See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/260188897

A methodology for the development of machining fixtures for components with

complicated geometry

Article in International Journal of Computer Integrated Manufacturing · September 2008

DOI: 10.1080/09511920701829907 · Source: DBLP

CITATIONS READS
12 312

4 authors, including:

Allan Hodgson Xun Chen

Loughborough University Liverpool John Moores University
74 PUBLICATIONS 471 CITATIONS 120 PUBLICATIONS 1,701 CITATIONS

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Autonomous vehicles and assisted driving View project

Micron Diamond Processing of Advanced Ceramics View project

All content following this page was uploaded by Allan Hodgson on 05 June 2014.

The user has requested enhancement of the downloaded file.

 International
Author manuscript, Journal
published of Computer
in "International Integrated
Journal Manufacturing
of Computer Integrated Manufacturing 21, 07 (2008) 848-856"
DOI : 10.1080/09511920701829907

Fo
rP
peer-00513398, version 1 - 1 Sep 2010

Responsive Manufacturing-A methodology for the

ee

development of machining fixtures

rR

Journal: International Journal of Computer Integrated Manufacturing

Manuscript ID: TCIM-2007-IJCIM-0035.R1

Manuscript Type: Special Issue Paper

ev

Date Submitted by the

02-Nov-2007
Author:
ie

Complete List of Authors: Wang, Yan; The University of Nottingham, School of Mech,
Materials & Man Eng'g
Hodgson, Allan; The University of Nottingham, School of Mech,
w

Materials & Man Eng'g

Chen, Xun; The University of Nottingham, School of Mech, Materials
& Man Eng'g
Gindy, N N Z; The University of Nottingham, School of Mech,
On

Materials & Man Eng'g

Keywords: FIXTURE DESIGN, ADVANCED MANUFACTURING TECHNOLOGY

Keywords (user): machining fixture, virtual simulations, CAD/CAM, FEA, kinematic

ly

URL: http://mc.manuscriptcentral.com/tandf/tcim Email:ijcim@bath.ac.uk

Page 1 of 25 International Journal of Computer Integrated Manufacturing

A methodology for the development of machining

fixtures for components with complicated geometry

*Y. Wang1, A. Hodgson2, X. Chen3, N. Gindy4,

124
School of Mechanical, Materials and Manufacturing Engineering, University

of Nottingham, UK
Fo

3
School of Computing and Engineering, University of Huddersfield, UK
rP

Abstract
peer-00513398, version 1 - 1 Sep 2010

ee

At present, fixture design in industry is largely an experienced-based, ad-hoc

process. In the case of the design of fixtures for complex components, this approach
rR

often results in a need for rework and consequent delays to production.

ev

Machining fixture design has been the subject of considerable research efforts;

however most research activities have addressed one or a small number of

iew

interactions between fixture and other manufacturing system elements.

In this paper, a novel fixture design methodology based on concurrent

On

engineering is described. This methodology models physical space, loads, stress,

deformation, thermal effects, vibration, etc., determines the loads and deflections
ly

arising from the locating, clamping and machining procedures, and estimates the

resultant effects on component quality.

To assist in the development of the methodology, the high technology industrial

partner originally provided the researchers with a range of complex machined

URL: http://mc.manuscriptcentral.com/tandf/tcim Email:ijcim@bath.ac.uk

1
 International Journal of Computer Integrated Manufacturing Page 2 of 25

components of various sizes, incorporating many different features. The

methodology has been tested against a range of successful and unsuccessful fixture

designs supplied by the industrial partner.

Key words: machining fixture, virtual simulations, CAD/CAM, FEA, kinematic

1 Introduction

A machining fixture is a device that holds components in a unique position rigidly

Fo

for machining operations, so that a batch of components, satisfying the quality

rP

requirements, can be produced. The machining fixture is a key element of the

component manufacturing system, but fixture design for components with

peer-00513398, version 1 - 1 Sep 2010

ee

complicated geometry is a lengthy process that includes conceptual design, detailed

design and validation. Often, the performance of a fixture is very difficult to predict,
rR

as it is influenced by a large number of direct and indirect factors including workpiece

shape, size, tolerance requirements, process plan, machining parameters (e.g.

ev

machining speeds and feeds), machining strategy, cutter paths and inspection
iew

strategy. The fixture design decisions rely on the designer’s assessment of the

effects of each factor on the specified product quality.

On

Typically, fixture development is still a method of trial-and-error. The more

experience the fixture designer has, the fewer fixture iterations will be required.
ly

However, even when employing very experienced specialised fixture designers, non-

optimal fixture performance occurs frequently. Without appropriate modelling tools

and a thorough understanding of the interactions of the fixture with the

manufacturing system of components, it is very unlikely that a fixture for a complex

URL: http://mc.manuscriptcentral.com/tandf/tcim Email:ijcim@bath.ac.uk

2
Page 3 of 25 International Journal of Computer Integrated Manufacturing

component can be correctly designed at the first attempt; this can only be achieved

by considering all relevant factors concurrently at an early stage of design.

The basic requirements of machining fixture are:

• Good loading repeatability, so that mass production of components can be

achieved
Fo
• Immobility: The components are held firmly such that they will not move under

machining forces
rP

• Minimal number of set-ups: The fixture makes full use of machine capability
peer-00513398, version 1 - 1 Sep 2010

and is designed in such way that the smallest number of set-ups is required whilst
ee

the fixture performance remains satisfactory. The reduction of the number of set-
rR

ups can significantly reduce machine time and error stack-up

• Accessibility: No collision is allowed between fixture, component, machine tool

ev

and machine
iew

• Good dynamic performance: No chatter or excessive vibration is allowed

during machining of component held by a fixture

On

• No excessive deformation of component occurs during fixturing and machining

ly

2 Background

A considerable amount of fixture design research employing advanced modelling

tools is currently taking place, typically taking account of one or two of the earlier-

stated requirements.

URL: http://mc.manuscriptcentral.com/tandf/tcim Email:ijcim@bath.ac.uk

3
 International Journal of Computer Integrated Manufacturing Page 4 of 25

With regard to loading repeatability and immobility, (Asada et al 1985) derived a

mathematical model of the necessary and sufficient conditions for deterministic

component locating. This model makes it possible to achieve an automated,

computer-generated optimised fixture locator layout. More recently, a significant

amount of research using kinematic analysis to achieve optimal fixture layout has

taken place. Conventionally, the contact between fixture and workpiece is often

modelled as point contact. a kinematic model of fixtures was developed with

Fo

consideration of underlying surface properties of locator-workpiece pairs, although

the fixture and workpiece were still treated as rigid bodies (Wang 2001). Further to
rP

this, (Wang 2002) provided a kinematic model to predict the contact force of the
peer-00513398, version 1 - 1 Sep 2010

ee

workpiece-fixture pair using a constrained quadratic optimisation by applying the

minimum norm principle. The model revealed that the passive contact force is
rR

history-dependent during a sequence of clamping and/or external force loading. A

method of fixture optimisation using multi-objective functions of accurate localisation

ev

and minimal but balanced locator contact force was proposed by (Pelinescu et al,

2002). The approach required a fixture layout with the minimum number of elements,
iew

i.e. six locators and a clamp for a three dimensional space, and friction was not taken

into consideration. (Ding et al 2001) presented a method for the automatic selection
On

of fixturing surfaces and fixturing points for polyhedral workpieces by employing the

constraints of form-closure and minimising the workpiece positioning error.

ly

Deformation and vibration analyses of fixture-workpiece pairs represent another

hotspot of the fixture research of last decade. Critical to the accuracy of these

analyses is the model of the contact relationships between fixture elements and the

workpiece. There are two commonly used methods: contact elasticity modelling (Li et

URL: http://mc.manuscriptcentral.com/tandf/tcim Email:ijcim@bath.ac.uk

4
Page 5 of 25 International Journal of Computer Integrated Manufacturing

al 1999 and Li et al 2001), where the workpiece is treated as an elastic body in the

contact zone and rigid elsewhere, and finite element methods (Liao et al 2001, Yeh

et al, 1999 and Zheng et al 2005). These modelling approaches can be used for the

optimisation of locator and clamping positions and clamping force magnitude

(DeMeter et al, 2001).

Accessibility analyses are typically based on the modelling of fixture space and its
Fo
interactions with tooling space to ensure that the tool has reasonable access to the

component during machining. (Kumar et al 2000 and Kow et al 2000) presented a

rP

computer-aided modular fixture design system in which the detection of machining

interference was realized using the cutter swept volume approach. Simulation was
peer-00513398, version 1 - 1 Sep 2010

ee

conducted to detect the collision statically.

rR

Although significant advances have been made, many of the problems have been

addressed by isolating individual fixturing system issues. A fixture design that

ev

optimally satisfies a subset of fixture design criteria is unlikely to provide an optimal

solution in terms of overall performance. It may result in an infeasible machining or

iew

inspection strategy. For example, the optimal locating position derived in isolation

might be on the datum surface for inspection or on the approach path of the machine
On

tool. Therefore, it is important to consider the relationships between fixtures and

other elements of the manufacturing system (in particular machining and inspection
ly

processes) in the early stages of fixture design, and to conduct the development of

the process plan concurrently, i.e. to take a concurrent engineering approach. The

final manufacturing system should be a trade-off between a range of parameters

including fixturability, machinability, testability and functionality, etc.

URL: http://mc.manuscriptcentral.com/tandf/tcim Email:ijcim@bath.ac.uk

5
 International Journal of Computer Integrated Manufacturing Page 6 of 25

As implied above, most research work to-date only considers a subset of the

fixture design problem, or treats fixture development as a serial process, i.e.

designing the fixture prior to devising the machining strategy. Research issues

concerning the interactions between manufacturing system and fixture are seldom

addressed for the design of an optimal manufacturing system.

3 A concurrent fixture development methodology

Fo

A concurrent fixture development methodology, developed at the University of

Nottingham, is described in this paper. The methodology enables the concurrent

rP

design of a fixture taking into account all key interaction factors (see later) and
peer-00513398, version 1 - 1 Sep 2010

machining strategies. The methodology uses space occupancy, kinematics and

ee

engineering error analysis. The industrial collaborator that supports this research
rR

produces many complex machined components with tight tolerances. The

collaborator recognises the need for a comprehensive concurrent approach to fixture

ev

design in order to ensure that delivery schedules and quality targets are met.
iew

The development of a single hit fixture (Figure 1 (c)), for holding cast turbine

blade (Figure 1(a)) to be machined to the final turbine blade (Figure 1(b)), is used as

case study in this paper.

On
ly

Figure. 1 Fixture development for turbine blades

URL: http://mc.manuscriptcentral.com/tandf/tcim Email:ijcim@bath.ac.uk

6
Page 7 of 25 International Journal of Computer Integrated Manufacturing

3.1 Key interaction factors

Good fixture design should of course ensure component quality and manufac-

turability. However, it should also contribute to the achievement of maximum

productivity, long tool life, low cost machining strategies, simple inspection strategies

and short lead times. As shown in Figure 2, a good fixture design requires an

understanding of the interactions between fixture, component, machine, tool,

Fo
process planning and inspection. The factors that should be considered are

discussed below.
rP
peer-00513398, version 1 - 1 Sep 2010

ee

Figure. 2 Interaction of key factors for the fixture design

rR

Component: Components are the key input for fixture design. Fixture design
ev

should take account of quantities of components, similarities within component

families, features to be machined, the geometry, size, material, etc. These factors
iew

will influence the type of fixture, the number of set-ups and the layout of the fixture.

On the other side, however, after satisfying the function and quality requirements,
On

component designs should take account of fixturability, machinability and testability

issues.
ly

Process planning: Process planning is the selection of processes, e.g. milling,

turning, drilling, grinding, and the plan for the process sequence. The selection of

processes is closely related to component material and the costs and efficiencies of

potential machining processes. The chosen process will directly lead to the selection

of a machine that is capable of that process. The selection of machine and process

URL: http://mc.manuscriptcentral.com/tandf/tcim Email:ijcim@bath.ac.uk

7
 International Journal of Computer Integrated Manufacturing Page 8 of 25

dictates the number of set-ups and the features that should be machined in a

specific set-up.

Machining strategy: It is important to consider the effects of machining strategy

on components and fixtures, in particular the potential component and fixture

deformations and vibration resulting from the machining forces and the

corresponding fixture clamping forces. In most current machining strategy work, the
Fo
relationship between machining strategy and fixture design is ignored. As a result,

the machining strategy is designed in favour of tool life and material removal rate,
rP

therefore, large machining forces leading to excessive component and/or fixture

deformation are often observed and, because large clamping forces are required to
peer-00513398, version 1 - 1 Sep 2010

ee

balance the machining forces, these clamping forces can result in further

deformations. The resultant combination of machining and clamping deformations

rR

may cause unacceptable profile errors. At the very least, tighter tolerances may

have to be applied to other error sources such as process error, component

ev

variation, etc., in order to compensate for the profile errors arising from machining
iew

parameter selection.

Machine: The selection of machine depends on the processes, machine cost (or
On

machine operating cost), component quality requirements, and availabilities of

machines in the workshop, etc. The machine envelope and the number of degrees of
ly

freedom of machine movement (e.g. five-axis machining centre) are important

influential factors on fixture set-up and rough space design of the fixture.

Tool selection: The factors concerning tool selection that are relevant to fixture

design include tool geometry, tool size, lead in and lead out distances, and approach

URL: http://mc.manuscriptcentral.com/tandf/tcim Email:ijcim@bath.ac.uk

8
Page 9 of 25 International Journal of Computer Integrated Manufacturing

direction. In current practice, the tool designer tends to select the tool type and tool

approach direction (to the component) to optimise tool life, without consideration of

fixture requirements. This approach often leads to unnecessarily constrained space

for fixture development; sometimes, this approach may result in collisions between

fixture, component and tool.

Inspection: It is important to take inspection requirements into account during

Fo
fixture design, in particular in-process inspection requirements. Sensors, machine

probes and CMMs (coordinate measurement machines) are increasingly used to

rP

enable inspection of components when they are loaded within fixtures. However, it is

seldom recognised that inspection space, tool space and fixture space share the
peer-00513398, version 1 - 1 Sep 2010

ee

same working space and should be conducted interactively with tool design and

fixture design. As a result, the locator or clamp may be put on or near the measure-
rR

ment datum surfaces of components. In this case, the fixture prevents in-process

inspection of certain areas, or limits the approach direction of the inspection device.
ev

The result may be longer inspection times, lower inspection accuracy or infeasible
iew

inspection strategy.

A range of virtual simulations to model the tooling, inspection and fixturing

On

processes and their interactions at an early stage of fixture design is paramount in

order to determine their impacts on component quality. A fixture development

ly

procedure with assistance of virtual simulations is shown in Figure 3. These

analyses of interactions are focused on three aspects: space occupancy, kinematics,

and engineering errors. The three forms of analysis are described in more detail in

the following sections. The virtual simulation tools include CAD/CAM (Computer

URL: http://mc.manuscriptcentral.com/tandf/tcim Email:ijcim@bath.ac.uk

9
 International Journal of Computer Integrated Manufacturing Page 10 of 25

Aided Design/Computer Aided Manufacturing) modelling, FEA (Finite Element

Analysis), and kinematic numerical analysis.

Figure. 3 Procedure of fixture development in relation to virtual simulations

Space occupancy analysis is conducted first, as it provides vital information for

Fo

set-up planning, enabling a decision as to the minimum number of fixtures that i

rP

required, the group of features to be machined using each fixture and the

approximate sizes of fixtures and tools. Following space occupancy analysis,

peer-00513398, version 1 - 1 Sep 2010

ee

kinematic analysis can be carried out based on the knowledge of the features to be

machined by each fixture, etc. Finally, deformation, stress, thermal and vibration
rR

analyses may be conducted depending on the circumstances.

ev

Armed with the results of the above analyses, the designer can propose a fixture

design that provides a satisfactory trade-off between the multiple requirements

iew

placed on the fixture. After fixture fabrication, a small number of components are

then machined to evaluate fixture performance and verify the fixture virtual

simulations.
On

For the case study (Single Hit fixture for turbine blades), Pro/Engineer is used for
ly

the space occupancy analysis, Matlab for the kinematics analysis, MSC.Patran for

the FEA Pre/Post Process and ABAQUS for the FEA Solver. Viper grinding is

chosen as the process for the turbine blades. Makino A55, a five axis machining

centre that is capable of grinding process, is thus selected tentatively, subjected to

URL: http://mc.manuscriptcentral.com/tandf/tcim Email:ijcim@bath.ac.uk

10
Page 11 of 25 International Journal of Computer Integrated Manufacturing

the collision check. CMMs (Coordinate Measurement Machines) are used for the

component quality measurement.

3.2 Space occupancy analysis

The primary tools used for space occupancy are CAD/CAM systems; in particular,

these are used to conduct accessibility analysis. The goal is to ensure that there are

no collisions between fixture, components, machine and tool. The space occupancy
Fo

analysis includes three steps:

rP

Step 1: Fixture space that is collision free with tool and inspection space
peer-00513398, version 1 - 1 Sep 2010

The fixture design for turbine blade is started from one set-up. Tool space,
ee

generated by sweeping grinding wheel along the tool trajectory, is modelled first for
rR

all the machined features. The grinding wheel is designed based on the best wheel

life. Similar to tool space, the inspection space is the space that sweeps the CMM
ev

probe along the probe path. The machined features on the turbine blade F1~F15 are

shown in Figure 4. The tool space and inspection space for the machined features is
iew

shown in Figure 5 (a). The trimmed remaining space, once the tool and inspection

space is subtracted from the total working space, is the space for fixture as shown in
On

Figure 5(b). An FEA of the rough fixture body designed within fixture space is

necessary to evaluate the rigidity of fixture space. If inadequate, the tool space or
ly

inspection space need to be compromised.

Figure. 4 Machined features of the HP T800 turbine blade

URL: http://mc.manuscriptcentral.com/tandf/tcim Email:ijcim@bath.ac.uk

11
 International Journal of Computer Integrated Manufacturing Page 12 of 25

Figure. 5: Fixture collision free space design for the T800 turbine blade (Geldart et
cal 2002)

Step 2: Fixture space modification constrained by the machine envelope.

Simulations of the movement of machine tool and fixture-workpiece pair are

conducted to check if any over travel problem is encountered. If yes, either tool

strategy has to be compromised or the fixture space has to be reduced. For the case
Fo
study, since five axis machining centre Makino A55 is used for the machine, the

fixture-workpiece pair are firstly rotated around A and B axis and then moved in the z
rP

axis and grinding wheel moves in X, Y planes. Once the tool path is generated

correctly for the features to be machine, over travel of machine can be evaluated.
peer-00513398, version 1 - 1 Sep 2010

ee

Step 3: Fixture detailed design and collision check with machine.

rR

This step should be undertaken in the detailed design stage of fixture after the

kinematical analysis and the engineering error analysis, which suggests the optimal
ev

fixture layout and geometry of locators, clamp and supports. With detailed modelling
iew

of machine and fixture, the collision check in step 3 will verify if fixture collides with

machine components. For the case study, the simulation includes the potential

collision check with coolant nozzle, wheel dresser etc on the Makino A55.
On

3.3 Kinematic analysis

ly

The impact of component geometry on the fixture layout is analysed by the

kinematic analysis, the assumption of which is that both fixture and component are

consumed to be rigid bodies.

URL: http://mc.manuscriptcentral.com/tandf/tcim Email:ijcim@bath.ac.uk

12
Page 13 of 25 International Journal of Computer Integrated Manufacturing

During the kinematic fixture layout analyses, a feasible fixture needs to satisfy

several basic requirements including repeatability, immobility and stability. Mathem-

atical simulation models (using Matlab software) are used to evaluate the constraints

placed on the workpiece by the fixture (including or excluding friction). Assuming

frictionless contact between fixture elements (locators and clamps) and components,

the component is in equilibrium under the action of clamping forces as expressed in

equation (1):
Fo

G ⋅ FL + C ⋅ FC = 0 (1)
rP

Where G is the so called fixturing matrix, G = [N 1L ,L , N iL ,L , N nL ] , N iL =[ n iL ,

peer-00513398, version 1 - 1 Sep 2010

ee

rLi × n iL ]’, n is the number of locators, (n is three for two dimensional space and six

for three dimensional space). n iL and rLi are the unit normal and positional vectors
rR

of the ith locator respectively, FL = [f L1 ,L, f Li ,L, f Ln ] ’, the f Li is the magnitudes of the
ev

reaction force on the ith locator. C is the clamping matrix and C= [ N 1C , L , N Cj , L , N mC ],

m is the number of clamps. N Cj = [ n Cj , rCj × n Cj ]’, and Where n Cj and rCj are the unit
iew

normal and positional vectors of the jth clamp pointing into the workpiece

respectively, FC = [f C1 ,L, f Cj ,L, f Cm ] ’, where f Cj is the magnitudes of the jth clamp on

On

component. Component repeatability with regard to the fixture means that the

workpiece is located in a unique position, namely ||G|| ≠ 0 and the maximum locating
ly

accuracy means that maximum ||G||:

||G|| ≠ 0 and Max (||G||2) (2)

URL: http://mc.manuscriptcentral.com/tandf/tcim Email:ijcim@bath.ac.uk

13
 International Journal of Computer Integrated Manufacturing Page 14 of 25

Immobility requires that the six degrees of freedom are fully constrained by fixture

elements (locators and clamps). For frictionless contact between fixture and

workpiece, the immobility condition is that the components of FC and FL in equation

(2) are positive:

FC>0 and FL>0 (3)

Stability means that the clamps push components to contact locators during the
Fo

entire machining process, namely the reaction force on the locator is positive. The

feasible clamp position is selected in terms of stability, the optimal clamping position
rP

is that where the clamps require the minimum clamping force to maintain stability.
peer-00513398, version 1 - 1 Sep 2010

The optimisation function is written as

ee

m
(f Cj )
rR

Minimise ∑
j =1

Subject to G·FL+C·FC+M·FM=0 (4)

ev

Bounds 0 < f Li (f Li ∈ FL ), 0< f Cj (f Cj ∈ FC ) , 0< f mn (f mn ∈ FM ) , 0 < f mt (f mt ∈ FM )

iew

Where M is the machining matrix of one point on the boundary of surface to be

machined, and M= [M n , M t ] and M n = [n nm , rm × n nm ] ’, M t = [n m

t t
, rm × n m ] ’.
On

n nm , n m
t
and rm are the unit normal vector, unit tangential vector and positional vector of
ly

the machining force respectively. FM= [f mn , f mt ] ’, f mn and f mt are the magnitudes of the

normal machining force and tangential force respectively.

Details of the fixture optimisation based on the requirements of repeatability,

immobility and stability are explained in (Wang et al 2006). Currently, the fixture

URL: http://mc.manuscriptcentral.com/tandf/tcim Email:ijcim@bath.ac.uk

14
Page 15 of 25 International Journal of Computer Integrated Manufacturing

layout optimisation for turbine blades is two-dimensional. Three optimal fixture

layouts are generated automatically from the kinematical modelling as shown in

Figure 6.

Figure. 6 Three optimal fixture layouts generated from the kinematic analyses (Wang
et al, 2007)
Fo

3.4 Engineering error analysis

rP

In order to help understanding the impacts of the machining, clamping and

peer-00513398, version 1 - 1 Sep 2010

ee

locating loads on component quality, finite element analyses (FEA) of static

deformation, dynamic deformation and natural frequency, friction etc. are employed.
rR

For the case study, only static deformation analysis is conducted at this stage
ev

The turbine blade (Figure 7(a)) is assumed to be elastic, and a rigid-deformable

contact model (Figure 7 (b)) has been built to represent the relationship between
iew

fixture and turbine blade. The deformation of the fixture is taken into consideration

by assigning a spring element to each of the locators and clamps, the stiffness of

which is calculated separately (e.g. Figure 7(c)). Since the clamping forces and
On

machining forces are often applied to the workpiece at different positions and

different times, multiple FEA steps are required. In order to simulate the clamping
ly

deformation and machining procedure on a machined feature of the component, four

FEA steps may be required: Step1: apply a small clamping force on the workpiece

to ensure that the workpiece gently contacts the fixture locators in order to ascertain

the relative position between fixture and components; Step 2: Apply the full clamping

URL: http://mc.manuscriptcentral.com/tandf/tcim Email:ijcim@bath.ac.uk

15
 International Journal of Computer Integrated Manufacturing Page 16 of 25

forces, the difference in the position of the component between Step 1 and Step 2

represents the clamping deformation; Step 3: Impose machining forces on a feature

of the component to be machined, the position deviation between Step 3 and Step 2

is then the machining deformation; Step 4: Release the machining force, and

maintain the clamping forces; Step 3 and Step 4 can be repeated if another feature

is to be machined using the same fixture.

Fo
Figure. 7 Static FEA deformation analysis of fixture-turbine blade pair (Wang et al
2007)
rP

FEA simulation of deformation can be used for the error decomposition and
peer-00513398, version 1 - 1 Sep 2010

ee

tolerance assignment. Surface error is defined as the maximum deviation between

the nominal machined surface and the actual machined surface. If Em is the
rR

resultant surface error arising from the deformation of both the workpiece and the

locators in the m direction; E mwp and E mloc i are the surface errors resulting from the
ev

deformation of the workpiece and the ith locators in the m direction respectively, and
iew

n is the number of locators, Em is written as shown in Equation (5) (Wang et al 2007)

n
Em= E mwp + ∑ (E mloc i ) (5)
On

i =1

In addition to calculating the clamping deformation and machining deformation,

ly

FEA can also be used for the machining strategy selection. For example, the

machining force magnitude and sequence is different for up-grinding and down-

grinding of the machining surface, the FEA output will suggest which one is better in

terms of deformation. The impact of friction (between the fixture element and

component) on component quality can be estimated by assigning a range of friction

URL: http://mc.manuscriptcentral.com/tandf/tcim Email:ijcim@bath.ac.uk

16
Page 17 of 25 International Journal of Computer Integrated Manufacturing

coefficients, which can be varied from zero (frictionless) to infinite. Details of the FEA

for fixture and turbine blade are explained in (Wang et al 2007)

3.5 Industrial evaluation of the methodology

The high technology collaborator company has provided the researchers with a

range of complex machined components of various sizes, incorporating many

different features, in order to evaluate the effectiveness of the methodology. This

Fo

has enabled all the analytical components of the methodology, i.e. space occupancy,

kinematics and engineering errors, to be thoroughly tested.

rP

The collaborator company has also provided the researchers with details of a
peer-00513398, version 1 - 1 Sep 2010

ee

number of its fixtures which had been developed via the conventional, experienced-

based approach; several of these had required rework, resulting in delays to

rR

production. The fixture development methodology was used to analyse these

fixtures. It highlighted a range of problems including (1) (single fit fixture) collisions
ev

requiring fixture modification (2) components moving within the fixture, and (3) large
iew

deformations. In the case of the reworked fixtures, the methodology identified most

of the problems that had occurred and, in the case of other fixtures, it also identified

areas of weakness which could result in reduced machining accuracy.

On

3.6 Current situation

ly

The fixture development methodology described in this paper is near completion.

It currently uses a range of commercially available software tools and is therefore not

at present packaged as an integrated toolset. However, it can be used as a set of

individual functions.

URL: http://mc.manuscriptcentral.com/tandf/tcim Email:ijcim@bath.ac.uk

17
 International Journal of Computer Integrated Manufacturing Page 18 of 25

Although the lack of integration is to some extent a disadvantage, the

methodology’s utilisation of commercial, industrially-proven tools makes it attractive

to our key collaborator, who has several of these tools. As a result, the company’s

fixture design expert is trialling the methodology, with the help of the researchers, on

the design of a fixture for complex machined components. The research team

awaits an industrial-based evaluation of the methodology based on this, following

which a decision will be made with regard to further exploitation.

Fo

4 Conclusions
rP

The machining fixture is a key contributor to the manufacturability of a

peer-00513398, version 1 - 1 Sep 2010

component, and should be designed to optimise the performance of the overall

ee

machining process (including in-process inspection). However, at the present time,

rR

industrial fixture development is still largely reliant on the experience of the designer

and a process of trial and error; this leads to unnecessary costs, delays and sub-
ev

optimal performance.
iew

The fixture development methodology described in this paper is novel in that it

enables the user to take account of machining strategy and all key interactions

between fixture, component and other system elements at an early stage. By

On

modelling and analysing a range of parameters including physical space, loads,

stress, deformation, thermal effects and vibration, the methodology enables the user
ly

to avoid many of potential problems of conventional fixture development and to

produce a near-optimal design prior to physical manufacture of the fixture.

Acknowledgements The authors would like to express their thanks to Rolls-Royce

plc for the company’s kind support.

URL: http://mc.manuscriptcentral.com/tandf/tcim Email:ijcim@bath.ac.uk

18
Page 19 of 25 International Journal of Computer Integrated Manufacturing

References

Asada, H., By, A. B.,1985, Kinematic analysis of workpart fixturing for flexible

assembly with automatically reconfigurable fixture, IEEE Journal of Robotics and

Automation, vol. RA-1, n 2, pp. 86-94.

Fo
De Meter, E. C., Xie, W., 2001, A model to predict minimum required clamp pre-

loads in light of fixture-workpiece compliance, International Journal of Machine Tools

rP

& Manufacture, vol. 41, n7, pp 1031-1054.

peer-00513398, version 1 - 1 Sep 2010

Ding, D., Liu, Y., Wang, M. Y, 2001, Automatic selection of fixturing surfaces and
ee

fixturing points for polyhedral workpiece, IEEE Transaction on Robotics and

rR

Automation. Vol. 17, no. 6, pp. 833-841.

Geldart, M., Chen, X. 2002, NC programming and grinding wheel accessibility for
ev

T800 HP turbine blade Single Hit Fixture, Rolls Royce UTC report, The University of
iew

Nottingham.

Kow, T. S., Kumar, A. S., Fuh, J. Y. H, 2000, Integrated approach to collision-free

On

computer aided modular fixture design, International Journal of Advanced

Manufacturing Technology, v16, n4, pp. 233~242.

ly

Kumar, A. S., Fuh, J. Y. H, Kow, T. S, 2000, An automated design and assembly of

interference-free modular fixture set-up, Computer-Aided Design, vol. 32, pp. 583-

596

URL: http://mc.manuscriptcentral.com/tandf/tcim Email:ijcim@bath.ac.uk

19
 International Journal of Computer Integrated Manufacturing Page 20 of 25

Li, B., Melkote, S. N., 1999, Improved workpiece location accuracy through fixture

layout optimisation, International Journal of Machine Tools & Manufacture, vol. 39,

pp. 871-883.

Li, B., Melkote, S. N., 2001, Optimal fixture design accounting for the effect of

workpiece dynamics, International Journal of Advanced Manufacturing Technology,

vol. 18, n10, pp. 701-707.

Fo

Liao, Y. G. and Hu, S. J. 2001, An integrated model of a fixture-workpiece system

for surface quality prediction, International Journal of Advanced Manufacturing

rP

Technology, vol. 17, pp. 810-818

peer-00513398, version 1 - 1 Sep 2010

ee

Pelinescu, M. D, Wang, M. Y, 2002, Multi-objective optimal fixture layout design,

Robotical and Computer Integrated Manufacturing, Vol.18, n5-6, pp 365-372.

rR

Wang, M. Y, 2001, A full-kinematic model of fixtures for precision locating

ev

application, Proceedings of the 2001 IEEE/RSJ, International conference on

Intelligent Robots and Systems, Maui, Hawaii, USA, pp.1135-1140.

iew

Wang, M. Y., 2002, Prediction of workpiece-fixture contact forces using the rigid

body model, Proceeding of DETC2002, 2002 ASME Design Engineering Technical

On

Conferences, pp13~19.
ly

Wang Y., Chen, X., Gindy, N., 2006, Optimisation of machining fixture layout under

multi-constraints, International Journal of Machine Tool and Manufacture, v 46, n 12-

13, October, pp. 1291-1300.

URL: http://mc.manuscriptcentral.com/tandf/tcim Email:ijcim@bath.ac.uk

20
Page 21 of 25 International Journal of Computer Integrated Manufacturing

Wang, Y., Chen, X., Gindy, N. 2007, Surface error decomposition for fixture

development, International Journal of Advance Manufacturing Technology, v 31, p

948-956.

Wang, Y., Chen, X., Gindy, N. 2007, Elastic deformation of a fixture and turbine

blades system based on finite element analyses, International Journal of Advanced

Manufacturing Technology, accepted, in press.

Fo

Yeh, J. H. and Liou, F. W., 1999, Contact condition modelling for machining fixture

set-up processes, International Journal of Machine Tools & Manufacture, vol. 39, pp.
rP

787-803.
peer-00513398, version 1 - 1 Sep 2010

ee

Zheng, Y., Rong, Y., Hou, Z., 2005, A finite element analysis for stiffness of fixture

units, Journal of Manufacturing Science and Engineering, vol. 127, issue 2, pp 429-
rR

432.
ev
iew
On
ly

URL: http://mc.manuscriptcentral.com/tandf/tcim Email:ijcim@bath.ac.uk

21
 International Journal of Computer Integrated Manufacturing Page 22 of 25

(a) Cast turbine blade (b) Final turbine blade (c) Single hit fixture for blades
Fo

Figure. 1 Fixture development for turbine blades

rP
peer-00513398, version 1 - 1 Sep 2010

ee

Components Process
• Component functionality • Process type and parameters
rR

• Component quality and quantity • Material stock and removal rate

• Material and density • Machining force, vibration …
• Geometry of raw and final
components
• Position & orientation of features
ev

Inspections
to be machined… • Device type, specification, geometry
and mounting position;
• Measurement position …
iew

Machine
• Machine envelope Fixture
• Degrees of freedom of • Number of set-ups and accessibility
machine movement
• Layout, clamping force, deformation
• Coolant nozzle position
… • Fixture space and geometry…
On

• Component functions are achieved

• Component quality requirements are met
ly

• The time and cost of manufacturing the component are minimised

…

Figure. 2 Interaction of key factors for the fixture design

URL: http://mc.manuscriptcentral.com/tandf/tcim Email:ijcim@bath.ac.uk

1
Page 23 of 25 International Journal of Computer Integrated Manufacturing

Process and
inspection
Problems fully Yes
Components understood? End

Fixture Yes
No
requirements
Design and Fixture final design Satisfactory?
Previous
experiences manufacture of
fixture prototype No
Machine Fixture
Experiments Lessons
Best practices evaluation learned
Fo

Virtual simulations (space occupancy, kinematics, and engineering errors)

Figure. 3 Procedure of fixture development in relation to virtual simulations

rP
peer-00513398, version 1 - 1 Sep 2010

ee

F10 F13 F15

F3 F4
Sa F5 F6
rR

F11
F2
F1 Aerofoil
ev

F7
F9 Sa F8 F14 F16
F12
iew

Orientation of machined feature

Orientation of collision-free fixturing space for one set-up

Figure. 4 Machined features of the HP T800 turbine blade

On
ly

URL: http://mc.manuscriptcentral.com/tandf/tcim Email:ijcim@bath.ac.uk

2
 International Journal of Computer Integrated Manufacturing Page 24 of 25

Tool/ Inspection space

Workpiece Fo

Collision-free Collision free fixture

Fixture space space
rP

(a) Tool and inspection space (b) Trimmed fixture space

peer-00513398, version 1 - 1 Sep 2010

Figure. 5: Fixture collision free space design for the T800 turbine blade (Geldart et
ee

cal 2002)
rR
ev

CL2(3L/2C)
Locator L2 Locator 1 CL2(3L/2C)
Locator 1
CL1(3L/2C)
iew

CL1(3L/2C)
Locator L1 CL(3L/1C) CL1(3L/2C) CL(3L/1C)
CL(3L/1C)
CL2(3L/2C)
Locator L3
On

Locator 3 Locator 3
Locator 2
Locator 2
(a) The 1st fixture layout (b) The 2nd fixture layout (c) The 3rd fixture layout
ly

Locator * Feasible clamping position

Optimal clamping position in case of three locators and two clamp (3L/2C)
Optimal clamping position in case of three locators and one clamp (3L/1C)

Figure. 6 Three optimal fixture layouts generated from the kinematic analyses (Wang
et al, 2007)

URL: http://mc.manuscriptcentral.com/tandf/tcim Email:ijcim@bath.ac.uk

3
Page 25 of 25 International Journal of Computer Integrated Manufacturing

K=Force/Deformation
Fo

(a) Mesh for turbine blade (b) Contact model between fixture and (c) Calculation of Fixture
turbine blade stiffness K
rP

Figure. 7 Static FEA deformation analysis of fixture-turbine blade pair (Wang et al

2007)
peer-00513398, version 1 - 1 Sep 2010

ee
rR
ev
iew
On
ly

URL: http://mc.manuscriptcentral.com/tandf/tcim Email:ijcim@bath.ac.uk

4
View publication stats