJSCR='temp-name.OBJSCR'
INIT OBJSCR LOGICAL=(OBJSCR(5000))

The numbers inside the parentheses are the maximum sizes of the DBsets in GINO blocks. By default,
one member with the same log-name as the DBset-name is assigned to each DBset. (For a description
of a GINO block, DBset-name and log-name, see the INIT statement). For example, the DBALL DBset
has one member also called DBALL. The dbs-name is determined from the dbs keyword on the nastran
command and temp-name is generated by the command procedure. The scratch DBsets are assigned
special names on a scratch disk. (For a description of the dbs-name and temp-name, see the Simcenter
Nastran Installation and Operations Guide).

The filenames above are those that would be typically generated on Linux computers. For example, if
the name of the input file is called MYJOB.DAT and the dbs keyword isn't specified, then the default
filenames for the permanent DBsets would be:

• MYJOB.MASTER

• MYJOB.DBALL

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 17-3

© 2024 Siemens
 17. Database Concepts

Also, they would have the same directory or path as the input file.

A physical filename is automatically created for any new DBsets specified by an INIT statement. For
example, to create a new DBset called DBUP for split database operations, specify:

INIT DBUP

The physical filename will then be: MYJOB.DBUP. For another example, to create two members for
DBALL, specify

INIT DBALL LOGI=(DB1 ,DB2)

Their physical filenames will be MYJOB.DB1 and MYJOB.DB2.

You can use the ASSIGN statement to override these default names. For example, to rename
MYJOB.DBUP to FENDER.DBUP, specify:

ASSIGN DBUP='FENDER.DBUP'
INIT DBUP

To rename MYJOB.DB1 and MYJOB.DB2, specify

ASSIGN DB1 ='FENDER.DBALL1'

ASSIGN DB2='FENDER.DBALL2'
INIT DBALL LOGI=(DB1,DB2)

The INIT statement may be used to change the maximum size of the predefined DBsets. (The size may be
specified in blocks, kilowords, megawords, kilobytes, megabytes, gigawords, or gigabytes.) For example,
to specify a size of 50,000 blocks for DBALL,

INIT DBALL LOGICAL=(DBALL(50000))

or 50,000 bytes,

INIT DBALL LOGICAL=(DBALL(50KB))

The INIT and ASSIGN statement may be used to specify more than one member for a DBset which may
exist on different physical devices.

ASSIGN DB1 ='physical file name of DB1 on disk 1'

ASSIGN DB2='physical file name of DB2 on disk 2'
INIT DBALL LOGICAL=(DB1(5000),DB2(5000))

17-4 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Delivery Database

Delivery Database
The Delivery Database contains the solution sequences. This database, in addition to the primary
database, is automatically assigned in order to execute a Simcenter Nastran solution sequence. For
example, the filenames of the database are

SSS.MASTERA
SSS.MSCOBJ
SSS.MSCSOU

If the solution sequence is not a Simcenter Nastran solution sequence, as in the case of a user's solution
sequence; then the ACQUIRE FMS statement may be used to select the appropriate database. ACQUIRE
NDDL selects the SSS database. The ACQUIRE statement is required when the DBLOAD FMS statement is
used with the ENDJOB FMS statement.

Deleting DBsets
There are several different ways that you can control what happens to permanent DBSets in the primary
database at the end of a solve.

• If scr=yes is specified on the nastran command or the INIT MASTER(S) statement appears in the FMS
Section, then Simcenter Nastran deletes all DBsets in the primary database.

• An individual DBset member is automatically deleted at the end of the run by specifying the TEMP
keyword on the ASSIGN statement. For example, to delete the USROBJ and USRSOU DBsets at the end
of the run, specify:

ASSIGN USROBJ='A' TEMP

ASSIGN USROBJ='B' TEMP

Note: If you use this method to delete a DBset, you can't recreate it for a future run.

• The DBSETDEL FMS statement provides a better method than the TEMP keyword on the ASSIGN
statement except that the DBsets are deleted at the beginning of the run. The deleted DBsets may be
recreated in a future run. For example:

DBSETDEL USROBJ,USRSOU

Database Autoassignment
When a database is created, the physical filenames of all DBset members are loaded into the MASTER
DBset. Therefore, when using DBLOCATE or RESTART, it is sufficient to assign the MASTER DBset member.
All other DBset members previously initialized, i.e., DBALL, DBUP, etc., are automatically assigned by the
program. For example, on restart, the following input will automatically assign the DBALL, USROBJ, and
USRSOU DBsets:

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 17-5

© 2024 Siemens
 17. Database Concepts

ASSIGN MASTER='filename of the MASTER DBset member'

RESTART

If DBLOCATE is used, then:

ASSIGN MSTR1 ='filename of the MASTER DBset member'

DBLOCATE LOGI=MSTR1

The autoassignment feature cannot assign the physical file of a DBset member that has been renamed.
This is because the MASTER DBset member contains the old physical filename. Therefore, an additional
ASSIGN statement is required to specify the new filename of the DBset member. If in the previous restart
example DBALL is renamed, then the following input is required:

ASSIGN MASTER='filename of the MASTER DBset member'

ASSIGN DBALL='new filename of the DBALL DBset member'
RESTART

In split database operations, it is often necessary to assign only some of the DBsets because the others
are “offline.” The NASTRAN AUTOASGN statement is used to specify the types of databases that will be
autoassigned:

1. AUTOASGN = 0: No databases will be autoassigned. This includes the primary database, Delivery
Database, and any located databases.

2. AUTOASGN = 1: All databases will be autoassigned. This is the default.

3. AUTOASGN = 2: Only the Delivery Database will be autoassigned.

4. AUTOASGN = 4: Only located databases will be autoassigned.

If some combination of the options above is desired, then sum their values. For example, if the delivery
and located databases are to be autoassigned, then specify NASTRAN AUTOASGN = 6.

Database Project and Version

All data stored on the primary database is assigned to a particular project identification string (ID) and
version identification number (ID). The current project ID is specified on the PROJECT FMS statement.
The current version ID is an integer value assigned by the program each time the database is attached as
the primary database. When a primary database is initialized or a new project is specified, then all items
stored on the database during the run are stored under a version ID of integer 1. Then, in subsequent
runs a new version is created with an ID equal to the last version ID incremented by one. If the RESTART
FMS statement is specified, then a new version of data is created which is equivalent to the last version.
In other words, all items in the old version are made available for reuse in the new version.

17-6 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Migrating Databases

Migrating Databases
Databases between versions in general aren't compatible and must be converted, or migrated, before
you can use them in a higher version of the software. This is required primarily because NDDL sequences
change. For example, some qualifiers may have been added to the paths of some data blocks in order to
support new capabilities in the solution sequences. Also, NDDL description of some data blocks change.

Importantly, a database created in nonlinear analysis, heat transfer analysis, or optimization analysis
cannot be fully migrated. This is because of changes in Bulk Data entry formats and/or data blocks. Also,
it would be inefficient to recreate because it is created by modules which perform the bulk of a nonlinear
analysis.

You can migrate a database in a single run which consists of two phases.

• In the first phase, the DBLOCATE FMS statement with the CONVERT and COPY keywords is used to
copy most of the data blocks and all of the parameters from the earlier version database to the
current version primary database.

• In the second phase, the remaining data blocks which cannot simply be copied into the current
version have to be regenerated. This involves the execution of a short solution sequence that
regenerates those data blocks: specify SOL DBTRANS (or SOL 190).

For example, the input file necessary to migrate a database created by one of the structured solution
sequences (SOLs 100 through 200) is:

ASSIGN MIG='physical filename of the MASTER DBset'

RFINCL DBTRANS
SOL DBTRANS
CEND

With the DBTRANS solution sequence, it is assumed that the early version database contains all of the
data from a completed and successful analysis. If not, then the solution sequences may fail and the
migration may not be successful. In this case, it is possible to alter DBTRANS with the ALTER statement to
obtain the appropriate data.

If the earlier version database wasn't created by a Simcenter Nastran solution sequence, then it may be
necessary to develop new DBLOCATE statements and a new migration solution sequence.

Database Archival, Compression, and Transfer

Database archival, compression and transfers are performed using the DBUNLOAD and DBLOAD FMS
statements. You can select all items on the database or selected items according to the WHERE clause.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 17-7

© 2024 Siemens
 17. Database Concepts

Database Archival and Retrieval

The DBUNLOAD FMS statement is used to archive the primary database. It writes out the data blocks and
parameters stored in the database to a sequential file in a binary format which is similar to that of the
OUTPUT2 module (see Simcenter Nastran DMAP Programmer's Guide). A database can only be unloaded
in a run separate from the run in which it is created or modified. The default action is to archive all data
blocks and parameters under all projects and versions in the database. The ASSIGN FMS statement is
also required to assign a FORTRAN unit on which to unload the database. The default unit number is 50.
This unit number and other default attributes are predefined under the logical keyword DBUNLOAD. For
example, a simple request to unload or archive a database is as follows:

ASSIGN DBUNLOAD='physical filename of archive file'

DBUNLOAD
ENDJOB

The DBLOAD FMS statement is used to read an archived database and copy it to the primary database.
The ACQUIRE statement is required to select the appropriate NDDL sequence. The ASSIGN FMS
statement is also required to assign a FORTRAN unit to the unloaded database. The default unit number
is 51. This unit number and other default attributes are predefined under the logical keyword DBLOAD.
For example, a simple request to reload the archived database above is as follows:

ACQUIRE NDDL
ASSIGN DBLOAD='physical filename of archive file'
DBLOAD
ENDJOB

The example above makes use of the ENDJOB statement, which means that only the FMS statements
will be processed in the run. A new BUFFSIZE may be assigned to the retrieved database with the
NASTRAN statement. SOL DBTRANS or DBTRANU must also be executed:

NASTRAN BUFFSIZE=xxxx
ASSIGN DBLOAD='physical filename of archive file'
DBLOAD
SOL DBTRANS
CEND

By default, the DBLOAD and DBUNLOAD statements will load and unload all the data under all projects
and versions found in the database. The WHERE clause may be used to be more selective according to
the project ID, version ID, item name, qualifier values, and/or DBset. Also, the CONVERT clause may be
used to modify the project ID, version ID, item name, qualifier values, and/or DBset.

Database Compression

When data is deleted from the database, its space is released for the storage of new data. This old
space will be reused before any new space is used. However, even if the space is not reused, the size
of the database is not reduced. After several restarts or the execution of the DBCLEAN FMS statement,
the database may contain a significant amount of released or “dead” space. There are two methods

17-8 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Transferring Databases Across Different Platforms

of removing dead space or “compressing” the database. The first method involves the DBUNLOAD and
DBLOAD FMS statements, and the examples shown above will also perform a database compression. The
second and simplest method involves the DBLOCATE statement specified with the COPY keyword. For
example, if the database was created with SOLs 100 through 200, then the input file is:

ACQUIRE NDDL
ASSIGN MYDB='physical file name of the MASTER DBset'
DBLOCATE LOGI=MYDB COPY
ENDJOB

It is also possible to compress an individual DBset and change its maximum size. For example, the input
file is:

ACQUIRE NDDL
ASSIGN MYDB='physical file name of the MASTER DBset'
DBLOCATE LOGI=MYDB COPY WHERE(DBSET='DBALL')
INIT DBALL LOGICAL=(DBALL(10000))
ENDJOB

Transferring Databases Across Different Platforms

You can also use the DBUNLOAD and DBLOAD statements to copy a database from one computer
platform to another.

• On the first computer, the DBUNLOAD statement converts the database to a “neutral” file on the
first computer. This neutral file is now suitable for a simple copy to the second computer or may be
accessed through a network if the two computers are so connected.

• On the second computer, the DBLOAD statement converts the neutral file back to a database. It is
also necessary to execute the solution sequence DBTRANS (or 190) as described under Database
Migration, because some data blocks cannot be converted to a neutral format. (In SOL DBTRANS;
data blocks EST, EMAP, ETT, PTELEM, SLT, KDICT, KELM, MDICT, MELM, BDICT, BELM, and ACPT are
regenerated.)

The ASSIGN statement is also required to assign the physical filename of the neutral file and specify
that the file is formatted since the default format (UNFORMATTED) is not transferable across dissimilar
computers.

The following example unloads a database created by one of the unstructured solution sequences on
the first computer to a neutral file:

ASSIGN DBUNLOAD='physical filename of neutral file' FORMATTED

DBUNLOAD FORMAT=NEUTRAL
ENDJOB

Then the neutral file is copied to, or accessed through a network by, the second computer:

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 17-9

© 2024 Siemens
 17. Database Concepts

ASSIGN DBLOAD='physical filename of neutral file' FORMATTED

DBLOAD FORMAT=NEUTRAL
SOL DBTRANU
CEND

With the DBTRANS solution sequence it is assumed that the neutral file contains all of the data from a
completed and successful analysis. If not, then the solution sequences may fail and the transfer may
not be successful. In this case, it is possible to alter DBTRANS with the ALTER statement to obtain the
appropriate data.

If the database was created in a heat transfer analysis then NASTRAN HEAT = 1 must be specified.

If the database on the first computer was not created by a Simcenter Nastran solution sequence, then it
may be necessary to develop a new transfer solution sequence.

Using the File Management Section with Databases

The File Management Section (FMS) is intended primarily for the attachment and initialization of
Simcenter Nastran database sets (DBsets) and FORTRAN files. For many problems due to the default
values, the FMS is handled automatically and is not required to be specified by you.

You can allocate the database file size in any of the following units:

• Bytes

• Words

• Megabytes (1,000,000 bytes)

• Megawords (1,000,000 words)

• Kilobytes (1,000 bytes)

• Kilowords (1,000 words)

• Simcenter Nastran GINO blocks (default)

• Mixture of the above units

Internally, Simcenter Nastran converts these units to GINO blocks. If an FMS statement is longer than
72 characters, it must be continued on the next line. You can have a maximum of 200 files (FORTRAN
and DBsets) assigned (online) at a time. This limit may be less for some machines due to their operating
system limit.

17-10 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Using the INIT Statement

Using the INIT Statement

INIT creates/initializes permanent and/or temporary DBsets. The INIT statement has two basic formats:
one for all the DBsets and one specifically for the SCRATCH DBsets.

Format (Simplified) for All DBsets Except SCRATCH DBsets

INIT DBset-name [LOGICAL=(log-name1(max-size1) ,

log-name2(max-size2),...log-namei(max-sizei) ,
...log-namen(max-sizen))]

DBset-name The logical name of the DBset being used (e.g., DBALL).

log-namei The i-th logical name for the DBset-name referenced in this INIT
statement. You can have up to 20 logical names for each DBset (1 ≤
n ≤ 20 ). An ASSIGN statement may be used to attach a physical file to
the i-th logical file.

max-sizei The maximum allowable size that may be written to the i-th file.

Example A

The following statement creates the DBALL DBset with a logical name of DBALL and the maximum size
of 50,000 Simcenter Nastran blocks instead of 250,000 blocks, which is the default value for DBALL:

INIT DBALL LOGICAL=(DBALL(50000))

Assuming that your input file is called “stat1.dat”, the physical file has the name stat1.DBALL unless an
ASSIGN statement is also used.

The following statement creates the DBALL DBset with logical names of DB1 and DBTWO:

INIT DBALL LOGICAL=(DB1(35000),DBTWO(60000))

The two physical files stat1.DB1 and stat1.DBTWO are created with a maximum of 35,000 and 60,000
Simcenter Nastran blocks, respectively.

Format (Simplified) for the SCRATCH DBset

INIT SCRATCH [LOGICAL=(log-name1(max-size1) ,

log-name2(max-size2),...log-namei(max-sizei)) ,
SCR300=(log-namei+1(max-sizei+1),...log-namen(max-sizen))]

Log-name1 through log-namei are allocated for regular scratch files as temporary workspace. This space
is not released until the end of the job. SCR300 is a special keyword that indicates the log-names are
members reserved for DMAP module internal scratch files. The space occupied by these files is for the

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 17-11

© 2024 Siemens
 17. Database Concepts

duration of the execution of the module. This space is released at the end of the module execution. You
can have up to a combined total of 20 logical names for the SCRATCH DBset (1 ≤ n ≤ 20).

Example B

The following statement creates the SCRATCH DBset with logical names of SCR1, SCR2, SCRA, and SCRB:

INIT SCRATCH LOGICAL=(SCR1(150MW),SCR2(100MW)) ,

SCR300=(SCRA(250MW),SCRB(300MW))

The two physical files stat1.SCR1 and stat1.SCR2 are created with maximum sizes of 150 and 100
megawords, respectively. These two files are regular scratch files. Two additional physical files stat1.SCRA
and stat1.SCRB are created with maximum sizes of 250 and 300 megawords, respectively. These last two
files are SCR300-type files.

Using the ASSIGN Statement

Use ASSIGN to assign a physical filenames to logical filenames or special FORTRAN files that are used by
other FMS statements or DMAP modules.

Format (Simplified) to Assign Logical Files

ASSIGN log-namei='filenamei' [TEMP DELETE]

log-namei The i-th logical name for the DBset created by the INIT statement.

TEMP Requests that filenamei be deleted at the end of the job.

DELETE Requests that filenamei be deleted if it exists before the start of the run. This
is optional; however, if this option is not used and the file exists prior to the
current run, then the job may fail with the following messages:

*** USER FATAL MESSAGE 773 (DBDEF)

THE FOLLOWING PHYSICAL FILE ALREADY EXISTS.
LOGICAL NAME = xxxx
PHYSICAL FILE = yyyy
USER INFORMATION: NO ASSOCIATED DEFAULT FILES OR ASSIGNED DBSETS
CAN
EXIST PRIOR TO THE DATA BASE INITIALIZATION RUN.
USER ACTION: DELETE THIS FILE AND RESUBMIT THE JOB.

DELETE is not a suggested option if you are using RESTART since you can delete your database
inadvertently. Manual deletion of unwanted databases is a safer approach.

17-12 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Example C

Example C

The following statements create a logical name DB1 for the DBset DBALL in the current directory:

ASSIGN DB1='sample.DB1'
INIT DBALL LOGICAL=(DB1(50000))

The physical file sample.DB1 is created in this case. Without the ASSIGN statement, the physical filename
created is called stat1.DB1, assuming once again that your input file is called stat1.dat.

The following statements create the two logical names DB1 and DB2 for the DBset DBALL:

ASSIGN DB1='/mydisk1/se/sample.DB1'
ASSIGN DB2='/mydisk2/sample.DB2'
INIT DBALL LOGICAL=(DB1(50000),DB2(40000))

DB1 points to a physical file called sample.DB1 that resides in the file system (directory) /mydisk1/se. DB2
points to a physical file called sample.DB2 that resides in the file system (directory) /mydisk2.

Format (Simplified) to Assign FORTRAN Files

ASSIGN log-key='filenamef' [STATUS={new,old},UNIT=u ,

FORM={FORMATTED,UNFORMATTED},DELETE]

log-key This is the logical keyword for the FORTRAN file being assigned. This file may
already exist. The default value depends on the keyword. Acceptable keywords are

DBC,DBMIG,INPUTT2,INPUTT4,OUTPUT2,OUTPUT4,
DBUNLOAD,DBLOAD, and USERFILE.

See the Simcenter Nastran Quick Reference Guide for detailed descriptions of these keywords and their
default values.

filenamef The physical name of the FORTRAN file.

STATUS Specifies whether the FORTRAN file is created (STATUS = new) or is an existing file
(STATUS = old).

UNIT Specifies the FORTRAN unit (e.g., UNIT = 12).

FORM Specifies whether the file written is in ASCII (FORM = FORMATTED) or binary
(FORM = UNFORMATTED) format. Note that if the log-key is OUTPUT2, the
FORMATTED output is in compressed ASCII format.

DELETE Requests that filenamef be deleted, if it exists before the start of the run.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 17-13

© 2024 Siemens
 17. Database Concepts

Example D

The following example creates a new FORTRAN file to be used for OUTPUT2 operations:

ASSIGN OUTPUT2='sample.out',STATUS=NEW,UNIT=11,FORM=FORMATTED

This FORTRAN file is in compressed ASCII format with a filename of sample.out and is assigned to unit
11. Note that a compressed ASCII file can be transferred directly across machines, but it should not be
edited.

Using the EXPAND Statement

Expand concatenates files into an existing DBset in order to increase the allowable disk space available
for Simcenter Nastran. The EXPAND statement is normally used in a restart run when you exceed the
disk space allocation in your previous run.

Format

EXPAND DBset-name LOGICAL=(log-namei(max-sizei),... )

DBset-name The logical name of the DBset to be expanded by the addition of new members
to an existing DBset previously defined with an INIT statement.

log-namei The logical name of the i-th member of the DBset. An ASSIGN statement should
be used to point this logical name to a physical file.

max-sizei The maximum size of the i-th member.

Only one EXPAND statement is allowed per run.

Example E

The original run creates a database with the name stat1.DBALL. However, this database was filled and
the job failed with the following error messages in the .F06 file:

*** USER FATAL MESSAGE 1012 (GALLOC)

DBSET DBALL IS FULL AND NEEDS TO BE EXPANDED.

For small to medium problems, it is best to rerun the job from the beginning with a larger file allocation.
For large problems, if rerunning the job is not practical, then the database can be expanded with the
following FMS statements:

RESTART
ASSIGN MASTER='stat1.MASTER'

17-14 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Using the RESTART Statement

ASSIGN DBADD='morespace.DB'
EXPAND DBALL LOGICAL=(DBADD(50000))

These statements assign an additional member, with a logical name of DBADD, to the existing DBset
DBALL. This member points to a new physical file called morespace.DB, which may contain up to a
maximum of 50,000 Simcenter Nastran blocks. You are restarting from “stat1.MASTER” in this case.

The EXPAND statement cannot be used for the scratch files since they are deleted at the end of each job.

Using the RESTART Statement

RESTART allows you to continue from the end of the previous run without solving the problem from the
beginning.

Format

RESTART [PROJECT='proj-ID', VERSION={version-ID,LAST},{KEEP,NOKEEP}]

proj-ID Project identifier used in the original run. It can have up to 40 characters. This
option is optional and is normally not used. The default proj-ID is blank.

version-ID The version number that you are restarting from.

KEEP If this option is used, then the version that you are restarting from is also saved
at the end of the current run.

NOKEEP If this option is used, then the version that you are restarting from is deleted at
the end of the current run.

Example F

The following statement causes the current run to use the last version in the database for the restart:

RESTART

At the end of the run, this last version that you are restarting from is deleted from the database. This is
probably the most commonly used form for RESTART.

The following statement instructs the current run (version 6 or higher) to use version 5 in the database
for restart:

RESTART VERSION=5,KEEP

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 17-15

© 2024 Siemens
 17. Database Concepts

At the end of the run, version 5 is also retained in the database. This format is used most often when
you want to ensure that a specific version is saved in the database (e.g., a large run from which you may
want to request additional data recovery in the future).

The following statement specifies that the current run (version 4 or higher with a proj-ID of xyz) uses
version 3 with a proj-ID of xyz in the database for restart:

RESTART PROJ='xyz' VERSION=3

At the end of the run, version 3 with a proj-ID of xyz is deleted from the database.

Using the DBCLEAN Statement

DBCLEAN deletes unwanted versions from the database. The DBCLEAN statement does not reduce the
size of the database. This statement deletes the data blocks for the specified versions to allow other data
blocks to reuse this space for subsequent restarts without potentially increasing the database size. Up to
ten DBCLEAN statements can be used in the FMS for each run.

Format

DBCLEAN VERSION={version-ID,*} [PROJECT={'project-ID',*}]

version-ID The version number that you want to remove from the database.

project-ID The identifier of the project to be deleted. The default is blank, which is the
most commonly used form.

* This is a wildcard command. It removes all versions or projects from the

database.

Example G

The following statements delete versions 3 and 6 with a blank proj-ID from the database in the current
run:

DBCLEAN VERSION=3
DBCLEAN VERSION=6

Using the DBDIR Statement

The DBDIR statement prints the database directory.

Format

DBDIR [VERSION={version-ID,*}] [PROJECT={proj-ID,*}][FORMAT={format-no}]

17-16 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Example H

version-ID Version number.

proj-ID Project number.

* This is a wildcard command. It refers to all versions or projects.

format-no This controls the type of directory information printed. The value is the sum of
the desired types listed below. The default value is 63 (sum of first six values),
which means the first six types will be printed.

1 = Project version table

2 = NDDL data blocks
4 = NDDL parameters
8 = NDDL empty data blocks
16 = NDDL data block trailers
32 = NDDL path value table for NDDL entries
64 = Scratch data blocks

Example H

The following statement causes the printing of the database directory for all versions of the current
project-ID in the database:

DBDIR

Using the INCLUDE Statement

The INCLUDE statement inserts an external file at the location where the INCLUDE statement is used. It is
not a pure FMS statement because it can be used anywhere in the input file, not just in the FMS section.

Format

Include 'filename'

filename Physical filename of external file to be inserted at this location.

The following run reads a file called sub1.dat with all the Case Control commands contained in it:

Sol 101
cend
include 'sub1.dat'
begin bulk
$
include 'bulk1.dat'
include 'bulk2.dat'
$

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 17-17

© 2024 Siemens
 17. Database Concepts

$ rest of bulk data file

$
.
.
enddata

This run also brings two additional files (bulk1.dat and bulk2.dat) into the Bulk Data Section. You may,
for example, want to include all your grid entries in file bulk1.dat and all your element connectivities in
bulk2.dat. As you can see, the INCLUDE statement can be a handy tool. For parametric studies, you can
potentially save a tremendous amount of disk space by using the INCLUDE statement instead of having
multiple files with duplicate input data.

Guidelines for Large Problems

In general, you can solve small to medium sized problems in Simcenter Nastran with very little
knowledge of FMS statements and database structure. However, when you're solving large problems,
knowledge of FMS statements and database structure can help you to allocate and optimize your
computer resources.

When running large problems, you should have some idea of the resource requirements—both disk
space and computer time—prior to engaging in the actual analysis. A utility program called ESTIMATE
provides a set of recommended resource values for running a particular job.

For most small to medium problems, the default values for all the file assignments should be more
than adequate. However, for a large problem, larger disk space allocation may be necessary. Several
examples are shown in this section that illustrate various options and consequences with each option.
The submittal command for each example is shown followed by a partial listing of the corresponding
input file. In each case, it is assumed that you are running on a Linux machine with “nastran” as the
name of the shell script executing Simcenter Nastran.

See also

• “ESTIMATE” in the Simcenter Nastran Installation and Operations Guide

Example I

The following submittal command and FMS statements allocate 300,000 and 5,000 Simcenter Nastran
GINO blocks for DBALL and MASTER, respectively:

nastran runa scr=no

$
$ filename - runa.dat
$
INIT DBALL,LOGICAL=(DBALL(300000))
INIT SCRATCH,LOGICAL=(SCRATCH(200000)) ,
SCR300=(SCR300(200000))

17-18 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Example J

$
$ THE REST OF YOUR INPUT FILE
$

Furthermore, the following physical files are created and saved in your current directory:

• runa.DBALL

• runa.MASTER

These statements also allocate 200,000 and 200,000 Simcenter Nastran GINO blocks for SCRATCH and
SCR300, respectively. These two files are allocated to the default scratch directory during the course of
the run and are deleted at the end of the run. Since the database is saved, this job is restartable. Other
than the allocation size of the databases, this is identical to the default setup.

When using the “scr = yes” option on the submittal command, the permanent data blocks that are
normally written to the DBALL DBset are now written to the SCRATCH DBset.

Example J

The submittal command and FMS statements below allocate 50, 500, and 800 megabytes for DBALL,
SCRATCH, and SCR300, respectively. Furthermore, by default, 5,000 Simcenter Nastran GINO blocks are
allocated each for the MASTER dbset. During the course of the run, all of these files are allocated to your
default scratch directory. Because you are using the “scr=yes” option on the submittal command, the
permanent data blocks that are normally written to the DBALL DBset are now written to the SCRATCH
DBset. In other words, you must allocate the disk space to the SCRATCH DBset, which you would
normally allocate to the DBALL DBset, when you use the “scr=yes” option on the submittal command. At
the end of the run, all of these files are deleted automatically. Since the database is not saved, this job is
not restartable.

nastran runb scr=yes

$
$ filename - runb.dat
$
INIT DBALL,LOGICAL=(DBALL(50MB))
INIT SCRATCH,LOGICAL=(SCRATCH(500MB)) ,
SCR300=(SCR300(800MB))
$
$ THE REST OF YOUR INPUT FILE
$

Example K

If you have sufficient disk space in your current and default scratch file systems (disk packs), the method
shown in “Example A” or “Example B” is the preferred procedure since it is the simplest method.
However, if you do not have sufficient disk space in these two file systems, but have other file systems

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 17-19

© 2024 Siemens
 17. Database Concepts

(e.g., /disk2, /disk3, and /disk4) mounted to your system, then the procedure listed below can be used.
You must have read and write privileges in order to use these file systems. For this example, the three
new file systems are /disk2, /disk3, and /disk4.

nastran runc scr=no

$
$ filename - runc.dat
$
ASSIGN DBALL1='runc.db1'
ASSIGN DBALL2='/disk2/user_guide/statics/runc.db2'
ASSIGN SCRATCH2='/disk3/user_guide/statics/runc.scrtch2'
ASSIGN SC3B='/disk4/user_guide/statics/runc.sc3b'
$
INIT DBALL,LOGICAL=(DBALL1(200000KB),DBALL2(100000KB))
INIT SCRATCH,LOGICAL=(SCRATCH1(100000),SCRATCH2(100000)) ,
SCR300=(SC3A(150000),SC3B(50000))
$
$ THE REST OF YOUR INPUT FILE
$

The permanent database is broken up into two separate logical files (DBALL1 and DBALL2). The above
run allocates 200,000 and 100,000 kilobytes for DBALL1 and DBALL2, respectively. The run also
allocates 5,000 GINO blocks for the MASTER dbset. DBALL1 is allocated to the current directory that you
are running on with the physical filename of “runc.db1". DBALL2 is allocated to the “/disk2/user_guide/
statics” directory with a physical filename of “runc.db2". The following additional physical file is also
created in your current directory:

runc.MASTER

These three files are saved at the end of the run and the job is restartable.

The regular SCRATCH and SCR300 files are each divided into two separate files. Two logical files
(SCRATCH1 and SCRATCH2) are assigned to the regular SCRATCH file. SCRATCH1 is assigned to the default
scratch directory requesting 100,000 GINO blocks. SCRATCH2 is assigned to the “/disk3/user_guide/
statics” directory with a filename of “runc.scrtch2" and requesting 100,000 GINO blocks. The SCR300
file is also divided into two files (SC3A and SC3B). SC3A requests 150,000 blocks and assigns this disk
space to the default scratch directory. SC3B requests 50,000 blocks and assigns this disk space to the
“/disk4/user_guide/statics” directory with a filename of “runc.sc3b”. All four of these scratch files are
deleted at the end of the run.

nastran rund scr=yes

Example L

Even though the setup in “Example D” looks virtually the same as that in “Example C,” the “scr=yes”
keyword causes the permanent data blocks-that are normally written to the DBALL DBset-to be written

17-20 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 BUFFSIZE

to the SCRATCH DBset. In other words, except for the amount used for overhead, the space that you
have allocated to DBALL1 and DBALL2 is not used at all. You must adjust the disk space allocation to the
SCRATCH DBset that you normally would allocate to the DBALL DBset when you use the “scr=yes” option
on the submittal command. The correct action in this case is to increase the disk space allocation for
SCRATCH1 and SCRATCH2, and decrease the disk space allocation for DBALL1 and DBALL2. Once again,
since the database is not saved, this job is not restartable.

$
$ filename - rund.dat
$
ASSIGN DBALL1='rund.db1'
ASSIGN DBALL2='/disk2/user_guide/statics/runc.db2'
ASSIGN SCRATCH2='/disk3/user_guide/statics/runc.scrtch2'
ASSIGN SC3B='/disk4/user_guide/statics/runc.sc3b'
$
INIT DBALL,LOGICAL=(DBALL1(200000KB),DBALL2(100000KB))
INIT SCRATCH,LOGICAL=(SCRATCH1(100000),SCRATCH2(100000)) ,
SCR300=(SC3A(150000),SC3B(50000))
$
$ THE REST OF YOUR INPUT FILE
$

BUFFSIZE

BUFFSIZE is the length of a Simcenter Nastran buffer in terms of words (32- or 64-bit words). It
controls the physical record size for data storage/transfer that is contained in many Simcenter Nastran
logical units. The default and maximum allowable BUFFSIZE is machine dependent. The default value is
recommended except for large problems. Each 32-, and 64-bit word contains 4 and 8 bytes, respectively.
This feature can be invoked by including the following NASTRAN statement in your input file.

NASTRAN BUFFSIZE=xxxxx

where xxxxx = (n · disk-block-size) + 1 ≤ limit and n is a positive integer.

A larger BUFFSIZE takes fewer requests of the computer operating system to transfer the same amount
of data, with each transfer involving a larger physical record size. This can reduce the I/O time which,
in turn, reduces the elapsed time. In general, the effect on CPU time is insignificant. A larger BUFFSIZE
decreases the number of I/O operations but may increase the database size. Since each data block uses
at least one buffer and the read/write operation requires a minimum of one buffer, the default BUFFSIZE
is recommended except for large problems. See Simcenter Nastran Installation and Operations Guide
for the default BUFFSIZE for various machines.

Disk space allocation is affected by the BUFFSIZE if the disk space units are in terms of GINO blocks. GINO
blocks are related to words by the following equation:

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 17-21

© 2024 Siemens
 17. Database Concepts

Equation 17-1.

“Example E” illustrates the use of a non-default BUFFSIZE.

Example M

Assume the machine that you are running on has a default BUFFSIZE of 2049. If you increase the
BUFFSIZE to 17291, using the equation in BUFFSIZE, this example allocates the same maximum amount
of physical disk space as “Example A,” which uses the default BUFFSIZE.

$
$ filename - rune.dat
$
NASTRAN BUFFSIZE=17291
INIT DBALL,LOGICAL=(DBALL(35536))
INIT SCRATCH,LOGICAL=(SCRATCH(23691)),SCR300=(SCR300(23691))
$
$ THE REST OF YOUR INPUT FILE
$

Assuming you have disk space, is there any other reason why you would not want to use a larger
BUFFSIZE for a large problem? The answer is “no” for most machines, provided you have sufficient
memory available on your machine. For some machines, optimum values are preselected, and you
should use these default values. See Reference 16. for further details regarding your particular machine.
Total memory allocation can be controlled by the “mem” keyword on the submittal line. The following
example allocates 100 mb of total memory for the run.

nastran stat1 mem=100mb

A portion of this total memory is allocated to the Executive System. The rest of the memory is then
available to the functional modules in Simcenter Nastran. Assuming that you are using the default
SCRATCH(MEM) and BUFFERPOOL, the portion of the memory that is available for your problem is
denoted as “User OPENCORE (HICORE)”, and can be estimated by Eq. 17-2.

Equation 17-2.

As you can see from Eq. 17-2, the amount of memory allocated to the Executive System is a function
of the BUFFSIZE. As the BUFFSIZE increases, a larger portion of the total memory is allocated to the
Executive System. A summary of the Simcenter Nastran memory utilization is printed near the beginning

17-22 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Example M

of the .F04 file. Some typical examples of the memory summary tables are shown in Listing 17-1,
Listing 17-2, and Listing 17-3. You can also let Simcenter Nastran estimate the amount of required
memory by using the following submittal command:

nastran stat1 mem=estimate

** MASTER DIRECTORIES ARE LOADED IN MEMORY.

USER OPENCORE (HICORE) = 1607959 WORDS
EXECUTIVE SYSTEM WORK AREA = 81854 WORDS
MASTER(RAM) = 30000 WORDS
SCRATCH(MEM) AREA = 204900 WORDS ( 100
BUFFERS)
BUFFER POOL AREA (EXEC) = 76183 WORDS ( 37
BUFFERS)
TOTAL Simcenter Nastran MEMORY LIMIT = 2000896 WORDS

Listing 17-1. Memory Summary Table(mem = 2MW, buffsize = 2049)

** MASTER DIRECTORIES ARE LOADED IN MEMORY.

USER OPENCORE (HICORE) = 3607831 WORDS
EXECUTIVE SYSTEM WORK AREA = 81854 WORDS
MASTER(RAM) = 30000 WORDS
SCRATCH(MEM) AREA = 204900 WORDS ( 100
BUFFERS)
BUFFER POOL AREA (EXEC) = 76183 WORDS ( 37
BUFFERS)
TOTAL Simcenter Nastran MEMORY LIMIT = 4000768 WORDS

Listing 17-2. Memory Summary Table(mem = 4MW, buffsize = 2049)

** MASTER DIRECTORIES ARE LOADED IN MEMORY.

USER OPENCORE (HICORE) = 1352015 WORDS
EXECUTIVE SYSTEM WORK AREA = 249516 WORDS
MASTER(RAM) = 30000 WORDS
SCRATCH(MEM) AREA = 1729100 WORDS ( 100
BUFFERS)
BUFFER POOL AREA (EXEC) = 640137 WORDS ( 37
BUFFERS)
TOTAL Simcenter Nastran MEMORY LIMIT = 4000768 WORDS
Listing 17-3. Memory Summary Table(mem = 4MW, buffsize = 17291)

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 17-23

© 2024 Siemens
 17. Database Concepts

17-24 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 18. Inertia Relief in Linear Static Analysis
Overview of Inertia Relief
Inertia relief is an advanced option that allows you to simulate unconstrained structures in the linear
static solution SOL 101, and in the linear static solution portions of SOLs 105 and 200. Typical
applications of inertia relief include modeling an aircraft in flight, an automobile on a test track, or
a satellite in space. The inertia relief method is a tool that you may not be familiar with. To help
understand the method, the first section provides a brief description of inertia relief. The last section
shows its implementation within Simcenter Nastran.

Although the internal processing is somewhat complex, the interface to the inertia relief option is quite
straightforward. There are two ways to invoke the inertia relief.

• The first method is to specify the SUPORT entry explicitly by including “PARAM,INREL,-1” in the Bulk
Data Section.

• The second method (recommended) is to let Simcenter Nastran select the SUPORT degrees of
freedom automatically by including “PARAM,INREL,-2” in the Bulk Data Section.

An optional “PARAM,GRDPNT,x”— where x is a grid point ID— can be specified in the model.
If “PARAM,GRDPNT,x” is used, the loads and accelerations will be summed about this point. If
“PARAM,GRDPNT,x” is not specified, then the loads and accelerations will be summed about the origin of
the basic coordinate system.

Description of Inertia Relief Using PARAM,INREL,-1

Static analysis by the finite element method assumes that the model contains no mechanisms and may
not move as a rigid body (strain free). If either of these conditions exists in a conventional finite element
analysis, the stiffness matrix for the model becomes singular. When Simcenter Nastran attempts to
decompose a singular matrix, a fatal message or unreasonable answers result.

Consequently, you can't perform conventional finite element static analysis on unconstrained structures.
However, a method called inertia relief is provided in Simcenter Nastran for analyzing these conditions.
A simple description of inertia relief is that the inertia (mass) of the structure is used to resist the applied
loadings, that is, an assumption is made that the structure is in a state of static equilibrium even though
it is not constrained. Two examples are a spacecraft in orbit or an aircraft in flight. In these cases, the
structure is in state of static equilibrium, although it is capable of unconstrained motion.

To invoke inertia relief, you must provide a SUPORT Bulk Data entry with a list of up to six non-redundant
degrees of freedom that describe the possible unconstrained motion. The easiest way to describe how to
use the SUPORT entry in static analysis is if you hold the SUPORT degrees of freedom constrained, there
is no possible rigid body motion. If all possible rigid body motion is not described on the SUPORT entry,
then the stiffness matrix is singular, and the problem either fails in decomposition or gives unreasonable
answers.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 18-1

© 2024 Siemens
 18. Inertia Relief in Linear Static Analysis

When you specify inertia relief, Simcenter Nastran calculates the forces that result from a rigid body
acceleration about the point specified on the “PARAM,GRDPNT,x” in the specified directions. Simcenter
Nastran then calculates the summation of all applied loadings in the same directions. Accelerations are
applied to the structure in the appropriate directions to “balance” the applied loadings. The structure is
now in a state of static equilibrium, i.e., the summation of all applied loads is 0.0. Since the problem is
not constrained, rigid body displacement is still possible.

Simcenter Nastran next constrains the SUPORT degrees of freedom to a displacement of 0.0 and provide
the relative motion of all other grid points with respect to that reference point. Hence, the term
“reference” degree of freedom is used to describe the SUPORT degrees of freedom in Simcenter Nastran.
The set of degrees of freedom described on the SUPORT entry belong to the r-set or reference set for
the solution. The computed solution is the correct one, and it is relative to any rigid body motion that is
occurring. A simple way to think of this is that the solution coming from Simcenter Nastran represents
the deformation of the structure you would see if you were standing at the SUPORT degrees of freedom.

Implementation of Inertia Relief Using PARAM,INREL,-1

Solution 101 requires that the model does not contain mechanisms. You specify the reference degrees
of freedom using the SUPORT entry as follows:

1 2 3 4 5 6 7 8 9 10
SUPORT ID1 C1 ID2 C2 ID3 C3 ID4 C4

Field Contents

IDi Grid or scalar point identification number.

Ci Component numbers.

In addition to the PARAM,INREL,-1 entry, a SUPORT entry is also needed. If PARAM,GRDPNT,x is specified,
the rigid body mass matrix is calculated about grid point x; otherwise, the rigid body mass is calculated
about the origin of the basic coordinate system. The inertia relief method requires that a realistic mass
distribution exists, and the degrees of freedom listed on the SUPORT entry must be connected elastically
to the model in all specified degrees of freedom. (For example, degrees of freedom 4, 5, and 6 on a grid
point with only solid elements attached cannot be used since solid elements have no stiffness in these
degrees of freedom.)

Whenever a SUPORT entry is used in static analysis, the epsilon and strain energy printed in the table
from UIM 3035 should be checked. The values printed for epsilon and strain energy in the UIM 3035
table should all be numeric zero. The strain energy printed in this table for matrix KLR represents the
strain energy in the model when the one SUPORT degree of freedom is moved 1.0 unit, while all other
SUPORT degrees of freedom are constrained. If the SUPORT degrees of freedom are properly specified,
the model should be capable of rigid body motion (strain-free) with no strain energy.

The values printed for the strain energy indicate the ability of the model to move as a rigid body. These
values should always be checked. If the structure is not constrained, the values should be numeric

18-2 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Implementation of Inertia Relief Using PARAM,INREL,-1

zero, but roundoff almost always results in a small nonzero value. Acceptable values are a function of
the units, size of the structure, and precision of the hardware; therefore, a recommended value is not
provided in this user's guide.

An additional feature allowed in Solution 101 is the solution of a problem under uniform acceleration.
This problem is posed using the DMIG,UACCEL entry in addition to the previously mentioned
requirements. In this case, uniform accelerations are applied to the model, and the solution is found.
Uniform accelerations are useful for situations, such as spacecraft liftoff and landing loadings, which are
often specified as static accelerations.

As an example of inertia relief, consider the three CBEAM model shown in Figure 18-1.

Figure 18-1. Inertia Relief Analysis of a CBEAM Model

The three CBEAM structure is to be analyzed as a free-free structure with a line load acting in the
Y-direction as shown and using inertia relief. To show the effect of the SUPORT point, two runs are
made, each with a different SUPORT grid point. The input file is given in Listing 18-1. Note that the
SUPORT information is shown for both runs; however, one set is commented out. For the first run,
“inertia1.dat”, the SUPORT point is located at grid point 1. All six degrees of freedoms for the SUPORT
point are placed on one grid point as required for Solution 101. The parameter INREL is set to –1, and
the parameter GRDPNT is set to the SUPORT point, which is grid point 1 in this case.

$ FILENAME - INERTIA1.DAT
ID LINEAR,INERTIA
SOL 101
TIME 5
CEND
TITLE = BAR WITH SUPORT ENTRY
LOAD = 1
DISP = ALL
SPCF = ALL
STRESS = ALL

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 18-3

© 2024 Siemens
 18. Inertia Relief in Linear Static Analysis

BEGIN BULK
PARAM POST 0
$
$
$ SUPORT ENTRY FOR INERTIA1.DAT
$
SUPORT 1 123456
PARAM GRDPNT 1
PARAM INREL -1
$
$ SUPORT ENTRY FOR INERTIA2.DAT
$
$SUPORT 3 123456
$PARAM GRDPNT 3
$PARAM INREL -1
$
GRID 1 0.0 0.0 0.0
GRID 2 10. 0.0 0.0
GRID 3 20. 0.0 0.0
GRID 4 30. 0.0 0.0
$
CBEAM 1 1 1 2 1. 1. 0.0
CBEAM 2 1 2 3 1. 1. 0.0
CBEAM 3 1 3 4 1. 1. 0.0
$ .5 .5 .5 .5
PBEAM 1 1 1. .667 .167 .1
.5 .1 1.5 -.1 -.5 .1 -.5 -.1
PLOAD1 1 1 FY FR 0. 1000. 1. 1000.
PLOAD1 1 2 FY FR 0. 1000. 1. 1000.
PLOAD1 1 3 FY FR 0. 1000. 1. 1000.
$
MAT1 1 1.+7 .3 7.43E-3
ENDDATA

Listing 18-1. Three CBEAM Model with Inertia Relief

A partial listing of the output is shown in Figure 18-2.

18-4 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Implementation of Inertia Relief Using PARAM,INREL,-1

Figure 18-2. Inertia Relief Output When Grid Point 1 Is the SUPORT Point

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 18-5

© 2024 Siemens
 18. Inertia Relief in Linear Static Analysis

Figure 18-3. Inertia Relief Output When Grid Point 1 Is the SUPORT Point (Continued)

18-6 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Implementation of Inertia Relief Using PARAM,INREL,-1

Figure 18-3. Inertia Relief Output When Grid Point 1 Is the SUPORT Point

The GPWG (Grid Point Weight Generator) output is calculated with respect to the grid point specified
on the parameter GRDPNT. The total mass of the structure in this model is 0.229, which is used by
Simcenter Nastran to develop the inertia loads. Inspection of the GPWG output should be part of your
routine model checkout.

The OLOAD RESULTANT contains seven sections—consisting of seven lines per section. The first section
is the resultant of the applied loads about the GRDPNT, which is the basic coordinate system in this case.
The remaining six sections are the loads necessary to impose unit acceleration about the SUPORT point.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 18-7

© 2024 Siemens
 18. Inertia Relief in Linear Static Analysis

Subcase 2 corresponds to the X-direction; Subcase 3 corresponds to Y-direction, etc. The first six lines of
each section represent the detailed contributions, whereas the seventh line represents the total.

Following the OLOAD output is User Information Message 3035 showing the strain energy and epsilon
due to the imposed unit accelerations about the SUPORT point. As discussed earlier, the epsilons and
strain energies should be small, which is the case for this example. Small epsilons and strain energy
tell you that you do not have any unwanted constraints or poorly defined MPCs causing a constraint in
your model. Always inspect User Information Message 3035 to make sure that the epsilons and strain
energies are small.

The intermediate matrix QRR is printed following the UIM 3035. The QRR matrix is the total rigid body
mass of the total structure. It is a 6 x 6 matrix measured about the PARAM,GRDPNT point in the global
coordinate system. Masses on the scalar points are not included. Following the QRR matrix is the QRL
matrix. The QRL matrix is the resultant of the “apparent reaction loads,” measured at the SUPORT point.
This resultant is equal and opposite to the OLOAD resultant shown earlier in the output, if the SUPORT
point is the same as the PARAM,GRDPNT point. There is one column for each loading condition. The last
matrix output is the URA matrix, which is the rigid body acceleration matrix that is computed from the
applied loads.

The displacement and stress output shown is the standard output as requested through the Case Control
Section. Note that the displacement at the SUPORT point is exactly 0.0. This SUPORT point should be
0.0 since this point is constrained for the solution, and the forces of constraint should be numeric zero.
The SPC forces are shown in the output to confirm that the loads are balanced at the SUPORT point. The
displacement of all of the other points in the model are relative to the SUPORT point.

The displacement solution calculated for the model is not a true deflected shape, but rather a solution
of the inertia relief calculations that balance the acceleration of the center of gravity with the applied
loads. In the “Automatic Inertia Relief” section, the model is run using the automatic inertia relief
method by specifying PARAM,lNREL,-2 instead of PARAM,INREL,-1 and removing the SUPORT entry. The
results shown in the figure Abridged Output Using the Automatic Inertia Relief Method indicate that
the “balance” displacement solution is in the direction of the applied load and is centered around the
center of gravity of the model.

For the PARAM,lNREL,-1 method, the “balance” solution is fixed at Grid 1, which is the grid point specified
on the SUPORT entry. To satisfy this constraint, the balance shape is rotated clockwise. Because the
model is free-free and the applied load is a constant pressure load along the entire beam, the true
deflected shape is a rigid body motion in the direction of the applied load. Such a solution produces no
bending and generates zero stress, which are confirmed by the results.

An interesting effect occurs when the SUPORT point is changed in the previous example. Suppose you
change the location of the SUPORT point from 1 to 3 (see the commented SUPORT entries in Listing
18-1).

The resulting displacement and stress output is shown in Figure 18-4.

18-8 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Implementation of Inertia Relief Using PARAM,INREL,-1

Figure 18-4. Inertia Relief Output When Grid Point 3 Is the SUPORT Point (Continued)

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 18-9

© 2024 Siemens
 18. Inertia Relief in Linear Static Analysis

Figure 18-4. Inertia Relief Output When Grid Point 3 Is the SUPORT Point (Continued)

18-10 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Implementation of Inertia Relief Using PARAM,INREL,-1

Figure 18-4. Inertia Relief Output When Grid Point 3 Is the SUPORT Point

Note that the displacement vector changes as the displacements are now relative to grid point 3, which
is the new SUPORT point. However, the stresses in the elements are the same because the stresses are
based on the relative displacement between the grid points, which is independent of the grid point used
for the SUPORT point.

A final comment on this simple beam model: suppose you used CBAR elements instead of CBEAM
elements for this model. The run would fail because the CBAR element does not have any torsional
inertia mass, whereas the CBEAM element does. This model is basically a one-dimensional structure so
that the rotational inertia about the X-axis is the mass inertia of the elements. When the inertia relief
method is used, the structure must have mass in all six directions. In general, however, there are no
restrictions on using a CBAR element with inertia relief. Most structures include CBARs that are not
co-linear. The example above was introduced merely to emphasize the need to have mass in all six
degrees of freedom.

Suppose you wish to impose a 10 g acceleration at grid point 2 of the beam structure shown in Figure
18-1. This acceleration can be applied using the DMIG,UACCEL Bulk Data entry as shown below.

1 2 3 4 5 6 7 8 9 10
DMIG UACCEL “0" “9" TIN
DMIG UACCEL L G1 C1 X1
G2 C2 X2 G3 C3 X3

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 18-11

© 2024 Siemens
 18. Inertia Relief in Linear Static Analysis

Field Contents

TIN Type of matrix being input.

L Load sequence number.

Gi Grid point identification number of a single reference point.

Ci Component number for Gi in the basic coordinate system.

Xi Value of enforced acceleration term in the basic coordinate system.

The input file for this example is shown in Listing 18-2.

$ FILENAME - UACCEL
ID LINEAR,UACCEL
SOL 101
TIME 5
CEND
TITLE = BAR WITH UACCEL INPUT
OLOAD = ALL
DISP = ALL
BEGIN BULK
PARAM POST 0
$
$
$ SUPORT INFORMATION
$
SUPORT 2 123456
PARAM GRDPNT 2
PARAM INREL -1
DMIG UACCEL 0 9 1
DMIG UACCEL 1 2 2 3864.0
$
GRID 1 0.0 0.0 0.0
GRID 2 10. 0.0 0.0
GRID 3 20. 0.0 0.0
GRID 4 30. 0.0 0.0
$
CBEAM 1 1 1 2 1. 1. 0.0
CBEAM 2 1 2 3 1. 1. 0.0
CBEAM 3 1 3 4 1. 1. 0.0
$
PBEAM 1 1 1. .667 .167 .1
.1 .1 .1 -.1 -.1 .1 -.1 -.1
$

18-12 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Implementation of Inertia Relief Using PARAM,INREL,-1

MAT1 1 1.+7 .3 7.43E-3

ENDDATA

Listing 18-2. Imposing a 10 g Acceleration Using the PARAM,UACCEL

The acceleration of 10 g is entered on the DMIG,UACCEL entry. When a DMIG (Direct Matrix Input) entry
is used, UACCEL, the first entry is the header entry, which is denoted by the 0 in field 3. Field 4 must be a
9, and field 5 is either a 1 for single precision input (the most common) or a 2 for double precision input.
For this example, the input is in single precision.

The second entry is where the applied acceleration is entered. Field 3 must be a 1, indicating that this
is the first load case (and the only load case in this model). The grid point component and value of the
enforced acceleration are entered in fields 6, 7, and 8. If acceleration is be to enforced in more than
one component, then the continuation entries are used. For this example, the acceleration of 3864.
in/sec2 is entered in field 8. The 3864.0 is 10 g in the English system-you must ensure that the units are
consistent.

When using the DMIG,UACCEL, the intermediate matrix URACCEL is output as shown in Figure 18-5.
This matrix is the rigid body acceleration that you input using the DMIG,UACCEL entry. Also shown is
the OLOAD output and the displacement vector. As can be seen, the SUPORT point is constrained, which
is what you would expect, and the total load applied to the structure corresponds to the mass of the
structure (see the GPWG output in Figure 18-5) times the enforced acceleration.

Although not commonly done, you can perform multiple inertia relief analyses in a single run. This
feature can be activated with the SUPORT1 entries instead of the SUPORT entry. Unlike the SUPORT
entry, which is automatically activated, the SUPORT1 entry must be called out by the SUPORT1 Case
Control command in order for it to be applied.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 18-13

© 2024 Siemens
 18. Inertia Relief in Linear Static Analysis

Figure 18-5. Partial Output of the UACCEL Example

Automatic Inertia Relief

In the previous sections, manual inertia relief is discussed where the SUPORT entry must be specified
explicitly along with PARAM,INREL,-1. The SUPORT entry lists the DOFs in the r-set, the reference set.
Some skill is required to use this entry correctly. When the r-set is chosen such that structure is not
constrained in a statically determinate manner, or if the stiffness attached to these points is inadequate,
the rigid body mode shapes may be poor, leading to low accuracy in inertial load calculations. A poor
static analysis solution may also result from the poor constraints, independent of the loading problems.
Both effects may result in poor overall accuracy. Poor results can occur with little warning when the
SUPORT entry is misused.

The automatic inertia relief method is the recommended inertia relief method. With the automatic
method, the specification of the SUPORT entry is no longer needed. To turn it on, simply add

18-14 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Automatic Inertia Relief

PARAM,INREL,-2 to the input file. The reference frame is selected automatically, in a manner that poor
solutions are unlikely because of the choice of reference frame variables.

The constraints associated with the reference frame are distributed to all points with mass. This means
that structures with modeling errors, such as a region of elements left out through oversight, will still
give reasonable results that can aid in diagnosing the modeling errors. A model in development may
contain many disjoint parts, inadvertently. Any part with three non co-linear points with mass will be
adequately constrained for solution.

In the manual support option, the free stiffness matrix Kaa is constrained by removing the r-set DOFs
from the a-set and imposing zero motion on them,

Equation 18-1.

Equation 18-2.

With the automatic inertia relief, the ur variables are no longer a subset of the ua variables. The
constraint equation (Eq. 18-2) is replaced by a more general constraint equation,

Equation 18-3.

The variables ur are a set of generalized coordinates that represent the average motion of the ua
variables, as weighted by the mass matrix. The zero value for this equation implies that the average
motion of all a-set points is zero, although the displacement variables are free to move relative to this
least-squares fit reference frame.

The manual style of inertia relief allows use of PARAM,GRDPNT to define the origin used in determining
the shape functions Da6. The presence of this parameter causes the output of the Grid Point Weight
Generator (GPWG) table, which lists the c.g. location of the model, its moments of inertia, and other
related data. This point in space is used as the reference point when computing rigid body mass and

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 18-15

© 2024 Siemens
 18. Inertia Relief in Linear Static Analysis

moments of inertia, and rigid body accelerations. If the parameter is not present, the basic origin is
used, in the basic coordinate system. It is good practice (but not required by Simcenter Nastran) to use
the same grid point on PARAM,GRDPNT and on the SUPORT entry.

Unlike the manual option, PARAM,GRDPNT is not used as the inertia relief reference point for the
automatic support option. The basic origin is used. If the PARAM,GRDPNT is present, it causes the
generation of the GPWG table, but does not influence the inertia relief calculations in any way.

The constraint forces caused by the distributed constraints are printed with the SPCFORCE case control
command. They are computational zeros for a well-defined model. PARAM,TINY (default value is 1.E-3)
is traditionally used to discard element strain energies with values less than “tiny”. The same parameter
requests that small SPC forces of the reference frame constraint type be replaced with binary zeros, for
the automatic support option only. This filtering is not used on conventional SPC forces from selected
SPC entries or AUTOSPC, and manual support DOFs.

Limitations for the Automatic Inertia Relief Method

Following are limitations for the Automatic Inertia Relief Method:

• The DMIG,UACCEL option is not supported.

• The Iterative Solver is not supported.

• There must be 6 rigid body modes (i.e. free-free).

EXAMPLE

Let's revisit the CBEAM model in the figure Inertia Relief Analysis of a CBEAM Model using the
automatic inertia relief method. This is accomplished by removing the SUPORT entry and replacing
PARAM,INREL,–1 with PARAM,INREL,–2. Figure 18-6 summarizes the displacements, spcforces, and
element stresses output.

The spcforces and stresses are both zeros, similar to the other two runs. As for the displacements, the
mass-weighted average motion of all a-set points must be zero. Since the mass is evenly distributed, ur
can be calculated for the T2 components as follows:

Equation 18-4.

See TAN 4002 for further details regarding the Automatic Inertia Relief Method.

18-16 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Limitations for the Automatic Inertia Relief Method

Figure 18-6. Abridged Output Using the Automatic Inertia Relief Method

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 18-17

© 2024 Siemens
 18. Inertia Relief in Linear Static Analysis

18-18 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 19. Contact
Overview of Contact
Simcenter Nastran provides a contact capability for a variety of analysis types, including both linear
static and multi-step nonlinear.

Solutions That Support Contact

You can create surface-to-surface or edge-to-edge contact in the following solution sequences:

• SOL 101 Linear Statics

• SOL 103 Real Eigenvalue, SOL 103 Response Dynamics, and SOL 103 Shock Response Spectra

• SOL 105 Linear Buckling

• SOL 111 Modal Frequency Response and SOL 111 Modal Frequency Response Functions

• SOL 112 Modal Transient Response

• SOL 401 Multi-Step Nonlinear (Simcenter Nastran and Simcenter Multiphysics) and SOL 401 STAR-
CCM+ Co-Simulation

• SOL 402 Multi-Step Nonlinear Kinematics (Simcenter Nastran and Simcenter Multiphysics)

Note:
Most of the contact inputs used by SOL 101 are supported by SOLs 401 and 402. For more
information, see the Multi-Step Nonlinear User’s Guide.

Surface-to-Surface Contact Summary

Surface-to-surface contact can be defined on the faces of the following elements.

• Shell elements CTRIA3, CTRIA6, CTRIAR, CQUAD4, CQUAD8, and CQUADR.

• 3-D Solid elements CHEXA, CPENTA, CPYRAM, and CTETRA.

Surface-to-surface contact conditions allow the solution to search and detect when a pair of element
faces come into contact. The contact conditions prevent the faces from penetrating and allow finite
sliding with optional friction effects. The surface-to-surface contact source and target regions consist of
shell and/or solid element faces. From element faces in the source region, a top and bottom normal is
projected. The software creates a contact element if:

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 19-1

© 2024 Siemens
 19. Contact

• Any of the source element normals intersect with an element in the target region.

• The distance between the two faces is equal to or less than the defined search distance.

Edge-to-Edge Contact Summary

Edge-to-edge contact can be defined on the edges of the following elements:

• Axisymmetric elements CTRAX3, CQUADX4, CTRAX6, CQUADX8.

• Plane stress elements CPLSTS3, CPLSTS4, CPLSTS6, CPLSTS8.

• Plane strain elements CPLSTN3, CPLSTN4, CPLSTN6, CPLSTN8.

Edge-to-edge contact conditions allow the solution to search and detect when a pair of element edges
come into contact. The software creates contact elements if the distance between the edges is equal
to or less than the search distance which you specify. The contact elements prevent the edges from
penetrating and allow finite sliding with optional friction effects.

The axisymmetric, plane stress, and plane strain elements can be defined in either the XZ plane or in the
XY plane. Edge-to-edge contact is supported in either orientation.

Contact Regions
Edge Contact Regions

19-2 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Contact Region Parameters

An edge contact region is a collection of axisymmetric, plane strain, and plane stress element free edges
in a section of the model where you expect contact to occur. These regions can be created using either
the BEDGE bulk entry or the BLSEG bulk entry.

• To define a contact edge using the BEDGE entry, you enter the element ID along with the corner GRID
IDs. You can define the edges on the BEDGE entry in any order.

• To define a contact edge using the BLSEG entry, you enter the GRID IDs in a continuous topological
order. You must include both midside and corner GRID IDs on the BLSEG entry.

All region IDs defined with the BEDGE, BLSEG, BSURF, BSURFS, BCPROP, or BCPROPS entries must be
unique.

Surface Contact Regions

A surface contact region is a collection of element free faces in a section of the model where you expect
contact to occur. These regions can be created using the shell elements (BSURF and BCPROP) and using
solid element free faces (BSURFS and BCPROPS).

• The BSURF entry is defined by its own unique ID and is a list or range of shell element IDs to include in
the region.

• The BCPROP entry is defined by its own unique ID and is a list of shell element property IDs. Shell
elements which use any of these listed property IDs will be included in the region.

• The BSURFS entry is defined by its own unique ID and is a list of solid element IDs each followed by 3
grid points defining which face of the 3-D element to include in the contact region.

• The BCPROPS entry is defined by its own unique ID and is a list of solid element property IDs. The free
faces of the solid elements selected with a property ID are automatically determined by the software.

A contact surface can be defined as any type of shell or any face of a solid element. Although parabolic
faces with omitted midside nodes are permitted, their use could affect accuracy.

All region IDs defined with the BEDGE, BLSEG, BSURF, BSURFS, BCPROP, and BCPROPS entries must be
unique.

Contact Region Parameters

Contact region parameters are defined using the BCRPARA bulk entry. The CRID field on BCRPARA must
match the ID used on one of the BSURF, BCPROP, BSURFS, BCPROPS, BEDGE, and BLSEG bulk entries to
be considered by the solution. The parameters SURF and OFFSET are supported with a surface-to-surface
contact definition. Only OFFSET is supported with an edge-to-edge contact definition.

• SURF is used to define the contact side of shell element regions. When SURF is "TOP", the contact side
is consistent with the shell element normal and when "BOT" the opposite. In a SOL 101 (including

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 19-3

© 2024 Siemens
 19. Contact

consecutive solutions 103, 111 and 112), SURF must be defined so that source and target contact
sides either face one another to represent a separation condition, or oppose one another to represent
an interference condition.

Understanding shell element normals, and making sure those in the same regions are consistent is
very important to ensuring that contact elements will be created as expected. See “Using Consistent
Normals” in the Simcenter Nastran User's Guide for information on using consistent shell normals.

The figure below shows examples of how the SURF option is used to modify the contact side when the
element normals vary. The arrow on each face shows the element normal. Source regions are shaded
while target regions are white.

• Use the OFFSET field to account for a rigid layer between contact face or edge regions. For example, a
model which has two metal surfaces coming into contact, and one of these has a ceramic coating. If
the ceramic material stiffness is not significant enough to be included in the analysis, it may not have
been specifically modeled, but the thickness it adds to the face of the metal may be important when
considering the contact problem.

You can also use the OFFSET field to analyze an interference fit problem if unconnected elements are
modeled coincident. The offset value in this example can represent the theoretical interference.

Contact Pairs
A contact pair combines two contact regions, source and target, in which contact will be analyzed
during the solution. Each contact pair can have its own unique friction value (if desired) and search
distance.

The BCTSET bulk entry is used to define both edge-to-edge and surface-to-surface contact pairs. The CID
field will need to match the value of ‘n' on the BCSET case control entry for the solution to recognize
this contact definition. The SIDi and TIDi fields are used to define source and target regions respectively
for a pair. As many pairs as desired can be included on a single BCTSET entry. Each pair can have a
unique friction value (optional), a minimum search distance, and a maximum search distance. You can
define the optional Coefficient of Friction field (FRICi) for each contact pair if you expect finite sliding to
occur. When contact is detected, the solver uses this value to calculate any tangential contact forces by
multiplying the normal contact force by FRICi.

19-4 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Contact Pairs

The regions you select with SID and TID depend on the type of contact:

• For edge-to-edge contact pairs, SID and TID are contact regions defined with the BEDGE and BLSEG
entries.

• For surface-to-surface contact pairs, SID and TID are contact regions defined with the BCPROP,
BCPROPS, BSURF, and BSURFS entries.

The minimum and maximum search distance fields (MINDi and MAXDi) define a range in which the
solver can initially determine if the distance between element edges or faces in a particular pair are
within the threshold for creating contact elements. These values are used once, at the beginning of SOL
101, to determine where contact elements need to be initially created. The minimum distance can be
negative if there is an interference condition modeled as overlapping regions.

Since SOL 101 assumes small deflection, the number of contact elements created in the initial step will
remain the same for the entire solution. Although, each contact element status can change during the
solution between active and inactive as the contact condition iterates.

Combining Contact Sets – BCTADD

You can optionally define multiple BCTSET/BCTPARM bulk entry sets, each set with unique contact set
IDs (CSID), and then combine them with a single BCTADD bulk entry. The multiple BCTSET/BCTPARM
bulk entry sets are created to adjust certain contact parameters locally. Contact parameters can also be
adjusted globally with a BCTPARM bulk entry having the same CSID as the BCSET case control command.

The following example demonstrates the inputs.

CASE CONTROL
$CSID on the BCSET case control matches CSID on BCTADD
BCSET = 108
...
BULK DATA
$Local Contact Set definitions
BCTSET 1 1 2 0.0 1.0
BCTSET 2 3 4 0.15 0.0 0.1
5 6 0.15 0.0 0.1
...
$Local Contact Parameters
BCTPARM 1 PENN 10 PENT 1
BCTPARM 2 PENN 1.0 PENT 0.1
...
$Local Contact Sets are combined with BCTADD
BCTADD 108 1 2
...
$Global Contact Parameters
BCTPARM 108 MAXS 30 NCHG 0.02

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 19-5

© 2024 Siemens
 19. Contact

See the section “Contact Control Parameters - BCTPARM” in this chapter for more information on contact
parameters.

Contact with Composite Solid Faces

Defining contact regions and pairs on composite solid faces which are perpendicular to the stack
direction (edge faces) may produce poor stress continuity. If the contact definition is between edge faces
belonging to different PCOMPS definitions, and if the number of plies on each PCOMPS definition is small
and the same, and the ply thicknesses are similar, the stress continuity should be fairly smooth. This also
applies to the results requested with the BCRESULTS case control command.

Additional Recommendations

When defining contact regions and pairs on geometry which is not tangent continuous, creating single
contact regions which cross corner transitions can result in non-uniform stress results around the
corners. It is recommended to break these areas into multiple regions and pairs as shown below.

When defining contact regions and pairs, it is recommended to not include the same element face in
multiple regions. In “A” below, an element is repeated in regions 2 and 3. In “B”, the same element only
exists in region 2. “B” is recommended. Repeating element faces multiple times in the same or different
regions can significantly increase memory requirements and degrade performance.

19-6 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Contact Control Parameters

Contact Control Parameters

The contact control parameters on the BCTPARM bulk entry (optional) can help you adjust the contact
algorithm when you are having problems getting a solution to converge and complete, or when the
contact results are not as expected. Also see the section Penalty Factor Adjustment and Contact
Stiffness below. For many solutions, the default settings are appropriate, and the BCTPARM entry is not
required.

• For surface-to-surface contact definitions, all of the parameters on the BCTPARM entry are supported.

• For edge-to-edge contact definitions, the parameters CTOL, MAXF, MAXS, NCHG, INIPENE, PENN,
PENT, PENTYP, RESET, CSTRAT, PREVIEW, TANSCL, and AUTOSCAL. The parameters REFINE, SHLTHK,
ZOFFSET, and INTORD are ignored.

You can optionally define multiple BCTSET/BCTPARM bulk entry sets, each set with unique contact set
IDs (CSID), and then combine them with a single BCTADD bulk entry. The multiple BCTSET/BCTPARM
bulk entry sets are created to adjust certain contact parameters locally. Contact parameters can also be
adjusted globally with a BCTPARM bulk entry having the same CSID as the BCSET case control command.
Also see the section Contact Pairs which includes an input example.

Global and local contact parameters have the following definition and rules:

• Global Contact Parameters

The BCTPARM bulk entry, which uses the same CSID entered on the BCSET case control command,
defines global parameters.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 19-7

© 2024 Siemens
 19. Contact

Any of the parameters on the BCTPARM bulk entry can be defined globally. A parameter’s default value
is used if it is not defined globally or locally.

• Local Contact Parameters

The BCTPARM bulk entries associated to individual BCTSET bulk entries, which are then combined with
a BCTADD bulk entry, define local parameters.

Following is a description of the input parameters. Local contact parameters have “*” next to their
names. See the BCTPARM bulk entry in the Quick Reference Guide for the full description and default
value for each parameter.

• AUTOSCAL*: Scales the automatically calculated penalty factors PENN and PENT either up or down.
Will not scale explicitly defined value of PENN or PENT. AUTOSCAL can be used to scale the stiffness of
specific contact pairs if convergence issues occur.

• TANSCL*: Scales the automatically calculated tangential penalty factor PENT either up or down. Will
not scale explicitly defined value of PENT. TANSCL can be used to scale the stiffness of specific contact
pairs if convergence issues occur.

• PENN*: Penalty factor for normal direction. PENN and PENT are automatically calculated by default.
When PENT is defined but PENN is undefined, PENN = 10 * PENT.

• PENT*: Penalty factor for transverse direction. PENN and PENT are automatically calculated by default.
When PENN is defined but PENT is undefined, PENT = PENN / 10.

• PENTYP*: Changes how contact element stiffness is calculated:

When PENTYP=1 (default), PENN and PENT have units of 1/(Length), and the contact element stiffness
is calculated by K = e*E*dA where e represents PENN or PENT, E is the modulus of the softer material
in each contact pair, and dA is area. A physical interpretation is that it is equivalent to the axial
stiffness of a rod with area dA, modulus E, and length 1/e.

When PENTYP=2, PENN and PENT become a spring rate per area Force/(Length x Area), and the
contact element stiffness is calculated as K=e*dA. The spring rate input is a more explicit way of
entering contact stiffness since it is not dependent on the modulus.

Note that when composites or orthotropic materials are used, the software computes an average
modulus E from these material definitions. This computation is the same for both contact and glue
conditions. See Average modulus computation under "Glue Control Parameters" in this guide.

Note that in NX Nastran 12, the automatic normal penalty calculation (PENN) was reduced by one
order. Since PENT = PENN / 10, this change also reduces the transverse direction penalty factor.
In addition, the contact friction algorithm was improved in NX Nastran 12. You can revert to the
previous (NX Nastran 11.2) penalty calculation and friction algorithm by defining the system cell
setting SYSTEM(655)=0. SYSTEM(655) is only supported when using the sparse solver. The element
iterative solver always use the previous penalty calculation and friction algorithm.

19-8 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Contact Control Parameters

• OPNSTF*: Open contact stiffness scale factor. The open contact stiffness is computed by OPNSTF *
closed stiffness.

• CTOL: The contact force convergence tolerance, found by comparing changes in traction between the
two contacting bodies. When the convergence tolerance is less than the CTOL, the contact condition
is assumed to be converged. (default = 0.01)

• MAXF: The contact algorithm iterates in an inner and outer loop. This parameter specifies the
maximum number of iterations of the inner or force loop. The inner loop forces a condition of zero
penetration between the contacting bodies. (default = 10)

• MAXS: Specifies the maximum number of iterations for the outer or status loop. This loop determines
which contact locations (or contact elements) are active. (default = 20)

• NCHG: Specifies a number of contact elements which can be active from one iteration to the next, yet
the software will consider the solution converged. If NCHG is a real number and is < 1.0, the software
treats it as a fraction of the number of active contact elements in each outer loop of the contact
algorithm. The number of active contact elements is evaluated at each outer loop iteration. If NCHG is
an integer ≥ 1, the value defines the allowable number of contact changes. If NCHG = 0, no contact
status changes can exist.

• SHLTHK: This is the shell thickness offset flag. If you set this value to 0, the contact surface is assumed
to be offset t/2 from the nodes defining a shell element. If the FE mesh represents the outer surface,
set this value to 1 so the thickness offset will be ignored. (default = 0)

• RESET: This is a flag to indicate if the contact status for a specific subcase is to start from the final
status of the previous subcase. If you set this value to 0, the contact status will start from a previous
subcase. If you set it to 1, the contact status will start from an initial state. (default = 0)

Note that when a bolt preload is defined on CBEAM or CBAR elements and contact is in the model, the
software runs an initial preload solution to compute the strain in the bolt and to compute a converged
set of contact status conditions before applying any service loads. The bolt preload behavior depends
on the RESET contact parameter setting.

Note that when contact conditions are defined and the set ID of any one of the BOLTLD, SPC, or
MPC case control commands changes from one subcase to the next, the contact conditions will be
recomputed in the new subcase. That is, the contact status in the new subcase will not be the final
contact status from the previous subcase. This will occur regardless of how the RESET parameter is
defined.

• When RESET=1, for the consecutive solutions in the same subcase as the preload and for any
consecutive subcases, the software applies the computed bolt strain from the initial preload
solution and it uses the final contact status from the initial bolt preload solution as the starting
contact status.

• When RESET=0 (default), for the consecutive solutions in the same subcase as the preload,
the software applies the computed bolt strain from the initial preload solution and it uses the
final contact status from the initial bolt preload solution as the starting contact status (same as

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 19-9

© 2024 Siemens
 19. Contact

RESET=1). For the solutions in any consecutive subcases, the software applies the computed bolt
strain from the initial preload solution, and it applies the final contact status from the end of the
prior subcase.

• When both RESET=0 or RESET=1, if the subcase which computes the bolt preload includes only the
bolt preload and no service loads, the two pass solution described above still occurs. That is, the
first solution computes the strain in the bolt and the converged set of contact status conditions.
Since there are no service loads for this scenario, the second solution is almost a repeat of the
first, although the contact status results from the second solution may be slightly different than
the contact status results from the first. The reason is that the second solution begins with the
final contact status from the first, but also continues to iterate on this contact condition until the
hard coded minimum number of contact iterations is reached. You can optionally adjust the contact
convergence criteria with the NCHG parameter on the BCTPARM bulk entry if you would like the
contact status between the first and second solutions to be more similar.

• INIPENE*: Controls definition of initial gap or penetration of the generated contact elements. If you
set this value to 0 or 1, the contact solution will use the value calculated from the grid coordinates.
If you set the value to 2, it will initially work the same as above, but if penetration is detected, the
penetration will be set to zero. If you set the value to 3, the gap and penetration will be zero for
all contact elements. Option 3 is very useful when the contact edges or surfaces are intended to be
coincident but due to irregularity in the meshes, some grids are not exactly on the edge or surface.
This can lead to contact pressures that have a few hot spots rather than a smooth distribution. This
option will ensure that a smooth distribution is obtained and in many cases the number of iterations
will be reduced along with run time. (default = 0 or 1)

• REFINE: Determines if the mesh on the source region is refined during the contact solution.
Refinement occurs when set to 2 (default). REFINE=0 turns the refinement off. REFINE=2 turns on
refinement.

• INTORD: Determines the number of contact evaluation points for a single element face on the source
region. The number of contact evaluation points is dependent on the value of INTORD, and on the
type of element face. A higher number of contact evaluation points can be used to increase the
accuracy of a contact solution. Inaccuracies sometimes appear in the form of nonuniform contact
pressure and stress results. There may be a penalty associated with using more evaluation points since
the time for a contact problem to converge may be longer. The table below shows how the number
of contact evaluation points is dependent on the element type, and how it can be adjusted using the
INTORD option. The “Face Type” column applies to shell elements, and to the solid element with the
associated face type.

Number of Contact Evaluation Points

Face Type INTORD=1 INTORD=2 INTORD=3
Linear Triangle 1 3 7
Parabolic Triangle 3 7 12
Linear Quad 1 4 9
Parabolic Quad 4 9 16

19-10 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Penalty Factor Adjustment and Contact Stiffness

• ZOFFSET: Determines if the shell element z-offset is included in the contact solution. By default,
ZOFFSET=0 and shell z-offsets are included. ZOFFSET=1 will prevent them from being included.

• CSTRAT: Under certain conditions, all of the contact elements could become inactive which may lead
to singularities. Setting the parameter CSTRAT=1 will reduce the likelihood of all contact elements
becoming inactive. By default, CSTRAT=0, and all contact elements can become inactive.

• PREVIEW: Requests the export of a bulk data representation of the element edges and faces where
contact elements are created.

Penalty Factor Adjustment and Contact Stiffness

The stiffness between contact surfaces is obtained from the values of PENN, PENT, and from PENTYP.
PENN and PENT, which can optionally be defined on the BCTPARM bulk entry, are automatically
calculated by the software by default. The penalty factors influence the rate of convergence, and to
a lesser extent, the accuracy of the contact solution. The automatic penalty factor calculation works well
for most instances, but manual adjustments may be necessary, particularly if a contact problem fails to
converge. The automatic penalty factor calculation is turned off if either PENN or PENT are defined.

The automatic penalty factor calculation estimates geometry characteristics using element edge lengths
in the vicinity of the contact regions. The software then uses the resulting estimated lengths to
automatically calculate contact penalty stiffness values PENN and PENT. The software calculated penalty
factor is included in the *.f06 output.

Below are some guidelines for manually setting the BCTPARM bulk entry inputs PENTYP, PENN and PENT,
along with examples to help determine characteristic geometry lengths.

PENTYP Description

When PENTYP=1 (default), PENN and PENT have units of 1/(length), and the contact element stiffness is
calculated by:

K = e*E*dA

where e represents the parameters PENN or PENT

E is the elastic modulus of the softer material in each contact pair.

and dA is the area associated with the particular contact point.

A physical interpretation is that it is equivalent to the axial stiffness of a rod with area dA, modulus E,
and length 1/e.

When PENTYP=2, PENN and PENT become a spring rate per area Force/(Length x Area), and the contact
element stiffness is calculated by:

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 19-11

© 2024 Siemens
 19. Contact

K=e*dA

where e represents the parameters PENN or PENT

and dA is the area associated with the particular contact point.

The spring rate input is a more explicit way of entering contact stiffness since it is not dependent on the
modulus.

When PENTYP=1, PENN and PENT have units of 1/(Length), and the following is recommended.

• For solid element regions and for edge regions:

PENN=10/L and PENT=PENN/10

where L is a length characteristic of the model.

• For shell element regions:

PENN=105t3/L4 and PENT=PENN/10

where t and L are the thickness and characteristic length of the softer of the 2 contacting surfaces,
respectively.

Following are some additional guidelines when PENTYP=1:

• Very high penalty factors (in excess of 1E4) aren't recommended and will cause numerical problems,
even for flat surfaces with very regular meshes.

• Very low penalty factors (less than 0.01) produce extremely slow convergence rates and aren't
recommended.

When PENTYP=2, PENN and PENT have units of Force/(Length x Area), and the following is
recommended.

• For solid element regions: PENN=10*E/L and PENT=PENN/10

where E is the elastic modulus of the softer of the 2 contacting faces, and L is a length characteristic
of the model.

• For shell element regions: PENN=105Et3/L4 and PENT=PENN/10

where E, t, and L are the elastic modulus, thickness, and characteristic length of the softer of the two
contacting shell surfaces.

The following examples may help you determine a reasonable characteristic length for penalty stiffness
calculations. Flexibility exists in the definition of the length.

19-12 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Penalty Factor Adjustment and Contact Stiffness

Figure 19-1. Example 1 – Use the Radius of the Contacting

Cylinder or the Radius plus the block Thickness as the Length

Figure 19-2. Example 2 – Use the Radius of the Inner or Outer Cylinder as the Length

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 19-13

© 2024 Siemens
 19. Contact

Figure 19-3. Example 3 – Use the Inner or Outer Radius of the Wheel as the Length

Figure 19-4. Example 4 – Use the Radius or Diameter of the Arch as the Length

Figure 19-5. Example 5 – Use the Length of One of the Cantilever Shells as the Length

19-14 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Contact Output Request

Contact Output Request

Contact results can be requested for edge-to-edge and surface-to-surface contact with the BCRESULTS
case control command. The output can be requested as PRINT, PLOT, and PUNCH for ALL grid points, or
for a set (n) of grids defined with the SET command. The available output requests are:

TRACTION: Contact pressure (scalar) and in-plane contact tractions (vector in basic coordinate system)
are output for each contact grid point.

FORCE: Contact force vector is output for each contact grid point.

SEPDIS: The separation distance output (SEPDIS) is a scalar quantity representing the source side normal
distance to the target. During the solution, the separation distance is known at the element integration
points, but is written to the grids when output. The result at each source grid is the value of separation
distance at the closest contact element. If there are two or more contact elements equidistant from the
grid, then the minimum value of separation distance is used at the grids rather than the average, since
the average gives unexpected results for coarse meshes. SEPDIS is only supported by solution 101.

The force output for edge-to-edge contact for axisymmetric elements is “Force / Radian”. The force
output for plane stress and strain elements is “Force”. This is consistent with the applied and reaction
force unit for these elements.

Contact Stiffness Output

You can output the final contact stiffness matrix in DMIG format. Request the contact stiffness matrix
by including the parameter KGGCPCH=1 in a SOL 101 input file. The solver writes the contact stiffness
matrix from the final contact iteration (not necessarily converged) into a PUNCH file in the format
required for DMIG (Direct Matrix Input at Grids).

This option is only available with the default sparse solver. The element iterative solver does not support
this option.

Note:
You can include differential stiffness conditions, which include contact stiffness, in a normal modes
solution (SOL 103) by using the STATSUB case control command.
Similarly, you can also use the new DMIG bulk entry option to add contact stiffness to a normal
modes solution.
The eigenvalues computed in a normal mode solution using the STATSUB option may be different
than the eigenvalues computed using the DMIG option. The STATSUB option includes both the
differential stiffness of the structure and the final contact stiffness, while the DMIG option includes
only the final contact stiffness.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 19-15

© 2024 Siemens
 19. Contact

Contact Preview Output

Simcenter Nastran internally creates contact elements using the contact regions and search distance you
define. Although these contact elements may become inactive as the solution iterates on the contact
conditions, it can be helpful to understand where they are initially created or perhaps not created.

The option to export a bulk data representation of the element edges and faces where contact elements
are created is available. Setting the PREVIEW parameter on the BCTPARM bulk entry to “1” makes the
request. For example:

$*
BCTSET 100 1 20.100000 0.00000.001000
BCTPARM 100 PREVIEW 1
$*

The software writes a bulk data file containing dummy shell element entries for face locations, and
dummy PLOTEL entries for edge locations. Dummy GRID, property and material entries are also written.
You can import the file into a preprocessor to display both source and target contact locations.

The preview file has the following naming convention:

<input_file_name>_cnt_preview_<subcaseid>_<contactsetid>.dat

For example, if an input file named test.dat includes a subcase numbered 101 and a contact set
numbered 201, the resulting preview file name would be:

test_cnt_preview_101_201.dat

At the end of a solution, some contact elements may become inactive because they did not participate
in the converged condition. To view the locations of the final active contact elements, you can request
contact tractions with the BCRESULTS case control command, and then view them in a post processor.

In the following simple example, the red mesh is forced vertically into the green.

19-16 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Contact Preview Output

Before the software applies any loading, the initial, open contact condition is evaluated. This is the point
in the solution in which the preview output is written.

The following illustration shows how the preview appears after you import the preview file into the NX
preprocessor. Note that the colors were manually modified in NX after importing the preview bulk data
file.

The traction results show the final contact condition after importing the results .op2 file.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 19-17

© 2024 Siemens
 19. Contact

Including CGAP Elements in a Contact Solution

When CGAP elements are included with linear contact, the system cell (412) OLDGAPS determines how
CGAPs are treated.

• When a BCSET case control command exists, CGAP elements are treated as linear contact elements
if the system cell (412) OLDGAPS is set to 0 (default). You can use CGAP elements this way with
surface-to-surface contact defined (BCTSET bulk entries exist), or without (no BCTSET bulk entries
exist). In the case where no surface-to-surface contact is defined, “n” on the BCSET case control
command can point to a BCTPARM bulk entry which optionally defines PENN, PENT or PENTYP for the
CGAP/linear contact elements, or to nothing if a BCTPARM bulk entry does not exist (an integer value
for “n” is still required in this case).

If you do not want the CGAP to be treated as a contact element and want it handled as in previous
releases as described in the Simcenter Nastran Element Library, set the system cell OLDGAPS = 1.

• If a CGAP is included in a model where contact is not defined (no BCSET card), it will be treated
as described in the Simcenter Nastran Element Library, regardless of the value of the system cell
OLDGAPS. In other words, the system cell OLDGAPS is only considered when a BCSET case control
command exists.

When CGAP elements are treated as contact elements, some of the properties on the CGAP and PGAP
bulk data entries are used to determine the CGAP/contact element stiffness. For contact in a linear
solution, the CGAP element will use U0, MU1, KA, and KT. The following rules apply to KA and KT:

• If KA is zero or blank, the value for KB is used for the normal stiffness as long as KB is nonzero.

• If KA and KB are both zero and/or blank, the normal stiffness is calculated from the PENN and PENT
fields on the associated BCTPARM. The automatic penalty factor calculation used for surface-to-surface
contact does not apply to CGAP elements.

19-18 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Contact conditions in modal, buckling, and dynamic solutions

• If KA and KB are both zero and/or blank, and PENN and PENT are undefined, the values PENN=10 and
PENT=1 will be used. PENTYP=2 should not be used in this case.

• If KT is zero or blank, the transverse stiffness is calculated from the PENT and PENTYP fields on the
associated BCTPARM.

Contact conditions in modal, buckling, and dynamic solutions

You can include contact stiffness in a normal modes solution (SOL 103), a linear buckling solution (SOL
105), or a dynamics solution (SOLs 108, 111, or 112) by defining multiple subcases.

1. You define a linear statics subcase which includes a contact set definition (BCSET case control
command).

2. You define a subcase for the normal modes, linear buckling, or dynamics computation which
includes a STATSUB command selecting the linear static subcase.

The contact stiffness applied in the normal modes, linear buckling, or dynamics subcase represents the
final contact condition from the linear statics computation around the contact interface. The resulting
contact edges and surfaces should appear attached when you view the normal modes, linear buckling,
or dynamics subcase results.

Note:
Contact conditions can be used with the element iterative solver. However, differential stiffness
conditions cannot be generated with the element iterative solver. Therefore, the default sparse
solver will always be used, even when the element iterative solver is requested.

During the linear statics computation, the software creates contact elements to prevent the contacting
edges and faces from penetrating, and allows finite sliding with optional friction effects. It does this
by augmenting the normal and tangential contact tractions without scaling the contact stiffness. As
a result, the contact stiffness can be adequate for the linear statics subcase although can be low for
the normal modes, linear buckling, or dynamics subcase. For example, you may see contact edges and
surfaces separating if the stiffness is low.

You can use the CNTSCL parameter to scale the contact stiffness for the normal modes, linear buckling,
or dynamics subcase. By default, CNTSCL=100.0, and the contact stiffness is scaled by 100.0. The
CNTSCL parameter must be defined in the bulk data section of the input file.

If you want to revert to the pre-Simcenter Nastran 2312 behavior, you can define CNTSCL=1.0, and the
software will not scale the contact stiffness.

The following examples demonstrate the input requirements for including contact stiffness in SOLs 103,
105, 108, 111 and 112. The examples include a PARAM,CNTSCL,75.0 definition, although this is not
required if you use the default contact stiffness scaling of 100.0.

Normal modes solution example (SOL 103):

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 19-19

© 2024 Siemens
 19. Contact

.....
SOL 103
CEND
SPC=5
$* Linear Statics Subcase
SUBCASE = 1
BCSET = 1
LOAD = 2
$* Normal Modes Subcase
SUBCASE = 2
STATSUB = 1
METHOD = 103
....
BEGIN BULK
PARAM,CNTSCL,75.0
...

Linear buckling solution example (SOL 105):

.....
SOL 105
CEND
SPC=5
$* Linear Statics Subcase
SUBCASE = 1
BCSET = 1
LOAD = 2
$* Buckling Subcase
SUBCASE = 2
STATSUB = 1
METHOD = 1
....
BEGIN BULK
PARAM,CNTSCL,75.0
...

Direct frequency response dynamics example (SOL 108):

.....
SOL 108
CEND
SPC=5
$* Linear Statics Subcase
SUBCASE = 1
BCSET = 1
LOAD = 2
$* Direct Frequency Response Subcase
SUBCASE = 2

19-20 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Contact conditions in modal, buckling, and dynamic solutions

STATSUB = 1
DLOAD = 22
FREQ = 100
....
....
BEGIN BULK
PARAM,CNTSCL,75.0
...

Modal frequency response dynamics example (SOL 111):

.....
SOL 111
CEND
SPC=5
$* Linear Statics Subcase
SUBCASE = 1
BCSET = 1
LOAD = 2
$* Modal Frequency Response Subcase
SUBCASE = 2
STATSUB = 1
METHOD = 1
DLOAD = 22
FREQ = 100
....
BEGIN BULK
PARAM,CNTSCL,75.0
...

Modal transient response dynamics example (SOL 112):

.....
SOL 112
CEND
SPC=5
$* Linear Statics Subcase
SUBCASE = 1
BCSET = 1
LOAD = 2
$* Modal Transient Response Subcase
SUBCASE = 2
STATSUB = 1
METHOD = 1
DLOAD = 22
TSTEP = 1
....
BEGIN BULK

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 19-21

© 2024 Siemens
 19. Contact

PARAM,CNTSCL,75.0
...

Contact Conditions in Linear Buckling (SOL 105)

Contact conditions can be included in a linear buckling solution (SOL 105), although there are important
considerations. The initial linear statics solution determines the differential stiffness (Kσ) and the final
contact stiffness (Kcontact). The final contact stiffness is added to the stiffness matrix for the buckling
subcase:

([K + Kcontact] + λ[Kσ]){D}=0

The buckling load is determined by multiplying λ by the applied load. When contact conditions are
included, this is valid when λ is close to "1.0", or when the contact conditions are insensitive to the
applied load. Although contact conditions can be included in linear solutions, they are iterative and
nonlinear conditions. Because contact conditions are nonlinear, the contact stiffness from the initial
statics solution may be nonlinearly dependent on the applied load. As a result, the contact stiffness
result which was valid for the original loading, may not be valid for the scaled load condition.

The buckling solution will report a warning if the lowest λ is not within 10% of 1.0 (0.9 - 1.1). When
the warning is issued, you can scale your load by λ, and then rerun both the linear statics solution with
contact conditions, and the buckling solution again. You will need to repeat this process until λ is close
1.0.

The inputs for a linear buckling solution with contact conditions require a subcase for the linear statics
subcase and the buckling solution.

In addition, the linear statics subcase must include the BCSET case control command. If the linear statics
subcase is the first subcase, then a STATSUB bulk entry is not needed. If it is not the first subcase, a
STATSUB bulk entry is needed in the buckling subcase to reference the linear statics subcase ID.

Non-converged Contact Option

The contact convergence tolerance is calculated by comparing changes in contact forces between the
two contacting bodies. A contact problem is considered converged when

• The calculated convergence tolerance is less than the value of CTOL (defined on the BCTPARM bulk
entry), and

• The number of contact changes from one iteration to the next is less than the value of NCHG (also on
BCTPARM).

Contact solutions iterate until these conditions have been met, or the number of iterations (MAXS field
on BCTPARM) has been exceeded. In either case, the solution continues, but produces a warning if
convergence does not occur.

19-22 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Contact Conditions with the Element Iterative Solver

The system cell 476 can optionally be used to force a non-converged contact solution to end with an
error.

By default, system cell 476 = 0, and a warning is reported in the *.f06 file for the non-converged contact
solution, the solution continues and results creation occurs.

If system cell 476 = 1, a fatal error is reported in the *.f06 file for the non-converged contact solution,
and the solution ends immediately with no results creation.

In most cases when a small percentage of contact elements fail to converge, they are typically not
sensitive to the quality of the stored results away from the contact regions.

Contact Conditions with the Element Iterative Solver

Contact conditions can be included when using the element iterative solver. The element iterative solver
is requested by including ITER=YES and ELEMITER=YES on the NASTRAN statement.

For example,

$* NASTRAN
$*
NASTRAN ELEMITER=YES, ITER=YES
$*
...

Preloaded bolts, glue conditions, and contact conditions can all be used simultaneously with the
element iterative solver.

You can include converged linear contact conditions in consecutive solutions 103, 111, and 112.
However, differential stiffness conditions cannot be generated with the element iterative solver.
Therefore, the default sparse solver is always used to generate these conditions, even when you request
the element iterative solver. See the section Contact conditions in modal, buckling, and dynamic
solutions for more information.

There are a few other things to consider.

• The penalty factors PENN and PENT on the BCTPARM bulk entry affect the convergence of both
the element iterative solver and the contact iterations. It is recommended to allow the software to
calculate the appropriate penalty factors, which occurs by default. The automatic calculation is turned
off if either of PENN or PENT are defined.

• Avoid using the element iterative solver when there is contact between solid and shell surfaces.
These models typically require reducing the convergence tolerance with the ITSEPS parameter on the
ITER bulk entry to 1.0E-10 or smaller, which usually increases the run time. As a result, the element
iterative solver may be less efficient than the default, sparse matrix solver.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 19-23

© 2024 Siemens
 19. Contact

• Contact between shell surfaces is supported, although the element iterative solver works best with
solid elements.

For the first iteration in each outer loop of the contact algorithm, the iterative solver convergence
tolerance defined with the ITSEPS parameter on the ITER bulk entry is used (default=1e-8). For the
remaining iterations in each outer loop, a tolerance 2 orders of magnitude lower is used since the
displacements being calculated are delta displacements. In addition, a minimum of 20 iterations in the
iterative solver are performed.

For more information on the Iterative Solutions, see:

• The chapter “Iterative Solution of Systems of Linear Equations” in the Simcenter Nastran Numerical
Methods User’s Guide.

• ITER in the Simcenter Nastran Quick Reference Guide.

Contact Conditions with Inertia Relief

Inertia relief is a Simcenter Nastran option that allows you to simulate unconstrained structures in a
static analysis. For example, an aircraft in flight is unconstrained, yet it can be analyzed in a static
analysis. With inertia relief, the mass of the structure resists the applied loadings such that the structure
is in a state of static equilibrium even though it is unconstrained in any or all degrees of freedom.

Contact conditions and inertia relief can be used together in a SOL 101 static solution.

The manual inertia relief is requested by including PARAM, INREL,-1 and the SUPORT bulk entry. The
manual approach allows you to constrain some DOF while leaving others unconstrained.

The automatic inertia relief option is requested by including PARAM,INREL,-2 in the bulk data section.
Since INREL=-2 requires that six rigid body modes exist (a free-free structure), active elements in the
contact source and target regions need to be in close proximity to each other.

If any portion of the elements participating in contact is unconstrained and the initial distance between
the active elements exceeds the distance allowed by small displacement theory, the applied contact
stiffness will produce a grounding effect, and the rigid body modes will no longer exist. This usually
results in all contact elements becoming inactive or in a solution with excessive penetration, which is a
solution in which the contact condition is not enforced.

See the “Inertia Relief in Linear Static Analysis” chapter in the Simcenter Nastran User’s Guide for more
information.

Contact with Static Condensation

Static condensation is an optional Simcenter Nastran method of partitioning and reformulating the
stiffness matrix with the goal of reducing the solution time. This method requires the model be
partitioned into the analysis set (A-set) and omitted set (O-set). In a static analysis, the results using

19-24 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Contact with Static Condensation

static condensation are numerically exact. The partitioned solution merely changes the order of the
operations of the unpartitioned solution.

An automatic static condensation option is available to use with models which include contact
conditions. To select this option, set the parameter CNTASET to “YES”. For example,

PARAM,CNTASET,YES

If you select this option, you should not manually select degrees of freedom with the ASET bulk entry.
When PARAM,CNTASET,YES is defined, Simcenter Nastran automatically places the degrees of freedom
which are part of the contact portion of the solution into the A-set. The remaining degrees of freedom
are then automatically placed into the O-set. The result is that a static condensation is performed to
reduce the full Kgg matrix to the Kaa matrix which contains only the contact degrees of freedom. The
contact iterations are then performed using the resulting Kaa matrix.

This solution option shows the best improvement when the number of contact degrees of freedom is
small compared to the overall number of degrees of freedom in the model. As the reduced A-set stiffness
matrix becomes larger and more dense, the performance benefit decreases.

This option is only supported in a linear statics solution (SOL 101). The reduced mass at the contact
degrees of freedom may not be an accurate representation of the original mass distribution. As a result,
the software will issue a WARNING and continue without the A-set reduction if you attempt to use the
PARAM,CNTASET,YES option in the dynamic solutions 103, 111, or 112.

Because PARAM,CNTASET,NO is the default option, automatic condensation does not occur by default.

Restrictions when PARAM,CNTASET,YES is defined

• If multiple subcases exist, the same contact set and the same constraint set must be used by all
subcases. This can be achieved by including the BCSET and SPC case control commands in the global
subcase. If the subcases use different contact and constraint sets, the software will continue without
the A-set reduction.

• If multiple subcases exist, only a single constraint set can exist, and the SPC case control command
must be in the global subcase.

• Bolt preload conditions are not supported.

• The iterative solver is not supported.

• You cannot select additional degrees of freedom with the ASET bulk entry to include in the Kaa matrix.
The software determines all A-set and O-set degrees of freedom.

• The dynamic solutions 103, 111, or 112 are not supported.

See Understanding Sets and Matrix Operations in the Simcenter Nastran User’s Guide for more
information on static condensation.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 19-25

© 2024 Siemens
 19. Contact

SOL 101 Contact Algorithm

There are several steps in the development of a contact algorithm:

• Kinematic equations describe the relative motion of two contacting surfaces.

• Equilibrium equations with boundary conditions and contact constraints are the basic equations that
are solved.

• To solve the contact problem for general bodies and surfaces, the kinematic and governing equations
are converted into equivalent finite element matrix equations.

• The assembled matrix equations are solved.

Kinematic Equations

Consider the motion of a point on a hitting surface relative to a target surface. At the point on the
hitting surface, the software constructs a set of Cartesian basis vectors, e1, e2, e3. The unit vectors, e1
and e2, are tangent to the hitting surface and the unit vector, e3, is perpendicular to the hitting surface.
The normal vector, e3, will also be designated as n.

The software assumes that the penetration of the hitting point into the target surface occurs at the
target point and is:

p = po + (uH - uT) ∙ n

where

po is the initial penetration determined by the underlying geometry

uH is the motion of the hitting point

uT is the motion of the target point

Note that the penetration is the negative of the gap separation.

In the figure, (A) is the target surface, and (B) is the hitting surface. (C) is the target point, and (D) is the
hitting point.

19-26 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Normal Contact Constraints

For calculating Coulomb friction forces, you must compute the relative tangential displacement
increment, which is given by:

Δut = (ΔuH – ΔuT) – [n ∙ (ΔuH – ΔuT)] ∙ n

This is the tangential component of the relative motion from one load case to the next. The path
dependence of friction is included by applying multiple load cases in sequential order.

Normal Contact Constraints

Contact constraints are imposed at the reduced integration points on the finite element faces of the
hitting contact region. These points are computed as needed during the solution.

The normal contact constraints between the two surfaces are given by:

p≤0

tn = – n ∙t ≥ 0

tnp = 0

• The first equation imposes the condition that the penetration of the hitting surface into the target
surface can't be greater than zero. Thus, surfaces can't interpenetrate.

The contact pressure, tn, is defined as the negative of the normal component of the surface traction.

• The second equation states that the contact pressure can't be less than zero or tensile. Or, normal
tractions between surfaces can't be tensile.

• The last equation imposes the condition that:

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 19-27

© 2024 Siemens
 19. Contact

p = 0 if tn ≥ 0 and tn = 0 if p ≤ 0

Coulomb Friction Contact Constraints

For coulomb friction, additional constraint conditions are required between the hitting and target
surfaces:

• The first equation imposes the constraint that the magnitude of the inplane friction traction, tt, can't
exceed the coefficient of friction μ, times the contact pressure. When the magnitude of friction force
reaches its maximum allowable value, the function Ф will be equal to zero.

• The second equation relates the relative tangential displacement increment between the hitting
surface and the target surface, Δut, to the magnitude of the relative slip increment, Δξ, which must
be a non-negative quantity.

• The final equation implies that if Δξ ≥ 0 (there is slipping between the surfaces) then Φ = 0 and if Δξ =
0 (the surfaces are sticking) then Φ ≤ 0.

Therefore, for Coulomb friction:

• The maximum possible tangential traction equals the coefficient of friction times the normal traction
force.

• Contacting surfaces will "stick" if the tangential traction is less than the coefficient of friction times the
normal traction force.

• Contacting surfaces will "slide" in the direction of the tangential traction if the tangential traction
equals the coefficient of friction times the normal traction force.

Finite Element Kinematic Equations

With the finite element method, the contacting bodies are discretized into elements, and their surfaces
are finite element faces. The contact constraint conditions are imposed at selected points on the
contacting surfaces. For gap elements, the constraint conditions are enforced at nodal points.

19-28 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Finite Element Kinematic Equations

In Simcenter Nastran, a contact element is defined to be a hitting point, a target point, and the
associated nodal degrees of freedom on the hitting and target finite element faces. Other data
associated with contact elements are coefficient of friction and a user defined rigid offset, if any, at
the location of the contact element. You can't directly construct the contact elements; they're formed
within the solver based on your input in the bulk data file.

In terms of finite element degrees of freedom, the penetration at a particular point is given by:

See the following figure:

The interpolation functions depend on the locations of the hitting and target points. Designating the
contact element nodal displacements on the hitting and target faces as the column matrix {u}, this
equation may be written as:

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 19-29

© 2024 Siemens
 19. Contact

where the row matrix, , is composed of interpolation functions times components of the surface
normal. Similarly, the finite element matrix equation for tangential slip is:

For all contact elements, these two equations may be written globally as:

{p} = [Qn]{U} + {Po} and {Δξ} = [Qt]{U} + {ξo}

Global Solution Strategy

There are several possible methods for solving this set of equations. The method used by Simcenter
Nastran is based on an augmented Lagrangian procedure, which offers some advantages over
competing methods. A penalty stiffness is added, but unlike pure penalty methods, the satisfaction
of the contact constraints may be achieved to almost any required accuracy through a series of contact
traction updates. The penalty number can, therefore, be lower than in a pure penalty method, resulting
in better conditioned equations. Pure Lagrange multiplier techniques require that all contacting bodies
be restrained without the presence of contact constraints.

For simplification, the frictionless problem will be considered. The augmented Lagrangian potential
function for the frictionless problem is:

π = 1/2 {U}T[K]{U} - {U}T{F} + {Tn}T([Qn]{U}+{Po}) + 1/2([Qn]{U} + {Po})T[εn]([Qn]{U} +Po})

where {Tn} is the vector of unknown normal tractions (or Lagrange multipliers) at the contact points and
[εn] is a diagonal matrix of normal penalty numbers. Taking the derivative of the potential function with
respect to the displacements and setting it equal to zero while holding the vector of normal tractions
constant produces:

([K]+[Qn]T[εn][Qn]){U} = {F}-[Qn]T({Tn}+[εn]{Po})

The term [Qn]T[εn][Qn] is a penalty stiffness and [Qn]T({Tn}+[εn]{Po}) is an additional force term resulting
from the contact pressure and any initial non-zero penetration. Note that if penetration is zero, the
penalty stiffness times displacement plus the force due to non-zero initial penetration sum to zero. So
when the contact constraints are exactly satisfied, the only thing added to the global equations is the
contact force. In the process of solving this set of equations, the contact pressure is iteratively updated
for each contact element to enforce the zero penetration constraint condition to within a specified
tolerance. The iterative update formula for the normal contact traction at a particular point is given by:

where p is the current value of penetration at the point and εn is the normal penalty number.

19-30 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Summary of Significant Aspects of Contact

The solution of this set of equations uses a double iteration loop. The inner force loop updates the
contact tractions, {Tn}, such that a zero penetration condition is enforced for all active contact elements
(to within a specified tolerance). Once the inner loop is converged, the status of all contact elements is
determined in the outer loop. A contact element is set inactive and removed from the global equation
if a tensile traction is required to close the contact element. If penetration occurs at an inactive contact
element, it's activated and included in the global equations.

The overall solution procedure requires these steps:

1. Form contact elements from user input of contact regions and pairs.

2. Compute stiffness for all elements including penalty stiffness for contact elements.

3. Compute loads for the current load case.

4. Begin contact outer loop iteration. Determine contact element status, depending on penetration
and contact tractions.

5. Assemble contact stiffness for active contact elements into the global stiffness matrix.

6. Begin contact inner loop iteration.

7. Update contact tractions and compute contact forces.

8. Check traction convergence:

a. If tractions have not converged, loop back to step 6.

b. If tractions have converged, loop back to step 4.

c. If tractions have converged and the number of contact status changes is not greater than the
specified allowed number of element status changes, proceed with the next load case.

Summary of Significant Aspects of Contact

1. The software assumes that penetration of the hitting point occurs along the direction of the hitting
face normal. Therefore, wedge-type problems with finite sliding may be poorly simulated using
linear contact analysis because contact elements are created only once using the initial geometry
of the problem.

For example, in the figure, (A) is the assumed contact point; (B) is the actual contact point if the
relative motion of the hitting point with respect to the target is in direction uH.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 19-31

© 2024 Siemens
 19. Contact

When two surfaces are separated by an initial gap and the direction of motion is significantly
different than the surface normal, then significant errors may occur (for example, wrong location
for target face stresses). If the initial gap is zero (or very small), the error may be negligible.

2. The initial penetration, Po, is computed from the FE geometry (isoparametric shape functions)
which can result in kinks or bumps at element edges. These bumps usually cause local stress
concentrations in a contact analysis. If possible, use the INIPENE parameter on the BCTPARM Bulk
Entry to minimize these effects.

3. Gaps may be mixed with surface-surface contact, thus letting you include beams and bars in
contact models. The gap element analysis has been reformulated as an augmented Lagrangian
procedure so that the same solution strategy is used for both contact and gaps.

4. The behavior of the iterative solution is influenced by the penalty numbers (normal and friction).
The defaults provide rapidly converging solutions for most problems. However, when the iterations
don't converge, you may need to adjust the penalty numbers. Some general trends are:

a. The softer the hitting or target surface(s), the smaller the normal penalty number, Pn

b. Larger penalty numbers usually speed convergence, but if penalty numbers are too large, ill
conditioned equations may result in MAXRATIO failures.

c. The friction penalty number is usually 10 to 100 times smaller than the normal penalty. When
many elements are sliding or sticking, a smaller penalty factor may speed convergence.

d. If an unrestrained portion of the model exhibits rigid body motions about a contact region,
reducing the penalty number may stabilize the solution.

The penalty numbers can be adjusted on the BCTPARM Bulk Entry.

19-32 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Summary of Significant Aspects of Contact

5. Contact can remove rigid body motions from the structure. However, use caution because tensile
contact elements are removed as the solution progresses. Therefore, if an insufficient number of
active contact elements are present to prevent rigid body motion, the stiffness matrix will become
singular and the solution will fail (or provide incorrect results).

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 19-33

© 2024 Siemens
 19. Contact

19-34 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 20. Gluing Elements
Overview of Gluing Elements
An option to glue elements together during a solution is available in Simcenter Nastran. Glue is a simple
and effective method to join meshes which are dissimilar. It correctly transfers displacement and loads
resulting in an accurate strain and stress condition at the interface. The grid points on glued edges and
surfaces do not need to be coincident.

Glue creates stiff springs or a weld like connection to prevent relative motion in all directions. For
discussion purposes, the internal elements created by the solver from glue definitions will be referred to
as “glue elements” in the documentation.

Solutions That Support Gluing

You can create glue definitions in all Simcenter Nastran solutions except SOL 401 Balanced Strain
Generation.

Glue Summary

The following table summarizes all supported glue conditions.

Type Description Solution Support

Edge-to-Edge Glue between the edges of 3D shell All solutions except 144 -146.
elements.
In heat transfer solutions, they are
Glue between the edges of the 2D solid treated as constant conductivity
axisymmetric, plane stress, and plane connections. See Glue in heat
strain elements. transfer solutions.
They cannot be used to represent
fluid-to-fluid glue connections in a
coupled acoustic analysis.
Edge-to-Surface Glue between 3D shell element edges All solutions except 144 -146.
and 3D shell or solid element faces.
In heat transfer solutions, they are
treated as constant conductivity
connections. See Glue in heat
transfer solutions.
They cannot be used to represent
fluid-to-fluid glue connections in a
coupled acoustic analysis.
Surface-to-Surface Glue between 3D shell and 3D solid All solutions except for SOL 144–
element faces. 146.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 20-1

© 2024 Siemens
 20. Gluing Elements

In heat transfer solutions, they are

treated as constant conductivity
connections. See Glue in heat
transfer solutions.
They can be used to represent
fluid-to-fluid glue connections in a
coupled acoustic analysis.

Note:
For SOLs 106 and 129, no glue condition updates when large displacements are requested
(LGDISP=1). By default the solution will end with a fatal error when glue conditions and large
displacements are requested. If the glue condition is in a location of the model that will not
experience large displacements, the glue condition should be reasonably accurate. For this case,
you can define SYSTEM(784)=1 to allow the solution to continue.

Defining Glue Conditions

• Select the glue condition with the BGSET case control command. See Defining and Selecting Glue
Pairs.

• Define source and target regions. See Glue Regions.

• Pair the source and target regions. See Defining and Selecting Glue Pairs.

• Optionally adjust the glue algorithm using glue control parameters. See Glue Control Parameters.

• Optionally request glue surface traction output. See Glue Output Request.

Edge-to-Edge Summary

You can define edge-to-edge glue between the edges of the following:

• 3D shell elements CQUAD4, CQUAD8, CQUADR, CTRIA3, CTRIA6, and CTRIAR

• 2D solid elements:

Axisymmetric elements CTRAX3, CQUADX4, CTRAX6, CQUADX8

Plane stress elements CPLSTS3, CPLSTS4, CPLSTS6, CPLSTS8

Plane strain elements CPLSTN3, CPLSTN4, CPLSTN6, CPLSTN8

You create edge regions with either the BEDGE or BLSEG bulk entries, then pair the regions using the
source and target fields on the BGSET bulk entry.

20-2 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Overview of Gluing Elements

Gluing the Edges of Axisymmetric, Plane Stress, and Plane Strain Elements

• The axisymmetric, plane stress, and plane strain elements can be defined in either the XZ plane or in
the XY plane. Edge-to-edge glue is supported in either orientation.

From the source region, the software searches in the outward normal direction. In addition, the
software searches a small distance in the inward normal direction in order to glue edges that may
interfere due to meshing irregularities.

A glue connection occurs when the following is satisfied.

A. When a projected normal hits the target side.
B. The distance between the edges is equal to or less than the search distance which you specify for
the glue pair on the BGSET entry.

The input files gluedg* are included in the installation_path\scnas\nast\tpl directory to demonstrate.

Gluing the Edges of Shell Elements

• A simplistic description of shell element edge-to-edge glue is that the software creates pseudo-faces
along the edges. From the glue points on the source side pseudo-face, the software projects a normal
in the outward normal direction. In addition, the software searches a small distance in the inward
normal direction in order to glue edges that may interfere due to meshing irregularities.

A glue connection occurs when the following is satisfied:

A. When a projected normal hits the pseudo-face on the target side.

B. The distance between the two pseudo-faces is equal to or less than the search distance in which
you specify for the glue pair on the BGSET entry.

The larger the angle is between neighboring shell elements, the less likely that a successful projection
can occur. For example, a glue connection cannot occur if the neighboring shell elements are
perpendicular. In this case, you could consider an edge-to-surface glue connection.

The image below helps to illustrate. The blue line represents the shell elements, and the red line
represents an internally created pseudo-face along the edges in the source and target regions. The
pseudo-face can be visualized by extruding the element edge from -t/2 to +t/2 in the parent element
normal direction.

Edge-to-edge glue definitions are supported in all solutions except SOLs 144 -146. In the heat transfer
solutions, they are treated as conductivity connections. The edges of shell elements that use the plane
strain option defined with MID2=-1 on the PSHELL entry are not supported in edge-to-edge glue.

The input files gluedg* are included in the installation_path\scnas\nast\tpl directory to demonstrate.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 20-3

© 2024 Siemens
 20. Gluing Elements

Edge-to-Surface Summary

• Edge regions are edges of the shell elements CTRIA3, CTRIA6, CTRIAR, CQUAD4, CQUAD8, and
CQUADR.

• Surface regions are the faces of the shell elements CTRIA3, CTRIA6, CTRIAR, CQUAD4, CQUAD8,
CQUADR and the faces of the solid elements CHEXA, CPENTA, CPYRAM, and CTETRA.

You create shell element edge regions with the BEDGE or BLSEG bulk entries, and you create shell or
solid element face regions with the BSURF, BSURFS, BCPROP, or BCPROPS bulk entries. You then pair the
regions using the source and target fields on the BGSET bulk entry. The source region in the pair must be
the edge region. The target region consists of shell or solid element faces.

A simplistic description of edge-to-surface glue is that the software creates pseudo-faces along the
edges in the source region. It then connects these pseudo-faces to the shell or solid faces in the target
region with weld like connections.

From the glue points on the source side pseudo-face, the software projects a normal in the outward
normal direction. In addition, the software searches a small distance in the inward normal direction in
order to glue edges and surfaces that may interfere due to meshing irregularities.

A glue connection occurs when the following is satisfied.

A. When a projected normal hits the face on the target side.

B. The distance between the two is equal to or less than the search distance which you specify for the
glue pair on the BGSET entry.

For example, the green line below represents shell or solid free faces in a target region, the blue line
represents shell elements, and the red line represents an internally created pseudo-face along the edges
in the source region. The pseudo-face can be visualized by extruding the element edge from -t/2 to +t/2
in the parent shell element normal direction.

Edge-to-surface glue definitions are supported in all solution sequences except solutions 144 -146.
In the heat transfer solutions, they are treated as conductivity connections. They cannot be used to
represent acoustic glue connections. The edges of shell elements that use the plane strain option
defined with MID2=-1 on the PSHELL entry are not supported in edge-to-surface glue.

The input files gluedg*.dat are included in the installation_path\scnas\nast\tpl directory to demonstrate.

20-4 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Overview of Gluing Elements

Surface-to-Surface Glue Summary

Surface-to-surface glue can be defined on the faces of the following elements.

• Shell elements CTRIA3, CTRIA6, CTRIAR, CQUAD4, CQUAD8, and CQUADR.

• Solid elements CHEXA, CPENTA, CPYRAM, and CTETRA.

You create shell or solid element face regions with the BSURF, BSURFS, BCPROP, or BCPROPS bulk entries.
You then pair the regions using the source and target fields on the BGSET bulk entry. From elements in
the source region, a normal is projected.

From solid face regions, the software searches in the outward normal direction. In addition, the software
searches a small distance in the inward normal direction in order to glue solid faces that may interfere
due to meshing irregularities.

From shell element face regions, the software searches in both the top and bottom normal directions.

The software creates a glue element if:

• Any of the source element normals intersect with an element in the target region.

• The distance between the two faces is equal to or less than the search distance which you specify for
the glue pair on the BGSET entry.

Surface-to-surface glue definitions are supported in all solution sequences except for SOL 144–146.
In the heat transfer solutions, they are treated as conductivity connections. Surface-to-surface glue
connections can also be defined between acoustic mesh faces. See Glue Conditions in Acoustics
Analysis.

Input files glue*.dat are included in the installation_path\scnas\nast\tpl directory to demonstrate.

Glue in heat transfer solutions

Glue is supported in the heat transfer solutions 153 and 159, and in a SOL 101 heat transfer/structural
combined solution.

The SOL 101 combined solution requires both a heat transfer and a structure subcase. See the HEATSTAT
parameter in the Quick Reference Guide for information. Since these subcases represent different
analysis types and glue has a different meaning in each, to define a unique set of glue parameters
(penalty values) for each subcase, you must create unique glue sets for each subcase. Specifically, you
must create a unique BGSET case control, BGSET bulk entry, and BGPARM bulk entry for each subcase.
You select the same glue regions in each of the glue sets.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 20-5

© 2024 Siemens
 20. Gluing Elements

Glue Regions
Edge Glue Regions

An edge glue region is a collection of shell, axisymmetric, plane strain, and plane stress element edges
in a section of the model where you expect glue to occur. These regions can be created using either the
BEDGE bulk entry or the BLSEG bulk entry, except for SOLs 401 and 402 which only support the BEDGE
entry. The BEDGE entry is recommended for all solutions.

• To define a glue edge using the BEDGE entry, you enter the element ID along with the corner GRID
IDs. You can define the edges on the BEDGE entry in any order.

• We recommend that you use the BEDGE entry over the BLSEG entry. If you decide to use the BLSEG
entry, you enter the GRID IDs in a continuous topological order. You must include both midside and
corner GRID IDs on the BLSEG entry.

An element should not appear more than once in the regions that are part of a glue/contact pair. All
region IDs defined with the BEDGE, BLSEG, BSURF, BSURFS, BCPROP, or BCPROPS entries must be unique.

Surface Glue Regions

A surface glue region is a collection of element free faces in a section of the model where you expect
glue to occur. These regions can be created using the shell elements (BSURF and BCPROP) and using
solid element free faces (BSURFS and BCPROPS).

• The BSURF entry is a list or range of shell element IDs to include in the region.

• The BCPROP entry is a list of shell element property IDs. Shell elements which use any of these listed
property IDs will be included in the region.

• The BSURFS entry is a list of solid element IDs each followed by 3 grid points defining which face of
the 3-D element to include in the glue region.

• The BCPROPS entry is a list of solid element property IDs. The free faces of the solid elements selected
with a property ID are automatically determined by the software.

An element should not appear more than once in the regions that are part of a glue/contact pair. All
region IDs defined with the BEDGE, BLSEG, BSURF, BSURFS, BCPROP, or BCPROPS entries must be unique.

Defining and Selecting Glue Pairs

A glue pair is a way to combine two regions, source and target, in which gluing will be analyzed during
the solution. Use the BGSET bulk entry to define each glue pair. The fields on the BGSET bulk entry are
described as follows:

20-6 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Defining and Selecting Glue Pairs

• GSID (glue set id) will need to match the value of ‘n’ on the BGSET case control entry for the solution
to recognize the glue definition.

• SIDi and TIDi (source and target id's) refer to regions created by the BSURF, BCPROP, BSURFS, BCPROPS,
BEDGE, and BLSEG entries, and are used to define source and target regions respectively for a pair.
For edge-to-surface glue pairs, the BEDGE and BLSEG IDs, which define the edge region, must be the
source region ID, and a shell or solid element face region ID as the target region ID. As many pairs as
desired can be included on a single BGSET bulk entry.

• SDIST (search distance) defines the distance in which the solver can initially determine if the distance
between element edges or faces in a particular pair are within the threshold for creating glue
elements. The default value of SDIST of 10 is large enough to handle most geometry situations,
but can be adjusted as needed. This value is used once, at the beginning of the solution, to determine
where glue elements need to be initially created.

Combining Glue Sets – BGADD

You can optionally define multiple BGSET/BGPARM bulk entry sets, each set with unique glue set IDs
(GSID), and then combine them with a single BGADD bulk entry. The multiple BGSET/BGPARM bulk entry
sets are created to adjust certain glue parameters locally. Glue parameters can also be adjusted globally
with a BGPARM bulk entry having the same GSID as the BGSET case control command.

The following example demonstrates the inputs.

CASE CONTROL
$GSID on the BGSET case control matches GSID on BGADD
BGSET = 108
...
BULK DATA
$Local Glue Set definitions
BGSET 1 1 2
BGSET 2 3 4
5 6
...
$Local Glue Parameters
BGPARM 1 PENN 80 PENT 80
BGPARM 2 PENN 110 PENT 110
...
$Local Glue Sets are combined with BGADD
BGADD 108 1 2
...
$Global Glue Parameters
BGPARM 108 REFINE 0GLUETYPE 1

See the section Glue Control Parameters – BGPARM Bulk Entry in this chapter for more information on
glue parameters.

Gluing Composite Solid Faces

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 20-7

© 2024 Siemens
 20. Gluing Elements

Defining glue regions and pairs on composite solid faces which are perpendicular to the stack direction
(edge faces) may produce poor stress continuity. If the glue definition is between edge faces belonging
to different PCOMPS definitions, and if the number of plies on each PCOMPS definition is small and the
same, and the ply thicknesses are similar, the stress continuity should be fairly smooth. This also applies
to the results requested with the BGRESULTS case control command.

Additional Recommendations

When defining glue regions and pairs on geometry which are not tangent continuous, creating single
glue regions which cross corner transitions can result in non-uniform stress results around the corners. It
is recommended to break these areas into multiple regions and pairs as shown below.

When defining glue regions and pairs, it is recommended to not include the same element face in
multiple regions. In “A” below, an element is repeated in regions 2 and 3. In “B”, the same element only
exists in region 2. “B” is recommended. Repeating element faces multiple times in the same or different
regions can significantly increase memory requirements and degrade performance.

20-8 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Glue Control Parameters

Glue Control Parameters

The glue control parameters on the BGPARM bulk entry (optional) can help you adjust the glue
algorithm. For most solutions, the default settings are appropriate, thus the BGPARM entry is not
required.

• For edge-to-edge glue, the parameters PENTYP, PENGLUE, and PREVIEW are supported.

• For edge-to-surface glue, the parameters PENTYP, PENGLUE, PREVIEW, and ESOPT are supported.

• For surface-to-surface glue, the parameters GLUETYPE, PENTYP, PENN, PENT, PENGLUE, INTORD,
REFINE, and PREVIEW are supported.

Note: See "Glue conditions" in the Multi-Step Nonlinear User's Guide for the SOL 401 and 402 glue
parameter descriptions.

You can optionally define multiple BGSET/BGPARM bulk entry pairs, each pair with unique glue set IDs
(GSID), and then combine them with a single BGADD bulk entry.

The multiple BGSET/BGPARM bulk entry pairs are created to adjust certain glue parameters locally. Glue
parameters can also be adjusted globally with a BGPARM bulk entry having the same GSID as the BGSET
case control command.

Global and local glue parameters have the following definition and rules:

• Global Glue Parameters

The BGPARM bulk entry, which uses the same GSID entered on the BGSET case control command,
defines global parameters.

Any of the parameters on the BGPARM bulk entry can be defined globally. A parameter’s default value
is used if it is not defined globally or locally.

• Local Glue Parameters

The BGPARM bulk entries associated to individual BGSET bulk entries, which are then combined with a
BGADD bulk entry, define local parameters.

The parameters GLUETYPE, PENN, PENT, PENTYP, and PENGLUE can be defined locally. A local
parameter definition overrides a global definition.

Glue Parameter Descriptions

• Following are general definitions for the parameters GLUETYPE, PENTYP, PENN, PENT, and PENGLUE,
along with details below.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 20-9

© 2024 Siemens
 20. Gluing Elements

GLUETYPE – Selects the glue formulation for surface-to-surface glue. (Default=2) Edge-to-surface
and edge-to-edge glue pairs always use GLUETYPE=2.

GLUETYPE = 1 - Normal and tangential springs will be used to define the connections.

GLUETYPE = 2 - A “weld like” connection will be used to define the connections.

PENTYP – Changes how glue element stiffness and “conductance” are calculated. (Default=1)

PENN – Penalty factor for normal direction when GLUETYPE=1. (Default=100)

PENT – Penalty factor for transverse direction when GLUETYPE=1. (Default=100)

PENGLUE – Penalty factor when GLUETYPE=2. (Default=1)

When gluing acoustic faces, GLUETYPE, PENTYPE, and PENT are all ignored. PENN is described in the
section “Glue Conditions in Acoustics Analysis”.

For structural solutions and for heat transfer solutions, GLUETYPE has unique penalty factor inputs.
These inputs and units are described below.

GLUETYPE=2 The glue penalty stiffness is defined by the PENGLUE penalty factor.
(default)
Structural solutions: PENTYP=1 PENGLUE is a unitless value (glue stiffness scale factor).
PENTYP=2 PENGLUE has the units of F/L2.
Heat transfer Always uses GLUETYPE=1.
solutions:

GLUETYPE=1 The glue penalty stiffness is defined by PENN and PENT.

(Heat transfer solutions always uses GLUETYPE=1)
PENTYP=1 Structural solutions: PENN and PENT have units of 1/
(length), and the glue element stiffness is calculated by
K = e*E*dA where e represents PENN or PENT, E is an
average modulus, and dA is area. See Average modulus
computation below. A physical interpretation is that it is
equivalent to the axial stiffness of a rod with area dA,
modulus E, and length 1/e.
Heat transfer solutions: PENT is ignored. PENN has
the units of 1/(length), and “conductance” at the glue
connection is calculated as
C = e*kavg*dA*100
where e represents PENN, kavg is an average of the thermal
conductivity (k) values for all source side elements in a
pair, and dA is area. A physical interpretation is that it is
equivalent to the axial "conductance" of a rod with area dA,

20-10 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Glue Control Parameters

conductivity kavg, and length 1/e. See Glue in heat transfer

solutions.
PENTYP=2 Structural solutions: PENN and PENT become a spring rate
per area Force/(Length x Area), and the glue element
stiffness is calculated as K=e*dA. The spring rate input is
a more explicit way of entering glue stiffness since it is not
dependent on the average modulus.
Heat transfer solutions: PENT is ignored. PENN has
the units of (thermal conductivity*length)/area, and the
“conductance” at the glue connection is calculated as C =
e*dA, where e represents PENN, and dA is the area. See
Glue in heat transfer solutions.

For glued coincident faces, there is little flexibility between the faces with default penalty factors.
Regardless of which GLUETYPE is used, the glue condition created between non-coincident faces
will not usually produce a local stiffness as accurate as using a conventional finite element for the
connection. The flexibility in the glue condition will depend on the GLUETYPE used, and the value of
the penalty factors. If you have non-coincident faces and the glue joint flexibility is important, then it
is recommended that you model this connection with conventional finite elements.

Glue in heat transfer solutions

Glue is supported in the heat transfer solutions 153 and 159, and in a SOL 101 heat transfer/structural
combined solution.

The SOL 101 combined solution requires both a heat transfer and a structure subcase. See the
HEATSTAT parameter for information. Since these subcases represent different analysis types and glue
has a different meaning in each, to define a unique set of glue parameters (penalty values) for each
subcase, you must create unique glue sets for each subcase. Specifically, you must create a unique
BGSET case control, BGSET bulk entry, and BGPARM bulk entry for each subcase. You select the same
glue regions in each of the glue sets.

Average modulus computation

• Generally, the modulus is averaged for the elements associated with the source side region.

• For elements using orthotropic and anisotropic materials, the element modulus value used in the
average modulus computation is computed as follows.

■ MAT2:
E = E' (1 - PR*PR)
where E' = (C11 + C22) / 2
PR = C12 / E'
and Cij are the values defined on the MAT2 entry.

■ MAT8:
E = (E1 + E2) / 2

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 20-11

© 2024 Siemens
 20. Gluing Elements

■ MAT9:
E = (E11 + E22 + E33) / 3
where Eii are the reciprocals of the first three diagonal terms of the compliance matrix obtained
by inverting the modulus matrix. The Cij defined on the MAT9 entry make up the upper right
triangle of the modulus matrix.

■ MAT11:
E = (E1 + E2 + E3) / 3

• For an element defined with a PSHELL entry, the modulus of the material selected in the MID2 field
on the PSHELL entry is used in the average modulus computation. If the MID2 field is undefined, the
material selected in the MID1 field is used.

• For elements defined as a shell composite defined with a PCOMP or PCOMPG entry, the software
creates a PSHELL entry which references software created MAT2 entries. The software created
PSHELL and MAT2 entries represent properties equivalent to the composite. The MAT2 entry
referenced by the MID2 field on the software created PSHELL entry is used to compute the average
modulus. This computation is the same as for the MAT2 described above.

• When a solid composite defined with the PCOMPS entry is included in a source glue region, the
average modulus is computed by averaging all modulus values defined on all materials defined in
the input file.

• INTORD and REFINE help to improve the accuracy of the glue solution. The number of locations where
normals are projected (glue points) from the source region is dependent on the value assigned to the
INTORD parameter, and on the element face type. The following table summarizes how the INTORD
value adjusts the number of glue points for a particular element face:

Number of Glue Points Used in Glue Element Evaluation

Face Type INTORD=1 INTORD=2 (default) INTORD=3
Linear Triangle 1 3 7
Parabolic Triangle 3 7 12
Linear Quad 1 4 9
Parabolic Quad 4 9 16

REFINE will increase the number of glue points by refining the mesh on the source region. Part of the
refinement process is to project element edges and grids from the associated target region back to the
source region. The resulting refinement on the source region is then more consistent with the target
side, which then gives a better distribution of glue elements. The refined grids and elements are only
used during the solution. The glue results are transferred back to the original mesh for post processing
results.

Refinement occurs when set to 2 (default). REFINE=0 turns the refinement off.

• You can use the PREVIEW parameter to request the export of a bulk data representation of the
element edges and faces where glue elements are created. See Glue Preview.

20-12 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Glue Conditions in Acoustics Analysis

• You can use the ESOPT parameter to change how the edge-to-surface glue stiffness is distributed on
the surface being glued. See Edge-to-Surface Glue Stiffness.

Glue Conditions in Acoustics Analysis

Surface-to-surface gluing can be used between acoustic mesh faces. The acoustic mesh, which uses the
solid elements CHEXA, CPENTA, CPYRAM, and CTETRA, can be a continuous 3-D mesh, or it can consist
of several independent meshes with adjacent acoustic faces glued. Similar to structural glue conditions,
the meshes between the acoustic glued faces can be dissimilar.

The acoustic-to-acoustic glue inputs are consistent with the structural-to-structural capability, except on
the BGPARM bulk entry, GLUETYPE, PENTYPE, and PENT are all ignored. PENN is used to calculate the
acoustics penalty matrix K:

where e is PENN, ρ is the average density of all fluid elements in the model, and dA is the surface area.
The K matrix for an acoustic element is defined by

For a small fluid column (tube of length L and cross-sectional area dA), the K matrix can be written as

Therefore, the penalty factor PENN (e) can be interpreted as 1/L.

Also see Simcenter Nastran Acoustics User's Guide for details on acoustics.

The input file glueac107a.dat is included in the installation_path\scnas\nast\tpl directory to

demonstrate.

Glue Output Request

An option to recover glue surface tractions for solid and shell elements is available. The BGRESULTS case
control command is used to request the glue traction output. The glue traction request is supported in
a linear static solution (SOL 101), in a normal modes solution (SOL 103) but only for a static preload
subcase if present, and in a linear buckling solution (SOL 105). The glue results from SOL 105 are a
result of the applied static loads which are not necessarily the loads at which buckling occurs.

The glue tractions, which are similar to the contact results, are calculated and stored at the grids
which are on the glue surfaces. The normal component of the tractions is a scalar while the in-plane

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 20-13

© 2024 Siemens
 20. Gluing Elements

(tangential) tractions are output in the basic coordinate system. For an edge-to-edge and edge-to-
surface glue pair, only forces and not tractions are recovered. Glue force recovered on axisymmetric
element edges is force per unit radian, which is consistent with applied force and reaction force for
these elements.

The output can be requested as PRINT, PLOT, and PUNCH for ALL grid points, or for a set (n) of grids
defined with the SET command.

Glue stiffness output

You can output the glue stiffness matrix in DMIG format. In your input file which includes a glue
definition, you can request the glue stiffness matrix by including the following parameter input.

PARAM,KGGLPCH,1

The solver writes the glue stiffness matrix into a PUNCH file in the format required for DMIG (Direct
Matrix Input at Grids). You can request the glue stiffness matrix in any solution except 144-146, and
401. This option is only available with the default sparse solver. The iterative solvers do not support this
option.

To include the glue stiffness in a consecutive solution:

• You include the K2GG case control command to select the DMIG entry.

...
Case Control
k2gg=KGGG
...

• You will include the DMIG in the bulk data section of your consecutive input file. You can optionally
use the INCLUDE entry. For example,

...
begin bulk
...
include name_of_dmig_punch_file.pch
...

Glue preview
Simcenter Nastran internally creates glue elements using the glue regions and search distance you
define. You can optionally export a bulk data representation of the element edges and faces where
glue elements are created. To do so, set the PREVIEW parameter on the BGPARM bulk entry to “1”. For
example:

20-14 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Glue preview

$*
BGSET 100 1 2 0.001000
BGPARM 100 PREVIEW 1
$*

The software writes a bulk data file containing dummy shell element entries for face locations, and
dummy PLOTEL entries for edge locations. Dummy GRID, property, and material entries are also written.
You can import the file into a preprocessor to display both source and target glue locations.

To view the glue locations, glue tractions can also be requested in solutions 101, 103, and 105. In the
solutions in which glue tractions are not supported, you can use the preview output.

The preview file has the naming convention:

<input_file_name>_glue_preview_<subcaseid>_<gluesetid>.dat

For example, if an input file named test.dat includes a subcase numbered 101 and a glue set numbered
201, the resulting preview file name is:

test_glue_preview_101_201.dat

In the following simple example, the red mesh is glued to the green.

Before applying any loading, the software creates the glue elements. This is the point in the solution in
which the preview output is written.

The following illustration shows how the preview appears after you import the preview file into the NX
preprocessor. Note that the colors were manually modified in NX after importing the file.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 20-15

© 2024 Siemens
 20. Gluing Elements

Edge-to-Surface Glue Stiffness

Shell theory does not account for changes in shell thickness or normal strains perpendicular to the plane
of the shell element. The ESOPT parameter gives you a choice of how to handle the linking of the zero
normal strains which exist in the shell element associated with the edge (colored blue in the example),
to the surface being glued (colored green in the example).

Either of the ESOPT options properly transfers all forces and moments through the glue connection. The
resulting difference between these options is the local strains on the surface being glued.

ESOPT=0 (default) The strains in the plane of the surface being glued in the direction
perpendicular to the edge are not constrained by the glue stiffness.

ESOPT=1 The strains in the plane of the surface being glued on the area
corresponding to the shell element thickness (t/2) in the direction
perpendicular to the edge, are constrained by the glue stiffness. This is
the glue stiffness behavior which existed before version 8.5.

20-16 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Edge-to-Surface Glue Stiffness

The following example illustrates the ESOPT options. It includes an equal compressive pressure load on
the top and bottom of the solid element mesh. The edge of the blue shell mesh is glued to the face of
the green solid mesh.

When ESOPT=0 (default), the deformations in the solid mesh demonstrate that the glue stiffness did
not prevent local strain at the glue interface.

When ESOPT=1, the deformations in the solid mesh demonstrate that the glue stiffness did prevent local
strain at the glue interface.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 20-17

© 2024 Siemens
 20. Gluing Elements

20-18 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 21. Plotting
Overview of Plotting in Simcenter Nastran
Simcenter Nastran can generate the following types of plots:

• X-Y graphs of transient response or frequency response.

• V-F and V-G graphs for flutter analysis.

X-Y Plots

Requests for structure plots or X-Y plots are made in the Case Control Section by submitting a structure
plot request or an X-Y output request.

The optional PLOTID command is considered to be part of the plot request although it must precede any
OUTPUT(PLOT), OUTPUT(XYOUT) or OUTPUT(XYPLOT) commands.

Plot requests are separated from Case Control by the OUTPUT(PLOT), OUTPUT(XYPLOT) or
OUTPUT(XYOUT) commands. Data above this command (except PLOTID) will not be recognized by the
plotter, even though it may have the same name (for example, the SET command).

See also

• “X-Y PLOT Commands” in the Simcenter Nastran Quick Reference Guide

• “PLOTID” in the Simcenter Nastran Quick Reference Guide

Flutter Analysis Plots

In Simcenter Nastran, you can use the aeroelasticity capabilities to perform a flutter analysis to
determine the stability of an aeroelastic system. You can then create graphs of your results.

See also

• “Flutter Analysis” in the Simcenter Nastran Aeroelastic Analysis User's Guide

Superelement Plotting
Plotting can occur at four different places in the superelement solution sequences. There are two plot
commands, SEPLOT and SEUPPLOT, used with other Case Control and PARAM commands to control the
type of plot to be prepared. A flow chart of the solution process is given in Figure 21-1, Figure 21-2,
and Figure 21-3.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 21-1

© 2024 Siemens
 21. Plotting

Undeformed structure plots are made early during Phase 0 in the structured solution sequences and
during Phase l in the unstructured solution sequences. They can be made for either one superelement
only, or for a superelement and all of its upstream members, as controlled by PARAM,PLOTSUP (see
Figure 21-1 or Figure 21-2). Plots are made for superelements selected by the SEMG command
(see Figure 21-2, block I), and listed in a plot request headed by SEPLOT SElD, where SEID is the
superelement identification number. In the structured solution sequences, the SEMG command is not
required and all superelements which appear on SEPLOT commands will be plotted (see Figure 21-1,
block 0). Undeformed plots can be used to check geometry and connectivity, and do not require the
presence of property or material commands. In the unstructured solution sequences, a branch to the
end of the loop immediately after the plot module can be made by using the PARAM,PLOT,-1 command.

In dynamic analysis, solution set XY-plots are requested by use of SEPLOT 0 (see Figure 21-3, block II).
In the data recovery phase, XY-plots and deformed structure plots for elements in one superelement
only are requested by the SEPLOT command (see Figure 21-3, block III). Deformed structure plots for a
superelement and all its upstream plots are requested with the SEUPPLOT command (see Figure 21-3,
block IV).

XY-plots and deformed structure plots are regarded as output requests and will result in automatic
execution of the data recovery loop as is required to produce the plots requested. For example, the
command SEUPPLOT 0 will result in data recovery being performed on the entire model, even in the
absence of any other output requests. For SEUPPLOT requests, if SUBCOM or SYMCOM subcases are
used, each superelement must have identical SUBCOM and SYMCOM structure.

21-2 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Superelement Plotting

Figure 21-1. Phase 0 and I Superelement Plot Control In the Structured Solution Sequences

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 21-3

© 2024 Siemens
 21. Plotting

Figure 21-2. Phase l Superelement Plot Control In the Unstructured Solution Sequences

21-4 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Superelement Plotting

Figure 21-3. Phase II, Ill, and IV Superelement Plot Control in All Solution Sequences

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 21-5

© 2024 Siemens
 21. Plotting

Post Processors
The plotps post processor reads plotting commands from a single Simcenter Nastran binary or neutral
format plot file and produces a file which can be printed or viewed on a PostScript device. Each image
will be oriented to best correlate the plot's page size with the printable page size (assumed to be 7.5
inches wide and 10.0 inches tall). Command line arguments can be specified in any order, command
keywords can be abbreviated by truncation.

On Linux systems, you can omit the plot file type (‘.plt' for binary files and ‘.neu' for neutral files).

Examples

• Translate a binary format plot file into PostScript:

plotps example.plt

• Translate a neutral format plot file into PostScript:

plotps example.neu

See also

• “PLOTPS” in the Simcenter Nastran Installation and Operations Guide

21-6 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 22. Linear Buckling
Overview of Linear Buckling
In linear static analysis, a structure is normally considered to be in a state of stable equilibrium. As
the applied load is removed, the structure is assumed to return to its original position. However, under
certain combinations of loadings, the structure may become unstable. When this loading is reached,
the structure continues to deflect without an increase in the magnitude of the loading. In this case,
the structure has actually buckled or has become unstable; hence, the term “instability” is often used
interchangeably with the term “buckling.”

Only linear buckling or elastic stability is considered in the discussion which follows; in other words,
assume there is no yielding of the structure and the direction of the forces do not change (i.e., follower
force effects are ignored). Other assumptions of elastic stability are discussed in “Linear Buckling
Assumptions and Limitations.” For a description of follower forces, see the chapter “Follower Stiffness”
in this User's Guide.

This chapter is organized into the following sections:

• Finite element approach

• Eigenvalue extraction method

• Assumptions and limitations of linear buckling analysis

• Examples

Finite Element Approach

In finite element analysis, the problem of linear buckling is addressed by including the effect of the
differential stiffness to the linear stiffness matrix. The differential stiffness results from including the
higher-order terms of the strain-displacement relationships. From a physical standpoint, the differential
stiffness matrix represents the linear approximation of softening (reducing) the linear stiffness matrix in
the case of a compressive axial load, and stiffening (increasing) the linear stiffness matrix in the case of a
tensile axial load. Since the differential stiffness either adds or subtracts from the linear stiffness, often it
is also called the incremental stiffness. The element linear and differential stiffness matrices are denoted
as [ka]i and [kd]i , respectively.

The differential stiffness matrix is a function of the geometry, element type, and applied loads. A look
at the differential stiffness matrix of a single planar bar element as shown in Figure 22-1 shows how a
linear buckling analysis is handled using a finite element approach. For clarity, only the y and θz degrees
of freedom at each end are retained for this example.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 22-1

© 2024 Siemens
 22. Linear Buckling

Figure 22-1. Single Planar Bar Element

From The Nastran Theoretical Manual, the differential stiffness for this planar bar element can be
represented as

Equation 22-1.

where Fxi is the axial force in the CBAR element. In this case, Pa (applied load) = Fxi since there is only
one element in the model and the applied load is in line with the element axis. In the general case, Fxi
is proportional to Paas long as the structure remains linear; in other words, if Pa is increased by scale
factor ai , then Fxi also increases by the same scale factor. The value “i” stands for the i-th element. Note
also that the differential stiffness matrix is dependent only on the element type, the applied forces, and
the geometry of the structure. This is the reason why the differential stiffness is also often called the
geometric stiffness matrix.

Eq. 22-1 can, therefore, be rewritten as follows:

22-2 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Finite Element Approach

Equation 22-2.

One can view αi as the distribution factor of the applied load to the i-th element. Each element in the
structure that supports differential stiffness (see “Linear Buckling Assumptions and Limitations”) has
an element differential stiffness matrix associated with it. Each of these differential stiffness matrices has
a scale factor similar to Eq. 22-2. In general, each αi is different for each element in the structure. The
value of each αi depends on the element type, the orientation of the element relative to the overall
structure, and the applied load.

The system linear stiffness matrix can then be represented as

Similarly, the system differential stiffness matrix can be represented as

In general, the individual [kd]i is more complicated than Eq. 22-1; however, the concept is the same. The
overall system stiffness matrix is represented by Eq. 22-3.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 22-3

© 2024 Siemens
 22. Linear Buckling

Equation 22-3.

The total potential energy is equal to

Equation 22-4.

In order for the system to achieve static equilibrium, the total potential must have a stationary value; in
other words, the relationship in Eq. 22-5 must be satisfied:

Equation 22-5.

where ui is the displacement of the i-th degree of freedom.

It is convenient to rewrite Eq. 22-5 as

Equation 22-6.

where and Pa is the applied load. In order for Eq. 22-6 to have a non-trivial solution, the
following relationship must be true:

22-4 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Finite Element Approach

Equation 22-7.

where || stands for the determinant of the matrix. Eq. 22-7 is only satisfied for certain values of Pa. These
values of are the critical buckling loads.

A real structure has an infinite number of degrees of freedom. The finite element model approximates
the behavior of the structure with a finite number of degrees of freedom. The number of buckling loads
obtainable for your finite element model is equal to the number of degrees of freedom of your model. In
other words,

Equation 22-8.

Eq. 22-7 can, therefore, be rewritten as

Equation 22-9.

Eq. 22-9 is in the form of an eigenvalue problem. Once you obtain the eigenvalues λi, the buckling loads
can then be obtained using Eq. 22-8. The values λi are the scale factors by which the applied load Pa is
multiplied to produce the critical buckling loads Pcri. As you can see from Eq. 22-8, the magnitude of the
applied load Pa is arbitrary for arriving at the correct Pcri . As an example, if Pa is increased by a factor of
10, then the calculated λi values in Eq. 22-9 are reduced by a factor of 10; in other words, their resulting
products Pcri remain the same.

In general, only the lowest buckling load is of any practical interest. The structure will fail prior to
reaching any of the higher buckling loads.

Solution 105

In Simcenter Nastran you can solve a linear buckling problem by using Solution 105 and following the
procedure listed below.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 22-5

© 2024 Siemens
 22. Linear Buckling

1. Apply the static loads to the first n subcases (n is usually equal to one) and treat them as static
analysis. The distribution of element forces due to these applied loads is generated internally. The
actual magnitude of these applied loads is not critical.

2. You can perform buckling analysis on any or all of the loading conditions used in Step 1. One
additional subcase is needed for each buckling analysis.

3. The n+1 to the n+m subcases must each request an eigenvalue method from the Bulk Data Section
to solve the eigenvalue problem shown in Eq. 22-9. In this case, m is equal to the number
of buckling analyses that you want to perform. Each buckling subcase may call out a unique
eigenvalue solution.

4. The differential stiffness matrix is automatically generated for each element that supports
differential stiffness. See “Linear Buckling Assumptions and Limitations” for a list of elements
that support differential stiffness.

5. You must then multiply the eigenvalues obtained in Step 3 by the appropriate applied loads to
obtain the buckling loads (Eq. 22-8) for each buckling analysis.

6. Each subcase may have a different boundary condition.

A typical input file used to calculate the buckling loads is shown in Listing 22-1. In most applications,
only one static and one buckling analysis is performed per run. Example 6 in “Buckling Examples”
contains an application of multiple static and buckling analyses.

$
SOL 105
TIME 10
CEND
TITLE = SAMPLE INPUT FOR BUCKLING ANALYSIS
SPC = 10
DISP = ALL
$
$ STATIC SUBCASE TO GENERATE INTERNAL ELEMENT FORCES
$
SUBCASE 1
LOAD = 10
$
$ EIGENVALUE CALCULATION TO OBTAIN SCALE FACTORS BY WHICH
$ THE APPLIED LOAD(S) IS MULTIPLIED BY
$
SUBCASE 2
METHOD = 20
$
BEGIN BULK
$

22-6 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Eigenvalue Extraction Methods

$ APPLIED LOAD IN THE STATIC SUBCASE - SUBCASE ONE

$
FORCE,10,100,,-100.,1.0,0.,0.
$
$ LANCZOS EIGENVALUE METHOD REQUESTING THE LOWEST EIGENVALUE
$
EIGRL,20,,,1
$
$ BRING IN THE REST OF THE BULK DATA ENTRIES
$
include 'bulk.dat'
$
ENDDATA

Listing 22-1. Input File for a Typical Buckling Analysis

You should also adhere to the following guidelines when solving a linear buckling problem in Simcenter
Nastran.

• The Case Control Section must contain at least two subcases.

• A METHOD command must appear in each buckling subcase to select the appropriate eigenvalue
extraction method (EIGRL or EIGB entry) from the Bulk Data Section. If there are two or more buckling
subcases, then a STATSUB = x command must be placed in each buckling subcase to select the
appropriate static subcase. If there is only one buckling subcase, then the STATSUB = x command is
optional if x references the first static subcase ID.

• A static loading condition must be defined with a LOAD, TEMP(LOAD), or DEFORM selection unless
all loading is specified by grid point displacements on SPC entries. All static subcases must be placed
before the first buckling subcase.

• Correct SPC sets must be selected for all subcases.

• Output requests that apply to only a particular subcase must be placed inside that subcase.

• Output requests that apply to all subcases may be placed above the subcase level.

• Some type of output request (e.g., disp(plot) = all) must be requested for at least one subcase.

“Buckling Examples” contains six example problems that illustrate some of these guidelines.

Eigenvalue Extraction Methods

A number of methods of real eigenvalue extraction are available in Simcenter Nastran:

• Householder

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 22-7

© 2024 Siemens
 22. Linear Buckling

• Modified Householder

• Enhanced inverse power

• Lanczos

Two of these methods can be used for linear buckling analysis:

• Enhanced inverse power

• Lanczos

A brief description of each eigenvalue method is presented in the following sections. For further details,
see the Simcenter Nastran Numerical Methods User's Guide.

Enhanced Inverse Power Method (SINV)

This method uses Sturm sequence logic to ensure that all modes are found within the specified
eigenvalue range. The Sturm sequence informs you of the number of modes below each trial
eigenvalue. See the Simcenter Nastran Numerical Methods User's Guide and the Simcenter Nastran
Basic Dynamic Analysis User's Guide for further details regarding the Sturm sequence check.

Lanczos Method

The Lanczos method overcomes the limitations and combines the best features of the other methods.
It is efficient, and if an eigenvalue cannot be extracted within the range that you specify, a diagnostic
message is issued. This method computes accurate eigenvalues and eigenvectors. Furthermore, it prints
meaningful user diagnostics and supports parallel processing computers.

Comparison of Methods

The best method for a particular model depends on four factors:

• The size of the model (the total number of degrees of freedom)

• The number of eigenvalues desired

• The available real memory on your computer

• How well you can estimate the range of your eigenvalues

The enhanced inverse power method (SINV) can be a good choice if the model is too large to fit into
memory, only a few modes are needed, and you have a reasonable idea of your eigenvalue range of
interest. It is useful for models in which only the lowest few modes are desired. This method is also
useful as a backup method to verify the accuracy of other methods.

22-8 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 User Interface

For medium to large models, the Lanczos method is the recommended method. Furthermore, the
Lanczos method takes full advantage of sparse matrix methods that can substantially increase
computational speed and reduce disk space usage. For overall robustness, the Lanczos method is the
recommended method.

User Interface

A METHOD command is required in a subcase of the Case Control Section to select the appropriate
eigenvalue extraction method in the Bulk Data Section. The Bulk Data entry is different depending on
whether you are using the enhanced inverse power (SINV), or Lanczos method. The EIGRL entry is used
for the Lanczos method, and the EIGB entry is used for the SINV methods.

EIGRL Format

The EIGRL entry has the following format:

1 2 3 4 5 6 7 8 9 10
EIGRL SID V1 V2 ND

The SID field is the set identification number, which is referenced by the METHOD command in the Case
Control Section. The V1 field defines the lower eigenvalue bound λl, and the V2 field defines the upper
eigenvalue bound λu. The ND field specifies the number of roots desired. Table 22-2 summarizes the
action Simcenter Nastran takes depending on the values specified for V1, V2, and ND.

Table. Number and Type of Roots Found with EIGRL Entry

Case V1 V2 ND Number and Type of Roots Found

1 x x x Lowest ND or all in range, whichever is smaller
2 x x All in range
3 x x Lowest ND in range [V1, + ∞]
4 x Lowest root in range [V1, + ∞]
5 x Lowest ND roots in [-∞, +∞]
6 Lowest root
7 x x Lowest ND roots below V2
8 x All below V2

EIGRL Example

1 2 3 4 5 6 7 8 9 10
EIGRL 10 1

The above example selects the Lanczos method requesting the lowest root for your model.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 22-9

© 2024 Siemens
 22. Linear Buckling

In order for this entry to be used, the METHOD = 10 command must be specified in the Case Control
Section.

EIGB Format

The EIGB entry has the following format:

1 2 3 4 5 6 7 8 9 10
EIGB SID METHOD L1 L2 NEP

The SID field is the set identification number, which is referenced by the METHOD command in the Case
Control Section. The METHOD field selects the desired eigenvalue extraction method (SINV). The L1 field
defines the lower eigenvalue bound λl, and the L2 field defines the upper eigenvalue bound λμ.

EIGB Example

1 2 3 4 5 6 7 8 9 10
EIGB 10 SINV 0.5 1.2

The above example selects the enhanced inverse power method. All eigenvalues between 0.5 and 1.2
are desired. In order for this entry be used, the METHOD = 10 command must be specified in the Case
Control Section.

Linear Buckling Assumptions and Limitations

The following assumptions and limitations apply to linear buckling analysis:

1. The deflections must be small.

2. The element stresses must be elastic.

3. The differential stiffness is supported for the following elements: CONROD, CROD, CTUBE, CBAR,
CBEAM, CBEND, CQUAD4, CQUAD8, CQUADR, CTRIA3, CTRIA6, CTRIAR, CSHEAR, CTRAX3, CTRAX6,
CQUADX4, CQUADX8, CHEXA, CPENTA, CPYRAM, and CTETRA.

4. A minimum of five grid points per half sine wave (buckled shape) is recommended.

5. The distribution of the internal element forces due to the applied loads remains constant.

6. Offsets should not be used on plate or shell elements (except CQUADR/CTRIAR) for buckling
analysis.

7. For 3-D buckling problems, the use of PARAM,K6ROT is recommended for CQUAD4 and CTRIA3
elements. A value of 100 is recommended.

22-10 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Buckling Examples

8. For structures that exhibit nonlinear material or large deflection deformations, the linear buckling
load obtained from Solution 105 may be different than the actual buckling load. For structures with
significant nonlinearities, it is recommended that you perform a nonlinear buckling analysis using
Solution 106.

See also

• “Performing Nonlinear Buckling Analysis” in the Simcenter Nastran Basic Nonlinear Analysis User's
Guide

Buckling Examples
This section contains six example problems pertaining to buckling. Each problem is chosen to illustrate
either a particular type of failure mode, the effect of an element type, the effect of a structural
configuration, or the method of performing multiple buckling analyses in a single run. The examples
are as follows:

1. Classical Euler beam buckling.

2. Lateral buckling.

3. Planar frame buckling.

4. Buckling of a stiffened panel with transverse shear flexibility.

5. Buckling of a cylinder under uniform axial load.

6. Multiple buckling analyses in a single run.

In each example, the results produced by Simcenter Nastran are compared with known theoretical
solutions.

Example 1 – Classical Euler Beam Buckling

Consider a classical Euler buckling problem. This problem is shown in Figure 22-2.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 22-11

© 2024 Siemens
 22. Linear Buckling

Figure 22-2. Euler Beam

The initial buckling occurs in the xz plane since I1 is smaller than I2.

Simcenter Nastran Results for Example 1

The input file used for this problem is shown in Listing 22-2. In this case, a 1000 N compressive load is
applied at the roller end to generate internal loads for the structure. The Lanczos eigenvalue extraction
method is used to request the first five modes. Either the CBAR or CBEAM elements can be chosen for
this problem; the CBAR elements are used in this case.

$ FILENAME - bukeuler.dat
ID EULER BEAM
SOL 105
TIME 10
CEND
$
TITLE = EULER BEAM
SUBTITLE = METRIC UNITS
SPC = 10
DISP = ALL
$
SUBCASE 1
LOAD = 10

22-12 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Simcenter Nastran Results for Example 1

$
SUBCASE 2
METHOD = 10
$
BEGIN BULK
$
PARAM POST 0
EIGRL 10 5
FORCE 10 11 -1000. 1. 0.
0.
$
CBAR 1 1 1 2 100
CBAR 2 1 2 3
100
CBAR 3 1 3 4
100
CBAR 4 1 4 5
100
CBAR 5 1 5 6
100
CBAR 6 1 6 7
100
CBAR 7 1 7 8
100
CBAR 8 1 8 9
100
CBAR 9 1 9 10
100
CBAR 10 1 10 11
100
$
GRID 1 0. 0.
0.
GRID 2 1. 0.
0.
GRID 3 2. 0.
0.
GRID 4 3. 0.
0.
GRID 5 4. 0.
0.
GRID 6 5. 0.
0.
GRID 7 6. 0.
0.
GRID 8 7. 0.
0.
GRID 9 8. 0.
0.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 22-13

© 2024 Siemens
 22. Linear Buckling

GRID 10 9. 0.
0.
GRID 11 10. 0.
0.
GRID 100 0. 0.
10.
$
MAT1 1 7.1+10 .33
2700.
$
$ RECTANGULAR SECTION OF DIMENSION .1m x .02m
$
PBAR 1 1 .002 6.667-8 1.667-6 2.332-7
$
SPC1 10 123 1
SPC1 10 23 11
SPC 10 1 4
ENDDATA

Listing 22-2. Input File for Euler Beam

As indicated on the EIGRL entry, the first five modes are requested for this problem, and the eigenvalue
table is shown in Figure 22-3. The first eigenvalue λ1 in this case is equal to 0.4672, while the applied
load in SUBCASE 1 is equal to –1000 N. Therefore, the lowest buckling load is equal to

Figure 22-3. Eigenvalue Table for a Euler Beam

A partial output that contains the first, second, and fifth eigenvectors is shown in Figure 22-4. The mode
shapes for the first five modes are plotted in Figure 22-5.

22-14 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Simcenter Nastran Results for Example 1

Figure 22-4. Eigenvectors for an Euler Beam

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 22-15

© 2024 Siemens
 22. Linear Buckling

Figure 22-5. Mode Shapes for an Euler Beam

Note that the first four system buckling modes are in the xz plane and the first buckling mode in the xy
plane is the fifth overall system mode. The lowest buckling load in the xy plane is, therefore, equal to

Equation 22-10.

Theoretical Results for Example 1

This Euler buckling load is calculated as follows (S.P. Timoshenko and J.M. Gere, Theory of Elastic
Stability, Engineering Societies Monograph Series, Second Ed., 1961):

22-16 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Example 2 – Lateral Buckling

For this particular problem, the buckling load in the xy plane is equal to

Table 22-6 compares the Simcenter Nastran results with theoretical results.

Table. Euler Beam Results Comparison

Mode Number Simcenter Nastran (N) Theoretical (N) % Difference

1 (XZ Plane) 467.2 467.2 0
5 (XY Plane) 11681.5 11681.4 0.001

As you approach the higher modes, the mode shapes become less smooth. This effect occurs because
for the same mesh density, the higher modes have fewer grid points per buckled sine wave than
the lower modes. If you are interested in these mode shapes, you should increase your model mesh
density to follow the general guideline of maintaining a minimum of five grid points per half sine wave.
However, these higher failure modes are seldom of interest in real world applications.

Example 2 – Lateral Buckling

The problem shown in Figure 22-6 illustrates an interesting case of lateral buckling. The model consists
of a cantilever beam with a vertical tip load applied at the free end. The dimensions of the beam
are identical to Example 1. In this example, the structure buckles in the lateral direction (xz plane)
perpendicular to the applied load rather than in the plane of the applied load (xy plane).

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 22-17

© 2024 Siemens
 22. Linear Buckling

Figure 22-6. Lateral Buckling

Simcenter Nastran Results for Example 2

Listing 22-3 contains the corresponding Simcenter Nastran model for this lateral buckling problem.
The CBEAM element is used in this case, although the CBAR element can also be used since
this beam consists of a constant cross section. A vertical 1000 N load is applied in the negative
y-direction to generate the element internal loads in Subcase 1. Once again, the Lanczos eigenvalue
extraction method is employed requesting two modes. Figure 22-8 contains the eigenvalue table and
corresponding eigenvectors. The static deflection and buckling mode shapes are shown in the figure
Frame Buckling.

It is interesting to note that the static deflection is in the plane of the applied load (xy plane), while
the buckling mode shape is in the plane perpendicular to the applied load (xz plane). Two modes were
requested on the EIGRL entry to illustrate another subtle point for this problem. The two eigenvalues
are identical in magnitude but different in sign. What is the significance of a negative eigenvalue in
a buckling analysis? In buckling analysis, a negative eigenvalue implies that the load that causes the
structure to buckle is opposite to the direction of the applied load.

Since both the structure and loading are symmetric, it does not matter whether the applied load is in
the +y or –y direction. This fact brings up another subtle point. If you specify V2 only on the EIGRL
entry, then Simcenter Nastran attempts to extract all the roots below V2. In this case, these roots
include all the negative eigenvalues. For a large model, calculating a large number of modes can be
expensive especially when you are only interested in the lowest modes. Note that these negative roots
are legitimate both from a physical and mathematical standpoint. In general, this situation only occurs
if the applied buckling load is in the opposite direction of the buckling load. For this particular problem,
there is no right or wrong direction. Therefore, in order to avoid this type of surprise, it is best to request
the lowest number of desired roots (ND) on the EIGRL entry when you are not sure of the buckling load
direction.

22-18 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Simcenter Nastran Results for Example 2

$ FILENAME - bucklat.dat
SOL 105
TIME 10
CEND
$
TITLE = LATERAL BUCKLING OF CANTILEVER BEAM
SUBTITLE = METRIC UNITS
SPC = 10
DISP = ALL
$
SUBCASE 1
LOAD = 10
$
SUBCASE 2
METHOD = 10
$
BEGIN BULK
$
EIGRL 10 2
$
FORCE 10 11 -1000. 0. 1.
0.
$
CBEAM 1 1 1 2 100
CBEAM 2 1 2 3
100
CBEAM 3 1 3 4 100
CBEAM 4 1 4 5
100
CBEAM 5 1 5 6
100
CBEAM 6 1 6 7
100
CBEAM 7 1 7 8
100
CBEAM 8 1 8 9
100
CBEAM 9 1 9 10
100
CBEAM 10 1 10 11
100
$
GRID 1 0. 0.
0.
GRID 2 1. 0.
0.
GRID 3 2. 0.
0.
GRID 4 3. 0.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 22-19

© 2024 Siemens
 22. Linear Buckling

0.
GRID 5 4. 0.
0.
GRID 6 5. 0.
0.
GRID 7 6. 0.
0.
GRID 8 7. 0.
0.
GRID 9 8. 0.
0.
GRID 10 9. 0.
0.
GRID 11 10. 0.
0.
GRID 100 0. 0.
10.
$
MAT1 1 7.1+10 .33
2700.
$
$ RECTANGULAR SECTION OF DIMENSION .1m x .02m
PBEAM 1 1 .002 6.667-8 1.667-6 2.332-7
$
SPC1 10 123456 1
ENDDATA

Listing 22-3. Input File for Lateral Buckling

22-20 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Simcenter Nastran Results for Example 2

Figure 22-7. Eigenvalue Table and Eigenvectors for Lateral Beam Buckling

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 22-21

© 2024 Siemens
 22. Linear Buckling

Figure 22-8. Static and Buckling Shapes

The lowest buckling load calculated by Simcenter Nastran is

Theoretical Results for Example 2

The theoretical buckling load can be calculated as follows (S.P. Timoshenko and J.M. Gere, Theory of
Elastic Stability, Engineering Societies Monograph Series, Second Ed., 1961):

22-22 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Example 3 – Planar Frame Buckling

Table 22-7 compares the Simcenter Nastran results to the theoretical results.

Table. Lateral Buckling Results Comparison

Simcenter Nastran (N) Theoretical (N) % Difference

218 218 0

Example 3 – Planar Frame Buckling

This example is a planar frame subjected to a compressive load at the top as shown in Figure 22-9.
At first glance, since both the structure and loading are symmetrical, you may be tempted to conclude
that the structure will buckle in a symmetric mode. As illustrated in Figure 22-10(a) (from Simcenter
Nastran) and S. P. Timoshenko and J. M. Gere, Theory of Elastic Stability, 1961, this isn't the case.
The lowest buckling mode is actually antisymmetric. The second buckling mode is symmetric (Figure
22-10(b)).

Figure 22-9. Planar Frame Buckling

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 22-23

© 2024 Siemens
 22. Linear Buckling

Figure 22-10. Frame Buckling

Simcenter Nastran Results for Example 3

Listing 22-4 shows a portion of the input file for this problem.

$
$ INPUT FILE - bukframe.dat
$
SOL 105 $
TIME 5
CEND
TITLE = FRAME BUCKLING PROBLEM
SUBTITLE = PLANAR MODEL
DISP = ALL
SPC = 1
$
SUBCASE 1
LABEL = STATIC LOAD CASE

22-24 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Theoretical Results for Example 3

LOAD = 1
$
SUBCASE 2
LABEL = BUCKLING CASE
METHOD = 10
$
BEGIN BULK
$
$ BRING IN REST OF BULK DATA FILE
$
INCLUDE 'bucklat.dat'
$
$ LOADS BULK DATA ENTRIES
$
FORCE 1 11 0 5000. -1.
FORCE 1 22 0 5000. -1.
$
$ CONSTRAINTS BULK DATA ENTRIES
$
SPC1 1 123456 1 33
$
$ PROPERTY AND MATERIAL BULK DATA ENTRIES
$
PBAR 1 1 3.46 1. 1.
MAT1 1 1.+7 .3
$
$ FOR CALCULATING BUCKLING MODES
$
EIGRL 10 3
ENDDATA

Listing 22-4. Input File for Frame Buckling

The lowest buckling load calculated by Simcenter Nastran is

Theoretical Results for Example 3

From Theory of Elastic Stability, the buckling load Pcr can be calculated as

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 22-25

© 2024 Siemens
 22. Linear Buckling

Equation 22-11.

where the value of k can be obtained by solving the following transcendental equation

Equation 22-12.

Solving Eq. 22-12, the lowest value of k is equal to

Equation 22-13.

Note that all the roots from the above transcendental equation only yield the antisymmetrical modes. In
this case, the lowest buckling load is

Table 22-8 compares the theoretical results to the Simcenter Nastran results for the antisymmetric
frame buckling.

Table. Frame Buckling Results Comparison

Simcenter Nastran (lb) Theoretical (lb) % Difference

7378 7377 0.01

22-26 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Example 4 – Buckling of a Stiffened Panel with Transverse Shear Flexibility

Example 4 – Buckling of a Stiffened Panel with Transverse Shear Flexibility

This problem is selected to illustrate the effect of the transverse shear flexibility in the buckling failure of
a stiffened panel. Figure 22-11 consists of a panel reinforced with stiffeners at both the top and bottom.
The panel is subjected to a distributed compressive load at the right edge. The failure mode of interest in
this case is the in-plane (xy plane) buckling of the stiffened panel.

Figure 22-11. Buckling of a Stiffened Panel

Simcenter Nastran Results for Example 4

The corresponding Simcenter Nastran input file for this problem is shown in Listing 22-5. The panel is
modeled with CQUAD4s. Since only the in-plane buckling failure mode is of interest, only the membrane
stiffness is requested for the CQUAD4s (MID1 only). The stiffeners are modeled with CRODs. Since this is
a planar model, the out-of-plane motion is constrained (3, 4, 5, and 6 DOFs).

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 22-27

© 2024 Siemens
 22. Linear Buckling

The panel is subjected to a distributed compressive load at the right end. As an alternative modeling
technique, an RBE3 element is connected to the grid points at the right edge so that the load can be
applied to a single grid point with the RBE3 spreading the loads to the other grid points.

The Lanczos eigenvalue extraction method is used in this case to request the lowest mode. The EIGRL
entry is used. For buckling analysis, fields 3 and 4 provide the range of the eigenvalue of interest.

$
$ FILENAME - stffqud4.dat
$
SOL 105
TIME 10
CEND
TITLE = PANEL WITH STIFFENERS
DISP = ALL
STRESS = ALL
SPC = 1
$
SUBCASE 1
LABEL = COMPRESSIVE LOAD
LOAD = 1
$
SUBCASE 2
LABEL = BUCKLING SUBCASE
METHOD = 10
$
BEGIN BULK
$
EIGRL 10 .7 .9
PARAM POST 0
$
RBE3 1000 1000 123456 2. 12 82 123 +
+ 164 205 287 328 369 410 2. 12345 +
+ 246 1.0 12 41 451
$
FORCE 1 1000 -1.0 100000.
$
$ THIS SECTION CONTAINS THE LOADS, CONSTRAINTS, AND CONTROL BULK DATA
ENTRIES
$
PSHELL 1 1 .2
PROD 2 1 .6
$
MAT1 1 1.+7 .32
$
$ BRING IN THE REST OF THE MODEL
$

22-28 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Simcenter Nastran Results for Example 4

include 'stffqud4.blk'
$
ENDDATA

Listing 22-5. Input File for the Buckling of Stiffened Panels

Figure 22-12. Eigenvalues for a Stiffened Panel

Figure 22-13. Stiffened Panel Buckling Mode

The lowest buckling load calculated by Simcenter Nastran is

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 22-29

© 2024 Siemens
 22. Linear Buckling

Theoretical Results for Example 4

If this problem is treated as an Euler beam, the critical buckling load Pe is equal to

Equation 22-14.

If the effect of the transverse shear flexibility is included, then it can be calculated as follows
(Timoshenko and Gere, Theory of Elastic Stability, 1961):

Equation 22-15.

If the transverse shear flexibility is ignored, then the buckling load deviates by 5.2%.

Table 22-9 contains a comparison between the theoretical results versus the Simcenter Nastran results
for the buckling of the stiffened panel. Note that the transverse shear flexibility of this structure is
automatically included when you are performing a buckling analysis in Simcenter Nastran. There is no
additional input required on your part.

Table. Stiffened Panel Buckling Results Comparison

Simcenter Nastran (lb) Theoretical (lb) % Difference

78,118 78,210 0.1

A review of the stresses also indicates that the structure will yield prior to reaching the linear buckling
load level. In other words, the critical failure mode may be due to yielding rather than to the linear
buckling of the structure. The knowledge of this linear buckling load level can still be of design

22-30 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Example 5 – Buckling of a Cylinder Under Uniform Axial Load

significance. If you are interested in detailed stresses for this problem, then a nonlinear analysis using
Solution 106 may be more appropriate in this case.

Example 5 – Buckling of a Cylinder Under Uniform Axial Load

The next example is the buckling of a cylinder subjected to a distributed compressive load at one end
and simply supported at the other end. The cylinder has a diameter of 20 inches and a length of 20
inches with a wall thickness of 0.03 inches as shown in Figure 22-14. This problem illustrates the
phenomenon of the buckling of a thin curved shell structure.

Figure 22-14. Buckling of a Cylinder Under a Compressive Load

Simcenter Nastran Results for Example 5

The CQUAD4 elements are used for the Simcenter Nastran model. A mesh of 20 elements along the
length and 72 elements around the circumference is used for this problem. The corresponding input file
containing the pertinent entries is shown in Listing 22-6.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 22-31

© 2024 Siemens
 22. Linear Buckling

$ FILENAME - buckcy20r.dat
$
ID CYLIN BUCKLING
SOL 105
TIME 200
CEND
TITLE = BUCKLING OF CYLINDER - SIMPLY SUPPORTED
SUBTITLE = 20" x 20" - t=.03" - CP IN CYL COORD SYSTEM - K6ROT = 100.
DISP = ALL
SPC = 1
$
SUBCASE 1
LABEL = STATIC LOAD
LOAD = 1
SPCF = ALL
STRESS = ALL
$
SUBCASE 2
LABEL = EIGENVALUE CALCULATION
METHOD = 1
$
BEGIN BULK
$
PARAM K6ROT 100.
PARAM POST -1
EIGRL 1 2
$
PSHELL 1 1 .03 1
$
MAT1 1 1.+7 .3
$
include 'full_cp1.blk'
$
$ THIS SECTION CONTAINS ALL DEFINED COORDINATE SYSTEMS
$
CORD2C 1 0 0.0 0.0 0.0 0.0 0.0 1. +
1. 0.0 0.0
$
RBE3 5000 5000 123456 1. 123 381 382 +
+ 383 384 385 386 387 388 389 390 +
+ 391 392 393 394 395 396 397 398 +
+ 399 781 782 783 784 785 786 787 +
+ 788 789 790 791 792 793 794 795 +
+ 796 797 798 1180 1181 1182 1183 1184 +
+ 1185 1186 1187 1188 1189 1190 1191 1192 +
+ 1193 1194 1195 1196 1197 1579 1580 1581 +
+ 1582 1583 1584 1585 1586 1587 1588 1589 +
+ 1590 1591 1592 1593 1594 1595
$

22-32 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Simcenter Nastran Results for Example 5

FORCE 1 5000 0 100000. -1.

$
ENDDATA

Listing 22-6. Input File for a Cylindrical Buckling Problem

Once again, an RBE3 element is used to distribute the load from a single grid point to the circumference
of the cylinder.

A static load of –100,000 lb is applied in Subcase 1 to generate the internal forces for the structure.
This load is applied to the RBE3, which, in turn, distributes the loads around the circumference of the
cylinder. The static deflection is shown in Figure 22-15.

The Lanczos eigenvalue extraction method is used in this case to request the lowest mode. The lowest
calculated eigenvalue is equal to 0.35986. The buckling load is, therefore, equal to

The corresponding buckling mode shape is shown in Figure 22-16. As you can see from the plots, there
are two grid points per half sine wave (four grid points per sine wave), which is below the recommended
value of a minimum of five grid points per half sine wave.

This problem is then remeshed with 40 elements along the z-direction keeping the same mesh density
in the circumferential direction. The new eigenvalue calculated in this case is equal to 0.34914. The
revised buckling load is, therefore, equal to

The corresponding mode shape is shown in Figure 22-17. In this case, there are three grid points per
half sine wave (six grid points per sine wave), which is still below the minimum requirement. To obtain
better accuracy, you can certainly further refine this model until you meet the minimum number of grid
points requirement. However, this mesh is sufficient for demonstrating the basic linear buckling features
of Simcenter Nastran.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 22-33

© 2024 Siemens
 22. Linear Buckling

Figure 22-15. Static Deflection of a Cylinder Due to a Compressive Load

Figure 22-16. Buckling Shapes of a Cylinder with 20 Elements Along the z-Direction

22-34 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Theoretical Results for Example 5

Figure 22-17. Buckling Shape of a Cylinder with 40 Elements Along the z-Direction

Theoretical Results for Example 5

The first buckling load (Timoshenko and Gere, Theory of Elastic Stability, 1961) can be calculated as
follows:

Table 22-10 contains a comparison between the results obtained with Simcenter Nastran versus the
theoretical results.

Table. Thin Cylinder Buckling Results Comparison

Mesh Density Simcenter Nastran (N) Theoretical (N) % Difference

72 x 20 34,262 34,225 .11
72 x 40 34,234 34,225 .03

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 22-35

© 2024 Siemens
 22. Linear Buckling

Example 6 – Multiple Buckling Analyses in a Single Run

So far, a single static analysis was considered followed by a single buckling analysis. Example 6 shows
you how to run multiple static and buckling analyses in a single run. In fact, what is done in this example
is to combine the Euler beam buckling (Example 1) and lateral buckling (Example 2) problem into a
single run. The corresponding input file is shown in Listing 22-7.

The model used is the same geometric beam model used in Examples 1 and 2. The first subcase
(Subcase 2) is a static subcase consisting of a cantilever beam with a vertical tip load applied to the
free end (see the figure Lateral Buckling). The second subcase (Subcase 5) is a static subcase consisting
of a simply supported beam with an axial load applied to the roller end (see the figure Euler Beam).
The third subcase (Subcase 11) is for a lateral buckling analysis. The Case Control command (STATSUB
= 2 in Subcase 11) tells Simcenter Nastran that you want to generate the differential stiffness matrix
from the first static subcase (Subcase 2). The fourth subcase (Subcase 21) is for a Euler beam buckling
analysis. The Case Control command (STATSUB = 5 in Subcase 21) tells Simcenter Nastran that you want
to generate the differential stiffness matrix from the second static subcase (Subcase 5). The Bulk Data
entries are similar to Examples 1 and 2. Note that different boundary conditions are allowed for different
subcases.

$ FILENAME - MULTBUCK.DAT
$
SOL 105
TIME 10
CEND
$
TITLE = LATERAL BUCKLING OF CANTILEVER BEAM
SUBTITLE = METRIC UNITS
DISP = ALL
$
SUBCASE 2
LABEL = CANTILEVER BEAM
LOAD = 10
SPC = 10
$
SUBCASE 5
LABEL = SIMPLY SUPPORTED BEAM
LOAD = 20
SPC = 20
$
SUBCASE 11
LABEL = LATERAL BUCKLING OF CANTILEVER BEAM
METHOD = 10
SPC = 10
STATSUB = 2
$
SUBCASE 21

22-36 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Example 6 – Multiple Buckling Analyses in a Single Run

LABEL = EULER BUCKLING OF SIMPLY SUPPORTED BEAM

METHOD = 10
SPC = 20
STATSUB = 5
$

Listing 22-7. Input File for Multiple Buckling Analyses (Continued)

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 22-37

© 2024 Siemens
 22. Linear Buckling

BEGIN BULK
$
EIGRL 10 2
$
FORCE 10 11 -1000. 0. 1. 0.
FORCE 20 11 -1000. 1. 0. 0.
$
CBEAM 1 1 1 2 100
CBEAM 2 1 2 3 100
CBEAM 3 1 3 4 100
CBEAM 4 1 4 5 100
CBEAM 5 1 5 6 100
CBEAM 6 1 6 7 100
CBEAM 7 1 7 8 100
CBEAM 8 1 8 9 100
CBEAM 9 1 9 10 100
CBEAM 10 1 10 11 100
$
GRID 1 0. 0. 0.
GRID 2 1. 0. 0.
GRID 3 2. 0. 0.
GRID 4 3. 0. 0.
GRID 5 4. 0. 0.
GRID 6 5. 0. 0.
GRID 7 6. 0. 0.
GRID 8 7. 0. 0.
GRID 9 8. 0. 0.
GRID 10 9. 0. 0.
GRID 11 10. 0. 0.
GRID 100 0. 0. 10.
$
MAT1 1 7.1+10 .33 2700.
$
$ RECTANGULAR SECTION OF DIMENSION .1m x .02m
$
PBEAM 1 1 .002 6.667-8 1.667-6 2.332-7
$
SPC1 10 123456 1
$
SPC1 20 123 1
SPC1 20 23 11
SPC 20 1 4
$
ENDDATA
Listing 22-7. Input File for Multiple Buckling Analyses

22-38 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 23. Follower Stiffness
Overview of Follower Stiffness
A follower stiffness is an adjustment to the standard linear stiffness matrix to account for a load's
dependency on the orientation. A common example to demonstrate is a bar in compression as a result
of an axial force, and the force direction remains tangential to the deformed axis:

Figure 23-1. Follower Force Example

Assuming that differential stiffness is also added in this example, the new stiffness matrix would become:

Ktotal = Klinear + Kdifferential + Kfollower

Follower stiffness due to pressure loads (PLOAD, PLOAD2, PLOAD4) and point loads (FORCE1, FORCE2,
MOMENT1, MOMENT2) were first introduced for modal analysis (SOLs 103 and 107) and buckling
analysis (SOL 105) and are applicable for both quadratic elements (QUAD8, TRIA6, HEXA, PENTA, TETRA)
and linear elements (QUAD4, TRIA3, HEXA, PENTA, TETRA). Follower stiffness was then extended to
dynamic responses for direct response analyses (SOLs 107, 108 and 109) and modal response analyses
(SOLs 110, 111 and 112). Centrifugal load (RFORCE) was implemented for buckling, dynamic analyses,
and cyclic symmetry analysis for modal and buckling (SOLs 115 and 116). Follower stiffness has been
further extended to include nonlinear analysis (SOL 106) for nonlinear static, nonlinear modal and
nonlinear buckling analyses.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 23-1

© 2024 Siemens
 23. Follower Stiffness

Benefits
The follower stiffness is generated if the structure is loaded by follower forces. Effects of follower
stiffness are prominent in dynamic and buckling analyses. For example, for a cylinder under external
pressure critical buckling load may be over-estimated and the natural frequencies in vibration may be
under-estimated in the absence of follower stiffness. To the contrary, this observations are reversed in
case of centrifugal loads.

For moderately geometric nonlinear analysis, exclusion of follower stiffness affects the rate of
convergence, but the converged solution is correct. For severely geometric nonlinear analysis, it may
not be possible to obtain a converged solution without including follower stiffness. As the geometric
nonlinearity intensifies, so is the effect of follower stiffness. Therefore, inclusion of follower stiffness
greatly enhances the convergence if the deformation involves severe geometric nonlinearity.

Input
The follower stiffness is primarily controlled by PARAM, FOLLOWK. By default, the follower stiffness is
included wherever applicable in all the solution sequences mentioned above. The effect of follower
stiffness may be ignored by specifying PARAM, FOLLOWK, NO. This parameter may be changed from
Subcase to Subcase.

You may also perform multiple dynamic analyses in a single run to study the effects of preload. For
this, you will need to specify multiple static Subcases (with static loads) in a dynamic analysis solution
sequence. For example, to study the effect of preloads on normal modes, you will need to:

1. Specify a static load subcase for each of the loading conditions, and

2. Specify normal modes subcases by specifying the METHOD command and selecting the pre-loading
condition by the STATSUB command.

In case of buckling analysis, two different STATSUB conditions may be selected: one for sustaining
preload by STATSUB(PRELOAD) and another for modulated preload by STATSUB(BUCKLE) for the critical
load in search.

In nonlinear analysis (SOL 106), the follower stiffness is included if PARAM, LGDISP, 1 is specified.
Although it is not recommended under normal circumstances, follower stiffness may be ignored in SOL
106 by PARAM, FOLLOWK, NO. When PARAM LGDISP is greater than 0, the differential stiffness of the
superelements are also computed and added to the nonlinear stiffness. This will result in somewhat
different solution than Version 70.5 if superelements are present. The follower stiffness in nonlinear
static solution improves convergence rate, but it does not alter the solution except for a small difference
due to round-off error.

Output
When a static subcase is specified for linear transient response analysis (SOLs 109 and 112), the data
recovery is controlled by PARAM, ADSTAT. By default (YES) the static solution will be superimposed

23-2 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Linear Dynamic Analysis with Follower Stiffness (SOLs 108, 109, 111, and 112)

on the dynamic response solution (displacement, stress and SPCForce). The relative solution can be
obtained in reference to the static solution point by PARAM, ADSTAT, NO. No provision is made
for frequency response analysis, because the static responses contribute only to the zero frequency
response.

Linear Dynamic Analysis with Follower Stiffness (SOLs 108, 109,

111, and 112)
If the external load changes as function of displacements, it is termed follower force, which creates
follower stiffness. The derivatives of the external load constitute a follower stiffness, i.e.,

This follower stiffness is included as a part of differential stiffness.

In direct transient response analysis (SOL 109), the following equation is solved

Equation 23-1.

with

In direct frequency response analysis (SOL 108) the following equation is solved:

or in the frequency domain, the complex equation

Equation 23-2.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 23-3

© 2024 Siemens
 23. Follower Stiffness

is solved by Gauss elimination in complex domain.

The modal transient response analysis (in SOL 112) reduces the system by normal modes before
integration, i.e.,

Equation 23-3.

If the damping is proportional, the above equation can be decoupled by

The modal frequency response analysis (in SOL 111) reduces the system by normal modes before Gauss
elimination in the complex domain, i.e.,

Equation 23-4.

In presence of follower forces, the differential stiffness in the above equations includes follower stiffness.
At this point, the solution is the one perturbed from the static solution under preload.

The follower stiffness also affects the data recovery in dynamic response analysis, controlled by PARAM,
ADSTAT. For linear dynamic response, the static solution can be superimposed after the dynamic
solution procedure. The preload effect is reflected only in the stiffness and the actual static load is
omitted in the dynamic response analysis. The total displacement can be obtained by superposing the
static solution to the transient response analysis:

Equation 23-5.

and

23-4 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Linear Dynamic Analysis with Follower Stiffness (SOLs 108, 109, 111, and 112)

Equation 23-6.

The stress output can be obtained by

Equation 23-7.

In case of Matrix method (PARAM, DDRMM, 0) for modal approach,

Equation 23-8.

is processed by the module DDRMM using augmented matrices:

and

which results in

with a static deformed shape as the first mode. The mode acceleration method also requires differential
stiffness in the recovery process, i.e.,

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 23-5

© 2024 Siemens
 23. Follower Stiffness

Then the mode acceleration is expressed as

Equation 23-9.

where K2GG is included in the modal analysis but not K2PP.

23-6 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 24. Laminates
Overview of Laminated Composite Materials
The term composite refers to an engineering material that is made up of more than one material.
Classical lamination theory takes the term composite another step to mean a material that is composed
of stacks of ply layers or lamina, with each lamina having its own properties. A laminate is a stack of
these individual lamina arranged with the individual lamina having different orientations of the principal
material directions. The laminae are bonded together with a thin layer of bonding material that is
considered to be of zero thickness.

Figure 24-1. Laminae Arranged to Form a Laminate

Each ply or lamina may be considered as a group of unidirectional fibers. The ability to orient the fibers
in particular directions lets you tailor the mechanical properties of a composite material to match the

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 24-1

© 2024 Siemens
 24. Laminates

loading environment. The principal material directions for the lamina are parallel and perpendicular to
the fiber direction. The principal material directions are referred to as:

• “Longitudinal” or “1-direction”, which corresponds to the fiber direction.

• “Transverse” or “2-direction”, which corresponds to the direction perpendicular to the fibers.

Figure 24-2 shows an exploded view of three cross-ply laminated plates. The n-laminae (n = 1,2,3,4) of
each of the three configurations are normal to the z-axis of the indicated coordinate system and the 1-
and 2-axes appended to the individual lamina denote principal material directions. The directions of the
principal material axes of each lamina alternate as implied by the use of the word "cross-ply" to describe
the configuration. The xy-plane of the coordinate axes is defined in the geometric middle plane of the
laminae.

Figure 24-2. Exploded View of Three Cross-Ply Laminated Plates

Plies are typically made of fibers bonded together by a matrix. If the ply is a tape, all the fibers are
oriented in the same direction. Cloth plies have fibers woven in two directions. Many different materials

24-2 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Modeling Laminates in Simcenter Nastran

can be used as fibers or matrices. Examples of fibers include graphite, glass, boron, silicon carbide, and
tungsten. Examples of matrices include epoxy and aluminum.

Modeling Laminates in Simcenter Nastran

Three methods are available for you to model composite laminates. The PSHELL and PCOMP methods are
used to model laminates with plate and shell elements like the CQUAD4, CQUAD8, CTRIA3, and CTRIA6
elements. The PSHELL and PCOMP methods are based upon classical lamination theory.

• You can use the PSHELL method to directly input membrane, bending, membrane-bending coupling,
and transverse shear constitutive relationships.

• You can use the PCOMP method to define the composite laminate on a ply-by-ply basis. When you use
the PCOMP method, the software computes the equivalent PSHELL and MAT2 entries for you.

The third method is the PCOMPS method. The PCOMPS method is used to model laminates with CHEXA
and CPENTA solid elements. Unlike the other two methods, the PCOMPS method is not based on classical
lamination theory. Instead, the element matrices are integrated layer-by-layer In the ply stack up
direction. This results in a new set of integration points for each laminate layer in the element. However,
the element interpolation functions do not change. For example if you model a solid composite using
a 20 grid CHEXA element, you still will have a parabolic variation of displacement in the element,
regardless of the number of layers. The PCOMPS method is useful for modeling very thick laminates
where interlaminar normal stresses may be important.

Each method allows you to model an entire stack of laminae with a single element because the material
properties of the stack are completely reflected in the matrices of elastic moduli for the element. With
the PCOMPS method, you can also optionally model the entire stack of laminae with multiple solid
elements. In all three methods, the software automatically calculates these matrices from user-supplied
definitions of the thickness, the material properties, and the relative orientation of these properties for
the individual lamina. Once the software calculates the matrices of elastic moduli, the analysis proceeds.

Because the material properties of the laminate are completely contained in the matrices of elastic
moduli, you can use standard data recovery methods to have Simcenter Nastran calculate:

• Stresses, strains, and appropriate failure indices in individual laminae.

• Interlaminar stresses and strains and a bonding failure index.

If the loading on a structure is sufficient to exceed the elastic limit of the material or if the material
behaves nonlinear elastically, then nonlinear methods are required.

The PSHELL, PCOMP, and PCOMPS methods are described further in the following sections.

Note: For three-dimensional composite analyses, you must develop the anisotropic material matrix
yourself. Typically, you would use the MAT9 bulk entry to define this material matrix.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 24-3

© 2024 Siemens
 24. Laminates

See also

• “MAT9” in the Simcenter Nastran Quick Reference Guide.

• “Understanding MAT9” in the Simcenter Nastran User's Guide.

Using the PSHELL Method

With the PSHELL method, you use the PSHELL entry with the appropriate material definitions and
geometric parameters to directly define a laminate. For example, you can use PSHELL to define
properties such as membrane thickness, material properties, and bending and transverse shear
parameters.

On the PSHELL entry, there are four different material ID (MID) fields as shown in Table 24-1.

Table. Material Properties on the PSHELL Entry

ID PSHELL Field Purpose

MID1 3 Membrane material property.

MID2 5 Bending material property.

MID3 7 Transverse shear stiffness material property.

MID4 4 (cont.) Material property used to define coupling between the

membrane and the bending deformation.

You can specify MAT1, MAT2, or MAT8 as the material type for any of the material (MIDi) fields. For
isotropic materials, use the same MAT1 identification for MID1 and MID2, and leave the MID3 and MID4
fields blank.

• Use MID1 if you want to include membranes only.

• Use MID2 (and optionally, MID3) if you want to include bending only.

• Use MID3 to have the software use thick plate theory, which includes the transverse shear flexibility,
to develop the element stiffness matrix. In general, for thin or curved surfaces, you should not use
MID3.

• Use MID4 to input a coupling relationship between the in-plane forces and bending moments. This
coupling only occurs in plates that are non-symmetric about the neutral plane or where the neutral
plane is offset from the grid points. Typical applications include reinforced skins and aluminum
bonded to fiberglass. You should leave the MID4 field blank if the element cross section is symmetric.

While the PSHELL method is sufficient for defining simple composites, determining the appropriate
material definitions can be difficult for more complex composites. Additionally, when you use the PSHELL

24-4 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Using the PCOMP Method

method, you can only recover smeared element data (running loads and moments, and equivalent
strains and curvatures).

See also

• “PSHELL” in the Simcenter Nastran Quick Reference Guide.

Using the PCOMP Method

The PCOMP entry provides a convenient way of entering appropriate material properties for a plate
consisting of layers of unidirectional fibers. The input on the PCOMP entry consists of the layer-by-
layer material definition. You can use PCOMP to define the thickness, orientation, and the material
identification number of each of the individual lamina.

When you solve your model, Simcenter Nastran calculates the membrane, bending, transverse shear,
and coupled membrane-bending material properties of the laminate as a whole from the data you
specified on the PCOMP entry. The software outputs these calculated properties in the form of one
equivalent PSHELL and four MAT2 entries, as shown in Figure 24-3. If you specify ECHO = PUNCH in the
case control section, the software prints the generated PSHELL and MAT2 entries in the PUNCH file after
the sorted bulk data echo.

Because the MIDi fields use more than eight digits, the software uses the large field format for the
generated PSHELL and MAT2 entries. You can then modify your input file with these new entries and
proceed with your analysis.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 24-5

© 2024 Siemens
 24. Laminates

Figure 24-3. Equivalent PSHELL and MAT2 Entries Are Generated

The ID numbers of the generated MAT2 entries are important.

• The MID1 material has an ID in the range of 100000000 to 199999999.

• The MID2 material has an ID in the range of 200000000 to 299999999, etc.

These MIDi ranges alert Simcenter Nastran that the material is part of a composite analysis. Therefore, if
you are using the equivalent properties in a future analysis instead of using the PCOMP entry, and you
are entering a thermal coefficient of expansion, don't change the ID numbers.

See also

• “PCOMP” in the Simcenter Nastran Quick Reference Guide.

Using the PCOMPS Method

The PCOMPS method is used to model laminates with CHEXA and CPENTA solid elements. Unlike the
other methods, the PCOMPS method is not based on classical lamination theory and is useful for
modeling very thick laminates where interlaminar normal stresses may be important.

Like the PCOMP method, the PCOMPS method allows you to define the laminate on a layer-by-layer basis.
Using the PCOMPS bulk entry, you can:

24-6 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Selecting Output Options

• Define the orientation, stacking, and thickness of each ply

• Assign MAT1, MAT9, or MAT11 material property entries to each ply

• Define reference temperature and damping coefficient for the laminate

• Specify ply and interlaminar failure theories and define allowable interlaminar shear and normal
stresses

Because the solid elements used to model a laminated structure have potentially numerous material
coordinate systems, but otherwise identical element properties, consider using MATCID bulk entries
to map the solid elements to a single PCOMPS property entry. By minimizing the number of PCOMPS
entries, the size of the input file is reduced and the model data is handled more efficiently.

See also

• “PCOMPS” in the Simcenter Nastran Quick Reference Guide.

• “MATCID” in the Simcenter Nastran Quick Reference Guide.

Selecting Output Options

For a shell element laminate defined with the PCOMP or PCOMPG entry, stress and strain output for
the individual lamina is supported in solutions 101, 103, 105, 106, 108, 109, 111, 112, 114, 129, 144
and 200. In all other solutions except for heat transfer SOLs 153 and 159, stress and strain can only be
recovered for the equivalent laminate. That is, output on the equivalent PSHELL created by the software.

The output you may request for elements referencing PCOMP or PCOMPG entries in solutions 101, 103,
105, 106, 108, 109, 111, 112, 114, 129, 144, and 200 includes:

• Stress and strain for the equivalent laminate (that is, output on the equivalent PSHELL created by the
software)

• Stress resultants, which are requested with the FORCE case control

• Stress and strain on the individual lamina including approximate interlaminar shear stress in the
bonding material

• Failure indices and strength ratios.

For a solid element laminate defined with the PCOMPS entry, stress and strain output for the individual
lamina is supported in solutions 101, 103, 105, 108, 109, 111, 112, and 401. The output you may
request for elements referencing the PCOMPS entry includes:

• Stress and strain on the individual lamina at the bottom, middle, or top of the plies.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 24-7

© 2024 Siemens
 24. Laminates

• Failure indices or strength ratios.

See the remarks on the PCOMP, PCOMPG, and PCOMPS bulk entry listings in the Quick Reference Guide
for information on these output requests.

For information on laminate output in the dynamic solutions 109 and 112, see Transient Response
Solution Control and Output in the Simcenter Nastran Basic Dynamics Guide.

For information on laminate output in the dynamic solutions 108 and 111 including random response,
see Frequency Response Laminate Output in the Simcenter Nastran Basic Dynamics Guide.

See also

• “STRESS” in the Simcenter Nastran Quick Reference Guide.

• “NOCOMPS” in the Simcenter Nastran Quick Reference Guide.

Classical Lamination Theory for Shell Composites (PCOMP,PCOMPG)

Classical lamination theory makes the following assumptions regarding the behavior of the laminae:

• The laminate consists of perfectly bonded laminae.

• The bonds are infinitesimally thin and nonshear-deformable; i.e., displacements are continuous across
laminae boundaries so that no lamina can slip relative to another.

• There is linear variation of strain through the laminate thickness.

Deformation in the X-Y plan of the plate at any point C at a distance z in the normal direction to plate
middle surface is

Equation 24-1.

Equation 24-2.

24-8 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Classical Lamination Theory for Shell Composites (PCOMP,PCOMPG)

where U, V, and W are the displacements along the X, Y, and Z directions in the element coordinate
system, and θx, θy are the rotations.

The strain-displacement-middle surface strain and curvatures relationship is given by:

Equation 24-3.

where the ϵ0's and χ's are the middle surface strains and curvatures, respectively.

The stress resultants for an N-layer laminate are obtained by integration of the stresses in each lamina
through the laminate thickness as:

Equation 24-4.

Equation 24-5.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 24-9

© 2024 Siemens
 24. Laminates

The stress resultant to strain relationship is:

Equation 24-6.

Equation 24-7.

where [G]k is the material matrix transformed from the laminate coordinate system into the lamina
coordinate system.

These relations can be written in the following form used to describe composite elements:

Equation 24-8.

where:

24-10 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Classical Lamination Theory for Shell Composites (PCOMP,PCOMPG)

[A] =

[B] =

[D] =

are named in composite element literature as the membrane, membrane-coupling, and bending
matrices, respectively.

In the shell element formulation in Simcenter Nastran (CQUAD4, CQUAD8, CTRIA3, and CTRIA6), these
relationships take the following form:

where:

Equation 24-9.

[A] = TG1 (3x3 matrix)

[B] = −T2G4 (3x3 matrix)
[D] =

{Q} =

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 24-11

© 2024 Siemens
 24. Laminates

{γ} =

T nominal plate thickness

Ts = effective transverse shear material thickness
[G3] = effective transverse shear material matrix

If you use the PSHELL entry, you can directly input G1, G2, G4, T, G3, and Ts (G1=MID1, G2=MID2,
G3=MID3, G4=MID4 on the PSHELL entry) If you use the PCOMP entry, you can have Simcenter Nastran
calculate the composite equivalent material matrices from the data you supply.

The thermal loads and moments are defined as:

The equivalent thermal properties are computed as follows:

MID1 (membrane) is computed as:

MID2 (bending) is computed as:

MID4 (membrane bending coupling) is computed as:

The thermal loads and moments from the homogeneous thermal membrane strain and curvature is
computed as:

24-12 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Classical Lamination Theory for Shell Composites (PCOMP,PCOMPG)

The following coefficients are used to determine equivalent thermal properties:

These coefficients are used to calculate the equivalent thermal properties as follows:

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 24-13

© 2024 Siemens
 24. Laminates

and

where {αε0}, {αχ}, and {αχε0} are the equivalent thermal properties for membrane, bending, and
membrane/bending coupling, respectively. Note that {αϵ0} is not directly calculated, but is determined
from {Gαϵ0} when the PCOMP input is used when the MID4 field on the PSHELL is > 400,000,000. Note
that {Gχξ0} cannot be input directly using PSHELL and {αϵ0} can be input only if [B] is invertible (which is
generally not true).

The thermal expansion relationships in the shell element formulation take the following form:

where α1, α2, and α3 are the thermal expansion inputs on the materials referenced by the MID1, MID2,
and MID4 fields on the PSHELL entry. If you use PCOMP, Simcenter Nastran automatically calculates
these relationships.

The terms G1, G2, and G4 are defined by the following integrals:

24-14 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Classical Lamination Theory for Shell Composites (PCOMP,PCOMPG)

Equation 24-10.

The limits on the integration are from the bottom surface to the top surface of the laminated composite.
The matrix of material moduli, [Ge], has the following form for isotropic materials:

Equation 24-11.

For orthotropic materials, the matrix, [Ge], is written as follows:

Equation 24-12.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 24-15

© 2024 Siemens
 24. Laminates

Here, ν1E2 = ν2E1 to satisfy the requirement that the matrix of elastic moduli be symmetric. In general,
the analyst may supply element properties with respect to a particular orientation which does not
necessarily correspond to the principal material axes. In this case, the analyst must also supply the value
of the angle, θ or material coordinate system that orients the element material axis relative to the side
G1-G2 of the element. The material elastic modulus matrix is then transformed by the program into the
element modulus matrix through the relation

Equation 24-13.

where:

Equation 24-14.

The finite element model for a structure composed of composite materials requires the evaluation of the
matrix of elastic moduli for each plate element of the model. The characteristics of the composite media
are totally contained in these matrices.

To illustrate evaluation of these matrices, consider the cross-ply laminates shown in the figure Exploded
View of Three Cross-Ply Laminated Plates. Here, all three configurations are represented by a single
quadrilateral plate element. The coordinate axes are coincident with the element coordinate axes. Then,
if we assume that each lamina of the n-ply laminates is of thickness T/n, where T is the total thickness
of each of three configurations, the matrices of elastic moduli may be evaluated from the following
relations:

24-16 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Classical Lamination Theory for Shell Composites (PCOMP,PCOMPG)

Equation 24-15.

Equation 24-16.

Equation 24-17.

These relations reflect the assumption that the xy-plane of the element coordinate system is coincident
with the geometric middle plane of the laminate. The xy-plane of the element coordinate system is
defined in the mean plane of the element so that any offset between the mean plane of the connected
grid points and the geometric middle plane of the laminate would be reflected in the integration limits
of the preceding relations.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 24-17

© 2024 Siemens
 24. Laminates

Note that I = T3/12 in the evaluation of [G2], i.e., the value (12I)/Γ2, will be assigned the default value of
1.0 on the plate element property entry.

The matrix of elastic moduli for transverse shear, [G3]m is defined as a two-by-two matrix of the form

Equation 24-18.

and the corresponding matrix transformed into an element coordinate system is given by

Equation 24-19.

where

The mean value of the transverse shear modulus for the laminated composite is defined in terms of
the transverse shear strain energy, U, through the depth

Equation 24-20.

A unique mean value of the transverse shear strain is assumed to exist for both the x and y components
of the element coordinate system, but for ease of discussion, only the evaluation of an uncoupled x
component of the shear moduli will be illustrated here. From Eq. 24-20 the mean value of transverse
shear modulus may be written in the following form

24-18 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Classical Lamination Theory for Shell Composites (PCOMP,PCOMPG)

Equation 24-21.

where G is an “average” transverse shear coefficient used by the element code and (Gx)i is the local shear
coefficient for layer . To evaluate Eq. 24-12, it is necessary to obtain an expression for (τzx(z)). This can
be accomplished by assuming that the x- and y-components of stress are decoupled from one another.
This assumption allows the desired equation to be deduced through an examination of a beam unit
cross-sectional width.

The equilibrium conditions in the horizontal direction and for total moment are

Equation 24-22.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 24-19

© 2024 Siemens
 24. Laminates

Equation 24-23.

Now, if the location of the neutral surface is denoted by and ρ is the radius of curvature of the beam,
the axial stress Ex may be expressed in the form

Equation 24-24.

Relation 20 may be differentiated with respect to x combined with Eq. 24-22 and Eq. 24-23. In a region
of constant Ex the result may be integrated to yield the following expression

Equation 24-25.

Eq. 24-29 is particularly convenient to use in the analysis of n-ply laminates because sufficient
conditions exist to determine the constants Ci (i = 1,2,...,n) and the “directional bending center” .
For example, consider the following laminated configuration

24-20 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Classical Lamination Theory for Shell Composites (PCOMP,PCOMPG)

At the bottom surface (i = 1, z = z0, and τxz = 0)

Equation 24-26.

and for the first ply at the interface between plies i = 1 and i = 2 (z = z1)

Equation 24-27.

At this interface between plies i = 1 and i = 2,

Equation 24-28.

and as (τxz)2 = (τxz)1 at z = z1,

Equation 24-29.

Then, in the ply z1<z<z2 the shear is

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 24-21

© 2024 Siemens
 24. Laminates

Equation 24-30.

In general, for any ply, Zi − 1 < z <zi, the shear is

Equation 24-31.

At any ply interface, zi, the shear is therefore

Equation 24-32.

where Tj = zj − (zj − 1).

Note that the shear at the top face, (τxz)n, is zero and therefore

Equation 24-33.

Eq. 24-33 proves that if is the bending center, the shear at the top surface must be zero.

24-22 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Classical Lamination Theory for Shell Composites (PCOMP,PCOMPG)

Eq. 24-31 could be substituted into Eq. 24-22 and integrated. A better form of Eq. 24-33, for this
purpose is

Equation 24-34.

where

Equation 24-35.

Substituting Eq. 24-34 into Eq. 24-25 and integrating the results, we obtain

Equation 24-36.

where

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 24-23

© 2024 Siemens
 24. Laminates

Equation 24-37.

This expression for the inverse shear modulus for the x-direction may be generalized to provide for the
calculation of each term in the two-by-two matrix of shear moduli.

Equation 24-38.

where:

k = 1,2
l = 1,2

Note that if no shear is given, [Gi]−1, and also that in Eq. 24-31

where [G2]* is calculated in the same manner as [G2] except that Poisson's Ratio is set to zero.

The moduli for individual plies are provided through user input because, in general, G12 ≠ G21 will be
used for the coupling terms. Finally,

Equation 24-39.

24-24 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Laminate Failure Indices and Strength Ratios

As an example, consider a single layer element. For this case let zi − 1 = −T/2, , fo = 0, and EI =
ET3/12. Evaluating Eq. 24-38 we obtain

Equation 24-40.

and

Equation 24-41.

which provides the correct factor for a nonuniform shear distribution in a plate.

The coefficients of thermal expansion derived for membrane-bending coupling, which appear in the
A1, A2, and A12 fields of MAT2 entry and correspond to the MID4 Field on the PSHELL, require special
interpretation. They are given by:

Equation 24-42.

To obtain the actual values of A1, A2, and A12 {αMAT2}, you must solve Eq. 24-42.

Laminate Failure Indices and Strength Ratios

When you use the shell composites (PCOMP/PCOMPG) or solid composites (PCOMPS), you can optionally
output a failure index and strength ratio for individual laminae and for the laminate bonding material.

Failure indices assume a value of one on the periphery of a failure surface in stress space.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 24-25

© 2024 Siemens
 24. Laminates

• If the failure index is < 1, the lamina stress is interior to the periphery of the failure surface and the
lamina is predicted to be safe.

• If the failure index is > 1, the lamina stress is exterior to the periphery of the failure surface and the
lamina is predicted to have failed.

The failure indices represent a phenomenological failure criterion in that only an occurrence of a failure
is indicated and not the mode of failure.

Strength ratio is a more direct indicator of failure than the failure index since it demonstrates the
percentage of applied load to the failure criteria. Strength ratio is defined as:

Strength Ratio (SR) = Allowable Stress / Calculated Stress

For example, a SR = 0.75 not only indicates that a failure has occurred, but also indicates that the
applied load is 25% beyond the allowable. A FI = 1.25 on the other hand does not represent a
percentage of failure; only that a failure condition exists.

See the remarks on the PCOMP, PCOMPG, and PCOMPS bulk entries in the Simcenter Nastran Quick
Reference Guide for the input requirements needed to request failure indices and strength ratio output.

Failure Theories for Shell Composites (PCOMP, PCOMPG)

You can use the FT field on the PCOMP or PCOMPG entries to request a failure theory which applies to all
plies. The allowable entries in the FT field are listed below.

“HILL” for the Hill Theory (HILL) for Shell Composites.

“HOFF” for the Hoffman Failure Theory (HOFF) for Shell Composites.

“TSAI” for the Tensor Polynomial Theory of Tsai-Wu (TSAI) for Shell Composites.

“STRN” for the Maximum Strain Theory (STRN) for Shell Composites.

Note:
If you specify a failure theory in the FT field on the PCOMP or PCOMPG entries, you must also
specify the allowable shear stress of the bonding material in the SB field.

Hill Theory (HILL) for Shell Composites

The Hill failure theory for orthotropic materials that have the same strength in tension and compression,
i.e., xt = xc and yt = yc can be expressed as:

24-26 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Failure Theories for Shell Composites (PCOMP, PCOMPG)

Equation 24-43.

X is allowable stress in 1-direction

Y is allowable stress in 2-direction
S is allowable stress in shear

and X = Xt if σ1 is positive or X = Xc if σ1 is negative and similarly for Y and σ2. For the interaction term
σ1σ2/X2, X = Xt if σ1 is positive or X = Xc if σ1 is negative.

A plot of the above equation obtained by setting the failure index to 1 on the σ1-σ2 plane yields an
ellipse and is the anisotropic yield criterion of Hill (modified later by Tsai, and hence also sometimes
known as the Tsai-Hill theory). Therefore, if the failure index so calculated is less than 1, the ply stresses
are inside the yield ellipse and the ply is said to be “safe”; conversely, if the failure index is greater than 1,
the ply stresses are outside the yield ellipse and the ply has failed.

Replacing the applied stress with (SR · applied stress), the Hill Failure Criteria can be rewritten in terms of
a strength ratio:

The above equation can be rearranged into quadratic equation format:

giving:

The value of a in the quadratic equation format matches the value for the Hill failure index (FI).

Substituting the values of a = FI, b = 0, and c = -1 into the general quadratic equation solution:

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 24-27

© 2024 Siemens
 24. Laminates

gives the Hill strength ratio:

Hoffman Failure Theory (HOFF) for Shell Composites

The Hoffman theory for an orthotropic lamina in a general state of plane stress with unequal tensile and
compressive strengths is given by

Equation 24-44.

The failure index is obtained by evaluating the left-hand side of the above equation.

Note that this theory takes into account the difference in tensile and compressive allowable stresses by
using linear terms in the equation.

To calculate the strength ratio, the following terms are defined:

F1 =

F2 =

F11 =

F22 =

F66 =

Substituting above terms into Hoffman FI equation and setting FI = 1:

24-28 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Failure Theories for Shell Composites (PCOMP, PCOMPG)

Replacing the applied stress with (SR · applied stress), the Hoffman Failure Criteria can be rewritten in
terms of a strength ratio:

Rearranging into quadratic equation format:

Using the general quadratic equation solution:

The Hoffman strength ratio is calculated using the following roots:

Tensor Polynomial Theory of Tsai-Wu (TSAI) for Shell Composites

The theory of strength for anisotropic materials proposed by Tsai and Wu specialized to the case of an
orthotropic lamina in a general state of plane stress with unequal tensile and compressive strengths is

Equation 24-45.

where:

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 24-29

© 2024 Siemens
 24. Laminates

F1 =

F2 =

F11 =

F22 =

F66 =

and F12 is to be evaluated experimentally.

The magnitude of F12 is, however, constrained by the following inequality called a “stability criterion”:

Equation 24-46.

The necessity of satisfying the stability criterion, together with the requirement that F12 be determined
experimentally from a combined stress state, poses difficulties in the use of this theory. Narayanaswami
and Adelman1 have suggested that F12 be set to zero and that use of Hoffman's theory or Tensor
Polynomial theory with F12 = 0 is a preferred alternative to the experimental determination of F12. If
you have a value for use with F12 in the theory, you can input that value in the MAT8 Bulk Data entry;
otherwise, with this failure theory, Simcenter Nastran sets F12 to 0.0. The left-hand side of the above
equation will be evaluated as the failure index by this theory.

Replacing the applied stress with (SR · applied stress), the Hoffman Failure Criteria can be rewritten in
terms of a strength ratio:

Rearranging into quadratic equation format:

1 Narayanaswami, R., and H. M. Adelman, “Evaluation of the Tensor Polynomial and Hoffman Strength Theories for
Composite Materials,” Journal of Composite Materials, Vol. II, 1977, p. 366.

24-30 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Failure Theories for Shell Composites (PCOMP, PCOMPG)

Using the general quadratic equation solution:

The TSAI-Wu strength ratio is calculated using the following roots:

Maximum Strain Theory (STRN) for Shell Composites

The midplane strains and curvatures are available in the element coordinate system. From these, the
stresses and strains in each individual lamina along the fiber direction and transverse direction can be
easily calculated. You can use the STRAIN case control command to request the output of lamina strains.

The maximum strain criteria has no strain interaction terms. The strain allowables specified on the MAT8
entry for each lamina include

Xt,Xc Allowable strains in tension and compression, respectively, in the

longitudinal direction.

Yt,Yc Allowable strains in tension and compression, respectively, in the

transverse direction.

S Allowable strain for inplane shear.

The failure index is calculated using

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 24-31

© 2024 Siemens
 24. Laminates

and i.e.,

where ε1, ε2,γ12 are the elastic strains (total strains minus

thermal strains).

Because you must understand which mode of failure index is critical; i.e., longitudinal (1), transverse
(2), or shear (12), Simcenter Nastran prints the mnemonic 1, 2 or 12 alongside the FP value to indicate
the critical direction.

Note: There is no change in the way the failure index is calculated for interlaminar shear stresses.

When you use the maximum strain theory, you may want to specify lamina stress allowables instead of
strain allowables on the MAT8. To do this, leave the STRN field on the MAT8 entry blank.

For this case, Simcenter Nastran calculates the failure indices using

and

that is,

To calculate the Maximum Stress (Strain) strength ratio, the failure index which is defined as FI =
Calculated Stress / Allowable Stress is set to unity, and the applied stress is replaced with (SR · applied
stress). The result is:

24-32 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Failure Theories for Solid Composites (PCOMPS)

Failure Theories for Solid Composites (PCOMPS)

To request that the software compute a ply and/or bonding failure index for solid composites, the
STRESS case control command must be present, SOUTi on the PCOMPS bulk entry must be set to “YES”,
the parameter SRCOMPS must be set to "NO" (default), and the following must be defined.

• For a stress or strain ply failure index, you must define the FTi field on the PCOMPS entry to request a
specific failure theory for each ply. In addition, the stress or strain allowables must be defined on the
referenced MATFT bulk entry. The allowable entries in the FTi field are as follows.

“HILL” for the Hill Theory for Solid Composites.

“HOFF” for the Hoffman Failure Theory for Solid Composites.

“TSAI” for the Tsai-Wu Failure Theory for Solid Composites.

“STRN” for the Maximum Strain Theory for Solid Composites.

“STRS” for the Maximum Stress Theory for Solid Composites.

“TS” for the Maximum Transverse Shear Failure Theory for Solid Composites.

To request that the software compute a ply strength ratio (SR) for solid composites, you must request
the failure indicies as described above, and you must define PARAM,SRCOMPS,YES.

Note:
When PARAM,SRCOMPS,YES is defined with solid composites, the software only computes the
strength ratio, and not both the failure index and strength ratio.

The failure theories are applied at each position that stress output is requested using the CPLYMID,
CPLYBT, or CPLYBMT describers on the STRESS case control command. Thus, for the solid composite,
you can evaluate the potential for failure at the top, middle, and bottom of each ply. By default, the
failure index output prints in the f06 file even when using the PLOT or PUNCH describers on the STRESS
and STRAIN case control commands. The parameter entry PARAM,NOFISR,1 can be used to turn off the
printing of the failure index output.

Hill Theory for Solid Composites

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 24-33

© 2024 Siemens
 24. Laminates

where,

X is the allowable stress in the 1-direction,

Y is the allowable stress in the 2-direction,

Z is the allowable stress in the 3-direction,

S12 is the maximum allowable in-plane shear stress,

S13 is the maximum allowable 23 shear stress, and

S13 is the maximum allowable 13 shear stress.

Hoffman Failure Theory for Solid Composites

24-34 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Failure Theories for Solid Composites (PCOMPS)

where,

Xt, Xc are the maximum allowable stresses in the 1-direction in tension and compression,

Yt, Yc are the maximum allowable stresses in the 2-direction in tension and compression,

Zt, Zc are the maximum allowable stresses in the 3-direction in tension and compression,

S12 is the maximum allowable in-plane shear stress,

S23 is the maximum allowable 23 shear stress, and

S13 is the maximum allowable 13 shear stress.

Tsai-Wu Failure Theory for Solid Composites

where,

Xt, Xc are the maximum allowable stresses in the 1-direction in tension and compression,

Yt, Yc are the maximum allowable stresses in the 2-direction in tension and compression,

Zt, Zc are the maximum allowable stresses in the 3-direction in tension and compression,

S12 is the maximum allowable in-plane shear stress,

S23 is the maximum allowable 23 shear stress S13 is the maximum allowable 13 shear stress,

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 24-35

© 2024 Siemens
 24. Laminates

F12 is the interactive strength constant for the 12 plane,

F23 is the interactive strength constant for the 23 plane, and

F13 is the interactive strength constant for the 13 plane.

Maximum Strain Theory for Solid Composites

The software reports the maximum from the following six computations.

1. If ε1>0,

If ε1<0,

2. If ε2>0,

If ε2<0,

3. If ε3>0,

If ε3<0,

4.

5.

6.

where,

ε and γ are the elastic strains (total strains minus thermal strains),

24-36 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Failure Theories for Solid Composites (PCOMPS)

Xt, Xc are the maximum allowable strains in the 1-direction in tension and compression,

Yt, Yc are the maximum allowable strains in the 2-direction in tension and compression,

Zt, Zc are the maximum allowable strains in the 3-direction in tension and compression,

S12 is the maximum allowable in-plane shear strain,

S23 is the maximum allowable 23 shear strain, and

S13 is the maximum allowable 13 shear strain.

Maximum Stress Theory for Solid Composites

The software reports the maximum from the following six computations.

1. If σ1>0,

If σ1<0,

2. If σ2>0,

If σ2<0,

3. If σ3>0,

If σ3<0,

4.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 24-37

© 2024 Siemens
 24. Laminates

5.

6.

where,

Xt, Xc are the maximum allowable stresses in the 1-direction in tension and compression,

Yt, Yc are the maximum allowable stresses in the 2-direction in tension and compression,

Zt, Zc are the maximum allowable stresses in the 3-direction in tension and compression,

S12 is the maximum allowable in-plane shear stress,

S23 is the maximum allowable 23 shear stress, and

S13 is the maximum allowable 13 shear stress.

Maximum Transverse Shear Failure Theory for Solid Composites

where,

S23 is the maximum allowable 23 shear stress, and

S13 is the maximum allowable 13 shear stress.

Bonding Material Failure Indices

To request the bonding failure index for shell composites (PCOMP,PCOMPG), you must define the
allowable bonding stress (SB field on the PCOMP or PCOMPG entry). Simcenter Nastran calculates the
failure index of the bonding material (FSB) as the maximum interlaminar shear stress divided by the
allowable bonding stress.

24-38 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Ply Stress and Strain Calculation for Shell Composites (PCOMP,PCOMPG)

As a result, the bonding failure index for the element is the largest value of the failure indices for all
plies of the element. See Calculating Interlaminar Shear Stress and Strains on Shell Composites
(PCOMP,PCOMPG).

To request the bonding failure index for solid composites (PCOMPS):

• You must define the ILFTi field on the PCOMPS entry to request the shear (SB) and/or normal (NB)
inter-laminar failure index.

• You must define the allowable shear and/or normal bonding stress (SB and NB fields on the PCOMPS
entry).

• You must include the CPLYBT or the CPLYBMT describers on the STRESS case control command to
request stress output for the top and bottom of the laminate plies.

Simcenter Nastran calculates the shear (FSB) and normal (FNB) failure indices of the bonding material as
follows.

Bonding Material Strength Ratios

To request the bonding material strength ratio (SR) for shell and solid composites, you must request the
failure indicies as described above, and you must define PARAM,SRCOMPS,YES. The bonding material
strength ratio is computed as follows.

Ply Stress and Strain Calculation for Shell Composites

(PCOMP,PCOMPG)
The parameter NOCOMPS determines if stress and/or strain recovery is at the composite ply layers
(default), on the equivalent PSHELL, or both. The STRESS and/or STRAIN case control commands are
required for any of these recovery options. When ply results are requested, stress and/or strain are
computed on every ply. To print the computed ply stress and/or strain results, the case control command
request must include the “PRINT” option, and SOUTi=YES must be defined on the ply definition.

When ply results are requested, the element strains are transformed to ply strains in the element
coordinate system by:

{εpxy} = {εe} – z{κe}

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 24-39

© 2024 Siemens
 24. Laminates

where:

{εpxy} = ply strains {εpx, εpy, γpxy}

{εe} = element strains {εx, εy, γxy}

z = distance from the neutral axis to the center of the ply

{κe} = element curvatures {κx, κy, κxy}

Note that the ply strains need to be rotated to the ply coordinate system using the following relation:

{εp} = [T]{εpxy}

where:

{εp} = ply strains in the ply coordinate system {ε1, ε2, γ12}

The ply stresses are computed from the ply strains using Hooke's Law and the ply material constitutive
matrix:

The resulting ply stresses and strains are used in the composite failure theories.

24-40 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Calculating Interlaminar Shear Stress and Strains on Shell Composites (PCOMP,PCOMPG)

Calculating Interlaminar Shear Stress and Strains on Shell

Composites (PCOMP,PCOMPG)
The two-dimensional plate theory used in the CQUAD4, CQUAD8, CTRIA3 or CTRIA6 elements doesn't
allow for the exact calculation of interlaminar stresses. High values of these interlaminar stresses can
lead to failures that are unique to composite materials.

Simcenter Nastran uses an approximate technique to determine the interlaminar shear stresses. The
basic assumption in this approximate technique is that the x- and y-components of stress are decoupled
from one another. The approximation consists of neglecting the effects of twisting moments in the
computation of interlaminar shear stresses. For the majority of composite material structural analyses,
this approximation generally yields satisfactory results for all elements in the model except those at the
edges of the structure.

The following equation is used to calculate (τxz)i for the i-th lamina:

Equation 24-47.

is calculated from the following equation:

Equation 24-48.

Similar expressions are used to calculate (τyz) and . For more information, see Classical Lamination
Theory for Shell Composites (PCOMP,PCOMPG).

The interlaminar shear strains are calculated by

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 24-41

© 2024 Siemens
 24. Laminates

and

The interlaminar shear stress and strain is reported in the .f06 output according to the ply number. In
the .f06 output, the interlaminar results reported for ply 1 are associated with the interface between
plies 1 and 2, the results reported on ply 2 are associated with the interface between plies 2 and 3, and
so on. The results reported on the final, or top ply will be zero since there is no interface.

Examples: Defining Laminate Material Properties

The following examples demonstrate how to use PSHELL and PCOMP to model a composite honeycomb
section.

Example: Using PCOMP to Model a Honeycomb Plate

As an example of the PCOMP entry, consider the cantilevered honeycomb plate shown in Figure 24-4.
Although the honeycomb structure is not considered a composite layup, it can be analyzed effectively
using the PCOMP entry.

24-42 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Example: Using PCOMP to Model a Honeycomb Plate

Figure 24-4. Honeycomb Cantilever Plate

The material properties of the honeycomb section are listed in Table 24-8. Note that the face sheets are
not of the same thickness. The result is a nonsymmetric plate (Figure 24-5), which can be noted by the
MAT2 entry for the MID4 field of the equivalent PSHELL.

Table. Honeycomb Material Properties

Modulus of Elasticity Tensile Limit(103 Compression Limit Shea rLimit

Material (106 psi) psi) (103 psi) (103 psi)
Aluminum Face 10.0 35 35 23
Sheets
Core 0.0001 0.05 0.3 0.2
Bonding Material – – – 0.1

Part of the input file for the honeycomb plate is shown in Listing 24-1. The load on the plate is a
4000-pound uniform load acting in the plane of the plate. The output requests include the element
force, laminae stress, laminae strain, and failure index table.

A selected portion of the output is shown in Figure 24-5.

$
$ FILENAME - PCOMP1.DAT
$

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 24-43

© 2024 Siemens
 24. Laminates

CQUAD4 1 100 1 2 6 5
CQUAD4 2 100 2 3 7 6
CQUAD4 3 100 3 4 8 7
CQUAD4 4 100 5 6 10 9
CQUAD4 5 100 6 7 11 10
CQUAD4 6 100 7 8 12 11
$
FORCE 1 12 0 1. 1000.0 0.0 0.0
FORCE 1 4 0 1. 1000.0 0.0 0.0
FORCE 1 8 0 1. 2000.0 0.0 0.0
$
SPC1 1 123456 1 5 9
$
PCOMP 100 100. STRN
120 0.05 0. YES 130 0.35 0.0 YES
120 0.02 0. YES
$
MAT1 120 10.+6 0.3
35.E3 35.E3 23.E3
MAT1 130 100. 0.3
50. 300. 200.

Listing 24-1. Honeycomb Plate

24-44 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Example: Using PCOMP to Model a Honeycomb Plate

Figure 24-5. Composite Output of the Honeycomb Plate

The plate results indicate that the plates are bending downward due to the axial load, which is expected
from a nonsymmetric cross section. The laminae stresses show that the face sheets are highly stressed
while the core has virtually no stress. The failure index table shows the failure for each laminae and the
bonding material to be less than 1, indicating that the structure will not fail. Note, however, that since
each lamina is assumed to be in a state of plane stress, the failure index doesn't take into account the
crushing of the core.

As a final example showing the use of the PCOMP entry, consider the model shown in Figure 24-6.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 24-45

© 2024 Siemens
 24. Laminates

Figure 24-6 shows an example of the data entries required to define a five layer composite material. A
typical element (with grid points 101, 102, 103, and 104) in the model is magnified to illustrate the
individual layers.

24-46 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Example: Using PCOMP to Model a Honeycomb Plate

Figure 24-6. Typical Composite Layup

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 24-47
© 2024 Siemens
 24. Laminates

In the example above, the direction of the x-axis of the material coordinate system (xm) is defined by the
projection of the x-axis of coordinate system 5 onto the element. Using this technique, the direction of
material coordinate system is independent of the shape of the element.

The material direction of all the elements in the model can be defined by referencing a single coordinate
system. The MCID field on the CQUAD4 Bulk Data entry defines the ID of this coordinate system. The
coordinate system is shown in the figure as the x5-, y5-, z5-axis and is defined on the CORD2R Bulk Data
entry.

• The x-axis of the material coordinate system can also be defined with an angle. If the THETA or MCID
field on the CQUAD4 Bulk Data entry is an integer, then the value is taken as a coordinate system. If it
is a real number , then the value is taken as an angle.

• The z-axis (zm) of the material coordinate system is in the direction of the z-axis of the element
coordinate system (ze).

• The direction of the y-axis (ym) follows the right-hand rule.

The CQUAD4 entry references the PCOMP entry, which defines the following layup:

Ply Material Thickness Angle

1 2 0.003 0.0
2 1 0.005 45.0
3 1 0.005 0.0
4 1 0.005 -45.0
5 3 0.007 90.0

The composite material is made of 5 plies. Ply number 1 is at the most negative zm. Note that since
zm is in the direction of ze, the element grid point connectivity order determines which is the top and
bottom surface of the composite and thus, the stacking direction. Ply angle 0 is in the direction of xm.
The positive sense of the ply angle is in the direction from xm to ym. The PCOMP entry references the
MAT8 data entries by the MID field.

The example uses three different materials. Note that strain allowables are specified on the MAT8 entry.
To let Simcenter Nastran know that these values are strain allowable, the STRN field must be set equal to
1.0.

• Material 1 is typical of a graphite epoxy tape. A tape has continuous fibers that all point in the same
direction. Note that E1 is ten times the value of E2.

• Material 2 is also a tape.

• Material 3 is a cloth since E1 equals E2. A cloth made of woven fibers has approximately the same
number of fibers going in the 1 direction as in the 2 direction.

24-48 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Example: Using PSHELL to Model a Honeycomb Section

Directions 1 and 2 refer to the ply coordinate system. The ply material properties on the MAT8 entry
are in the ply coordinate system.

• The 1-axis of the ply coordinate system is in the direction of the ply angle.

• The 3-axis is in the direction of zm (or ze).

• The 2-axis follows the right-hand rule.

Ply results are output in the ply coordinate system.

Example: Using PSHELL to Model a Honeycomb Section

If you're only interested in the overall behavior of the honeycomb composite section, then you can
use an equivalent PSHELL/MAT2 or PSHELL/MAT1 combination to model the PCOMP/MAT8 effect. This
representation yields the correct equivalent stiffness and, consequently, the correct displacements.
However, the stresses in general are meaningless since the information regarding the individual layers
are not available to Simcenter Nastran with the PSHELL/MATi combination. If you want to obtain the
stresses, you should use the PCOMP/MAT8 combination.

This example shows how to model the honeycomb section shown in Figure 24-7 using PSHELL/MAT1.
The assumptions are that the membrane and bending loads are carried by the facesheets and the
transverse shear loads are carried by the honeycomb core. The cross section consists of two 0.01 in thick
aluminum facesheets sandwiching a 0.75 in thick honeycomb core. The properties of the section are
shown in Figure 24-7.

Figure 24-7. Honeycomb Cantilever Plate Represented by PSHELL/MAT1

The calculation of the equivalent properties is shown below:

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 24-49

© 2024 Siemens
 24. Laminates

The transverse shear moduli (G13 and G23) are different in the two transverse directions, which is
typical of honeycomb core. Since the MAT1 entry only allows a single shear modulus, an average Gcore is
used.

A more accurate representation of G13 and G23 is to use the MAT2 entry instead of the MAT1
entry. Listing 24-2 and Listing 24-3 show the input files using the PCOMP/MAT8 and PSHELL/MAT1
entries, respectively. As shown in the corresponding abridged output files in Figure 24-8 and Figure
24-9, the displacements calculated by using the PSHELL/MAT1 combination are comparable with the
displacements obtained using the PCOMP/MAT8 combination.

$ filename - pcomp2.dat
ID LINEAR,PCOMP2
SOL 101
TIME 5
CEND
TITLE = HONEYCOMB COMPOSITE CANTLEVER PLATE
SUBTITLE = SYMMETRIC - .01/.75/.01
$
SPC = 1
DISP = ALL
$
SUBCASE 1
LABEL = AXIAL LOAD

24-50 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Example: Using PSHELL to Model a Honeycomb Section

LOAD = 1
$
SUBCASE 2
LABEL = VERTICAL SHEAR LOAD
LOAD = 2
$
SUBCASE 3
LABEL = VERTICAL TWIST LOAD
LOAD = 3
$
BEGIN BULK
$
PARAM POST -1
$
GRDSET 6
GRID 1 0.0 -10.
0.0
GRID 2 10. -10.
0.0
GRID 3 20. -10.
0.0
GRID 4 30. -10.
0.0
GRID 5 0.0 0.0
0.0
GRID 6 10. 0.0
0.0
GRID 7 20. 0.0
0.0
GRID 8 30. 0.0
0.0
GRID 9 0.0 10.
0.0
GRID 10 10. 10.
0.0
GRID 11 20. 10.
0.0
GRID 12 30. 10.
0.0
$

Listing 24-2. Honeycomb Sandwich Using PCOMP (Continued)

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 24-51

© 2024 Siemens
 24. Laminates

CQUAD4 1 100 1 2 6 5
CQUAD4 2 100 2 3 7 6
CQUAD4 3 100 3 4 8 7
CQUAD4 4 100 5 6 10 9
CQUAD4 5 100 6 7 11 10
CQUAD4 6 100 7 8 12 11
$
FORCE 1 12 0 1. 1000.0 0.0 0.0
FORCE 1 4 0 1. 1000.0 0.0 0.0
FORCE 1 8 0 1. 2000.0 0.0 0.0
$
FORCE 2 12 0 1. 0. 0. 1.
FORCE 2 4 0 1. 0. 0. 1.
FORCE 2 8 0 1. 0. 0. 2.
$
FORCE 3 12 0 1. 0. 0. 1.
FORCE 3 4 0 1. 0. 0. -1.
$
SPC1 1 123456 1 5 9
$
PCOMP 100
120 0.01 0. YES 130 0.75 0.0 YES
120 0.01 0. YES
$
MAT8 120 10.+6 10.+6 0.3 3.85+6
MAT8 130 1000. 1000. 100. 1.+5 0.5+5
ENDDATA

Listing 24-2. Honeycomb Sandwich Using PCOMP

$ pshell2.dat
ID LINEAR,PSHELL2
SOL 101
TIME 5
CEND
TITLE = EQUIVALENT HONEYCOMB COMPOSITE CANTLEVER PLATE
SUBTITLE = SYMMETRIC - .01/.75/.01
$
SPC = 1
DISP = ALL
$
SUBCASE 1
LABEL = AXIAL LOAD
LOAD = 1
$
SUBCASE 2

24-52 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Example: Using PSHELL to Model a Honeycomb Section

LABEL = VERTICAL SHEAR LOAD

LOAD = 2
$
SUBCASE 3
LABEL = TWIST LOAD
LOAD = 3
$

Listing 24-3. Honeycomb Sandwich Using PSHELL/MAT1 (Continued)

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 24-53

© 2024 Siemens
 24. Laminates

BEGIN BULK
PARAM POST -1
$
GRDSET 6
GRID 1 0.0 -10.
0.0
GRID 2 10. -10.
0.0
GRID 3 20. -10.
0.0
GRID 4 30. -10.
0.0
GRID 5 0.0 0.0
0.0
GRID 6 10. 0.0
0.0
GRID 7 20. 0.0
0.0
GRID 8 30. 0.0 0.0
GRID 9 0.0 10.
0.0
GRID 10 10. 10.
0.0
GRID 11 20. 10.
0.0
GRID 12 30. 10.
0.0
$
CQUAD4 1 100 1 2 6 5
CQUAD4 2 100 2 3 7 6
CQUAD4 3 100 3 4 8 7
CQUAD4 4 100 5 6 10 9
CQUAD4 5 100 6 7 11 10
CQUAD4 6 100 7 8 12 11
$
FORCE 1 12 0 1. 1000.0 0.0 0.0
FORCE 1 4 0 1. 1000.0 0.0 0.0
FORCE 1 8 0 1. 2000.0 0.0 0.0
FORCE 2 12 0 1. 0. 0. 1.
FORCE 2 4 0 1. 0. 0. 1.
FORCE 2 8 0 1. 0. 0. 2.
FORCE 3 12 0 1. 0. 0. 1.
FORCE 3 4 0 1. 0. 0. -1.
$
SPC1 1 123456 1 5 9
PSHELL 100 100 .02 100 4332. 200 38.5 +PSLL
+PSLL -.385 .385
MAT1 100 10.+6 0.3

24-54 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Example: Using PSHELL to Model a Honeycomb Section

MAT1 200 1000. .75+5

$
ENDDATA

Listing 24-3. Honeycomb Sandwich Using PSHELL/MAT1

Figure 24-8. Displacement Outputs for Honeycomb Sandwich Using PCOMP

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 24-55

© 2024 Siemens
 24. Laminates

Figure 24-9. Displacement Outputs for Honeycomb Sandwich Using PSHELL/MAT1

24-56 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 25. Cyclic Symmetry
Overview of Cyclic Symmetry
Many structures, including pressure vessels, rotating machines and antennas for space communications,
are made up of virtually identical segments that are symmetrically arranged with respect to an axis.

When you're performing analyses on symmetrical structures, you can exploit that symmetry to reduce
greatly reduce the time needed to create and solve an FE model. One advantage is that you will
be required to provide an analytical model for only the “fundamental regional” of a structure, and
this fundamental region is generally substantially smaller than the analytical model required by more
traditional analysis techniques. Analyzing only the fundamental region lets you:

• Reduce the input data describing the finite element model to just that of the fundamental region

• Restrict the computation of g-size stiffness and mass matrices to just the fundamental region

• Exploit the orthogonality of symmetrical component sets, such as the symmetrical and
antisymmetrical sets in reflective symmetry, to reduce the cost of matrix operations

For example, if a body consists of N identical segments in the sense that the entire body can be
generated through reflections and/or rotations of one of these N segments (the fundamental region) ,
you only have to define a finite element model for the fundamental region and provide a list of the
boundary grid points for this fundamental region. With Simcenter Nastran's cyclic symmetry option
(the CYSYM bulk data entry), the software performs all these tasks automatically for each of the three
supported subclasses of cyclic symmetry: rotational symmetry (ROT), dihedral symmetry (DIH), and
axisymmetry (AXI).

Supported Solution Sequences

In Simcenter Nastran, you can use cyclic symmetry techniques in the following solution sequences:

SOLution 114 Superelement Static Analysis

SOLution 115 Superelement Normal Modes Analysis
SOLution 116 Cyclic Buckling
SOLution 118 Superelement Direct Frequency Response

Comparing Cyclic Symmetry and Traditional Reflective Symmetry Techniques

The following table compares the major steps in the analysis procedures required with traditional
reflective techniques and with cyclic symmetry techniques and highlights some of the many advantages
of cyclic symmetry.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 25-1

© 2024 Siemens
 25. Cyclic Symmetry

Reflective Symmetry Cyclic Symmetry

Analytical Reduces the size of the analytical Reduces the size of the model to a
Model model by one-half for each plane “fundamental region,” which is generally
of reflective symmetry. substantially smaller than the entire model

Boundary In the general case of p planes You supply a list of the grid points
Conditions of reflective symmetry, you must on the boundary of the fundamental
explicitly define 2p sets of region. Boundary conditions are automatically
boundary conditions with SPCi Bulk imposed by the program.
Data entries.

Applied You must decompose the various Applied loads are defined to act at arbitrary
Loads loading conditions into symmetric locations throughout the entire structure
and antisymmetric components. respective of whether any or all of these
locations are in the explicitly defined
fundamental region.

Computer For each set of defined boundary A static condensation is performed once to
Solution conditions, a lengthy matrix reduce the size of the analysis set. In static
decomposition and equation analysis, all but the boundary degrees-of-
solution are required. Static freedom should be omitted. In real eigenvalue
condensations aren't generally analysis, buckling analysis, and dynamic
economically feasible in static analysis, the boundary degrees-of-freedom
analysis. and selected internal points are kept in the
analysis set. The static condensation involves
a lengthy decomposition which should be less
lengthy than any one of the decompositions
required by traditional techniques because
it only involves the fundamental region.
Several additional decompositions of relatively
trivial magnitude are required after the static
condensation.

Data You must manually combine the Simcenter Nastran automatically provides the
Recovery results for each boundary condition response for the entire structure without any
with SUBCOM or SYMCOM special user action. SUBCOM and SYMCOM
commands to obtain the complete entries aren't necessary and shouldn't be used.
response of the structure.

Understanding Types of Symmetry

There are a number of different types of geometric symmetry, as shown in Figure 25-1.

Cyclic symmetry is a subclass of the general class of geometric symmetry. Cyclic symmetry is defined by
the rotation of a fundamental region about an axis. Within the category of cyclic symmetry, there are a

25-2 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Understanding Types of Symmetry

number of different subtypes. You can use Simcenter Nastran to perform cyclic symmetry analyses on
the following types of geometric symmetry:

• Rotational symmetry

• Dihedral symmetry (a combination of reflective plus rotational symmetry)

• Axisymmetry (with and without skew)

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 25-3

© 2024 Siemens
 25. Cyclic Symmetry

Figure 25-1. Types of Geometrical Symmetry

Most of the types of symmetry shown in Figure 25-1 are subclasses of cyclic symmetry. There are inter-
relationships between these different subclasses. For example, translational symmetry in one direction,
see Figure 25-1(d), can be considered to be a limiting case of rotational symmetry in which the axis
of symmetry is infinitely far away. Likewise, axisymmetry without skew is a limiting case of dihedral
symmetry as the number of segments is increased without limit, and axisymmetry with skew is a
limiting case of rotational symmetry.

The relationship of the different classes of symmetry to the general class called cyclic symmetry is
summarized in Table 25-3. With few exceptions, all important types of symmetry are subclasses of
cyclic symmetry, and they can, therefore, be analyzed with the cyclic symmetry capability in Simcenter
Nastran.

25-4 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Understanding Types of Symmetry

Table. Types of Symmetry

Subclass of Cyclic
Type Symmetry
Reflective Symmetry
One Plane Yes
Two Planes Yes
Three Planes No
Rotational Symmetry (Repetition about an axis) Yes
Dihedral Symmetry (Reflective symmetry plus rotational symmetry) Yes
Translational Symmetry
One-D Yes (special case)
Two-D No
Three-D No
Axisymmetry
With Skew Yes
Without Skew Yes

Note that in more general cases, a model may be susceptible to both dihedral and rotational symmetry.
For example, consider the three regular N-gons: the equilateral triangle, the square, and the pentagon.

In these figures, the internal solid lines in conjunction with the edges define fundamental regions, e.g.,
A0B, that may be permuted through rotational symmetry operations. The axis of rotation is normal to
the plane of the page and passes through the point 0.

In addition, the dotted internal lines such as a0 may be used to, for example, reflect A0a into a0B. Also,
a0B may be reflected through B0 to obtain B0b, and so on around each N-gon until the identity of
A0a is reached. Clearly, the fundamental region A0a for the dihedral symmetry operations is smaller

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 25-5

© 2024 Siemens
 25. Cyclic Symmetry

than the fundamental region A0B for rotational symmetry operations. Importantly, both of these
fundamental regions are substantially smaller than the fundamental region required by traditional
reflective symmetry techniques.

Cyclic Symmetry Theory

Simcenter Nastran's cyclic symmetry capabilities rely on the development of transformation relations
that let you manipulate the mathematical representation of the fundamental region in accordance with
the appropriate symmetry operations. The following operations are performed:

1. You create a Simcenter Nastran input file for the fundamental region.

2. You define the mathematical representation of the model in terms of physical quantities, such as
[K]{u} = {P}

3. The software transforms the mathematical representation of the model to cyclic quantities, such as
[K]{u} = {P}

4. The software solves the problem in terms of cyclic quantities.

5. The results are transformed from the cyclic quantities to physical quantities.

Steps 3 and 5 both depend on the existence of the necessary transformations. Step 4 is also of interest
in that it must be demonstrated that problem solution procedures in terms of cyclic quantities are valid
and require less computer time than more traditional solution procedures. These matters are addressed
in the following sections.

Symmetrical Components

Consider an arrangement of N identical objects symmetrically disposed about a normal. The angle
between these objects is

Equation 25-1.

and, in addition, let U1 (r,t) be any variable displacement, force, stress, temperature, etc. in object 1,
the fundamental region. The variables r and t denote distance and time respectively. The corresponding
quantity in the n-th object is Un (r,t) and as a single valued solution is required

25-6 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Symmetrical Components

Equation 25-2.

The various Un's are, in general, unsymmetrical quantities that are not related in a simple manner.

Fortescue's symmetrical component theory states that a general vector field on N-vectors with arbitrary
phase relationships may be resolved into N-sets of N-symmetrically phased vectors of equal magnitude.
For N = 3, the vectors may be illustrated by the following diagram.

Now, in accordance with the theory of symmetrical components, the following vector sums may be
stated.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 25-7

© 2024 Siemens
 25. Cyclic Symmetry

Equation 25-3.

Here, the Un's represent physical quantities and the xkn terms represent symmetrical components. The
superscripts k denote the order of the symmetric components. The subscripts n are taken to be the
number of a particular vector in the vector field. In the general context of cyclic symmetry, the subscripts
n denote the number applied to each of the N identical objects under consideration. The object n = 1 is
arbitrarily taken as the fundamental region.

Understanding Rotational Symmetry

In rotational symmetry, as shown in the figure Types of Geometrical Symmetry (b), the structure is
comprised of identical segments that are symmetrically arranged about an axis. Common examples of
structures with rotational symmetry are skew slabs and many types of rotating machinery including
gears, electric motors, turbines, and helicopter rotors.

With rotational symmetry, if the structure has only two segments, the second segment is created by
rotating the first segment through 180 about an axis A. In general, the n-th segment is created by
rotating the fundamental region (i.e., the first segment) through 360(n-1)/N degrees about axis A where
N is the total number of segments.

Mathematically, a structure has rotational symmetry if each of its relevant physical properties Q repeats
at even angular intervals around an axis A. Thus, referring to the structure shown in Figure 25-2,

Equation 25-4.

where:

θn =

25-8 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Transforming Physical Components to Cyclic Components (ROT)

Figure 25-2. Conforming Boundaries for the ROT Option

You can apply rotational symmetry by using the following procedure:

1. Model one segment. Segment boundaries may be curved surfaces.

2. Ensure each segment has its own coordinate system that rotates with the segment. The local
displacement coordinate system must conform at the joining points.

3. Specify a paired list of points on side 1 and side 2 that are to be joined.

Transforming Physical Components to Cyclic Components (ROT)

These transformations for the cyclic symmetric components in terms of the general field vectors may be
written in summation form for the general case of N vectors.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 25-9

© 2024 Siemens
 25. Cyclic Symmetry

Equation 25-5.

where a is given by the following equation:

Equation 25-6.

For more information, see Symmetrical Components.

Transforming Cyclic Components Back to Physical Components (ROT)

After the solution phase of the problem, it will normally be necessary to transform the results from cyclic
components back into physical components.

The following equation

Equation 25-7.

can be written in the following form.

25-10 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Transforming Cyclic Components Back to Physical Components (ROT)

Equation 25-8.

where,

and

If the exponentials in the bracketed terms in the above summation are replaced by the relations e±ix =
cosx ± sinx, the typical bracketed term takes the following form:

The purpose for the somewhat arbitrary introduction of Eq. 25-8 is now recognized, and the typical
bracketed term assumes the following form:

The final form of the transformation from cyclic components to physical components is therefore,

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 25-11

© 2024 Siemens
 25. Cyclic Symmetry

Figure 25-3.

In Eq. 25-3, Un represents physical quantities such as displacement force, stress, temperature in the n-th
segment in terms of the cyclic components in the fundamental region. The following points should be
noted:

1. and xN/21 is nonexistent when N is odd.

2. when N is even.

3. The notation for cyclic components of order k = 0 and k = N/2 is distinct from the notation for
cyclic components of order 1 ≤ k ≤ kL because of the somewhat peculiar nature of these two
components.

Boundary Conditions and Problem Solution (ROT)

The transformation equations of Cyclic Symmetry Theory are assumed to allow one to transform a
problem that is mathematically defined in terms of physical components into a set of mathematical
statements written in terms of cyclic components. Specifically, the mathematical description of the n
replications of the fundamental region may be written in the form

Equation 25-9.

This expression represents the mathematical description of the system in terms of physical coordinates.

The application of previously defined transform equations may now be introduced to obtain the desired
mathematical description of the system in terms of cyclic components. First, however, consider the
matrix representation of the transform and the final form of the transformation from cyclic
components to physical components. The matrix representation of the transform assumes the form

25-12 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Boundary Conditions and Problem Solution (ROT)

Equation 25-10.

and the final form of the transformation from cyclic components to physical components assumes
the form

Equation 25-11.

It is easily demonstrated that

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 25-13

© 2024 Siemens
 25. Cyclic Symmetry

Return now to Eq. 25-9, the mathematical representation of the system in terms of physical coordinates,
and represent the energy of the system in the form

Equation 25-12.

From Eq. 25-9, Eq. 25-11, and Eq. 25-12

Equation 25-13.

The system energy in terms of cyclic components may be written as

Equation 25-14.

Since the energy must be invariant to the transformations, Eq. 25-14 and Eq. 25-15 may be equated

Equation 25-15.

25-14 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Boundary Conditions and Problem Solution (ROT)

That is, the equations of the system in terms of physical components are transformed into a set of
equations in terms of cyclic components. As there is no coupling between the N identical objects of the
system, the matrix Yn may be written so that it consists of N identical blocks along the diagonal. These
blocks represent, in general, the equations of motion of each of the N identical objects of the system.

As an example, assume N = 4 (KL = 1), and expand Eq. 25-15. Note that Y1 = Y2 = Y3 = Y4.

In general, for N objects the above equation may be written

These equations imply that the equations of motion in terms of cyclic components are either N or N/2
times the corresponding equations written in terms of physical components. This argument provides an
explanation for the presence of the coefficients that appear in the terms of Eq. 25-10.

To this point, it has been found that except for boundary conditions, the equations for motion in
terms of cyclic components (k = 0, 1, ..., N/2) may be developed from the equations of motion of the
fundamental region (n = 1). In essence, an arbitrarily large physical system may be investigated in terms
of a fundamental region which has been transformed into a number of models defined in terms of cyclic
components. The boundary conditions between cyclic models must now be defined.

The only connection between the n identical objects in question exists along the boundaries between
adjacent segments. Specifically, consider three of the N objects as indicated in the following figure.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 25-15

© 2024 Siemens
 25. Cyclic Symmetry

These objects are displaced from one another to more clearly define the boundary conditions between
the objects. Note the designation of the two sides of each of the elements as side 1 and side 2 as one
proceeds in the counterclockwise direction. From the requirement of displacement compatibility, it is
necessary that the typical boundary degrees-of-freedom, u, satisfy the equations

Equation 25-16.

These physical degrees-of-freedom may be transformed into cyclic components of order K using the
matrix representation of the transform.

Equation 25-17.

25-16 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Boundary Conditions and Problem Solution (ROT)

A comparison of Eq. 25-17 provides the result

Equation 25-18.

for 1 ≤ k ≤ kL. For k = 0,

Equation 25-19.

and for ,

Equation 25-20.

The cyclic component of Eq. 25-20 may be written in terms of the variables Ukc and Uks in accordance
with the definitions.

Eq. 25-18 may now be introduced to yield the equations

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 25-17

© 2024 Siemens
 25. Cyclic Symmetry

Upon the replacement of the exponentials with Euler's equation, the appropriate boundary conditions
may be represented as follow:

Equation 25-21.

These boundary conditions are of the form of constraint equations in which the variables on side 2 are
dependent and the variables on side 1 are independent. Indeed, these equations may be viewed as MPC
equations that couple kc and ks terms of side 1 to the corresponding terms of side 2.

The equations of motion in terms of cyclic components may be written in the following form.

Equation 25-22.

From the previous discussion on boundary conditions, the above equation may be rewritten in terms of
the following disjoint problems.

25-18 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Understanding the Harmonic Index

Equation 25-23.

The boundary Eq. 25-19, Eq. 25-20, and Eq. 25-21 must be introduced to remove the dependent
variables that are defined to be on side 2 of the region. If you also use the OMIT feature, the only
independent degrees-of-freedom in the above disjoint problems are the degrees-of-freedom on side 1
that have not previously been declared dependent through MPC, SPCi, and/or OMITi Bulk Data entries.
Thus, the matrices in the above disjoint problems, although dense, are of small size relative to, say, the
matrices required to perform the same analysis by more traditional methods.

Understanding the Harmonic Index

The harmonic index k plays an important role in the analysis of structures with cyclic symmetry. The
harmonic index k has the values k = 1, 2, ..., N/2. If the number of segments is large, the higher
harmonics may contribute very little to the total solution. In Simcenter Nastran you can select the
harmonics (values of k) to use in the analysis.

If the only nonzero symmetrical load components are for a particular value of k, then the only nonzero
symmetrical response components are for the same value of k, provided only that the structure is
linear and satisfies the equation in Understanding Rotational Symmetry. This fact leads directly to an
efficient method of analysis in which each symmetric load component is computed from Eq. 25-25 and
applied to a finite element model that is very much smaller than the complete structure.

The symmetrical components are evaluated from the physical components by means of the following
summations that replace the Fourier integrals used in axisymmetric analysis. The argument (p) is
dropped for convenience.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 25-19

© 2024 Siemens
 25. Cyclic Symmetry

Equation 25-24.

Note that stiffness damping, and mass matrices become uncoupled for ūk.

Eq. 25-24 also applies to the expansion of applied loads into symmetrical component sets. Thus

Equation 25-25.

The implied coordinate systems in the physical segments remain fixed in the segments and are rotated
with them. To satisfy continuity between segments, the grid point on side 2 of segment 1 must coincide

25-20 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Rotational Symmetry Example

with the grid point on side 1 of segment 2 and their coordinate systems must be aligned (see the
figure Conforming Boundaries for the ROT Option). As a result, you must align the grid points on
sides 1 and 2 of the fundamental region so that their positions and displacement coordinate systems
differ only by a rotation through the angle a = 2π/N about the axis of symmetry. This is most easily
accomplished by using a cylindrical coordinate system for the grid points on sides 1 and 2. If this is
done, the boundary conditions are satisfied by imposing the following constraints on the symmetrical
components of displacement at corresponding points on sides 1 and 2 (indicated by subscripts).

Equation 25-26.

Rotational Symmetry Example

As an example of the application, consider the simple problem illustrated in Figure 25-4. The skew slab
is divided into two identical segments. The displacement coordinate system for the second segment
is rotated 180° from the first. A cylindrical coordinate system is used (at least along sides 1 and 2)
to ensure the compatibility of displacement components at the points in the boundaries between
segments.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 25-21

© 2024 Siemens
 25. Cyclic Symmetry

Figure 25-4. Example of Rotational Symmetry

Since N = 2, there are two symmetrical component sets ū0 and ūN/2 = ū1. The symmetrical load
components, as computed from the following equation.

25-22 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Rotational Symmetry Example

Equation 25-27.

are as follows:

Equation 25-28.

and are applied to separate copies of the finite element model for the fundamental region
labeled the K = 0 model and the K = 1 model. The boundary condition linking all six degrees of freedom
at corresponding pairs of grid points on sides 1 and 2, as computed from the following equation,

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 25-23

© 2024 Siemens
 25. Cyclic Symmetry

Equation 25-29.

are

Equation 25-30.

These boundary conditions can be supplied in Solution 101 with multipoint constraints; however, they
are provided automatically in Simcenter Nastran's cyclic symmetry capability.

The K = 0 and K = 1 models are solved in the usual way to obtain the symmetrical response components
ū0 and ū1. As a final step, using the following equation,

25-24 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Dihedral Symmetry

Equation 25-31.

the physical responses are evaluated as follows:

Equation 25-32.

Note: You could use the static condensation procedure in this problem. With that procedure, all the
interior degrees of freedom are placed in the uo set and eliminated, resulting in a greatly reduced
stiffness matrix for the remaining degrees of freedom on the boundary (sides 1 and 2). Since this only
needs to be done once, there is a net saving of computer time.

See also

• “Static Condensation (Guyan Reduction)” in the Simcenter Nastran User's Guide

Dihedral Symmetry

Dihedral symmetry, as shown in the figure Types of Geometrical Symmetry (c), is a special case of
rotational symmetry that combines rotational and reflective symmetry, since each segment has its own
plane of reflective symmetry. Thus, in that figure, half-segment 1L is the mirror image of half-segment
1R (the fundamental region), and these two halves of segment 1 are rotated together through 120° to

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 25-25

© 2024 Siemens
 25. Cyclic Symmetry

form segment 2 and through 240° to form segment 3. Water towers and reinforced domed roofs are
familiar examples of structures with dihedral symmetry.

Dihedral symmetry (DIH) can be useful because:

• DIH takes advantage of reflective symmetry to reduce the fundamental region by one-half relative to
the size of the fundamental region required in the ROT option.

• You can use the DIH option for analyses that would normally be handled with traditional reflective
symmetry techniques.

Figure 25-5 and Figure 25-6 show two different examples of dihedral symmetry and the notation
system used for the DIH option in Simcenter Nastran.

Figure 25-5. Dihedral Symmetry

25-26 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Dihedral Symmetry

Figure 25-6. Dihedral Symmetry

The fundamental region, for which you create a finite element model, is the right half of the first
segment (1R). With dihedral symmetry, all segment boundaries must be planar.

• The plane that forms the boundary between the right and left halves of segment 1 is labeled side 2.

• The other boundary of the fundamental region, which must also be planar, is labeled side 1.

Note that:

• Local displacement systems axes, associated with the inter-segment boundaries, must be in the plane
or normal to the plane.

• In the specific case of a structure with a single overall plane of reflective symmetry (i.e., the complete
structure is represented by segments 1R and 1L) the reflective plane may be defined as side 1 or side
2.

• The two halves of a segment that join at a reflective plane are denoted as the right, “R”, and left, “L”,
halves. The L half segments are mirror images of the R half segments.

• R halves use a right-handed coordinate system and L halves use a left-handed coordinate system.
Therefore, at each boundary a degree-of-freedom is associated with its mirror image on the adjacent
segment so that compatibility always involves a left- and a right-hand coordinate system. The
degrees-of-freedom on the boundaries are further categorized into odd and even components in
accordance with the following table.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 25-27

© 2024 Siemens
 25. Cyclic Symmetry

Component Degrees-of-Freedom

Even Displacement vectors parallel to the boundary and rotation vectors normal to the
boundary.

Odd Displacement vectors normal to the boundary and rotation vectors parallel to the
boundary.

This has important implications for how you input of loads data and interpret the output. The
coordinate systems for other half-segments in Figure 25-5 are either rotated, or rotated and reflected
as indicated in Table 25-5.

Table. Coordinate Systems for Half Segments

Rotation About the Axis of

Half-Segment Type of Coordinate System Symmetry

1R Right-handed 0°

1L Left-handed 0°

2R Right-handed 120°

2L Left-handed 120°

3R Right-handed 240°

3L Left-handed 240°

The typical procedure that you follow when using dihedral symmetry is:

1. You model one half-segment (1R).

2. At a boundary between segments or half-segments, the displacement components must be

parallel and perpendicular to the boundary. The boundaries are planes passing through the axis
of symmetry.

3. You provide separate lists of the points on sides 1 and 2 which must be joined.

As in rotational symmetry, the response quantities at corresponding points in the physical segments can
be expressed as a truncated Fourier series with the polar angle of the n-th segment θn as the argument
where N is the number of segments. The series for response quantities in the right half-segments is

25-28 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Dihedral Symmetry

Equation 25-33.

where kL is the largest integer less than or equal to N/2 (depending on whether n is odd or even) and
ūkc, ūkc*, ūks, and ūks* are symmetrical component sets. The series for response quantities in the left
half-segments is

Equation 25-34.

where, in addition to previously defined terms, a = 2π/N.

The symmetrical component sets are evaluated from the physical components as follows:

Equation 25-35.

Equation 25-36.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 25-29

© 2024 Siemens
 25. Cyclic Symmetry

where, in addition to previously defined terms, δ = 1/2 for k = 0 and N/2, and δ = 1 for all other values of
k.

The number of symmetrical components sets is equal to the number of half-segments in the structure
(neglecting the identically null sets ūks and ūks* for k = 0, N/2). As in the case of rotational symmetry,
the symmetrical load components for a particular value of k produce only the symmetrical response
components for the same value of k, or, to put it another way, the analyses for the different values of k
are uncoupled.

Furthermore, the analyses for the starred (*) and unstarred symmetrical components are uncoupled. For
K = 0 and k = N/2, the complete finite element model for the unstarred terms consists of a single copy
of the fundamental region (1 R). For 0 < k < N/2, the complete finite element model consists of two
copies of the fundamental region with coupling between their boundaries. The boundary conditions are
different for each value of k. The finite element models for the starred (*) terms are identical to those for
the unstarred terms for k > 0. Only the symmetrical load components, which are computed by Eq. 25-35
and Eq. 25-36, are different.

As an example, consider the case of a structure with two planes of reflective symmetry as shown in
the figure Reflective Symmetry with One or Two Planes of Symmetry–A Special Case of Dihedral
Symmetry. Applying dihedral symmetry to this case, you can see that

• The number of segments N = 2.

• The segment angles a = n.

• θ1 = 0.

• θ2 = π

• The harmonic indices k = 0, 1.

Substitution of these values into Eq. 25-35 and Eq. 25-36 gives

25-30 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Summary of Transformations, Boundary Conditions, and Problem Solution (DIH)

Equation 25-37.

Since all of the signs are positive in the equation for ū0c, it follows that ū0cis symmetrical with respect
to both side 1 and side 2 in the figure Reflective Symmetry with One or Two Planes of Symmetry–A
Special Case of Dihedral Symmetry. The other nonzero symmetrical components are antisymmetrical
with respect to one or both of the sides as shown in Table 25-6.

Table. Symmetrical Components for Reflective Symmetry Example

Symmetry
Symmetrical Component Side 1 Side 2
ū0c S S
ū1c S A
ū0c* A A
ū1c* A S

Regardless of the number of segments, the starred terms are always antisymmetrical, and the unstarred
terms are always symmetrical with respect to side 1 of the fundamental region. You can take advantage
of this to reduce the amount of calculation when it can be determined beforehand that the applied
loading has a plane of symmetry or antisymmetry that coincides with side 1 of the fundamental region,
or, in other words, if the problem has an overall plane of symmetry. The amount of computation can be
reduced still further if the problem has a second overall plane of symmetry perpendicular to the first. The
second overall plane of symmetry can exist only if there is an even number of segments.

Summary of Transformations, Boundary Conditions, and Problem Solution (DIH)

The form of the transformations between cyclic and physical components and the boundary conditions
between the segments of the object in question depend somewhat on the notation system.

When you select the DIH option of cyclic symmetry, you can take advantage of overall planes of
symmetry when the applied loads are symmetrically and/or antisymmetrically disposed with respect to
these planes. In these cases, you only need to specify loads for segments in the first half plane when

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 25-31

© 2024 Siemens
 25. Cyclic Symmetry

there is one plane of overall symmetry and in the first quadrant when there are two planes of overall
symmetry.

You specify symmetrical or antisymmetrical boundary conditions on the DSYM Case Control command.
You can only include one DSYM command per subcase. A single plane of reflective symmetry must be
with respect to Side 1, and if a second plane of reflective symmetry exists, it must be perpendicular
to Side 1. In the case of two planes of symmetry, the model must contain an even number of whole
segments.

The following sections describe several of the equations of interest for the DIH option, along with the
equation number for the corresponding expression under the ROT option. In the following equations,
the starred (*) terms donate anti-symmetric motions and unstarred terms denote symmetric motions.
The subscripts R and L refer to the right and left halves respectively.

See also

• “DSYM” in the Simcenter Nastran Quick Reference Guide

Transforming Physical Components to Cyclic Components (DIH)

Equation 25-38.

Here,

Equation 25-39.

and the positive sign is associated with the unstarred terms.

25-32 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Transforming Cyclic Components Back to Physical Components (DIH)

The following is the corresponding equation for the ROT option:

Figure 25-7.

For more information, see Transforming Cyclic Components Back to Physical Components (ROT).

Transforming Cyclic Components Back to Physical Components (DIH)

Equation 25-40.

These relations are derived from the following equation for the ROT option:

Equation 25-41.

For more information, see Boundary Conditions and Problem Solution (ROT).

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 25-33

© 2024 Siemens
 25. Cyclic Symmetry

Boundary Conditions and Problem Solution (DIH)

Side 1, EVEN

Equation 25-42.

Side 1, ODD

Equation 25-43.

Side 2, EVEN

Equation 25-44.

Side 2, ODD

25-34 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Boundary Conditions and Problem Solution (DIH)

Equation 25-45.

In static analysis with cyclic symmetry, the equation

is solved in SSG3 (Static Solution Generator – Part 3) for {ūa}K

and

are formed in CYCLIC3 (Cyclic Symmetry Transformation – Module 3).

The symmetrical components of the loads and are subsets of which is obtained in
CYCLIC3 by multiplying the load vector formed in SSG1 by the GFORE matrix formed in CYCLIC1 module

where {Pl} is formed in SSG2 (Static Solution Generator – Part 2) and the transformation matrix [GFORE]
is formed in CYCLIC1 module.

The cyclic component constraint data (CYCD) used in CYCLIC3 for the formulation of the transformation
matrices [Gck] and [Gsk] are generated in CYCLIC2 (Cyclic Symmetry – Module 2).

The solution {ua}K is transformed to symmetric components in CYCLIC4 as follows:

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 25-35

© 2024 Siemens
 25. Cyclic Symmetry

The symmetric components for each harmonic index are appended to form{ū}x in the CYCLIC4 module.
The physical segment displacements are obtained as follows:

where the transformation matrix is formed in the CYCLIC1 module.

In the case of vibration analysis with cyclic symmetry the following equation is used in READ to
determine the eigenvalues and eigenvectors,

where is formed in a manner similar to in CYCLIC3. The symmetrical components of the

eigenvectors are recovered in CYCLIC4 and the physical segment data is recovered using the GBACK1
matrix created in CYCLIC4 module.

Since all the transformation matrices are extremely sparse, none of the matrix multiplications indicated
above will require large amounts of computer time. The most significant operation is the triple
product associated with the determination of . In the usual application this matrix is dense and
approximately a-size. The computer time for the triple product can be estimated as ten matrix packing
operations on a full matrix of a-size.

See the following equation for the corresponding boundary conditions for the ROT option:

Figure 25-8.

25-36 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Reflective Symmetry

For more information, see Transforming Cyclic Components Back to Physical Components (ROT).

Reflective Symmetry

Reflective symmetry with one or two planes of symmetry is a special case of dihedral symmetry (Figure
25-10).

• Reflective symmetry with one plane of symmetry is identical to dihedral symmetry with one segment.

• Reflective symmetry with two planes of symmetry is identical to dihedral symmetry with two
segments.

Figure 25-9. Reflective Symmetry with One or Two

Planes of Symmetry–A Special Case of Dihedral Symmetry

Mathematically, a structure has reflective symmetry with respect to a yz plane if each of its relevant
physical properties Q (such as mass density and elastic modulus) satisfies the equation

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 25-37

© 2024 Siemens
 25. Cyclic Symmetry

Equation 25-46.

or in other words, if the physical properties are identical at the image points on the two sides of the yz
plane. The plane of symmetry divides the structure into two halves that are arbitrarily called the right
half and the left half. The right half is also called the fundamental region because, in analysis, it is the
half for which a finite element model is prepared, whereas the finite element model for the left half is
only implied.

The physical properties included in Q do not include the loads applied to the structure or its responses
(displacements, internal forces, or stresses). Let the symbols ur, and ul represent the values of some
particular response u at some point in the right half and at its image point in the left half of the
structure. ur and ul don't necessarily satisfy the reflective symmetry property Eq. 25-46, but their
average us clearly does. Thus,

Equation 25-47.

so that

Equation 25-48.

and

Equation 25-49.

25-38 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Reflective Symmetry

Since the right-hand sides of Eq. 25-48 and Eq. 25-49 are equal, it follows that

Equation 25-50.

The quantity us is called the symmetrical part of the response. The antisymmetrical part of the response
ua is defined as one-half of the difference of ur, and ul. Thus,

Equation 25-51.

The antisymmetrical part of the response has the property that

Equation 25-52.

Eq. 25-47 and Eq. 25-51 show how to obtain the symmetrical and antisymmetrical parts of the response
from the values of the response for the left and right halves of the structure. They may be written more
compactly in matrix form

Equation 25-53.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 25-39

© 2024 Siemens
 25. Cyclic Symmetry

The vector

is called the vector of symmetrical components, and the vector

is called the vector of physical components. The physical components may be found from the
symmetrical components by solving for them from Eq. 25-53.

Equation 25-54.

Eq. 25-53 and Eq. 25-54 apply to any response quantity, and they also apply to the applied loads. Thus,
any particular applied load P has symmetrical components Ps and Pa evaluated as follows:

Equation 25-55.

In order to include vectors as well as scalars in u and P, it is necessary to select the coordinate systems
in the right and left halves to be mirror images of each other as shown in Figure 25-10. The coordinate

25-40 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Reflective Symmetry

system in the right half is right-handed, to conform with standard Simcenter Nastran practice. Note,
however, that the coordinate system in the implied left half is left-handed. This means, for example, that
loads applied indirectly to the left half, by listing the images of their points of application in the right
half, must be expressed in the particular left-hand coordinate system shown in Figure 25-10.

Figure 25-10. Reflective Symmetry

If the loads applied to a symmetrical structure are symmetrical, the response is symmetrical. The
responses also include the displacements in the plane of symmetry, i.e., u, v, w, θx, θy, and θz at x
= 0. Since the positive directions of u, θy, and θz are opposite in the two halves, continuity requires
that they be zero at x = 0 in a symmetrical solution. Thus, u = 0, θy = 0, and θz = 0 are boundary
conditions that can be applied with SPCs if only the right half is represented by a finite element model.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 25-41

© 2024 Siemens
 25. Cyclic Symmetry

The boundary conditions for v, w, and θx are natural, i.e., their reaction forces in the plane of symmetry
are zero by symmetry, and they may be left unconstrained.

In like manner, if the loads applied to the structure are antisymmetrical, the response is antisymmetrical.
By similar reasoning it can be shown that the antisymmetrical boundary conditions for a finite element
model of the right half are v = w = θx = 0.

If the applied loads are neither symmetrical nor antisymmetrical, the response may be obtained by the
following procedure (which is only valid if the response is a linear function of the applied loads):

1. Compute the symmetrical and antisymmetrical load sets Ps and Pa by means of Eq. 25-55.

2. Apply Ps to a finite element model of the right-hand side with symmetric boundary conditions. The
response us is obtained.

3. Apply Pa to a finite element model of the right-hand side with antisymmetric boundary conditions.
The response ua is obtained.

4. Find the physical responses ur and ul by means of Eq. 25-54.

Everything that has been said about one plane of symmetry can be extended to two or three orthogonal
planes of symmetry. The steps that were outlined are performed automatically in Simcenter Nastran's
cyclic symmetry capability, using the identity between reflective symmetry with one or two planes
of symmetry and dihedral symmetry with one or two segments, respectively. The steps may also be
performed using basic Simcenter Nastran procedures but with considerable nonautomated work.

See also

• “Symmetry” in the Simcenter Nastran User's Guide

Axisymmetry

In axisymmetry, as shown in the figure Types of Geometrical Symmetry (e), the object is generated by
rotating the fundamental region continuously about an axis, thereby forming a solid of revolution. There
are two subclasses of axisymmetry:

• In the first subclass, the fundamental region is a plane surface. This subclass is known as axisymmetry
without skew.

• In the second subclass, the fundamental region is a skewed surface. This subclass is known as
axisymmetry with skew.

These two subclasses are distinguishable only if the mechanical properties of the fundamental region
are anisotropic. Examples of structures with skewed axisymmetry may be found in nature (e.g., the
trunks of pine trees) and in spiral-wound reinforced shells.

25-42 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Axisymmetry

Mathematically, a structure is axisymmetric (axisymmetric without skew) if each of its relevant physical
properties Q (such as mass density and elastic modulus) satisfies the equation

Equation 25-56.

where r, θ, and z are the components of position in a cylindrical coordinate system. The cross section for
θ = 0 is the fundamental region whose properties are mapped into the entire structure by Eq. 25-56.

The response of the structure and its applied loads do not necessarily satisfy Eq. 25-56. Let u(r, θ, z) be
some particular response quantity (displacement, internal force, or load). If u is a component of a vector
, let the coordinate system that expresses be the cylindrical coordinate system used in connection with
Eq. 25-56. Then the Fourier series representation of u(r, θ, z) may be written as

Equation 25-57.

or, dropping the explicit dependence on r and z,

Equation 25-58.

The harmonic response coefficients ū0, ūks, ūkc and may be regarded as symmetrical component sets
that replace u(θ) in structural analysis. They are evaluated from the physical component u(θ) by the
following Fourier integrals:

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 25-43

© 2024 Siemens
 25. Cyclic Symmetry

Equation 25-59.

Applied loads P(θ) can also be expanded in a Fourier series. The symmetrical component load sets are,
by analogy with Eq. 25-59,

Equation 25-60.

Fourier series are often used in the analysis of axisymmetric structures. The first step is to find the
harmonic load coefficients from Eq. 25-60. The next step is to find the harmonic response coefficients
as functions of the harmonic load coefficients. It can be shown from a consideration of the properties
of trigonometric functions that, in linear analysis, ū0 depends only on that the harmonic response
coefficients for a particular harmonic ūks and ūkc depend only on the harmonic load coefficients
and for the same harmonic index k. Computational effort can be further reduced by noting that
higher harmonics where is problem dependent do not contribute significantly to the solution and need
not be computed. Once the harmonic response coefficients are found, the physical components can be
computed by means of Eq. 25-57.

25-44 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Overview of Performing a Cyclic Symmetry Analysis

Using Cyclic Symmetry

Overview of Performing a Cyclic Symmetry Analysis

In a cyclic symmetry analysis, your total model consists of N identical segments which are numbered
consecutively from 1 to N. In your input file, you use the CYSYM bulk data entry to specify the type of
cyclic symmetry (rotational, dihedral, or axisymmetric) and the number of segments in the model.

You then generate the model for one segment, using regular elements and standard modeling
techniques. You can use any element in the Simcenter Nastran Element Library, except axisymmetric
elements.

Simcenter Nastran automatically rotates all other segments and their coordinate systems to equally
spaced positions about the polar axis. The segment boundaries must be conformable; i.e., when
the segments are put together, the grid points and the displacement coordinate systems of adjacent
segments must coincide. This is easiest to ensure if you use a cylindrical or spherical coordinate system.

The two boundaries are called Sides 1 and 2.

• In the case of rotational symmetry, Side 2 of segment n is connected to Side 1 of segment n+1, as
shown in the figure Conforming Boundaries for the ROT Option.

• In the case of dihedral symmetry, Side 1 is on the boundary of the plane of overall reflective
symmetry, and Side 2 is on the midplane or reflective plane of the segment, as shown in the figure
Dihedral Symmetry.

When you create your input file, you specify the grid point numbers on Sides 1 and 2 on the CYJOIN Bulk
Data entry. You also list grid points that lie on the axis of symmetry on the CYAX Bulk Data entry.

Specifying Loading and Constraints

Complete generality in the definition of loading conditions is achieved by allowing you to independently
specify loads, enforced motions, and/or temperatures on each of the N segments, even though the
fundamental region is the only segment explicitly represented in the analysis. Output requests are
handled with the same degree of generality in that solutions are provided for each of the N segments.

If your loading has the same symmetry as the structure, only a single analysis is required for a
symmetrical subregion of the structure with symmetrical boundary conditions. If your loading is
unsymmetrical, the following additional tasks must be performed:

1. Separate applied loads into symmetry sets (e.g., symmetrical and antisymmetrical sets for a
structure with one plane of symmetry).

2. Apply separate constraint sets to a subregion of the structure in order to model the boundary
conditions corresponding to each symmetry condition.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 25-45

© 2024 Siemens
 25. Cyclic Symmetry

3. Compute separate solutions for each symmetry condition.

4. Combine the results for each symmetry condition to obtain the complete solution for the complete
structure.

Simcenter Nastran also includes a way of handling the special case of unsymmetrical loads on a
structure with one or two planes of symmetry. With this method:

• Tasks 1 and 2 (listed above) you perform by hand by hand.

• Task 3 is accomplished by creating subcases for each symmetry condition, with different SPC sets and
loading conditions for each subcase.

• Task 4 is accomplished with SYMCOM and SYMSEQ Case Control commands.

In a cyclic symmetry analysis, you apply loads to the model with the following bulk data entries:

• LOADCYH (harmonic load)

• LOADCYN (physical segment load)

• LOADCYT (AXI option) Bulk Data entry

You only need to define a single subcase is needed for each loading condition. That subcase must
contain a LOAD, TEMP(LOAD), or DEFORM command that selects a LOADCYH, LOADCYN, or LOADCYT
Bulk Data entry (see Figure 25-11). The LOADCYH and LOADCYN entries in turn reference FORCE,
MOMENT, PLOAD, SLOAD, SPCD, DEFORM, and TEMP entries. The LOADCYH entry may also reference an
RFORCE or GRAV entry in order to specify centrifugal or gravity loading, respectively. The LOADCYT entry
is used only with the AXI option to specify loads as a function of azimuth angle.

Importantly, you can't use the SUBCOM/SUBSEQ commands to combine loads in a cyclic symmetry
analysis. To combine loads, use the LOAD Bulk Data entry.

25-46 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Specifying Loading and Constraints

Figure 25-11. Cyclic Symmetry Flowchart

There are no restrictions on using single-point constraints, multipoint constraints, or the OMIT feature.
The software automatically applies constraints between segments to the degrees-of-freedom at grid
points specified on CYJOIN Bulk Data entries which aren't otherwise constrained. You can also use the
SPCD Bulk Data entry to vary the magnitude of enforced displacements for each of the segments.

In static analyses, you can use the OMIT feature to remove all degrees-of-freedom at internal grid
points without losing accuracy. Since this reduction is applied to a single segment prior to the
symmetry transformations, it can greatly reduce the amount of subsequent calculation. In the case
of vibration analysis, you can use the OMIT feature to reduce the size of the analysis set with the typical
approximations. You can't use the SUPORTi entries for the free bodies in a cyclic symmetry analysis.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 25-47

© 2024 Siemens
 25. Cyclic Symmetry

Specifying Harmonics

You specify the harmonics to use in the cyclic symmetry analysis on the HARMONICS Case Control
command. The HARMONICS command must be present in Case Control and it must appear above the
subcase level. In buckling analysis, the program limits the static preload to the zero harmonic and the
information on the HARMONICS command is used only for the eigenvalue analysis.

Bulk Data Entries Specific to Cyclic Symmetry

Table 25-7. lists bulk data entries that are unique to cyclic symmetry analyses

Table. Bulk Data Entries Unique to Cyclic Symmetry

Bulk Data Entry Function

CYJOIN Defines the grid points on sides 1 and 2 of the fundamental region.

CYSYM Specifies the number of segments in the entire model and the type of symmetry.
Allowable symmetry types are ROTational, DIHedral, and AXIsymmetric.

CYAX List of grid points that lie on the axis of symmetry.

CYSUP List of supported components at a single grid point that are “constrained” in
order to prevent rigid body motion for free bodies.

LOADCYN Defines applied loads in terms of physical components.

LOADCYH Defines applied loads in terms of their harmonic components.

LOADCYT Specifies loads as a function of azimuth angle for the AXI symmetry type.

These entries are described below.

Using CYSYM

The CYSYM entry lets you define important parameters for cyclic symmetry analysis. For example, you
use it to select the type (STYPE) of cyclic symmetry to perform in the analysis, such as dihedral (DIH).
The format of the CYSYM entry is:

1 2 3 4 5 6 7 8 9 10
CYSYM NSEG STYPE

25-48 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Using CYJOIN

Field Contents

NSEG Number of segments that comprise the total structure.

STYPE Symmetry type (“ROT”, “AXI”, “DIH”).

• For the ROT and AXI options, the polar angle θ subtended by the element model for the fundamental
region, must be such that θ = 360°/NSEG exactly. For the AXI option, the NSEG value should be large
enough (typically 360 segments) to approximate true axisymmetry.

• For the DIH option, θ must equal 180°/NSEG exactly if there are any points on side 1. In this case, the
polar angle θ is smaller because the fundamental region is the right half of segment 1. If there are no
points on side 1, θ must be less than or equal to 180°/NSEG in the DIH option because otherwise the
segments would overlap.

• To properly apply loads and interpret output data, you must be aware of the orientation of the
coordinate system associated with the individual segments. For the ROT and AXI options, the
coordinate system used to define the model of the fundamental region is rotated through the
symmetry operation that “generates” each of the other identical segments. For the DlH option,
the coordinate system used to define the model of the fundamental region (segment 1 R) is
rotated through the necessary symmetry operations that generate the R-half of each of the identical
segments. The coordinate system for the L-half of a segment is the mirror image of the coordinate
system associated with the R-half of the segment in question; thus, the L-half of any segment has a
left-handed coordinate system.

See also

• “CYSYM” in the Simcenter Nastran Quick Reference Guide

Using CYJOIN

The boundary points of a segment in cyclic symmetry problems using the CYJOIN entry are as follows:

1 2 3 4 5 6 7 8 9 10
CYJOIN SIDE C G1 G2 G3 G4 G5 G6
G7 G8 G9 -etc.-

Field Contents

SIDE Side identification.

C Type of coordinate system used on boundaries of dihedral or axisymmetry

problems.

Gi Grid or scalar point identification numbers.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 25-49

© 2024 Siemens
 25. Cyclic Symmetry

One CYJOIN entry plus all necessary continuation entries are used to specify the GRID and/or scalar
points (SPOINT), on side 1 of the fundamental region. A second CYJOIN entry plus all necessary
continuation entries are used to specify the GRID and/or scalar points (SPOINT) on side 2 of the
fundamental region. As will be shown, points that lie on the axis of symmetry appear on a CYAX Bulk
Data entry but not on a CYJOIN entry.

• For the ROT option, the two CYJOlN lists must be of the same length and, since the boundary
conditions for the fundamental region involve these two lists, it is required that the two lists be
ordered. That is, the n-th point on side 1 is related through boundary conditions to the n-th point
of side 2. Side 2 is related to side 1 as indicated in the figure Conforming Boundaries for the ROT
Option.

Note that displacement compatibility must be maintained at the boundaries. This requirement
necessitates that the displacement coordinate systems for the boundary points be either cylindrical or
spherical.

• The AXI option is a subset of the ROT option in which all the grid points in the fundamental region
lie either on side 1 or on side 2. The rules for the CYJOIN entry are identical for both the ROT and AXI
options of cyclic symmetry.

• For the DlH option, side 1 denotes the boundary between segments, and side 2 denotes the side
that is contained in the plane of reflective symmetry of the segment. Since the boundary conditions
do not relate sides 1 and 2, it is not necessary that the two lists be of equal length. It is, however,
necessary under the DlH option to provide an entry in the third field for the parameter C. The
appropriate entries are listed in Table 25-12.

Unlike the ROT option (in which case, one and only one logical CYJOIN entry must exist for each
of side 1 and side 2), any number of CYJOIN entries may be used to define the points on a side
of the fundamental region in the DIH option. Thus, each point along a boundary may appear on a
separate CYJOIN entry if, for example, each point has a different coordinate system. For the DIH option,
the displacement coordinate systems associated with boundary points must be oriented so that the
components of motion are normal and parallel to the boundary planes.

Table. Parameter C for the DIH Option

Type of Point C Description

Grid T1 T1 component of translation motion normal to boundary.

T2, R, or C T2 component of translation motion normal to boundary.

T3 or S T3 component of translation motion normal to boundary.

25-50 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Using CYAX

Type of Point C Description

SCALAR T1 If the user-defined translation is normal to the boundary or if

the user-defined rotation vector is in the plane containing the
boundary.

Blank, T2, or If the user-defined translation is parallel to the boundary or if

T3 the user-defined rotation vector is normal to the boundary.

See also

• “CYJOIN” in the Simcenter Nastran Quick Reference Guide

Using CYAX

The CYAX Bulk Data entry lets you list points on the axis of symmetry. The format of the CYAX entry is:

1 2 3 4 5 6 7 8 9 10
CYAX G1 G2 G3 G4 G5 G6 G7 G8
G9 G10 -etc.-

Field Contents
Gi A list of grid points on the axis of symmetry.

• The coordinate system that locates points on the axis of symmetry and the displacement coordinate
system for these points must be rectangular with the z-component aligned with the axis of symmetry.
In addition, if the DlH option is selected, the y-axis must be perpendicular to side 1 of the
fundamental region.

• Grid points on the axis of symmetry may not appear on a CYJOlN Bulk Data entry.

• Grid points on the axis of symmetry may not be referenced as dependent points by MPCs and/or rigid
elements.

• Grid points on the axis of symmetry may have SPCs applied in accordance with the following rule that
is necessary to satisfy symmetry. If NSEG > 3, SPCs must be applied to both components 1 and 2 or to
neither. The same rule applies to components 4 and 5.

• Grid points listed on CYAX Bulk Data entries must be in the a-set.

See also

• “CYAX” in the Simcenter Nastran Quick Reference Guide

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 25-51

© 2024 Siemens
 25. Cyclic Symmetry

Using CYSUP

The CYSUP bulk data entry lets you define rigid body supports for cyclic symmetry analysis. All supported
degrees-of-freedom for a rigid body must be specified at a single grid point. The format of the CYSUP
entry is:

1 2 3 4 5 6 7 8 9 10
CYSUP GID C

Field Contents

GID Grid point identification number.

C Component numbers.

The total structure can have, at most, six rigid body degrees of freedom. Supports that remove only the
rigid body motions are applied to the k = 0 and k = 1 harmonics of the components of motion listed
on this entry. These components may not be constrained by SPCs, may not be referenced as dependent
degrees of freedom by MPCs and/or rigid elements, and may not be referenced on OMIT or OMIT1 Bulk
Data entries.

Rigid body motion is possible only for harmonics k = 0 and k = 1. Translation along and rotation about
the axis of symmetry are the only possible rigid body motions for k = 0, while translations normal to
the axis of symmetry and rotations about these translational directions are the only possible rigid body
motions for k = 1. The Simcenter Nastran Quick Reference Guide lists special rules relative to the proper
support of rigid body motion for CYSUP.

See also

• “CYSUP” in the Simcenter Nastran Quick Reference Guide

Using LOADCYN

The LOADCYN entry lets you identify the segments on which referenced loads, defined in terms of
physical components, are to be applied. The format of the LOADCYN entry is:

1 2 3 4 5 6 7 8 9 10
LOADCYN SID S SEGID SEGTYPE S1 L1 S2 L2

Field Contents

SID Load set identification number.

S Scale factor.

25-52 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Using LOADCYH

Field Contents

SEGID Segment identification number.

SEGTYPE Segment type.

Si Scale factors.

Li Load set ID numbers.

The loads are referenced by listing the identification numbers of one or two LOAD Bulk Data entries.
Scale factors on the LOADCYN Bulk Data entry are used to form a linear combination of the load sets
referenced on the LOAD Bulk Data entries. Alternatively, the LOADCYN Bulk Data entries may directly
reference load sets comprised of the types of static load definition entries shown in Table 25-19.

Table. Static Load Definitions for Cyclic Symmetry

Point Loads Distributed Loads Temperature Loads Enforced Displacement

FORCE PLOAD1 TEMP SPCD
FORCE1 PLOAD2 TEMPP1
FORCE2 PLOAD4 TEMPRB
MOMENT TEMD
MOMENT1
MOMENT2
SLOAD

The loads listed on these entries are applied at the images in the segment referenced in field 4 of the
LOADCYN entry of the grid points and elements listed on these entries. The coordinate systems for
image points are rotated and/or reflected from the coordinate system in the fundamental region.

Note that GRAV and RFORCE Bulk Data entries are not included in the above list. These static load entries
are handled automatically in a very straightforward fashion through the LOADCYH Bulk Data entry.

See also

• “LOADCYN” in the Simcenter Nastran Quick Reference Guide

Using LOADCYH

The LOADCYH entry provides for the identification of referenced loads that are defined in terms of
harmonic components. The format of the LOADCYH entry is as follows:

1 2 3 4 5 6 7 8 9 10
LOADCYH SID S HID HTYPE S1 L1 S2 L2

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 25-53

© 2024 Siemens
 25. Cyclic Symmetry

Field Contents

SID Load set identification number.

S Scale factor.

HID Harmonic component.

HTYPE Harmonic type.

Si Scale factor on Li.

Li Load set identification number.

This option is particularly useful for the definition of gravity (GRAV) and centrifugal loads (RFORCE)
because you only need to specify GRAV or RFORCE under the HTYPE field on the entry and the correct
harmonic components of the specified load are automatically generated. The component of gravity
parallel to the axis of symmetry produces harmonic loads of the order k = 0, and components of gravity
perpendicular to the axis of symmetry produce loads of the order k = 1. Components of centrifugal force
produce harmonic loads of the order k = 0 if the spin axis coincides with the axis of symmetry, and of
the order k = 0, k = 1, and k = 2 if it does not.

See also

• “LOADCYH” in the Simcenter Nastran Quick Reference Guide

Using LOADCYT

You use the LOADCYT Bulk Data entry with the AXI option to specify loads in terms of physical
components as a tabular function of the azimuth angle (measured in degrees). The format of the
LOADCYT entry is:

1 2 3 4 5 6 7 8 9 10
LOADCYT SID TABLEID1 LOADSET1 METHOD1 TABLEID2 LOADSET2 METHOD2

Field Contents

SID Load set identification number.

TABLEIDi Table ID for table load input for load set Li.

LOADSETi Load set Li.

METHODi Method of interpolation.

25-54 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Case Control Commands Specific to Cyclic Symmetry

Field Contents

0 interpolate the load with the Fourier coefficients specified in the

table up to the specified number of harmonics.

1 interpolate the magnitude of the load at corresponding grid points

in all segments.

This capability is implemented by referencing a set of load definition entries under the heading LOADSET
on the LOADCYT entry.

These entries define a magnitude for the load. The azimuthal variation of the applied load is specified
on the TABLEDi entry referenced under the heading TABLEID. The net applied load is the product of the
magnitude given on the load definition entry times the appropriate value on the TABLEDi entry. The
heading METHODi on the LOADCYT entry is utilized to specify the type of interpolation to be used in
assigning loads to degrees of freedom.

See also

• “LOADCYT” in the Simcenter Nastran Quick Reference Guide

Case Control Commands Specific to Cyclic Symmetry

Table 25-24 lists Case Control commands that are unique to cyclic symmetry analyses.

Table. Case Control Commands Unique to Cyclic Symmetry

Command Function

HARMONICS Specifies harmonics for which the computation is desired.

DSYM Provides for either one or two planes of overall symmetry under the DIH
option.

NOUTPUT Controls requests for the output of physical components.

HOUTPUT Controls requests for the output of harmonic components.

With cyclic symmetry, note that:

• References to SPC and/or MPC sets in the Case Control Section must appear above the subcase level,
i.e., before the first subcase.

• You must have one subcase per loading condition. Loads, enforced deformations, and enforced
displacements are requested in the standard manner, i.e., through LOAD, TEMP(LOAD), DEFORM
and SPC set selections. As described above, loads can be input in terms of physical components

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 25-55

© 2024 Siemens
 25. Cyclic Symmetry

and/or harmonic components. With the AXI option, you can also specify loads in terms of physical
components that are tabulated as a function of the azimuth angle (in degrees).

Using HARMONICS Case Control Command

A HARMONICS command must appear above the subcase level (i.e., before the commands that describe
the first subcase) to specify the harmonics for which solutions are to be computed. The HARMONIC
command cannot reference a set definition that specifies “ALL”. If you need to use all the harmonics,
specify:

HARMONICS = ALL

See also

• “HARMONICS” in the Simcenter Nastran Quick Reference Guide

Requesting Output

The NOUTPUT and HOUTPUT Case Control commands let you request output from a cyclic symmetry
analysis.

• You can use the NOUTPUT command to request physical output within a subcase for all output
quantities, such as forces and displacements, requested in the subcase.

• You can use the HOUTPUT command to request harmonic output within a subcase for all output
quantities requested in the subcase.

These commands let you specify the segments for which you want output and, with HOUTPUT, the cyclic
components for which output is desired.

In the case of thermal or deformation loading, you can only request element force and stress output
that's consistent with the form of the loading. In other words, you can only request physical output if
you defined the loads with a LOADCYN Bulk Data entry and harmonic output if you defined the loads
with a LOADCYH Bulk Data entry.

Using a single subcase for each loading condition introduces the need to refer to “coded” subcases for
SORT2 output and plotting purposes. Since a single subcase now refers to NSEG segments (in the case
of ROT and AXI options) and two times NSEG segments (including Right and Left halves, in the case of
DIH option), you need a unique coded subcase ID to refer to a particular segment of the structure for
plotting purposes in PLOT commands wherever a subcase ID is to be supplied. This coded ID is obtained
as follows:

Coded Subcase ID for a Particular Segment

= SUBCASE ID · 10000 + SEGMENT ID · 10 + zero (for ROT or AXI option)

25-56 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Typical Case Control Section Examples

See also

• “HOUTPUT” in the Simcenter Nastran Quick Reference Guide

• “NOUTPUT” in the Simcenter Nastran Quick Reference Guide

Coded Subcase ID for a Particular Harmonic

Note:

• If a subcase contains TEMP(LOAD) and/or DEFORM commands, irrespective of the presence of a LOAD
command, element force and stress output are only correct in terms of physical components if the
TEMP(LOAD) and the DEFORM loads are defined in terms of physical components. Similarly, if the
TEMP(LOAD) and DEFORM data are provided in terms of harmonic components, element force and
stress output are only correct in terms of harmonic components.

• If, under the DIH option of cyclic symmetry, a DSYM command is explicitly present under a subcase,
the output is, by default, limited to the segments in the first half plane if there is one plane of overall
symmetry, and to the segments in the first quadrant if there are two planes of overall symmetry. This
default applies irrespective of the segments listed on NOUTPUT commands.

You can request undeformed and deformed plots of the fundamental region.

Typical Case Control Section Examples

To illustrate some of the rules described previously, typical Case Control sections for the cylinder (Figure
25-12) and square plate (Figure 25-13) problems are shown in Listing 25-2 and the listing Input File
for Five Degrees Strip of Cylinder, respectively.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 25-57

© 2024 Siemens
 25. Cyclic Symmetry

Figure 25-12. Cylinder Example

25-58 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Typical Case Control Section Examples

Figure 25-13. Square Plate Example

TITLE=CYCLIC SYMMETRY-STATICS
SUBTITLE=FIVE DEGREE SEGMENT OF CYLINDER
LABEL=AXISYMMETRIC LOADING
SPC=1
SET 1=0
SET 2=1
HARMONICS=1
NOUTPUT=2
DISP=ALL
STRESS=ALL
FORCE=ALL
SUBCASE 1
LABEL=BAND LOADING
LOAD=1
SUBCASE 2
LABEL=CENTRIFUGAL LOADING
LOAD=2
BEGIN BULK

Listing 25-1. ROT or AXI Symmetry Type

TITLE=CYCLIC SYMMETRY-STATICS
SUBTITLE=SIMPLY-SUPPORTED SQUARE PLATE
SPC=1
HARMONICS=ALL
DISPLACEMENT=ALL

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 25-59

© 2024 Siemens
 25. Cyclic Symmetry

NOUTPUT=ALL
DSYM=SS
SUBCASE 1
LABEL=LOAD AT CENTER OF PLATE
LOAD=1
SUBCASE 2
LABEL=UNIFORM PRESSURE LOAD
LOAD=2
SUBCASE 3
LABEL=TEMPERATURE LOAD
TEMP(LOAD)=3
SUBCASE 4
LABEL=INPLANE GRAVITY LOAD
LOAD=4
DSYM=SA
BEGIN BULK

Listing 25-2. DIH Symmetry Type

The following comments apply to the Case Control sections above:

• In Listing 25-1, the HARMONICS command specifies that solutions are to be computed only for the
zero harmonic. If the applied loads are axisymmetric, no higher harmonics need to be considered.
For Listing 25-2, HARMONICS = ALL specifies that solutions are to be computed for all meaningful
harmonics.

• A DSYM command appears in all four subcases of Listing 25-2. The first three subcases specify that
the boundary conditions for both side 1 and side 2 are to be symmetric. This specification has two
effects: 1) loads need only to be specified for the first quadrant of the structure (the fundamental
region in this case), and 2) output is only processed for the segments in the first quadrant of the
model irrespective of the analyst's requests for output in other segments. The particular lateral gravity
condition considered in the fourth subcase is symmetric with respect to side 1 of the fundamental
region and antisymmetric with respect to a plane perpendicular to side 1 of the fundamental region.
It must be emphasized that the appropriate boundary conditions for each of the symmetry conditions
involved in this example are internally applied by the program.

• Note that the specifications of loading conditions in the Case Control Section conform to the
procedural rules for static analysis in Simcenter Nastran.

• The NOUTPUT Case Control command controls output requests for selected segments. For Listing
25-1, output requests are limited to the first segment, i.e., the fundamental region. For Listing 25-2,
output requests for all four subcases are limited to segments in the first quadrant because of the
presence of the DSYM commands, even though output for ALL right-handed segments is requested.
The specification of requests for specific output quantities in the Case Control Section conform with
the procedural rules for static analysis in Simcenter Nastran.

25-60 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Performing Axisymmetric Analysis

Performing Axisymmetric Analysis

You can use Simcenter Nastran's cyclic symmetry capabilities to solve axisymmetric problems without
defining any special axisymmetric elements.

Solving Axisymmetric Problems

Use the following procedure to solve static load problems:

1. Select the AXI symmetry option on the CYSYM Bulk Data entry.

2. Select a large number of segments (for example, 180) on the CYSYM Bulk Data entry.

3. Construct a model with two rows of grid points and a single row of elements in the meridional
direction.

4. Define a set of harmonics on a SET Case Control command. The set of harmonics should be
reasonably small for efficiency, but sufficient in size to yield reasonably accurate results.

5. Select the defined set of harmonics on the HARMONICS Case Control command. Convergence
of the harmonic output towards zero for the higher harmonics will verify that the number
of harmonics specified on the HARMONICS Case Control command and used in the analysis is
sufficient.

6. Select a static loading condition with the LOAD, TEMP(LOAD), and DEFORM Case Control
commands which reference LOADCYN, LOADCYH, or LOADCYT Bulk Data entries.

7. Place static loads only on Side 1 or on elements. The boundary conditions will take care of loading
on Side2.

8. Prepare the referenced TABLED1 entries if LOADCYT entries are used to specify loads as a function
of azimuth angle. The number of x-y pairs in a given table should be sufficient to produce
reasonably accurate results when linearly interpolated.

9. Select the physical output for a reasonably small number of segments with the NOUTPUT Case
Control command or ALL of the harmonic output with the HOUTPUT Case Control command.

Use the following procedure to obtain the vibration modes of an axisymmetric structure:

1. Select the AXI symmetry option on the CYSYM Bulk Data entry.

2. Select a large number of segments (for example, 90) on the CYSYM Bulk Data entry.

3. Construct a model with two rows of grid points and a single row of elements in the meridional
direction.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 25-61

© 2024 Siemens
 25. Cyclic Symmetry

4. Define a set of harmonics on a SET Case Control command where the number of harmonics is
reasonably small for efficiency, but sufficient to cover the frequency range of interest.

5. Select the defined set of harmonics on the HARMONICS Case Control command.

6. Select all the harmonic output with the HOUTPUT Case Control command. This will give the
eigenvectors, in a concise form, which may be interpreted by the formula

where φ is the azimuth angle measured from Side 1 of the fundamental region and K is the
harmonic index.

Axisymmetric Example

The following cyclic symmetry example problem uses the same cylinder model as Example 5 – Buckling
of a Cylinder Under Uniform Axial Load.

A five degrees strip, as shown in Figure 25-14, is modeled using the axisymmetric features.

25-62 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Axisymmetric Example

Figure 25-14. A Five Degrees Strip of a Cylinder Using Axisymmetry

A copy of the input file is shown in the figure Reflective Symmetry. Since only a five degrees strip
is modeled, seventy-two segments are required to generate the full cylinder model internally. The
CYSYM entry accomplishes this purpose. Furthermore, the boundary points for segment one are
defined on the CYJOIN entries—one each for side one and side two. Note that with axisymmetry, no
interior points are allowed. The applied load is a uniform compressive load applied at the top of the
cylinder, using the LOADCYT/TABLED1/FORCE combinations. The TABLED1 defines the spatial distribution
around the circumference of the cylinder-which in this case is evenly distributed. Since the structure is
axisymmetric, and the load is constant around the circumference, only the zeroth harmonic is required
for the analysis. The displacements at the top of the cylinder and stresses at the top and bottom of the
cylinders are shown in Figure 25-15. The same output using the full model is shown in Figure 25-16 for
comparison.

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 25-63

© 2024 Siemens
 25. Cyclic Symmetry

$ filename - cyclic1.dat
$
SOL 114
TIME 600
CEND
SUBCASE 1
TITLE= A 5 DEG STRIP
SET 100 = 381,382
SET 200 = 1,343
SET 1000 = 1
SET 1 = 0
HARMONICS = 1
NOUTPUT = 1000
LOAD = 1
SPC = 1
DISPLACEMENT= 100
SPCFORCES=ALL
STRESS= 200
BEGIN BULK
PARAM POST -1
PARAM
,SNORM,
20.
$ Elements and Element Properties for region : pshell.1
CYSYM,72,AXI
CYJOIN,1,T2,1,20,39,58,77,96
,115,134,153,172,191,210,229,248,
,267,286,305,324,343,362,381
$
CYJOIN,2,T2,2,21,40,59,78,97
,116,135,154,173,192,211,230,249,
,268,287,306,325,344,363,382
$
SPC1,1,123,1,2
SPC1,1,12,381,382
$
LOADCYT,1,10,11,1
TABLED1,10
,0.,1.,360.,1.,ENDT
FORCE,11,381,1,1388.889,,,-1.
$
PSHELL 1 1 .03 1 1
CQUAD4 1 1 1 2 21 20 0. 0.
CQUAD4 19 1 20 21 40 39 0. 0.
CQUAD4 37 1 39 40 59 58 0. 0.
CQUAD4 55 1 58 59 78 77 0. 0.
CQUAD4 73 1 77 78 97 96 0. 0.

25-64 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Axisymmetric Example

CQUAD4 91 1 96 97 116 115 0. 0.

CQUAD4 109 1 115 116 135 134 0. 0.
CQUAD4 127 1 134 135 154 153 0. 0.
CQUAD4 145 1 153 154 173 172 0. 0.
CQUAD4 163 1 172 173 192 191 0. 0.
CQUAD4 181 1 191 192 211 210 0. 0.
CQUAD4 199 1 210 211 230 229 0. 0.
CQUAD4 217 1 229 230 249 248 0. 0.
CQUAD4 235 1 248 249 268 267 0. 0.
CQUAD4 253 1 267 268 287 286 0. 0.
CQUAD4 271 1 286 287 306 305 0. 0.
CQUAD4 289 1 305 306 325 324 0. 0.
CQUAD4 307 1 324 325 344 343 0. 0.
CQUAD4 325 1 343 344 363 362 0. 0.
CQUAD4 343 1 362 363 382 381 0. 0.
$

Listing 25-3. Input File for Five Degrees Strip of Cylinder (Continued)

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 25-65

© 2024 Siemens
 25. Cyclic Symmetry

MAT1 1 1.+7 3.84+6 .3 0. 0. 0. 0.

$ Nodes of Group : group5
GRID 1 1 10. 0. 0.1
GRID 2 1 10. 5. 0.1
GRID 20 1 10. 0. 1.1
GRID 21 1 10. 5. 1.1
GRID 39 1 10. 0. 2.1
GRID 40 1 10. 5. 2.1
GRID 58 1 10. 0. 3.1
GRID 59 1 10. 5. 3.1
GRID 77 1 10. 0. 4.1
GRID 78 1 10. 5. 4.1
GRID 96 1 10. 0. 5.1
GRID 97 1 10. 5. 5.1
GRID 115 1 10. 0. 6.1
GRID 116 1 10. 5. 6.1
GRID 134 1 10. 0. 7.1
GRID 135 1 10. 5. 7.1
GRID 153 1 10. 0. 8.1
GRID 154 1 10. 5. 8.1
GRID 172 1 10. 0. 9.1
GRID 173 1 10. 5. 9.1
GRID 191 1 10. 0. 10.1
GRID 192 1 10. 5. 10.1
GRID 210 1 10. 0. 11.1
GRID 211 1 10. 5. 11.1
GRID 229 1 10. 0. 12.1
GRID 230 1 10. 5. 12.1
GRID 248 1 10. 0. 13.1
GRID 249 1 10. 5. 13.1
GRID 267 1 10. 0. 14.1
GRID 268 1 10. 5. 14.1
GRID 286 1 10. 0. 15.1
GRID 287 1 10. 5. 15.1
GRID 305 1 10. 0. 16.1
GRID 306 1 10. 5. 16.1
GRID 324 1 10. 0. 17.1
GRID 325 1 10. 5. 17.1
GRID 343 1 10. 0. 18.1
GRID 344 1 10. 5. 18.1
GRID 362 1 10. 0. 19.1
GRID 363 1 10. 5. 19.1
GRID 381 1 10. 0. 20.1
GRID 382 1 10. 5. 20.1
$
CORD2C 1 0. 0. 0. 0. 0. 1.
+ A

25-66 Simcenter Nastran User's Guide, Simcenter Femap 2406 Series

© 2024 Siemens
 Axisymmetric Example

+ A 1. 0. 0.
ENDDATA

Listing 25-3. Input File for Five Degrees Strip of Cylinder

Figure 25-15. Abridged Output Using Axisymmetry

Figure 25-16. Abridged Output Using Full Model

Simcenter Nastran User's Guide, Simcenter Femap 2406 Series 25-67

© 2024 Siemens