A STUDY OF UNDERGROUND

POWER CABLE INSTALLATION METHODS

EPRI EL-374
(Research Project 7824)
ERDA E(49-18)-1824

Final Report

April 1977

Prepared by

Arthur D. Little, Inc.

Acorn Park
Cambridge, Massachusetts 02140

--------------------- NOTICE----------------------
Principal Investigator This report was prepared as an account of work
sponsored by the United States Government. Neither
James IMicol the United States nor the United States Ene
Research and Development Administration, nor any
their employees, nor any of their contractors,
subcontractors, or their employees, makes any
warranty, express or implied, or assumes any legal
liability or responsibility for the accuracy, completeness
or usefulness of any information, apparatus, product or
process disclosed, or represents that its use would not
infringe privately owned rights.

Prepared for

Electric Power Research Institute

3412 Hillview Avenue
Palo Alto, California 94304
and
Energy Research and Development Administration
20 Massachusetts Avenue
Washington, D. C. 20545

obtrbution *' rns oecuvem *

 DISCLAIMER

This report was prepared as an account of work sponsored by an

agency of the United States Government. Neither the United States
Government nor any agency thereof, nor any of their employees,
makes any warranty, express or implied, or assumes any legal liability
or responsibility for the accuracy, completeness, or usefulness of any
information, apparatus, product, or process disclosed, or represents
that its use would not infringe privately owned rights. Reference
herein to any specific commercial product, process, or service by
trade name, trademark, manufacturer, or otherwise does not
necessarily constitute or imply its endorsement, recommendation, or
favoring by the United States Government or any agency thereof. The
views and opinions of authors expressed herein do not necessarily
state or reflect those of the United States Government or any agency
thereof.

D IS C L A IM E R

Portions of this document may be illegible in electronic image

products. Images are produced from the best available
original document.
 LEGAL NOTICE

This report was prepared pursuant to an act of Congress. Publications of the findings
and recommendations herewith should not be construed as representing either an
approval or disapproval by the Energy Research and Development Administration.
The purpose of this report is to provide information and alternatives for further
consideration by the Energy Research and Development Administration and other
federal agencies.

Furthermore this report was prepared by Arthur D. Little, Inc., on account of work
sponsored in part by the Electric Power Research Institute Incorporated which does
not make any warranty of representation with respect to the accuracy, maintenance
or usefulness of the information obtained in this report and does not assume any
liabilities with respect to the use of or damages resulting from the use of any
information disclosed in this report.
 ABSTRACT

It has been determined by the utility industry that the costs of installing under­
ground high voltage transmission cable can be 50% of the total circuit in service
cost. This study was directed towards underground cable systems in service and
those types that can be commercially available in the near future. Detailed exa­
mination of the various installation steps was made and a range of costs were pre­
sented for each. After analyzing those higher cost steps for each cable type
studied, various technologies were identified that could reduce these costs if
utilities adopted these methods for use in other areas of construction.

The result of these analyses was to recommend research and development in three
general areas of installation procedure: route planning and design, pavement and
soil removal, and pipe installation methods. Recommendations in the route planning
and design area were: improved computer routing; subsurface preliminary surveying;
thermal analysis of in-situ soils, backfill material, and how cable circuits are
geometrically arranged in the trench. In pavement and soil removal, recommenda­
tions for improved methods yielded topics in the areas of: microwave, plasma torch
and water jet concrete cutting; improved guided boring systems; shoring techniques;
and soil removal systems. The installation of pipes yielded R&D recommendations
for: automatic welding techniques, integrated transmission systems, and standardi­
zation of accessories and equipment among utilities.

iii
 ACKNOWLEDGEMENTS

The staff of Arthur D. Little, Inc., and in particular the Program Manager,
Dr. James Nicol, wish to express to the members of the Steering Committee their
appreciation for the guidance and support provided them during the program. The
members of the Steering Committee were:

Mr. William Ryan, Boston Edison Company, Boston, Massachusetts,

Chairman.
Mr. William Smith, Energy Research and Development Administration,
Washington, D. C.
Mr. William Wilson, Detroit Edison Company, Detroit, Michigan.
Mr. Irving Wilson, Northeast Utilities, Berlin, Connecticut.
Mr. Aldo Zanona, Commonwealth Edison Company, Chicago, Illinois.
Mr. A. K. Akay, Pacific Gas and Electric Company, San Francisco,
California.
Mr. Herman Brooksieker, Cleveland Electric Illuminating Company,
Cleveland, Ohio.
Mr. Ralph Samm, Electric Power Research Institute, Palo Alto,
California.
Mr. Thomas Rodenbaugh, Electric Power Research Institute, Palo Alto,
California.

In the course of this program, the staff of Arthur D. Little, Inc., have received
information, assistance, and advice from many sources. It has not been possible,
within the body of the report, to mention explicitly all these contributions.
We, therefore, wish to record at this point our gratitude to all of those indivi­
duals who have helped. Should the following list be incomplete, we hope we shall
be excused.

N. Akay, Pacific Gas and Electric Company, San Francisco, California.

W. Anderson, Vice President, Geophysical Survey Systems, Billerica,

Massachusetts.
H. Brooksieker, Cleveland Electric Illuminating Company, Cleveland,
Ohio.
J. Cheung, Flow Reserch, Inc., Kent, Washington.
L. Chirstides, President, Sub-Search Corp., Rochester, New York.
L. Clipp, Exotech, Inc., Gaithersburg, Maryland.

v
J. Donnelly, Boston Edison Company, Boston, Massachusetts.
D. Duffy, American Water Blaster, Houston, Texas.
A. Eberle, Columbia Gas System Service Corp., Columbus, Ohio.
T. Ekelson, I-T-E Imperial, Spring House, Pennsylvania.
J. Giorello, Burndy Corp., Norwalk, Connecticut.
E. Geary, Boston Edison Company, Boston, Massachusetts.
A. Guzdar, Foster-Miller Associates, Waltham, Massachusetts.
E. Kazan, Kaiser Aluminum Company, Portsmouth, Rhode Island.
E. Isaac, Pacific Gas and Electric Company, San Francisco, California.
J. Jacobson, Flow Research, Inc., Kent, Washington.
L. Kilar, I-T-E Imperial, Spring House, Pennsylvania.
T. Linnert, Cleveland Electric Illuminating Company, Cleveland, Ohio.
G. Long, Institute of Gas Technology, Chicago, Illinois.
P. Long, Pacific Gas and Electric Company, San Francisco, California.
H. Maddock, Boston Edison Company, Boston, Massachusetts.
R. McElroy, Department of Public Works, Cambridge, Massachusetts.
R. Mott, Cleveland Electric Illuminating Company, Cleveland, Ohio.
J. H. Olsen, Flow Research, Inc., Kent, Washington.
W. Pauley, Ingersol1-Rand Research, Inc., Princeton, New Jersey.
S. Poulas, Geotechnical Engineers, Belmont, Massachusetts.
W. Ryan, Boston Edison Company, Boston, Massachusetts.
J. Stoneberger, UTEC Constructors, Inc., South Lynnfield, Massachusetts.
J. Thompson, Northeast Utilities, Berlin, Connecticut.
W. P. Wilson, Detroit Edison Company, Detroit, Michigan.
E. Yie, Institute of Gas Technology, Chicago, Illinois.
A. Sanona, Commonwealth Edison Co., Chicago, Illinois.

vi
 CONTENTS

Section Page

1 SUMMARY AND RECOMMENDATIONS 1-1

Summary 1-1
Recommendations 1-2

2 COSTS OF PRESENT INSTALLATION METHODS 2-1

Introduction 2-1
Methodology ■ 2-2
Civil Engineering Cost Estimates 2-4
Route Surveying, Exploration and Test Borings 2-4
Surface Removal and Restoration 2-4
Excavation and Shoring 2-4
HPOF Pipe-Type Cable 2-5
Compressed Gas Insulated Cable 2-6
Extruded Dielectric Cable 2-16

3 A STUDY OF SELECTED INSTALLATION PRACTICES AND ASSOCIATED

TOPICS 3-1
Introduction 3-1
Route Planning and Exploration 3-2
Survey Systems and Procedures 3-3
Route Planning 3-7
Cable Trench Design 3-9
Thermal Considerations 3-9
Trench Configuration 3-12
Construction Activities 3-22
Surface Removal 3-22
Excavation Techniques 3-26
Manhole Design and Installation 3-41
Pipe and Cable Handling 3-43
Underground Cable Installation in Rural Areas 3-45
Cable Section 3-47
Cable Environment 3-48
Installation Methods 3-51

vii
Section Page

Standardization 3-55
HPOF Cable Systems 3-55
Inventories 3-56
Testing Procedures 3-59

4 RECOMMENDATIONS FOR RESEARCH AND DEVELOPMENT 4-1

Program A: Route Planning and Design 4-3
Project Al: Computerized Route Planning 4-3
Project A2: Subsurface Surveying System 4-3
Improved Cable Thermal Environment 4-4
Project A3: Exploration of Cable Thermal Environment 4-4
Project A4: New Thermal Materials 4-5
Project A5: Thermal Analysis of Cable Environment 4-5
Project A6: The Thermal Interaction of Cables Arranged
One Above the Other 4-6
Program B: Pavement and Soil Removal Techniques 4-6
Project Bl: High-Power Microwave Pavement Breaking
Systems 4-7
Project B2: Plasma Torch Pavement Breaking System 4-7
Project B3: Water Cannons for Surface Removal 4-8
Project B4: Improved Trenching Machines 4-9
Project B5: Water Jet Excavation Techniques 4-9
Project B6: Improved Boring Systems 4-10
Project B7: Automated Soil Removal System 4-11
Project B8: Improved Shoring Techniques 4-11
Project B9: Temporary Shoring System 4-12
Program C: Installation Methods 4-12
Project Cl: Automatic Welding and Testing of Pipe Joints 4-12
Project C2: Reduction of Manhole Costs 4-13
Project C3: Integrated Transmission Cable Systems for
Underground Rural Use 4-14
Project C4: Testing Procedures 4-15
Project C5: Standardization 4-15

REFERENCES R-l

viii
 FIGURES

Figure Page

2.0 Sequential Diagrams for Cable Installation 2-3

2.1a Sketches for HPOF Cable Installations - Single Circuit 2-9
2.1b Sketches for HPOF Cable Installations - Double Circuit 2-12
2.2 Installation of SF^ Cables 2-13
6
2.3 Installation of Directly Buried Extruded Dielectric Cable 2-14

3.1 Isotherms for a Line Source Below an Isothermal Plane 3-15

3.2 Isotherms for a Line Source Between Two Isothermal Planes 3-16
3.3 Thermal Field for Two Thermal Source-Horizontal Configuration 3-19
3.4 Thermal Field for Two Thermal Sources-Vertical Configuration 3-20
3.5 Soil Removal System-Conceptual Design 3-30
3.6 Boom Trencher Limitations 3-30
3.7 U-Shaped Trench Shields 3-40
3.8 Possible Shield Connectors 3-40
3.9 Box Excavator Shield 3.40

ix
 Section 1

SUMMARY AND RECOMMENDATIONS

SUMMARY

The costs of installing an underground high-voltage power transmission cable con­

tribute significantly to the total cost of the system. For example, it was
indicated in 1972 that for a typical HPOF pipe type 345 kv cable the installation
costs were 50% of the total cost.* The study examined the opportunities for re­
ducing the cost of underground power transmission by improving power cable designs
or by introducting new technologies of power transmission. The present report
concentrates on the practices and procedures involved in the installation phase to
determine if significant cost savings may be obtained by the application of new
or improved technologies.

At the suggestion of the Steering Committee, the study was focused on underground
cable systems now in service in the United States and those types which may be
commercially available in the near future (such as 230 kV extruded dielectric
cable). Since emerging technologies, such as cryoresistive and superconducting
cables, are unlikely to enter commercial service until the late 1980s or early
1990s there will be adequate opportunity in the future for the development of
economical and appropriate installation methods.

Following identification of the many steps involved in installing HPOF pipe type
cable, extruded dielectric cables and SF insulated systems. Section 2 contains a
D
detailed examination of each of the steps. Using agreed upon assumptions, which
are made explicit, the costs (or a range of costs if appropriate) are presented
for each item in the installation process. The cost data are assembled in Tables
2.1 through 2.3.

In the subsequent phase of the study, the high cost items identified in Section 2
were analyzed for their cost-reduction potential. In some cases cost reductions
could be achieved immediately if a utility were to adopt technologies already in

♦"Underground Power Transmission," a study for the Electric Research Council by

Arthur D. Little, Inc., Cambridge, Mass. ERC Pub. 1-72.

1-1
use elsewhere; others, such as splicing of paper-tape cables showed little poten­
tial for savings. Section 3 contains a detailed discussion of all of the oppor­
tunities examined with preliminary suggestions as to how savings may be achieved.
Much of the discussion is directed to urban installations. Section 3 also contains
a review of the differences between urban and rural installations, and thus, of
how their R&D needs may differ.

RECOMMENDATIONS

It is pointed out in Section 4 that savings may be achieved by actions that are
not generally regarded as technical R&D. These include improved communication
between utilities, better planning technique, and standardization of components
and testing techniques. Task forces of utility engineers may be a convenient means
for examining some of these topics. The principal content of Section 4, however,
is a presentation of the recommended Research and Development Program on Installa­
tion Methods. It consists of three parts:

• Program A: Route Planning and Design

• Program B: Pavement and Soil Removal Techniques
• Program C: Installation Methods

These programs are composed of the following individual research projects.

• Program A:
— Project Al: Computerized Route Planning
— Project A2: Subsurface Surveying System
— Project A3: Exploration of Cable Thermal Environment
Project A4: New Thermal Materials
— Project AS: Thermal Analysis of Cable Environment
Project AS: The Thermal Interaction of Cables Arranged
One Above the Other

Program B:
— Project Bl: High-Power Microwave Pavement Breaking System
— Project B2: Plasma Torch Pavement Breaking System
Project B3: Water Cannons for Pavement Breaking
— Project B4: Improved Trenching Machines
— Project B5: Water Jet Excavation Techniques
— Project B6: Improved Boring Systems
— Project B7: Automated Soil Removal System
Project B8: Improved Shoring Techniques
— Project B9: Temporary Shoring System

1-2
 Program C:
— Project Cl: Automatic Welding and Testing of Pipe Joints
— Project C2: Reduction of Manhole Costs
— Project C3: Integrated Transmission Cable System for
Underground Rural Use
— Project C4: Testing Procedures
— Project C5: Standardization

All of these research projects are described in Section 4 along with a brief justi­
fication of the program, the directions of development that appear promising, and
and estimate of the level of effort required, at least through the first phase of
the project. The total effort so estimated amounts to 48 man-years.

1-3
 Section 2

COSTS OF PRESENT INSTALLATION METHODS

INTRODUCTION

Phase I of this investigation was a cost analysis of all the tasks involved in
installing three selected types of cables and the isolation of key high-cost items.
The information provided a necessary basis for Phase II, the study of areas where
significant cost reductions may be achieved through research and development.

It is important to emphasize that the purpose of Phase I was to establish the

relative costs of each task associated with a given type of cable installation.
The data collected were not intended to provide reference figures against which
bid prices of proposals and/or contracts could be compared. The range of figures
encountered for most items, reflecting contractor efficiency, regional geographic
differences and urban/suburban conditions, was so large that a listing of price
spread for each item would have defeated the purpose of Phase I. Therefore, with
several specific exceptions (which will be discussed shortly) the figures shown in
the tables are simply representative of each item. In general, the representative
figure is an average of the high and low costs we have encountered.

The cost data presented in the accompanying tables represent the combined input of
utility personnel, cable installation contractors, cable manufacturers and the ADL
project team. The general policy followed has been to show a range of costs only
where physical parameters may be prime determinants of the cost. As an example,
excavating sand is certainly much cheaper than excavating extremely close-packed
rocky soil or shale. In another case, the route planning phase may be more expen­
sive in congested urban areas than in relatively uncrowded suburban areas. Explana­
tions for each item represented by a range of costs are usually indicated directly
in the table for that item.

Cost information was developed for those cable systems that are expected to be in
most common usage during the next ten years. These cable types are (a) high
pressure oil filled pipe-type cable, up to 345 kv, (b) gas insulated (SF ) cable
b

2-1
up to 500 kv, and (c) extruded dielectric cables up to 345 kv. Some of the types
are not presently in use, but estimates of installation costs at 1974 rates are
presented where possible. Information on how the costs would change with voltage
level changes is also an important factor. For this reason three voltage levels
were studied for each cable type.

METHODOLOGY

The installation procedure for each cable type was broken down into its many in­
dividually identifiable steps, and presented in a sequential diagram (Figure 2.0).
The present costs of each of these steps or groups of steps were identified as
closely as possible. For the most part, the costs of materials were omitted; in
a few cases costs of materials were included as special costs, and appropriately
identified on the tables which will now be described. The specific steps for the
different installations and the costs associated with them are given in Tables 2.1a,
2.1b, 2.2, and 2.3. Those steps that are associated with only one or two cable
types are labelled by number. The steps that are in some way common to all under­
ground installations are labelled by letters A through L. The units of the numbers
are generally in dollars/mile for all installations. Costs that are total dollar
amounts per circuit are indicated directly on the tables in each instance.

Typical cable installation projects may involve a choice of alternative techniques

depending on the nature of the route to be followed. The soil conditions, the
desired power rating of the circuit, and the nature and number of obstructions are
typical additional factors which must be considered. No one single listing of pro­
cedures can cover all the alternatives without being repetitious. For this reason,
many of the alternatives are covered in the single headings and an average or range
of costs is given.

One item not directly considered, and frequently a significant cost component of
urban cable installation is the relocation of existing utility lines. The cost for
these activities, performed by the appropriate utilities, is kept to a minimum by
adjustments in the new cable route by reducing it to zero is unrealistic. The number
(zero to 10/mile) and cost of necessary relocations has too wide a range to be
included in the tables. The cost of each relocation is, in general, greater than
the original cost of installation but how much greater will depend on the specific
circumstances. Improved exploration techniques may result in more route flexibility
and thus lower relocation costs.

2-2
 TRIFURCATORS INSTALL OIL
INSTALL VACUUM
PIPE JOINT AND PRESSURE G
ELECTROLYSIS. TREAT
INSTALLATION 'WELDING •►TERMINATION*
PIPE
(l-«) 15-10) MONITORING
EQUIPMENT (10
MAKEUP
(42)
(31-371

CLOSE
MANHOtE TRENCH
INSTALLATION

TRENCHING

to
I
00

INSTALL PRESSURE VACUUM CHARGE

PIPE t CABLE JOINT WELDING
^ELECTROLYSIS ,
GAS INSULATED (SF) CABLE 'INSTALLATION •TERMINATIONS' TERMINATIONS TEST - * SYSTEM
(1-2) (3-6) MONITORING
EQUIPMENT (7)
(0-9) (ID (13)

cable CABLE JOINT TERMINATION

EXTRUDED DIELECTRIC CABLE laying * MAKEUP - MAKEUP '
(t) (2-7) (7-12)

Figure 2.0 Sequential Diagrams For Cable Installation

CIVIL ENGINEERING COST ESTIMATES

Route Surveying, Exploration and Test Borings

Route surveying includes the traditional field work and analysis of electric and
other utility records and plans. The extent of the analysis required depends on
the nature of the installation, ranging from minimal for new suburban developments
under construction to substantial for older urban areas, densely interconnected
with many utilities. A range of costs is presented based on the effort which may
be required for the two cases mentioned.

Some of the installation items are so intimately connected that the costs could
not be presented separately. An example of this is the combination of exploratory
excavation and test borings, where a single price is given for both items. Again,
it is almost impossible to arrive at a generalized figure for a type of installation
because the exploratory activity and number of borings required vary with the
geology, degree of urbanization and prior experience in the area. In some areas,
test borings may be considered as an optional activity; an obvious case is where
much is already known about local soil characteristics. In other areas, borings
may help to decide if shoring will be required, or if the local excavated material
may be used for backfill or thermal sand. Thus, the figures shown are an attempt
to incorporate a reasonable estimate for these activities into the tables.

Surface Removal and Restoration

Surface removal costs typically would fall between the two figures indicated. The
lower end of the range is for removal of macadam surface, and the upper end is for
macadam over a concrete base. These numbers make the simplifying assumption that
the whole trench is of one particular type; for any particular installation, the
surface being removed could vary through the whole range. It could also include
sidewalks, curbing and park land. Surfaces must be cut out in a rather precise
manner in order to avoid undue interference with traffic and to permit ease of
restoration. Existing underground installations must also be protected. The costs
of resurfacing include the costs of the paving materials because it is normally a
completely subcontracted function.

Excavation and Shoring

The most common methods of soil removal in present use are the open cut and cut-
and-cover techniques. The excavation costs shown here are representative of costs
associated with common backhoe-type equipment. Shoring of the trench walls is done

2-4
only when soil conditions and trench depth require it. The average trench depths
considered in this study (4 and 5 feet) are not required by the Occupational Safety
and Health Act (OSHA) to be shored. However, in practice the actual trench depth
can vary greatly depending on obstructions. Poor soil and water conditions could
make shoring necessary at reduced trench depths. A very common method of shoring
presently used is driving 2' x 8' lengths of lumber into the soil along the trench
walls. The costs of shoring installation given in the tables are the costs of
wooden shoring on both sides of a 5 foot deep trench for one foot of trench length.

HPOF PIPE-TYPE CABLE

Two tables are presented for the HPOF pipe-type cable. Table 2.1a is for a single
circuit installation, and 2.1b is for two parallel circuits per installation. Al­
though the additional costs for the double circuit relative to the single circuit
are large, the savings that can be realized by putting in two pipes at the same
time rather than waiting are significant. A second cable can be installed many
years after the first, or both cables can be installed at the same time. The im­
proved system reliability available from the double circuit is obvious. For these
reasons, the HPOF pipe-type cable is usually installed as a double circuit as in­
dicated in Table 2.1b. Table 2.1a for the single circuit is included for com­
pleteness. Parallel single circuits (in two separate trenches) are sometimes in­
stalled in order to increase the thermal ratings of the individual circuits above
their double circuit capacities.

The particular assumptions made regarding trench dimensions for each voltage level
(single and double circuit) are shown specifically in Figures 2.2 and 2.3. These
dimensions are representative of typical practice; it is recognized that some util­
ities may favor shallower burial of cables to avoid shoring and interference pro­
blems, and to reduce excavation costs. Conversely, other utilities may have to
work at significantly greater depths.

For the HPOF pipe-type cable, the assumption was made that each cable is approxi­
mately 1/2 mile long. This value is limited by the length of cable that can be
delivered to the site and installed, and will vary somewhat with the voltage and
power rating of the circuit. This limitation on cable length implies two joints
per mile; thus, two manholes per mile are required.

A range for manhole installation is indicated for the double circuit, HPOF pipe-
type cable. Physical considerations would lead one to expect some increase in

2-5
manhole costs for the double circuit manhole compared with the single circuit, but
it is surprising that it is as great as the range shown. These are costs that have
been paid by utilities in the recent past.

Several alternative methods are available for removing the heat from the conductor
in the HPOF pipe-type cable. The oldest is natural cooling where the heat is con­
ducted away by the surroundings. In this case, the cable requires only oil pres­
surizing equipment (Item 38 of Table 2.1). It is also possible to force cool the
system and several different geometries are possible. The oil can be circulated
through a return pipe laid parallel to the pipe containing the circuit. This re­
quires a return pipe (Item 39) and an oil circulating pump (Item 41). Some heat
exchange with the surroundings occurs while the oil is in the return pipe (similar
to naturally cooled), but usually a separate heat exchanger is put in the circuit
(Item 40). For double circuits, it is desirable to keep the oil circulating sys­
tems separate so that a leak in one line will not affect the other. However, under
some conditions it may be desirable to have the two oil circulating systems con­
nected in parallel with a common 5 inch (or larger, based on pressure drop calcu­
lations) oil return line and appropriate isolating valves. Older, naturally cooled
double circuits can be converted to forced cooled without the requirement of in­
stalling oil return pipes by circulating the oil through the two pipes in series.
Because of the varieties of possible configurations, the individual cost items in
Table 2.1 are indicated as optional. It was assumed that the oil systems for the
individual circuits in the double circuit installation were independent of each other.

COMPRESSED GAS INSULATED CABLE

The cost estimates presented in Table 2.2 for pipe placement and joint welding of
the rigid gas insulated (SF ) cables reflect the following conditions. Cable-pipe
o
lengths are pre-assembled at the factory and then delivered to the installation
site. Two types of SF systems have been installed, one using multiple finger
6
tulip type conductor connectors, and the second using slip fit connectors. This
latter type requires the welding of the center conductors as well as the pipe: it
is more complicated and expensive than the multiple finger type of connector and is
not now seriously considered. In any case, only a few short installations of either
type of SF cable system have been made with the result that the costs shown do not
6
represent costs that may be realizable on a mass production basis. As welding
crews become more proficient and approach the lower end of the learning curve, the
costs associated with welding should be reduced. Both pressure tests and vacuum
tests are used; the vacuum test cleans out all moisture and other volatile foreign

2-6
TABLE 2.1a - INSTALLATION PROCEDURE AND COSTS FOR HPOF PIPE TYPE CABLE - SINGLE CIRCUIT - See Figure 2.1a

138 kV 230 kV 345 kV Costs shown are in $/mile of 30 circuit and

$/mi 1 e $/ini 1 e $/mi 1 e materials are excluded; exceptions in either
case are specifically noted.

2,500 — 10,000 A. Survey (field surveying and blueprint analysis)

8. Exploration (including test pits around congested
11,700 14,700 sites)
C. Test Boring (to determine soil characteristics)

5,300 — 15,800 7,000 — 21,100 0. Surface Removal (range is from macadam to macadam
over a concrete base)
14,600 — 58,700 19,600 — 78,200 E. Excavation (range is from sand to rocky soil)
10,600 14,100 F. Disposal (assumes no suitable storage area on site)

$30/lineal foot of trench G. Shoring (price assumes reusable wooden planks

forced into the trench bottom as needed)
1. Field Fabrication
30,000 35,000 2. Testing Pipe Installation
50,000 3. Bending
4. Placement
5. File & Clean
6,000 8,000 6. Rough & Prime Paint Pipe Joint Preparation
10,500 7. Bitumastic and Finishing
Use of 47 foo t (nominal) pi pe implies 113 joints/mile. 8. Clamp & Finish
11,000 14,000 19,000 9. Meld Pipe Joint ^ t e: Welding and X-raying
11,000 11,000 17,500 10. X-ray Weld ^ ep place in practice after
5 above.
10,000 11,000 17,000 H. Manhole Installation, 2 per mile assumed
800 11. Electrolysis/Monitoring Equipment Installation
21,200 28,200 I. Thermal Sand, Installation
13,500 —- 16,900 17,600 — 22,000 material, 8-10 $/yd3 Close Trench
14,100 18,800 J. Backfill* Installation
5,900 — 9,500 7,800 — 12,500 material , 5-8 $/yd^
K. Resurface Trench (this figure is for a typical
66,000 79,000 urban installation; it Includes macadam and
concrete base materials).
 TABLE 2.1a - INSTALLATION PROCEDURE AND COSTS FOR HPOF PIPE TYPE CABLE - SINGLE CIRCUIT (Confd.)

* = Optional Expense
138 kV 230 kV 345 kV Costs shown are in $/mile of 30 circuit and
$/mile $/mile $/mile materials are excluded: exceotions in either
ease are specifically noted.
12. Set up Cable Reels
13. Set up Winch and Pulleys
18,000 19,000 21,000 14. Blow "Mouse" through Pipe
15. Install Leader Cable
Prices assume 2 pulling opera tions per 16. Install Oiling Tube and Discs Cable
mile as an av erage figure. 17. Attach Pulling Cable Installation
18. Pull Oil Discs/Pulling Cable
19. Attach Cable Header to Cable
20. Pull Cable

21. Install Nitrogen Barrier

22. Prepare Cable/Instal1 Connector
11,500 14,000 16,000 23. Finish Connector
24. Apply Paper Tapes 1
It is assume: that there are 2 joints per cable 25. Apply Shield Tape, Ground Strap
per mile, and that each circu it consists of 3 cables. 26. Remove Nj Barrier, Apply Bond Wire Cable
27. Install Spacer ("Spider") Joint
28. Bind Joints, Connect Bond Makeup
29. Position Joint Sleeve Halves J
30. Weld Joint Sleeve

31. Install Trifurcators with Riser Pipes

8,000 Total Per Circuit (2 sets/circuit)

32. Seize and Cut off Skid Wire

13,000 15,000 18,000 33. Install Check Plate m
Total Total Total 34. Remove Copper Shield Tape,Sealing Tape,
Per Circuit Per Circuit >er Circuit Prepare Cable Ends, Install Connector,
Clean Connector &S
It is assumed that 6 intermin ations are required 35. Apply VC Tape, Build Stress Cone, jj 2;
for each 30 c ircuit. Apply Shielding Tape —o
36. Continuity Check and Phase Check J «
37. Install Pothead

6,000 Total Per Circuit 38. Install Oil Pressurizing Equipment

47,000 *39. Install 2 inch Oil Return Pipe

falono with cable oioe)
4,000 *40. Install Heat Exchanger
5,000 *41. Install Oil Circulating Pumo
3,500 4,500 6.000 42. Vacuum Treat Pipeline (once per circuit)
4,000 7,000 12,000 43. Fill Pipeline and Potheads with Oil
(labor only)

2,500 2,500 2,500 L. Acceptance Test

267.7 336.8 286.2 — 351.3 361.1 — 450.4 Totals ($10i/mile) exclusive of costs associated with shoring,
forced cooling, terminations and startup.

i
 138 KV 230 kV 345 kV

SURFACE SURFACE SURFACE

BACKFILL BACKFILL BACKFILL

THERMAL THERMAL THERMAL

SAND SAND SAND

10 8"l.D. 3
I 6 I.D.
IXI
6
V
3'

*It is recognized that some utilities may prefer to use 8 inche I.D. pipe
for large diameter conductors; in this case, the data for 230 kV pipe
installation items would apply.

Figure 2.1a Sketches For HPOF Cable Installations - Single Circuit

TABLE 2.1b - INSTALLATION PROCEDURE AND COSTS FOR HPOF PIPE TYPE CABLE - DOUBLE CIRCUIT - See Figure 2.2b

138 kV 230 kV 345 kV Costs shown are In $/m11e of 30 circuit and

S/mHe t/mile $/m1 le materials are excluded; exceptions 1n either
case are specifically noted.

2,500 — 10,000 A. Survey (field surveying and blueprint analysis)

B. Exploration (Including test pits around congested
18,000 21,000 sites)
C. Test Boring (to determine soil characteristics)

8,800 — 26,400 10,600 — 31,700 D. Surface Removal (range 1s from macadam to macadam
over a concrete base)
24,000 — 97,800 29,300 —117,300 E. Excavation (range Is from sand to rocky soil)
17,600 21 ,100 F. Disposal (assumes no suitable storage area on site)

$30/lineal foot of trench length G. Shoring (price assumes reusable wooden planks
forced Into the trench bottom as needed)
1. Field Fabrication
50,000 60,000 90,000 2. Testing Pipe Installation
3. Bending
4. Placement

11,000 13,500 19,000 5. File & Clean

6. Rough i Prime Paint Pipe Joint Preparation
Use of 47 fot t (nominal) pi pe Implies 113 joints 7. Bitumastic and Finishing
per mile per circuit. 8. Clamp & Finish
20,500 24,000 35,000 9. Weld Pipe Joint NoteWelding and X-raying
21,000 21,000 34,000 10. X-ray Weld place In practice after
5 above.
10,000 — 23,200 12,000 — 27,800 H. Manhole Installation, 2 per mile assumed
1.600 11. Electrolysis/Monitoring Equipment Installation
35,200 42,300 I. Thermal Sand, Installation -■
23,500 — 29,300 25,700 — 32,100 material, 8-10 $/yd3 close Trench
23,500 28,200 J. Backfill, Installation
9,800 — 15,700 11,800 — 18,900 materl al, 5-8 $/yd3 -*
K. Resurface Trench (this figure Is for a typical
92,000 106,000 urban Installation, It Includes macadam and
concrete base materials).
 TABLE 2.1b - INSTALLATION PROCEDURE AND COSTS FOR HPOF PIPE TYPE CABLE - DOUBLE CIRCUIT (Cont'd.)

* ■ Optional Expense
138 kV 230 kV 345 kV Costs shown are in $/mile of 3? circuit and
$/mile $/mile $/mile materials are excluded; exceptions in either
case are specifically noted.
12. Set up Cable Reels
13. Set up Winch and Pulleys -i
32,000 34,000 38,000 14. Blow "House" through Pipe
15. Install Leader Cable
Prices assu me 4 pulling ope rations per 16. Install Oiling Tube and Discs Installation
mile as an average figure ( two each per circuit) 17. Attach Pulling Cable
18. Pull Oil Discs/Pulling Cable
19. Attach Cable Header to Cable
20. Pull Cable

21. Install Nitrogen Barrier

22. Prepare Cable/Instal1 Connector
22,000 27,000 31,000 23. Finish Connector -i
24. Apply Paper Tapes
It is assum ed that there ar e 2 joints per 25. Apply Shield Tape, Ground Strap Cable
cable per m lie, and that ea ch circuit 26. Remove N? Barrier, Apply Bond Wire Joint
consists of 3 cables. 27. Install Spacer ("Spider") Makeup
28. Bind Joints, Connect Bond
29. Position Joint Sleeve Halves
30. Weld Joint Sleeve
IT-:

31. Install Trifurcators with Riser Pipes

16,000 - Total (both circuits) (2 sets/circuit)

32. Seize and Cut off Skid Wire

26,000 30,000 36,000 33. Install Check Plate
Total Total Total 34. Remove Copper Shield Tape Sealing Tape, Tjb
Per Circuit Per Circuit Per Circuit Prepare Cable Ends, Install Connector, „ ^4
Finish Connector n> a
It is assum ed that 6 termin ations are 35. Apply VC Tape, Build Stress Cone, £ EL
required fo r each 3f) circui t. Apply Shielding Tape “g
36. Continuity Check and Phase Check J «•
37. Install Pothead

1 2,000 (Total , 2 circuits) 38. Install Oil Pressurizing Equipment

*39. Install two 2 inch Oil Return Pipes

90,000 (along with cable pipe)

8.000 *40. Install Two Heat. Fxrhanaers

10.000 *41. Install Two Oil 01 r cti 1 a t i no Pumos
7.000 9.000 12.000 42. Vacuum Treat Pinelines (once ner circuit!

8,000 13,000 22,000 43. Fill Pipelines and Potheads with Oil
(labor only)
4,500 L. Acceptance Tests

423. —► 546.8 j 446. — 569.8 559.4 705. Totals ($10^/mile) exclusive of costs associated with shoring,
forced cooling, terminations and startup.
 SURFACE SURFACE

BACKFILL BACKFILL

5
THERMAL SAND THERMAL SAND

L---------30"----------
2 ‘

°Ep Q)
-12

L ----------------------- 6'----------------------- J

138 kv: D = 6" I. D. 345 kV= D= 10"l.D.

230 kv: D= 8" I. D.

Figure 2.1b Sketches For HPOF Cable Installations - Double Circuit

 230 kV SURFACE

‘BACKFILL

2 Vz'

345 kV SURFACE

BACKFILL

THERMAL SAND

SURFACE
500 k V,

BACKFILL

THERMAL SAND

Figure 2.2 Installation of SF^ Cables

2-13
 SURFACE

BACKFILL

ii
THERMAL SAND

3' j.irj
O O

3'

NOTE: Identical trench dimensions and cable spacing have

been assumed for all voltages from 138 kV to 345 kV;
for larger trenches, refer to Table 2.1a (4 foot width)
and Table 2.2 (5 foot width).

Figure 2.3 Installation of Directly Buried Extruded Dielect Cable

2-14
 TABLE 2.2 - INSTALLATION PROCEDURE AND COSTS FOR GAS INSULATED (SF>;) CABLES - See Figure 2.2

230 kV 345 kV 500 kV Costs shown are in $/mile of 30 circuit and

$/mile $/mile $/mile materials are excluded; exceptions in either
case are specifically noted.
2,500 —- 10,000 2,500 —11 ,000 2,500 — 13,000 A. Survey (field surveying and blueprint analysis)
8. Exploration (including test pits around congested
14,700 21,000 27,000 sites)
C. Test Boring (to determine soil characteristics)

8,000 — 26,400 1 1 ,400— 34,300 13,200 — 39,600 D. Surface Removal (range is from macadam to macadam
applied over a concrete base)
19,600 — 78,200 31 ,800—127,100 37,700 —147 ,000 E. Excavation (range is from sand to rocky soil)
14,100 22,900 26,400 F- Disposal (assumes no suitable storage area on site)
$24/l.f.o.t. $30/lineal foot of trench G. Shoring (assumes reusable wooden planks as in 2.1a)

79,200 83,200 88,000 1. Pipe Placement

2. Leveling Installation
69,000 78,000 97,000 3. Install Conductor Grading and Pipe Shields
It should be not :d that these fi gures are pure estimates, 4. Connect and Weld Joints T Joint
since no "long" transmission 1in as U z 1 mile) have been 5. X-ray Joints Assembly
built at present 6. Apply Protective Coating at Joints
-15

Not Standardized; see text. H. Manhole Installation, 2/circuit

7. Electrolysis/Monitoring Equipment Installation
1,500 Total 1,800 Total 2,100 Total 8. Install Riser Pipes, 2 sets of 3 per circuit

15,000 Total 18,000 Total 21 ,000 Total 9. Install Cable Terminations, 2 sets of 3 per
circui t
23,800 44,700 46,200 I. Thermal Sand, Installation
1 5,900 1 9,800 29,800— 37,300 30,800 — 38,500 material , 8-10 $/yd^
11 ,700 11 ,700 11,700 J. Backfill, Installation
4,900 — 7,800 4,900— 7,800 4,900 — 7,800 material, 5-8 $/yd^
92,000 12,000 25,000 K. Resurface Trench (urban installation, materials incl.)
12,000 13,000 15,000 10. Install Gas Equipment
2,500 2,500 2,500 11. Pressure Test
2,500 2,500 2,500 12. Vacuum Test
2,000 3,200 5,550 13. Charge System
5,000 5,000 5,000 L. Acceptance Test
369.4 — 460.7 170.1 — 607.2 531. —► 687.8 1 Totals ($10'7mile) exclusive of costs associated with shoring,
1 forced cooling, terminations and startup.
material present in the system before it is finally charged and the pressure test
is a more effective way of finding leaks.

The need for manholes to carry out pulling and splicing of cables is well estab­
lished for HPOF and solid dielectric cables, but there is no similar need for gas-
insulated cable. Under certain conditions, such as along urban installation,
there may be a need to put gas pressurization equipment and/or reservoirs under­
ground rather than on the surface; and the dimensions required for the enclosure
will reflect the particular needs at a given site. The need for "manholes" depends
on specific installations, and the dimensions are similarly tied to specific needs.
We have, therefore, not inserted any particular costs in the table for installation
of manholes, but left this item unspecified.

EXTRUDED DIELECTRIC CABLE

There are presently no installations of 230 kV and 345 kV extruded dielectric cable
in the United States. The costs presented in Table 2.4 are estimates for future
installations at these voltages. It is difficult to directly bury a cable a half-
mile long in a city street because of obstructions from other utilities and the
maximum length of trench that can be open at any one time. Consequently, the cable
may be ducted or installed in a pipe with costs that closely resemble those of in­
stalling an HPOF pipe-type cable.

Extruded dielectric cables are most useful where obstructions are not a problem
and they can be installed without ducts. Under some conditions the cable is laid
alongside a water pipe to improve its thermal environment. Table 2.3 presents the
costs of a directly buried system, that is without ducts. It should be noted,
however, that at least one large midwestern utility is increasingly turning to the
use of plastic ducted extruded dielectric cable primarily for economic reasons for
power transmission in the 115 kV region. They selected this type of system for the
installation of approximately 50 circuit miles, because they can demonstrate cost
savings on the order of $50,000 per mile for the total initial cost. Additional
advantages are that less pulling tension is required, and that greater ease in
handling results from a cable weight much less than LPOF cable.

For extruded dielectric cables, the assumption was made in our analysis that each
cable is approximately 1/2 mile long. The actual value is limited by the length
of cable that can be delivered to the site and installed, and will vary somewhat
with the voltage and power rating of the circuit. In general, temporary manholes
are required for this type of cable. The two exceptions arise if surge diverters

2-16
 TABLE 2.3 - INSTALLATION PROCEDURES AND COSTS FOR DIRECTLY BURIED EXTRUDED DIELECTRIC CABLES - See Figure 2.3

138 kV 230 kV 345 kV Costs shown are in $/mile of 3f) circuit and materials
$/m1lB $/m11e $/m11e are excluded; exceptions in either case are
specifically noted.
2,500 10,000 A. Survey (field surveying and blueprint analysis)
8. Exploration (Includes test pits at congested
11,700 locations)
C. Test Boring (done to determine soil characteristics)

5,300 — 15,800 0. Surface Removal (range is from macadam to macadam

over a concrete base)
14,600 — 58,700 E. Excavation (range is from sand to rocky soil)
10,600 F. Disposal (assumes no suitable storage area on site)

$30/11neal foot of trench G. Shoring (price assumes reusable wooden planks as in

2.1a)
6,600 7,000 7,500 1. Cable Laying
2. Remove Jacket, Shield
3. Pencil Insulation, Clean Apply Connector Cable
LT-

6,000 7,000 8,000 4. Apply Tapes Joint

5. Apply Shielding, Braid and Bond Wire Makeup
6. Apply Outer Tapes 2/mile
7. FinishJoint
14,000 H. Manhole Installation (2/circutt)
21,200 I. Thermal Sand, Installation
13,500 — 16,900 material, 8-10 $/yd^
14,100 J. Backfill, Installation
5,900 — 9,500 material , 5-8 $/yd3

66,000 K. Resurface (price 1s for an urban Installation, and

includes materials)

Termination
7. Remove Jacket, Shield Tape, Semi-con Tape '

(2 sets)
Makeup
10,200 12,000 15,000 8. Terminate Conductor
Total Total Total 9. Pencil Insulation, Build Stress Cone
Per Circuit Per Circuit Per Circuit 10. Apply Shielding, Braid, Install Bond Wire
11. Apply Outer Tapes
12. Make Final Connection J
2,500 L. Acceptance Tests
192. —*■ 261 . 1 193.4 —» 262.5 194.9 —- 264. Totals ($103/mile) exclusive of costs associated with shoring,
forced cooling, terminations and startup.
are to be utilized for cross-bonding purposes and at terminations. In these ex­
ceptional cases, permanent enclosures are appropriate.

2-18
 Section 3

A STUDY OF SELECTED INSTALLATION

PRACTICES AND ASSOCIATED TOPICS

INTRODUCTION

The information on the costs of installation developed and presented in the previous
section provided a guide to the performance of the next phase of the study—the
examination of those steps in the installation process in which research and devel­
opment offered the hope of significant cost reduction. The investigation included
extensive discussion with major urban utilities on the Atlantic and Pacific coasts
and in the Midwest. These had three objectives:

• To review present local installation practices,

• To search for any problem area that had been overlooked, and
• To receive comments upon ideas for later R&D now being explored.

In addition to serving these functions, the discussion, however, brought into focus
other aspects of the installation problem. Some of these have already been stated,
but they bear repeating.

• Virtually all underground transmission has been and for some time
in the future will continue to be installed in the already built-up
areas of a city. Accordingly, our primary emphasis in this study
is on urban installations. It is recognized, however, that the
number of cable installations in suburban and even rural areas is
expected to increase in the future. We have, therefore, incorpor­
ated an analysis of non-urban cable installations in a separate
section within this section.
• The very high costs of excavation in the city stem from several
factors including the existence of previously installed and im­
perfectly charted water and sewer lines, other cable lines, etc.;
the highly variable soil and fill in old cities which have gone
through various phases of construction and reconstruction; the
difficulties of working in congested city streets and so on. All
of these tend to emphasize the pragmatic approach to urban excava­
tion: any techniques used must be flexible and easily varied; the
long unvaried operation so beloved of mass-production can rarely
be used successfully.

• Local history and geology have a strong influence upon techniques

used in the installation of underground transmission cables. There
is therefore a wide variation between cities. It is not always
clear why techniques used successfully in one city are rejected
elsewhere.

3-1
 • The explicit cost of an individual step in the installation process
may not reflect accurately its actual cost to the completed task.
For an example, route exploration may not appear to be an expensive
element, but the effects of inadequate or imprecise location of
obstacles in the path of the cable system can add substantially to
the costs of excavation (either actual or necessitated by substan­
tial contingency allowances).

ROUTE PLANNING AND EXPLORATION

The installation of underground cables in urban areas is often complicated by the

presence of many underground obstacles. If the density of urban obstructions such
as other transmission cable systems, water pipes, steam lines, gas mains and other
buried items is high enough, the planning phase of cable installation alone may be
difficult and costly requiring updated blueprints of known utility lines and struc­
tures foundations. Suburban and rural route planning are progressively less de­
manding in terms of underground obstacles. Steam lines and complex intersections
are rarely, if ever, encountered; normal obstacles may be occasional crossings of
other utility lines, rock ledges, and poor soil areas such as swamps.

Continuing, in urban areas, even with updated blueprints of known utility lines
and structure foundations, questions of clearance for proposed pipe-type systems
may arise; the usual solution is exploratory excavation. These "test pits" usually
require slow and costly manual excavation because of the density of obstructions.
Although the figure may vary with local labor rates, a nominal cost estimate for
this type of manual excavation would be around $35 per cubic yard. If a parti­
cular cable installation requires a large number of test pits, their cost may be
a significant fraction of the total cost.

Blueprint analysis and planning is a cost item which generally decreases as one
moves from older congested urban areas to newer urban areas to suburban environ­
ments. The planning of cable installations in heavily congested urban underground
areas, however, is sufficiently expensive to warrant exploration of cost-reducing
techniques. In a separate section entitled Route Planning, we will discuss a
proposed method for utilizing computer capabilities to reduce expenses in this phase
of cable installation. The following discussion is centered upon acquiring the
data necessary for detailed route planning.

Before proceeding to an analysis of current underground exploratory systems, it is

necessary to have an accurate understanding of the environment wherein these systems
must operate. Urban areas often present a very difficult environment for remote

3-2
sensing systems. Soil densities and conductivities may vary within a wide range
in a given area. This will affect the elastic sound speed and the electromagnetic
wave velocity which are crucial to radar-like systems. In addition, a typical
street may have layers of macadam, concrete or reinforced concrete, crushed stone,
water-bearing strata or various fill materials. These will all complicate the
task of obstacle location and identification. Suburban and rural areas are much
more amenable to effective radar-like system surveys.

The remaining problems include the nature of the obstacle material and the funda­
mental limitations of each system. Obstructions may consist of such materials as
rock, steel, concrete, cast iron, fired clay, fiberglass, plastics, and asphaltic
fiber-reinforced compounds. The response of each material to a given sensing sys­
tem is different. For example, the responses of plastic and fiberglass materials
to eletromagnetic radiation are similar, but different from concrete or cast iron.
The limitations of several systems, such as depth capability and accuracy of loca­
tion, and the material they are best suited to detect will be discussed in the
following sections.

Survey Systems and Procedures

Pulsed High-Frequency Electromagnetic Systems. These systems, which fall into the
broad category of radar systems, can respond well to high density gradients (such
as between soil and rock, ferrous objects, fired clay and water-bearing strata).
A general purpose unit which responds well to these discontinuities will not res­
pond as well to buried transite, fiberglass and PVC objects. Some specialized
systems, however, can locate these materials rather well when they assume a long,
thin cylindrical shape; they will be discussed shortly.

The depth of accurate exploration is inversely proportional to the frequency em­

ployed, while the accuracy of spatial positioning of objects is directly propor­
tional to the pulsed frequency employed. Soil attenuation characteristics establish
an upper frequency limit in the neighborhood of 1000 MHz, while the positioning
accuracy requirements of suitable detecting systems imply a lower limit of around
200 MHz.

The preceding discussion implies that either a compromise mid-range frequency such
as 300 MHz should be employed, or that a dual-frequency device would be appropriate.
The latter would have a depth capability of as much as 22 feet at a lower frequency

3-3
of 200 MHz, but its reolution would be 30 inches at best.^ Under the same
soil conditions, a frequency of 800 MHz would result in depth capability of around
9 feet and a resolution of approximately 7 inches.

A typical general purpose downward-looking pulsed radar system in operation today

has an average depth capability of 8 feet, and a stated positioning accuracy of
(2 )
± 6 inches. This particular system employs a pulsed (approximately 2 nano­
seconds) broadband signal (300 MHz) and relies upon the different reflective
"signature" of each material to produce information at the single receiving-trans­
mitting antenna. Interpretation of the reflected signals may incorporate computer­
ized data processing, and is usually done in conjunction with existing utility maps
and drawings which help to identify reference objects. It is particularly hampered
by street surfaces containing reinforced concrete, and areas where overhead metallic
structures exist. The steel mesh in the concrete may effectively stop the signal,
and overhead structures contribute unwanted signal reflections to the receiving
antenna which travels a few inches above ground.

The Electroscience Laboratory at Ohio State University has developed a system

(3)
particularly oriented toward detecting and locating plastic gas pipelines.
This system takes advantage of the long, thin cylindrical geometry of pipe-lines
and the fact that a radar reflective cross section for the geometry is considera­
bly enhanced when the electric field of the pulse is parallel to the pipeline
direction. A highly specialized antenna has been designed to respond well to this
signal. Additional processing of the received signal will enable the operator to
determine the type of material, the depth and lateral location. Since the inven­
tors have applied for a patent on the device, other details of operation are closely
guarded. Performance figures are given as a depth capability of 10 feet and a
resolution of 6 inches in both horizontal and vertical positioning.

To summarize, under "reasonable" conditions in urban/suburban areas present under­

ground radar systems can provide immediate advantages in terms of direct labor cost
reductions, reduced traffic disruption while engaged in subsurface street explora­
tion and the elimination of damage of adjoining utility lines incurred during
exploratory excavation. Reasonable in this contestt implies no steel reinforcing
mesh in the concrete paving, an essentially homogeneous type of soil or fill mater­
ial, and relatively wide spacing (> 6 inches) between parallel pipes. Serious
system limitations will result from the presence of steel mesh and a widely varying
mixture of soils; parallel pipes closely spaced will register as a single large
object. Rural areas where the soil is essentially homogeneous would permit the
optimal use of radar systems.

3-4
The advantages of these systems rely upon a careful and complex analysis of the
recorded pulse reflections which often contain multiple reflections of relevant
objects and a high percentage of clutter. Our recommendations for future research
in this area are directed toward simplifying the task of translating these elec­
tronic images, hence speeding up the overall process and lessening the need for
highly trained interpretation specialists. Specific suggestions include the
following:

• Increased development of information derived from the polarization

of reflected signals
• Continued development of the synthetic aperture approach for general
purpose radar systems
• Continued development of the pulsed radar system developed solely
for pipe detection
• Continued development of computer processing techniques to retrieve
information from dispersion data, through Fourier analysis, and to
reduce clutter
• Possible development of multi-antenna interferometric systems to
enhance positioning accuracy and produce additional subsurface
propagation data.

Continuous Low-Frequency Electromagnetic Systems. Several varieties of this gen­

eral system are in common use. The most popular type relies upon the ability of
transmitted low-frequency electromagnetic waves to induce a current in buried
metallic objects. Besides this obvious limitation of excluding transite, fiber­
glass, and clay pipes, these systems generally suffer from poor resolution. Their
major virtues are ease of operation (both transmitter and receiver are lightweight
and used on or above the surface) and the fact that a relatively quick but crude
survey of an area can be obtained.

A more sophisticated system has been designed to locate buried metallic objects
and pipes. This system (commonly referred to as the GRM system) requires that two
electrodes be driven into the ground near the area to be mapped. Such a procedure
is more time consuming than the previously mentioned system, but the GRM system has
higher positioning accuracy (± 2 inches). This superior accuracy is due in part
to the fact that both null and peak signal detection may be employed, and in part
to refinements in circuit elements. Considering the relatively low operating
frequency of around 3 kHz, its claimed resolution (the maximum sseparation distance
observable between two parallel objects) of 8 inches is excellent. Improvements
in this system would result primarily from increasing the size of the antenna,
which would reduce the ease of application. It thus appears that further research
into this system would not result in significant improvements.

3-5
Acoustic Systems. Two particular areas of acoutistic application led us to con­
sider the use of acoustic systems for underground surveying: sonar and acoustic
holography. The latter technique has shown particularly good performance in pro­
viding detailed imaging of massive opaque objects. Unfortunately, each technique
suffers from a major physical limitation when applied to the task of locating
passive buried objects, particularly those buried beneath city streets. This
limitation is due to the inhomogeneity in density of the soil or materials covering
the objects of interest. Changes in density produce changes in the local sound
speed. A partial list of materials and sonic speeds is presented below; it should
be noted that these are nominal speeds, since each material is subject to local
. (5)
variations.

Material Speed of Sound in fps

Water 'MSOO
Soil (Texas) ^5000
Concrete 10000
Clay Rock 11420
Marble 12500
Slate 14800
Granite 19700

Density changes also produce deviations from straightline propagation. The com­
bination of uncertainty in return pulse timing analysis (due to varying sound
speeds) and unknown propagation paths implied that precise positioning cannot be
established. Knowledge of the soil density variations in the area of interest may
only be gained through exploratory excavation and testing; excavation, of course,
is the costly procedure we are trying to reduce or eliminate.

Another, equally complicating factor is the physical construction of most roadbeds.

There are three material interfaces in conventional roadbeds and there are acoustic
impedance mismatches at all three. The three interfaces are 1) the interface
between the free air and the concrete road surface; 2) the interface between the
concrete and the gravel roadbed; and 3) the interface between the gravel roadbed
and the subsurface soil. Because of the impedance mismatches, both the layer of
concrete and the gravel roadbed will act like acoustic ducts (waveguides) in which
much of the acoustic energy will be trapped. This trapped energy will propagate
essentially in the horizontal direction but it would suffer multiple reflections
before reaching the receiver. These signals will be stronger than those signals
being reflected off the underground obstacles, and the total received signal would
often be indecipherable. If one bores through both pavement and roadbed, so as to
couple the source (and similarly, the receiver) directly to the soil, the acoustic

3-6
conditions would be optimized at the cost of prohibitively expensive and time-
consuming application effort (boring many holes in the city street and refilling).

Our overall opinion of acoustic underground surveying systems is that such a system
is not practical at present. In addition, the theoretical limitations imply that
research into improving this system will not produce significant benefits to a
utility in terms of underground obstacle determination and location.

Other Obstacle Detection Techniques. At this time we should emphasize that our
prime purpose in this report is to examine methods and potential methods for re­
ducing the cost of underground cable installations. The underground surveying
systems we have just discussed were chosen because it appears that two of the three
have near-term potential for actually reducing installation costs. We examined
acoustic systems for their cost-reduction potential based on successful applications
in other fields (sonar and holography). Our conclusion was that these systems
would not be so successful in utility application.

A detailed examination of all feasible underground surveying methods was the subject
of a Stanford Research Institute report ("The Use of Advanced Technologies for
Locating Underground Obstacles," SRI Project PYU-2660, government contract number
14-01-0001-1570). The topics covered included radar and electromagnetic methods,
acoustic methods, infrared and nuclear activation methods. Of those methods not
already discussed here, only passive magnetic detection appeared to have some promise,
according to the SRI report. We suspect that this type of system, which relies
upon perturbations of the earth's magnetic field, is of questionable value to
utilities because of its sensitivity to local extraneous magnetic material.

Route Planning

Finding a satisfactory cable routing through an urban street intersection can be

a major problem both in terms of complexity in planning and exploratory excavation
expense. Although the preceding sections were devoted to examining underground
surveying systems which would help reveal additional obstacle data, the effect on
route planning was somewhat indirect. By combining known data from layout plans
and surveys with the additional data (or verifications) gained through the under­
ground surveying systems, the route planner would presumably be able to determine
if a power cable system would be threaded through the intersection. It is this
phase of the installation that we will now examine.

3-7
Current routing procedures rely upon the careful analysis of existing survey data
and plans. Depending on the complexity of the intersection, much effort may be
expended on calculations and clearance checks. If a proposed route has been cleared
through most of an intersection and then runs into insurmountable obstacles, it
may sometimes be necessary to shift the whole projected route starting from the
entrance side. This implies a new round of calculations and projections.

The preceding method of approach would seem to be amenable to the use of computer­
ized data and analysis. Once the data from surveys and plans have been translated
into a computer's memory, several procedures are possible. If a large-X-Y plotter
is available, a drawing of the input data may be executed as a check against the
original. Following this verification, a planner may then feed data on a proposed
new route into the computer. Utilizing programs which are being developed for this
general purpose' '(e.g., the layout of complex plant piping schemes) the program
output may indicate whether the proposed route is satisfactory or where there will
be some conflict. In the first case, the satisfactory route may then be added
automatically to the new drawing. In the second case, the planner may select a new
proposed routing and quickly determine its acceptability through computerized cal­
culations. The skills of a route planner may thus be directed toward experienced
estimates of successful routes without the burden of extensive calculations.

At present, the main area where developmental effort is required is in the initial
process of transferring line drawings and comments from survey drawings into digi­
tizers and alpha-numeric data suitable for computer processing. Several approaches
are possible, ranging from manual operations which translate line drawings into
punched card decks (the least capital-intensive approach) to automatic digitizers
and optoelectronic drawing processors.

It is anticipated that future programs may be written to incorporate the minimum

bending radii of HPOF cable systems, the incremental costs of raising or lowering
the projected route, and even estimates of total cost including relocation of
obstacles if specified. In an ultimate form the system may be able to accept under­
ground survey data directly. These are, of course, only suggestions for more ad­
vanced programs. For the present, it appears that programs under development for
laying out complex industrial piping schemes may be of significant benefit in re­
ducing the cost of route planning.

3-8
CABLE TRENCH DESIGN

Thermal Considerations

When current flows in power cables, heat is generated as it is in any electrical

resistor. Although the cable conductors may have an extremely low resistance, the
high currents carried in transmission cables may cause a significant amount of heat,
which must be conducted away to avoid excessive cable temperatures. Unfortunately,
materials that provide good electrical insulation also provide a high level of
thermal insulation, so that the path from the electrical conductor to the sin:round­
ing soil has a high thermal resistance. Once the heat reaches the surrounding
soil, it must flow to a heat sink as efficiently as possible.

This section will investigate two general methods of reducing the costs of under­
ground cable installations from the point of view of the thermal environment:
1) the reduction of direct installation costs, and 2) increasing the power rating
of existing power transmission systems, either previously installed or proposed,
to reduce the need for new transmission capacity. The latter approach would reduce
the cost of the system but not necessarily the cost of installing the individual
cable.

Thermal Sand. Most thermal backfill materials presently in use are granular (e.g.,
sand). These granular materials must be installed and tamped into position and
lifts to have good compaction and little settling. "Puddling" or compacting with
water has been found to be less satisfactory.

The thermal resistivity of a granular material is governed by the contact between

soil grains and the fluid contained in the pores between the grains. The wider the
range of grain sizes, the greater the number of grain-to-grain contacts, since the
smaller grains can be positioned in the spaces between the larger grains. Tamping
increases the compaction and the number of grain-to-grain contacts. While the
grain-to-grain resistivity may be low, the resistance is inversely proportional to
the contact area and may be high. Water will fill the spaces near the grain con­
tacts and increase the effective area. Because the layer of water is very thin, its
resistance is low.

Solid quartz has a thermal resistivity of ll°C-cm/watt; uniform, dry crushed quartz
has a thermal resistivity of 500°C-cm/watt. When water (resistivity of 160°C-cm/watt)
is added to crushed quartz until the material is 30% saturated, the thermal resisti­
vity of the material becomes 45°C-cm/watt. This resistivity is an order of magnitude

3-9
less than the resistivity of dry, crushed quartz and almost a factor of 4 less
than the resistivity of water. Lower thermal resistivities are available in widely
. . moist
graded, . . sands.
. (ID

Once a moist thermal sand is in position around the cable, maintaining the proper
moisture content becomes important. The water close to the pipe is hotter and has
a higher vapor pressure than does the water a short distance away. For this reason,
the water will tend to migrate away from the heated circuit. If this process pro­
ceeds to the point of seriously increasing the thermal resistance, a hot spot will
develop in the cable.

Devising some method of retaining the moisture uniformly in thermal sand when the
circuit is operating at high capacity could lead to significant savings by increas­
ing the power rating of the circuit. Two methods of reducing the tendency of water
to migrate should be considered: 1) lowering the water's vapor pressure by use of
an additive, and 2) increasing the ability of water to adhere to the sand particles
by use of wetting agents. However, neither of these approaches will overcome gross
variations in the overall thermal environment.

Use of Native Soils for Backfill. The least ejqpensive material that can be used
as thermal backfill is the original material from the trench. Use of this material
in its native condition can reduce costs in two ways: 1) by reducing the width of
the required trench and thereby reducing the overall excavation and restoration
costs, and 2) by reducing the costs of purchasing special materials and the trans­
portation and disposal of spoils.

In some regions the spoils are readily usable as thermal backfill. However, deter­
mining when these conditions exist, before excavation begins, is a problem. The
determination should be included as part of surveying and exploring the route.

A thermal probe exists for measuring the thermal conductivity of the soil before
excavation. It is most useful in suburban or rural installations where there is
easy access to the actual soil surface. In urban areas the resistivity of the soil
beneath an unpaved area close to the proposed route may not be an accurate guide to
the thermal resistivity along the route. To be valid, the survey should be performed
at the season of the year when the cable load is highest and at least twice to avoid
being misled by events such as recent rainfall.

3-10
The effect of variable drainage patterns and depth relative to the water table can
be so great that the cost of using special "thermal sand" may be difficult to jus­
tify. For this reason, several utilities contacted during this study do not always
use "thermal sand" for backfill, and install local variable-grade sands which have
no particularly good thermal characteristics. But under appropriate conditions a
thermal survey may reduce the uncertainties sufficiently to justify the use of
imported thermal sand in a narrower trench over a portion of the route, thus in­
creasing the cable capacity.

If the thermal survey leads to a 33% reduction in the trench width over 25% of the
circuit length (HPOF—single circuit), the excavation and backfill costs would be
reduced by $5-10,000/mile. This reduction would pay for the costs that might be
incurred by the survey. The savings would increase if the percentages increased.
The savings mentioned do not include savings in surface removal or restoration be­
cause the method would be most appropriate for use in the suburban or rural areas
where the right-of-way may be unpaved. The approach would not be appropriate in
urban areas because of the difficulty in performing a thermal survey and also be­
cause of problems associated with the likely future installation of other utility
lines along the transmission route.

The concept of measuring the thermal properties of the soil before excavation is
not new to the utility industry. Previous efforts to develop a thermal probe, how­
ever, were hampered by maintenance problems. This fact does not invalidate the
concept, it merely illustrates the high level of reliability that the equipment
must have before it would be used by the utility industry. Should the native soil
be unsuitable for use as a thermal backfill, it could be physically modified; pre­
liminary cost estimates indicate that this process would be very expensive.

Liquid Thermal Backfill. The most reliable method for providing the required
thermal contact between grains is to cement sand particles in a solid matrix.
Clearly, this matrix cannot be solid when it is installed; it must be in either a
granular or a liquid form. Two liquid backfills which solidify in place are pre­
sently used in the United States: concrete and a limestone slurry. The concrete
sets to a hard solid, while the limestone slurry sets to a solid which can be
easily broken. While both these materials are more expensive than thermal sand to
purchase, when the cost of tamping the latter into place is included, the total
costs are roughly comparable. Both the concrete and the slurry are relatively easy
to install using the bottom and the walls of the trench as forms.

3-11
Concrete is generally less satisfactory than limestone slurry for the thermal en­
velope of electric transmission lines. The set concrete has a low thermal resis­
tivity of approximately 50°C-cm/watt, which is acceptable as a thermal envelope.
In addition, concrete's ability to provide physical protection in areas where
"dig-ins" are likely, justifies the increased costs compared with thermal sand.
However, the hardness of concrete impedes repair that requires access to the
circuit.

The slurry is a mixture of limestone powder, cement, and water which is made up and
delivered to the site in a cement mixer. It sets in about three hours and provides
an envelope with a thermal resistivity of 40-50°C-cm/watt with 1% moisture content.
Since limestone powder is abundant, the combined materials and installation can be
less expensive than thermal sand. Unlike concrete, the limestone slurry is not
hard after it sets and access to the circuit presents no problem for a repair crew.

Concrete and the limestone slurry represent only two of the possibly satisfactory
solid thermal backfill materials that can be installed as liquids to reduce the
dependence on the moisture content. A material that (1) could be installed as a
liquid, (2) would harden to a solid, (3) has a low thermal resistivity when set,
and (4) is made up of materials that are plentiful and inexpensive in all parts of
the country (e.g., fly-ash is plentiful and chemically similar to concrete) would
provide a reduction in overall installation costs. More importantly, the threat of
"thermal run away" would be significantly reduced. The criterion for determining
cable capacity would be the maximum cable temperature independent of the variable
moisture content of the soil. Even if the use of these materials does not lead to
a lower installation cost, it may provide a reduced cost per unit transmission
capacity by increasing the cable capacity.

Trench Configuration

Calculation of Optimal Trench Configuration. The trench configurations presently

used for the different types of underground cable installations are the result of
detailed analysis by many authors. Prior to 1958, empirical fits to experimental
data were used to determine the configuration of the thermal sand around the cable
necessary to ensure efficient heat transport to the environment. In 1959, Fink
and Smerke performed the first determination of the proper trench configuration
using a calculation based on a set of assumed physical conditions.U2) Most cal­
culations performed since that time have followed this pattern.

3-12
The thermal field considered by Fink and Smerke was a linear heat source at a con­
stant distance, Xq, below an infinite isothermal surface. It is implicit in their
assumptions that the temperature of the soil is everywhere higher than the surface
temperature and only approaches the surface temperature as the distance from the
heat source (cable) becomes infinite. Under these assumptions, they found that
essentially all the heat goes to the surface. The trench configuration was deter­
mined by locating the surfaces which have a constant heat flux crossing them. That
surface which enclosed the highest densities of heat flux from the pipe to the
surface was used to determine the trench configuration. The volume inside the sur­
face is filled with thermal sand to ensure that the high density paths for heat
flow have low resistance to the effective heat sink.

While other authors have used computational techniques quite different from those
used by Fink and Smerke, and have applied them to more complicated situations (e.g.,
multiple sources), the physical assumptions appear to have remained much the same.
For example, in 1974, Winders calculated the effect of discontinuities in the
thermal conductivity of the surrounding soil on the ampacity of three directly
buried cables. He assumed an isothermal plane at the soil surface, but made no
explicit assumption about the soil temperature. However, his calculation implies
that the soil temperature is everywhere higher than the surface temperature and
approaches the surface temperature at infinity. It is possible that this implicit
assumption has led to a reduction of the circuit's ampacity with higher surface
temperature that was greater than necessary under all circumstances.

In 1958, most electric utilities were experiencing peak demand for power during the
winter months when the surface temperature was significantly lower than the ambient
subsurface soil temperature. This deviation from the assumptions of the analysis
deformed the thermal flux lines, so that more of the thermal flux to the surface
passed through the volume filled by the thermal sand than was calculated. The cal­
culated trench configuration therefore was conservative.

In recent years, many utilities have experienced peak demands during the summer
when the surface temperature may be higher than the soil temperature just three
feet below the surface. Since most underground transmission cable is laid below
macadam, an efficient absorber of solar radiation, the effective surface temperature
can reach 40°C in much of the United States. Meanwhile, the subsurface temperature
remains relatively constant at about 10°C, depending on the climatology of the area.
Under these conditions, most of the heat will flow into the soil and will not be appre­
ciably aided by the presence of the thermal sand. The proper trench configuration
should be the one that is suitable for use during the time of peak power demand.

3-13
The effect of the increased surface temperature on the ampacity of the underground
transmission lines has not been ignored and is discussed in the literature. How­
ever, the method of derating the cables with temperature, has been limited by the
absence of detailed calculations of the shape and effect of the subsurface heat
sink.

Using the surface temperature as the only criterion can lead to a higher derating
of the circuits than is necessary. Figure 3.1 shows the isothermal planes around
a source for the condition of a 40°C surface temperature and soil temperature of
40°C at infinity for a single line source at x^. Figure 3.2 shows the isothermal
planes for the condition of a thermal source at x^, a 40°C surface temperature and
a 10°C isothermal plane located at 3x below the surface. Neither of these situa-
o
tions truly represents reality: the 10 C isothermal plane should be a curve simi­
lar to the dotted line in Figure 3.2.

When the positions of the isotherms were calculated for Figures 3.1 and 3.2, the
heat sources were assumed to be just large enough to provide a 60°C isotherm at
the points y = 0, x = (1 ± 1/16) x and y = ± x /16, x = x . The heat source re-
o o o
quired for use under the conditions shown in Figure 3.2 was 50% higher than the
heat source required for use under the conditions shown in Figure 3.1. This implies
that the power capacity of the cable that is calculated using the wrong assumptions
is roughly 20% lower than necessary.

The heat source that could be accommodated for conditions similar to those in
Figure 3.1, but for which the surface temperature is 0°C, is more than three times
higher than when the surface is 40°C. This implies that the calculated cable
capacity would be 170% higher in the winter than in the summer. The allowable heat
source calculated for 0°C isotherm at the surface and 10°C a distance 3x from the
o
surface is only 2% lower than the simpler model (0 C at the surface). Using the
simpler model has (as expected) little effect on the winter rating but a signifi­
cant effect on the summer rating.

Two questions arise when one examines the assumption about the theoretical thermal
environment: 1) Is the present trench configuration providing the best, low thermal
conductivity environment for a cable under conditions that exist during maximum
load; and 2) Is the ampacity of the cable being correctly estimated by use of the
present theoretical model? If the situation shown in Figure 3.2 is closer to
reality during maximum load than the situation shown in Figure 3.1, then the answer
to the first question appears to be "No" (dotted line indicates present trench

3-14
 40° C

Figure 3.1 Isotherms For A Line Source Below An Isothermal Plane.

 I
3-16

es
Figure 3.2 Isotherms For A Line Source Between Two Isothermal Plan
configurations). However, obtaining even a partial answer to the second question
is considerably harder because of the variations in the conditions along the cir­
cuit route.

Both the experimental determination of the proper operating capacity of existing

cables or the ability to improve the ampacity of short sections of existing cable,
and thereby the whole circuit, would potentially reduce the transmission systems
costs. A 20% increase in capacity for existing cables would postpone the need for
new transmission systems for roughly three years (7%/year demand growth). The total
savings from such a postponement would be roughly 40% of the capital cost of a new
system (fixed charges per year = .15 x capital costs). If the demand growth per
year were to become 4%, the new construction would be postponed about 5 years with
a total savings of roughly 65% of the capital cost of the new circuit. Obviously,
increasing existing cable ratings will be extremely beneficial.

Trench Configurations for Multiple Circuits. Multiple circuits are usually installed
in underground systems because of the increased overall system reliability and be­
cause of the reduced costs relative to two single circuits each with the same ampa­
city. However, the presence of the second circuit distorts the thermal environ­
ment of the first and reduces its ampacity.

Since it is presently assumed that all the heat generated in the underground cir­
cuits will go to the surface, all multiple circuits are installed in a side-by-side
configuration to give the heat from each circuit equal access to the heat sink.
Placing two circuits two feet apart in the same trench leads to a derating of each
cable of roughly 15%. This derating increases circuit costs by demanding that each
circuit have a larger conductor size than required for its normal ampacity; the
reduced installation costs compensate for the increased cable cost. The optimum
trench configuration has been calculated in the same manner as the optimum configu­
ration for the single circuit and has the same initial assumptions.

In urban installations, there are many intersections where it is difficult and ex­
pensive to dig a trench wide enough to accommodate a double circuit. Sometimes
the problems are great enough to force an expensive rerouting through less crowded
intersections. Under these conditions, it may be profitable to consider installing
the circuits one about the other in a narrower trench, as is sometimes done with
distribution cable. While the vertical configuration may cause interference with
utility lines that otherwise would have been unaffected, the concept is worth ex­
ploring because it may result in more route flexibility and thus lower installation
costs.

3-17
Before seriously considering the construction problems involved in placing circuits
one above the other, it is necessary to consider changes in the thermal environment
that result from the change in configuration. When it is assumed that the primary
heat sink at peak demand is the surface, the vertical configuration will cause an
increase of the upper circuit rating and a decrease in the lower circuit rating
relative to their ratings in the side-by-side configuration. The opposite is true
when the primary heat sink is the soil. Whether the overall change under specific
peak demand conditions is higher or lower than for the side-by-side configuration
can only be answered after a detailed thermal analysis. Preliminary calculations
of the magnitude of the two allowable heat sources side-by-side, using the surface
as the heat sink, confirms the 15% derating of each circuit relative to the single
circuit configuration. The difference between the allowable heat sources of the
horizontal configuration and the allowable heat sources for the vertical configu­
ration is only 5%; or a 2% difference in power rating.

An empirical, graphical method, the flow net, for solving Laplace's Equation (e.g.,
heat flow equation) for extremely complicated situations is now widely used by
soils engineers to calculate the seepage rate of water through structures made up
of soils of different porosities. The technique is capable of yielding equipoten-
tials (e.g., isotherms) accurate to within 5%, and flow rates (e.g., heat flow)
accurate to approximately 10%. Figures 3.3 and 3.4 are thermal flow net constructed
for two circuits in the horizontal and vertical orientations, with the soil surface
at 40°C and a subsoil surface at 10OC. These figures show that the overall flow
rate (i.e., space between flow paths) from the vertical configuration is approxi­
mately the same or greater than from the horizontal configuration under these thermal
conditions for the same magnitude heat source. While we do not offer these dia­
grams or the preliminary analysis as final proof of the thermal acceptability of
the vertical configuration, we believe that further investigation would be fruitful.

The problems that may be encountered in installing a double circuit in the vertical
orientation are varied and, depending on the circumstances prevailing at the con­
struction site, they may be significant. If the upper pipe is to maintain its
position at roughly five feet below the surface, the lower pipe would be seven feet
below the surface. This site would then require shoring in a situation where shields
are essentially unusable and the cost of shoring could cancel out other savings.
However, if the site already requires shoring, the increased shoring costs could be
less. Another possibility is to raise the upper circuit to a three-foot depth and
the lower circuit to a five-foot depth and possibly have no shoring. This may not

3-18
-19

Figure 3.3 Thermal Field For Two Thermal Source-Horizontal Confiruration.

 ‘10°C isotherm
-20

Figure 3.4 Thermal Field For Two Thermal Sources-Vertical Configuration.

be physically possible in a crowded intersection; the thermal considerations do not
pose a significant problem in many localities and some utilities already install
some of their circuits at the three-foot depth.

The actual installation steps for a vertical orientation will be complicated and
lengthy. When a granular thermal backfill that requires layered compaction is
used, the only practical approach is to install the lower circuit first, compact
the thermal sand and then install the second circuit. All jointing and bending
would be done as with the single circuit.

If the thermal backfill being used is installed in a liquid form (e.g., concrete
or a limestone slurry) requiring no tamping, other methods of installation could
be considered. In very complicated intersections, it is often necessary to joint
the duct or pipe sections together outside the trench and winch them into position.
In the side-by-side configuration, each pipe is winched separately. For the ver­
tical configuration, the pipes could be banded together with spacers and the back­
fill could be installed all at once.

One problem with the vertical configuration would be difficult access to the lower
pipe by the repair crew. However, digging past the top pipe and gaining access to
the lower pipe from a horizontal position should not be too difficult.

While the vertical configuration might be most useful in crowded intersections,

when liquid thermal backfill is used it could also reduce the costs of installation
under less difficult conditions. A five-foot x five-foot x one mile trench has a
volume of soil requiring excavation of 4,900 cubic yards under 2,900 square yards
of ground-level surface; a seven-foot x three-foot x one mile trench has a volume
of soil requiring excavation of 4,100 cubic yards under 1,800 square yards of
ground-level surface. This orientation, therefore, reduces essentially all civil
engineering costs except shoring. If shoring is already required because of water
or soil conditions, the increase would be only 40% of existing shoring costs
($64,000/mile), approximately twice the decrease in the surfacing costs. For the
reduced trench depths, the three-foot x five-foot x one mile trench would have
a volume of 2,900 cubic yards and a surface of 2,900 square yards. The volume of
the five-foot x three-foot x one mile trench would be the same, but the surface
would be 1,800 square yards. The surface removal and resurfacing costs would be
reduced by over $30,000/mile. If cables installed in this manner interfere with
other lines, e.g., gravity feed sewer lines, conventional installation techniques
should be used.

3-21
CONSTRUCTION ACTIVITIES

Surface Removal

Pavement breaking may be a substantial fraction of the excavation costs for a cable
installation. This would particularly be the case if reinforced concrete paving
had to be broken up; asphalt paving is not a problem in comparison. Conventional
pavement breaking techniques encompass a wide range of equipment, but noise and/or
shock is still a major problem. Governmental noise regulations, from maximum city
levels to federal standards, may also limit the choice of high production breaking
techniques. Following a brief assessment of current procedure, we will discuss
advanced pavement (and rock) fracturing techniques and submit suggestions for future
research areas. It should be emphasized that we are discussing initial rough
breaking, and not the more precise pavement trimming of resurfacing.

Conventional Pavement Breaking. It appears that asphalt paving can be removed

relatively quickly and easily by simple mechanical techniques. A typical procedure
is to slit the paving using a cutting wheel or knife edge mounted on a backhoe (or
a similar piece of equipment), expose an edge of the paving, and use the shovel to
both create chunks of convenient size and load them into a truck. Consequently,
we see no need for research programs on asphalt removal.

The removal of concrete is a major problem. Conventional methods employ hand-held

pneumatic paving breakers, hydraulic or pneumatic powered machine-mounted breakers,
rotary saws, crane and wrecking ball combinations, and blasting (where applicable).
Hand-held breakers are perhaps the slowest implements, but they can operate with
fair precision. The primary disadvantages of the large devices are noise and
potential shock (vibration) damage adjacent to the break-up area. The smaller
pieces of equipment (jackhammers and saws) are both noisy and labor-intensive.

Mufflers are available to reduce sound levels, but they reduce production rates.
Machine-mounted breakers offer a significant advantage in that much faster breaking
rates may be achieved. Manufacturers claim that one unit can do the work of 10 or
more hand-held breakers. The noise level of a hydraulic unit is less than the
noise level for the equivalent pneumatic unit. Mufflers are available for all
pneumatic breakers.

One potential disadvantage of machine-mounted breakers is that the high impact

and vibration levels can adversely affect adjacent utility lines, plumbing,

3-22
foundations and structures located near the equipment. The same disadvantage may
apply to wrecking balls; this technique is also at a disadvantage when physical
obstructions hamper the positioning or use of a crane. Blasting has always been
highly restricted, and its disadvantages are well known.

Concrete saws, cooled with a water spray, are occasionally enployed under certain
conditions to slit paving and thus permit easier breakup of slabs. For example,
if the concrete is relatively thin and utility lines closely underlie the surface,
the saw can cut to a relatively precise depth. This would allow slabs to be cut
out and removed (by other equipment) with minimal chance of damage to the utility
lines.

Cost figures on a unit basis (per square yard) depend upon the type and depth of
paving, the equipment employed, and local labor costs. For trench widths of
3' to 5' and pavement (asphalt with concrete base) depths of 9" to 12" a typical
figure will be between $6 and $11 per square yard.

To summarize these methods, it appears that machine-mounted breakers can achieve

the highest production rates for breaking up concrete, and that these rates are
significantly higher than those achieved with 10 to 80 men using hand-held breakers
(manufacturers' estimates). The major drawback is potential damage to the machine
applying the tool, adjacent utility lines and adjacent structures; shock and
vibrations are the basis of this potential damage.

The Plasma Torch, The plasma torch is intended to cut a slit through material;
subsequent thermal cracking produces chunks and rubble which may be handled by
conventional means. It has the additional virtue that steel reinforcing rods
embedded in the concrete pose no special problem, since they are melted along with
the concrete. The Institute of Gas Technology investigated the technique and ex­
perienced problems with both electrode life and noise levels. The experimental
model would cut a 4" deep slit in concrete at the expense of slow travel; around
(7)
1/2" per minute.

At present, a commercial company has developed a prototype unit that will produce
a cut 8" deep and 3" wide in granite or basalt at the rate of 4" per minute.^
Its input power is around 300 kW. Higher power models have been planned for
tunneling applications, which would consume around 600 kW.

3-23
Several areas relating to plasma torch efficiencyf material lifetime and noise
shielding will require additional R&D effort before one can make realistic com­
parisons between this system and its closest current alternatives. It appears
that this device would be most effective in breaking up reinforced concrete paving
and foundations when high shock levels would be potentially hazardous.

Concentrated Microwave Heating. This unusual approach is based on early (pre-1962)

work done in Russia and England on cracking rock via high energy microwaves.
Follow-up experimental work by IGT using two 5-kW units resulted in pavement break-
C9)
ing rates of 0.3 to 1.0 ft/min. in concrete 9 to 5 inches thick, respectively.
The test frequency was 2450 MHz. Although microwave heating has succeeded in
cracking concrete pavement, the cracks are not satisfactory from a surface removal
viewpoint. This is because the cracks are not "linear," but rather crooked in
nature; consequently, an interlocking type of crack network develops, which hampers
straightforward removal of the material. It appears that microwave heating may be
applied more successfully in conjunction with high pressure water jets, which will
be discussed shortly. Conventional microwave tubes and systems appear to be
satisfactory for developing a microwave heating system of sufficient RF power.

A doubling of power such that two 10 kW units (operating at a frequency of 970 MHz)
might be applied in field use has been proposed. The lower frequency should result
in deeper penetration. Field tests at Consolidated Edison in New York City showed
that the experimental set-up was very quiet, produced no harmful level of microwave
(9)
radiation, and did not produce local television or radio interference. It is
anticipated that additional shielding and system design will permit the same satis­
factory operation at the higher power level.

To summarize, microwave heating has succeeded in cracking pavement quietly albeit

slowly; the form of the crack is a drawback, however, in that the interlocking
edges hamper rapid removal of material. The cracks are nevertheless capable of
being utilitized rather well by water cannons, the next topic for analysis.

Ultra High Pressure Water Jets. Although water cannons have been developed pri­
marily for piercing ability based on ultra-high impact pressures, their use in a
slightly different application seems quite efficient. If one uses the water jet
as a wedge to split pre-cracked concrete, most of the impact on the concrete will
be in the form of tension. Since the tensile strength of concrete is much less than
its compressive strength (approximately one-tenth), the procedure exploits the
prime weakness of concrete.

3-24
Several types of water cannons are being developed in at least three or­
ganizations for the purpose of unassisted rock and concrete breakup. At present,
only very basic conceptual sketches of the devices appear in the literature, and
dimensions are not specified. Visual comparison with typical backhoes and forklift
trucks seems to indicate that the size of the water jet units is roughly comparable
to present pneumatic and hydraulic breakers.

Two regimes of operating pressure are being considered, resulting from two differ­
ent pulse-generation techniques. The first is commonly known as the impact-extru­
sion method; it can produce a maximum jet impact pressure of approximately 100,000
psi, and is capable of operating several times per second. The second technique
relies upon annular cumulation of pressure via shock compression. This method is
capable of creating jet pressures of greater than 1,000,000 psi; its major draw­
back is a low recycling rate. One firm is attempting to construct a hybrid device
which will both develop pressures in the cumulation regime and maintain a favorable
recycling rate. The company estimates that its prototype for utilities (now nearing
completion) will be able to crack concrete 6" thick at the rate of 10 feet per
minute.Other firms have slated tests for their models in the immediate
future. Considering the current state of development, it is not possible to make
an economic comparison with conventional pavement breaking systems. The water jet
systems claim quietness as a prime virtue, but they must be environmentally satis­
factory and relatively close to conventional operating costs (or lower) for ready
acceptance by construction companies or utilities. Since the systems use very
minute "charges" of water for each pulse (vO.006 cubic feet) it is not anticipated
that muck or pools of water will be a serious problem in trenches.

Non-Impact Breaker. The particular device we discuss here is called the "Nibbler."
It employs twin hydraulic cylinders in a J-shaped forging and operates by putting
the edges of pavement slabs under high localized tension. The slabs fracture into
conveniently-sized pieces without the continual noise and vibration common to im­
pact devices. Field tests in Britain, where the device was recently developed,
indicate that mesh-reinforced concrete slabs up to 10" thick can be broken at the
rate of 570 square feet per hour. Assuming a trench width of 4', this area rate
transforms to an equivalent lineal rate of 2.4 feet of trench length per minute. ^

One particular drawback of the Nibbler is that saw cuts are required at the edges
if a precise width of paving is to be removed. This is due to the random direction
of fractures which emanate from the fulcrum point. If saw cuts are made, however,
it is likely pieces the width of the trench will be broken off in each "bite";

3-25
this would permit faster breakup than would be the case without saw cuts at the
edge. If mounted on the end of a hydraulic boom, this device can also grasp the
broken slab segment and load it into a nearby truck. Thus in addition to the ad­
vantages of minimal audible noise and the absence of impact generated vibration,
the device doubles as a loader of the rubble.

Summary

All three of the pavement-breaking systems discussed previously have advantages

over current high production rate systems in certain applications where noise levels
are quite restrictive and the potential for shock damage exists. The plasma torch
system has the greatest theoretical applicability in that it cannot only fracture
concrete paving, but also be employed to break up reinforced concrete structures
without the need for an oxyacetylene torch to cut the bars. Problem areas requir­
ing additional research include: (1) reducing the noise level during operation,
(2) developing improved torch tip materials and design for greater longevity, and
(3) improving the overall efficiency of energy transfer to minimize the need for
large electric power supplies.

It appears that more development than research is needed to create a high-power

microwave system suitable for practical pavement cracking. The previous recom­
mendations of lowering frequency and raising the power level do not depend on
future developments in microwave systems. Despite the interlocking nature of the
fractures, this system is extremely quiet compared to any conventional technique
and it has zero potential for shock damage to nearby utility lines.

The water cannons just described are currently undergoing testing and development
programs. They offer significant advantages in terms of decreased noise level
during operation and reduced potential for shock damage. When used in a cutting
mode, as opposed to fracture expansion, they may be able to provide the smooth
edge needed for the permanent pavement patch. Erosion of certain key design
components is one problem area at present, as is the difficulty in upgrading
prototypical designs to reliable, commercially acceptable units. The proprietary
nature of current development precludes establishing more specific R&D guidelines
at this time.

Excavation Techniques

Excavation cost is a major portion of the total cost of any underground transmis­
sion line, as shown in Section 2. We will how review the excavation methods

3-26
presently available and present some way to reduce costs, such as increased digging
speeds for common techniques and modifications of potentially less expensive
techniques.

In this section we discuss each of the following methods:

• Backhoe excavation,
• Trenching machines <
• Plowing,
• Horizontal boring and augering,
• Soil removal around obstacles by water, and
• Shoring-

Backhoe Excavation. Open-cut trenching with the familiar backhoe is the simplest
and most commonly used method of installing underground cable systems. The width
of trench that can be excavated in one pass varies with the size of the bucket
(up to 6 feet in width). Maximum trenching rates available with such machines are
around 75 feet/day in cities and about 200 feet/day in suburban areas.

In urban, suburban and downtown areas, where most underground transmission lines
are located, machine digging is limited to relatively small equipment which allows
a high degree of flexibility for digging around obstacles. When medium-size rocks
or boulders are encountered, they are exposed and either lifted out or broken up
and then excavated. Backhoes are also available with rotating heads that can dig
a hole underneath the main vehicle in relatively soft soil. These provide the
ability to dig short tunnels underneath roads or other obstacles or change the
grade of a trench once it is excavated.

Although the backhoe is a great improvement over the previous generation of excava­
tors (e.g., the power shovel), it has built-in inefficiencies: about 75% of its
time is spent lifting soil out of the trench onto a truck; only 25% of the time is
spent actually digging. Equipment manufacturers are improving the control systems
to operate in automatic modes that might reduce the time spent lifting. But the
largest portion of the inefficiency is inherent in the need to lift the bucket out
of the trench. A separate machine for delivering the removed soil from the trench
bottom (e.g., a conveyer belt) would improve the efficiency and speed of the backhoe
without sacrificing its flexibility.

3-27
Trenching Machines. Excavating an open trench with semi-automated continuously
operating equipment is called "trenching." There are at least two distinct types
of trenchers; the wheel type, and the more conventional chain (or boom) type.
These machines all operate on basically the same principle: a series of buckets
or similar devices, mounted on a moving closed-loop chain or wheel, penetrates and
removes the earth continuously.

The wheel type of trencher has buckets, with integral teeth, mounted around the
circumference of a wheel that is mounted on a crawler tractor for power and control.
The equipment usually has bucket cleaners and power conveyers for collecting and
moving the dirt out of the excavation. The frost-cutter modification, with the
buckets replaced by rugged bucket-shaped teeth is used for hard and firm types of
earth, ice, frozen soil, and clay.

The more conventional chain machine is a very specialized piece of equipment with
high efficiency but little versatility. Consequently, a trenching machine can ex­
cavate an unobstructed trench of a given width and depth at less cost than any
other method. Today's machines are capable of digging trenches from 8 inches to
6 feet wide, and to 25 feet depths. Trenching rates with chain-type machines de­
pend on soil conditions, but speeds of advance from a few feet per minute to more
than 40 feet per minute are possible. However, present trenching machines cannot
cut through rock, large boulders, pavements, or similar obstacles; most trenchers
require level ground with a clearance of several feet on both sides, and sharp
bends take extra time. Whenever other digging machinery must also be used, the
economies of trenching are degraded. Therefore, trenchers are used only after
careful survey has shown very few natural or man-made obstacles. Thus, they are
used infrequently in cities, and their limitations can also cause expensive detours
for a rural installation.

Modifications to trenching machines would allow them to be used more economically

under a greater variety of conditions. One company has proposed designs that would
use proven mining techniques, using pressures greater than 10K psi at the tool
cutting edges. Because 10K psi is higher than crushing resistance of concrete or
soft rock, such a new machine could be used for surface removal, after initially
sawing the concrete.

Trencher turning limitations occur because of the size of the digging element.
Modifications to reduce this length would reduce the problem and make the machine

3-28
more flexible. However, the only way to reduce the digging length of a wheel
trencher is to reduce the diameter of the wheel. An entirely different type of
support system and power train would be required to allow excavations to present
depths.

Chain trenchers boom lengths could be modified to reduce the minimum turning radius
and thus increase the machine flexibility. A boom is usually about 30° from the
vertical; varying the angle varies the trench depth. The length of the boom along
the trench is the actual length times sin 30°. If the boom were vertical (0°), then
the effective length would be much smaller, and it would be easier to make turns.
Changing the boom angle to the vertical would also make it possible to dig closer
to obstructions and therefore make the rencher more useful for suburban installa­
tion projects. Clearly, if a boom is so modified that it excavates vertically, the
total machine design must include a new method of controlling the excavation depth.
One proposed system is made up of two separate elements: a vertical digging ele­
ment and a following soil remover (mucker). Such a design calls for the mucker to
work at an angle. If this design could include the ability to rotate the elements
to compensate for a surface grade trencher, flexibility would be increased.

Many modifications to trenching machines are possible to make the equipment more
flexible and useful while retaining the present speed and efficiency. The modifi­
cations discussed, however, would still not make it possible to use trenchers in
urban installations where existing utility lines are the primary obstructions. The
use of any machine that completely cuts the trench in one pass could create a
problem. If an obstruction follows the trench for any significant distance, avoid­
ing it would leave a great deal of material to be removed by some other means, thus
reducing savings.

The main advantage of a trenching machine is its speed, for a single unit can dig
up 1000 feet/day; but most cities have a limit on the length of trench that can be
open at any one time. If this permissible length is less than the length the
trencher can dig in a day, then either all of the other installation steps must be
completed in less than one day or else the trencher must sit idle part of the time.
If the allowable trench length is twice the machine's capacity, then all other
steps must be completed in less than two days. Depending on the type of system
being installed, such schedule interdependence would severely restrict the usable
productivity of the trencher. This is less of a problem in rural areas.

3-29
Figure 3.5 Soil Removal System^-Conceptual Design

3-30
Plowing. Plowing machines are designed to open the earth, plant the cable, and
cover it with one piece of equipment. The design of the cable plow is intended to
disturb the surrounding earth as little as possible. The feed mechanism for intro­
ducing the cable immediately behind the plowshare is integral with the plow. Thus,
plows are unsuitable for use with pipe-type or other rigid cable systems, or with
any system requiring field assembly operations or special backfill materials.
Obstructions, man-made or natural, must be avoided or passed over. The extra cable
joints required to go beneath an obstruction are not acceptable with transmission
lines, but may be acceptable for distribution cables.

At present, there are no plows suitable for installing cables of the size ana
rating considered in this report. In particular, no plows are capable of providing
the proper thermal environment. However, plowing has turned out to be very econom­
ical for distribution cable, telephone cable, and gas lines. This economy results
from specialization and from the integration of the three operations of digging,
placing the cable, and burying the cable. Consequently, an integrated cable system
could reduce the cost of any rural underground electric power transmission that
might be required. (This question will be addressed in greater detail in the
section on rural installations.)

Horizontal Boring and Augering. Horizontal boring and augering are processes of
mechanical removal of soil and rocks in a horizontal direction without a continuous
surface cut. Once the horizontal hole is in place and stabilized to prevent cave
ins, an electric transmission cable can be installed. There are several benefits
that can accrue from this method of excavation but there are also some severe pro­
blems. This section will describe the principal issues and indicate their relative
importance.

All presently available boring techniques are considerably more expensive than open-
cut trenches and are used only when an obstacle, e.g., an airport runway, railroad
switchyard, river, etc., is too large to avoid economically. The boring machine
must be lowered into a position in line with the hole to be bored, after which the
hole is driven beneath all obstructions. The greatest cost component is associated
with the initial setup and alignment of the equipment and the incremental cost of
increasing the length or diameter of the hole beyond nominal values (about 200 feet
or 1 foot in diameter) is not proportional to the additional material removed. The
longer the borehole, the lower the average cost per foot of excavation. However,
the longer the installation, the greater the problem of accurate hole placement
and the greater the problem of using boring techniques in urban excavations.

3-31
The accuracy with which a horizontal borehole (not a tunnel) can be placed using
presently available, unguided equipment is between 1 and 2%. This means that a
1000 foot long borehole will have an end point located within 10-20 feet of the
desired end point; an unacceptable error in the urban environment. There are four
sources of this error: initial alignment, variations in soil hardness along the
route, torque on the drill string and gravitation. The initial alignment of a
borer is difficult, time consuming, and subject to error. The second source of
error occurs because all present boring methods involve mechanical removal of
soils; each change in the hardness of the soil causes the cutting edge to be de­
flected slightly. The torque applied by the drill string tends to cause horizontal
deflection and gravity tends to cause vertical deflection. None of these sources
of error can be eliminated, but techniques to minimize or compensate for their
effect can be considered. A guided tunneler, as will be discussed, could reduce
the errors significantly.

Once a horizontal borehole is excavated and the cable installed, the extra space
must either be backfilled with thermal sand or some other method of cooling the
cables must be provided. Like the initial excavation, backfilling the hole is
very expensive. At the present time, a common method of backfilling with thermal
sand at the proper level of compaction is to blow the sand into place down a plas­
tic tube. Interviews with personnel who have supervised this type of operation
indicate that it is difficult, time consuming and consequently very costly. A
liquid or slurry backfill material which could be poured into the hole and later
harden in place would reduce the cost of backfill for naturally cooled cables.
Installing a solid backfill during the boring process would prevent the retraction
of the excavator in the event of equipment failure and should not be considered.

Installing thermal backfill is not the only method of facilitating the dissipation
of the waste heat; other methods can be and are used. One alternative cooling
technique is the once-through circulation of water around or near the cable for
the purpose of removing the heat. The forced cooling of cables by refrigerating
and circulating the oil dielectric in the cable has been treated in depth in the
literature and represents a second cooling alternative. A third alternative is the
reduction of the cable ampacity (conversely, the increase of the conductor cross-
section with no increase in the cable ampacity) so that the required thermal dis­
sipation can be provided by the air surrounding the cable. Forced cooling of the
oil replaces the initial high cost of backfilling with increased operating costs
(energy costs for oil cooling); derating the cable increases the cable costs.
Water cooling and cooling the oil and the previously mentioned thermal slurry

3-32
appear to promise significant decrease in the cost of providing the appropriate
thermal environment for cables installed in long horizontal boreholes compared to
the present cost of installing thermal backfill. But these techniques should also
be considered for use in open cut trenches.

Unguided boring using present techniques appears to be economical only when it

significantly reduces the overall circuit length and when its poor accuracy is
acceptable; but even this cost reduction can prove to be an illusion. Once the
cable is installed in a horizontal borehole, the nature of the obstacle under which
the cable is installed could provide just as large an obstruction to cable repair
as it did to installation. Airport managers are unlikely to approve the excavation
of a runway for either installation or repair. A river would be a significant
obstacle for repair crews. The obstacles can impede repair to such an extent that
sometimes the only feasible repair method is the removal, repair and re-installation
of the circuit. The cost of this procedure could erase any of the cost benefits
that might accrue from the reduction in the circuit length; this is true irrespec­
tive of the method of boring.

The two problems of horizontal boring are the high cost and low accuracy with which
the hole can be placed is the more important to the electric utilities. The vast
majority of underground transmission cables are presently installed in the urban
and suburban environments and this situation is unlikely to change significantly
during the next decade. Even if the cost of new boring systems were to be less
than the cost of open-cut trenching, the poor accuracy would make the equipment
unusable in the most probable installation environment. For this reason, and others
ERDA and EPRI sponsored a study by Ingersoll Rand of the accuracy that might be
achieved by the successful development of a guided tunneler.(14)

The two guidance techniques that show the most promise of providing equipment capa­
ble of use in the urban environment are acoustic and inertial. Of these two, the
acoustic technique shows the greatest promise; the acoustic guidance equipment will
probably be more rugged and less costly than an equally acceptable, high precision
inertial guidance system. However, it is still too early to state with certainty
that the acoustic systems will be satisfactory and both approaches should be con­
sidered further.

An inertial guidance system used with a horizontal boring machine would consist of
two gyroscopes to measure the vertical and horizontal orientation of the boring
equipment as a function of time and a wire reel to measure the total tunnel length

3-33
as a function of time. This information can be manipulated to yield the position
of the boring equipment at any time and then guide excavations. A 1000 foot long
borehole should have an end point located within five feet of the desired end
point.

The acoustic guidance technique described in the ERDA/EPRI sponsored report is

made up of a sonic generator mounted on the tunneler and four sonic detectors
placed on the surface at known positions near the desired tunnel path. The time
that it takes the acoustic signal to travel from the tunneler to each of the detec­
tors can yield information on the location of the tunneler relative to its desired
position and then the operator can guide future excavations. The average local
speed of sound in the area could also be acquired as part of the process.

This approach to sonic guidance systems solves many of the problems with sonar
systems described in the section on Route Surveying but some problems remain. The
tunneler is not a passive reflector of sound energy but a significant source, and
the problem of sorting out multiple reflections from multiple objects in a medium
in which the speed of sound is unknown does not exist. But, the method assumes
that the average speed of sound along each of the four paths is a constant; an
assumption that is not always true. The problem of coupling the sound energy into
the soil will not be as severe with the generator located underground as it was
with the generator located at the surface, but the problem of coupling the sound in
the soil to the receivers at the surface will be the same as previously described.
In this case, however, it will be economically feasible to break the road surface
in order to gain direct access to the soil; the holes will be small and few in
number and the cost of excavation and restoration should be relatively small. This
system could verify the location of obstacles along the route and guide the tunneler
simultaneously. However, obstacle location is more difficult than guidance so the
development of the system for guidance purposes should take priority.

The number of different methods that could be used to actually drill the horizontal
(14)
hole are quite large, and are described in detail in the ERDA/EPRI report.
One unique device for boring the hole is the "Subterrene," presently being investi­
gated at the Energy Research & Development Administration's Los Alamos Scientific
Laboratory. Essentially, the device melts its way through rock and soil with a
molybdenum-coated tip heated to 1800°C. The waste material can be either pressed
into a tunnel liner of hard, obsidian-like glass, or it can be extruded through the
center of the bit back to the machine, or some combination of both of these methods
can be used. This bit is unlikely to be deflected by changes in any of the physical

3-34
properties of the materials but requires very accurate initial alignment. The use­
fulness of this method depends on the thermal properties of the wall material. How­
ever, it may be used to provide a duct for directly buried, water-cooled cables.

The largest hole yet drilled by the Subterrene is about 4 inches in diameter, bored
at the rate of 5 feet/hour. The proponents of the device feel that it has considera­
bly greater potential for development. However, in its present form, it could pro­
vide greater accuracy for existing straight line borers. If the Subterrene, using
a guidance system from the open end, could produce a small accurate hole, a laser
guidance system could be mounted at the exit end to guide a mechanical boring
machine to enlarge the hole. Another possibility is to use the primary hole for
the drill string and then pull the boring head through.

Once the borehole is in place and lined the cable must be installed, but deciding
which cable type should be used under these conditions is not clear. It may be
that directly buried cables are more suited for this type of installation than HPOF
Pipe Type Cables.

Installing a somastic-coated pipe for an HPOF Pipe Type Cable in a 2000 foot long
Horizontal hole, without destroying the coating is a difficult task. It will cer­
tainly be necessary to support the pipe in the borehole to protect it during the
pull and this support may be enough to center the pipe in the hole before any
thermal backfill is installed. If the hole liner can be used as the pipe this pro­
blem would not exist, but then the only way to modify the natural thermal environ­
ment would be the forced cooling of the oil in the pipe.

The Pirelli Cable Company has developed a technique for installing directly-buried
cables in trenches. The technique may be easily modified for installing cable in
a long horizontal hole with rough walls. A wire rope is hung from supports installed
along the trench; the electrical cable is suspended from this rope using special
hangers and pulled along the route, following the trench without touching the ground.
Once the length of cable is deployed, the hangers are removed and the cable is laid
in the trench in the appropriate position. If this method could be modified for use
in the horizontal borehole, the primary problem would be in the installation of the
wire rope or ropes and an appropriate number of supports or spacers. This equip­
ment would be lighter, more flexible, and more rugged than the pipe or cable, and
so this task should be relatively easy to accomplish. Unlike the present method
used in open trenches, the hangers could not be removed after installation. The
acceptability of leaving the metal hangers in place would be determined by the

3-35
effect that this would have on the electrical properties of the cable. There should
be no electrical problem if the HPOF pipe is hung in this manner, so long as the
somastic coating is left intact. The probable effect on directly-buried cables is
unclear.

The total installation costs estimated by Ingersoll-Rand in the ERDA/EPRI report

using some of the new boring techniques vary between $211 thousand and $343 thousand
per mile for a 2000 foot borehole 2 feet in diameter. These costs include the costs
of placing the hole, removing the spoils, and lining the hole. The total cost for
the equivalent operations for open-cut trenching and the tasks related to open-cut
trenching that would not be performed if the hole were bored, varies between $164
thousand and $250 thousand per mile, excluding the cost of shoring. The costs asso­
ciated with horizontal boring (2000 foot hole) and open-cut trenching would be com­
parable if 570 feet (28%) of the trench required shoring or if the borehole were
smaller. If these costs can be realized in practice, they will be low enough to
allow improved boring techniques to be considered as an alternative to open-cut
trenching.

Soil Removal by Water. Presently, whenever obstacles are encountered during ex­
cavation, the machine digs to within a safe distance of the obstruction and the
remaining solid is removed by a crew with shovels at roughly $35/cy. While the
removal of rock requires the use of other, more expensive excavation techniques,
the most serious problem is from existing utility lines which must not be damaged.
These are embedded in relatively light, backfilled soil. Use of a water jet or an
air/water jet to remove the soil would alleviate some of the dependence of costly
manual labor.

Water jets and water/air jets are well known in various parts of the construction
industry. They are presently used for cleaning and preparing rock and concrete
surfaces at new dam sites. Also, the water cannon is used for water boring and
has been considered for surface removal. In general, the water pressures that can
be used for soil removal will be limited by the nature of the obstruction; the
higher the pressure, the higher the digging speed. To have the optimum digging
speed under all conditions, a new device or a new application of available devices
should have a variable pressure control.

The principal advantage of a water jet system is to reduce the cost of soil removal.
If the cost of water management under specific site conditions should become too

3-36
high, the technique should not be used. Situations where the problems of water
management already exist, and are being dealt with represent the most promising
area for initial water-jet applications.

For an air and water system at a maximum pressure of 200 psi, the soil removal rate
will vary between 4 cubic yards/hour for loose soils to 1 cy/hour for soils that
have been compacted long enough that some cementation has occurred. The combination
air and water jet increases the turbulance of the stream and thereby the soil re­
moval rate; the combination also reduced the amount of water needed. The cost for
removing the soil will vary between $5/cy and $22/cy, including machine rental and
an operator, costs that are very similar to machine excavation rates. The amount
of water used is roughly 10-20 gallons/minute, which should pose no disposal problem
in city or suburban streets using normal storm drainage system, although filtration
of the particulants will be required by the cities.

While a water jet at 200 psi may damage some underground obstacles, other obstacles
may not be affected at all. No single water jet system will be ideal (digging
speed vs. protection of obstacles) under all conditions. The question of when to
use the water jet will require additional study.

A water system could possibly undercut trench walls. In an urban or downtown site,
water splash might annoy nearby pedestrians and property. Temporary, lightweight
water shields could be used both in the trench and on the pavement next to the
trench at minimal cost.

Shoring. Generally, there are two reasons for shoring trench walls: (1) to protect
the workers in the trench, and (2) preserve the load bearing capability of the
surrounding soil. Under OSHA, regulation shoring is required for the entire trench
wall whenever trench depths exceed five feet. Under the worst of conditions, whole
sections of trench must be shored and it is impossible to dig to the required depth
without cave-ins, this occurs when the soil is very unstable or is below the water
table. A system that reduces the costs of shoring under difficult conditions, while
keeping the costs for simpler conditions the same or less would be ideal.

Under moderate shoring conditions (over 5 feet trench depth or where there is a
small but significant chance of cave-in), trench shields which satisfy OSHA require­
ments seems to represent the most economical approach but they introduce other
problems. The shields are either braced steel plates that can be dropped into the
trench, with the braces snapped into place or prefabricated openend boxes. When

3-37
the workis completed, the shields are removed and moved forward and the trench is
backfilled and compacted.

Trench shields, as presently configured, cannot be used under most conditions. They
cannot be used when there are obstructions of any kind in the trench; and since the
shields are relatively long, their use is ruled out in many urban installations.
However, they can be used effectively where obstructions are widely spaced.

One of the biggest problems with trench shields is cross bracing. Once in position,
these braces are just as much a hindrance to futher construction as any other ob­
struction. For HPOF pipe-type cable or ducted extruded dielectric cable, this
problem can be solved by assembling the pipe or duct into relatively long lengths
and pulling it into place over or under the bottom braces with a winch. The bottom
braces have to be either separable from the shield or half way up the shields to
allow retrieval of the shield. Thus, trench shields are most useful where there
are no obstructions, but their use requires fabrication techniques that are used
where there are many obstructions. Drop-in shoring, like trench shields, which
avoids the need for one brace of box end, would be very useful.

One possible design that could be considered for new trench shields consists of two
shielding plates bolted to a U-shaped assembly in three sections. The two plates
which actually hold back the soil would be bolted to the upright sections. The
bottom section could be designed to come in various lengths for different trench
widths. After the cable pipe has been laid, the various pieces of the shield sys­
tem would be disconnected and removed for reuse. The length of the section should
allow easy installation without a special crane. A simple method of connecting
adjacent shields for rigidity and to prevent the ingress of soil is also required.
The plates' edges will have to penetrate the soil far enough to prevent the soil
in the trench bottom from shifting appreciably.

Trench shields are expensive and are justified only under those conditions where
they can be recovered for reuse. Many situations require that the trench be at
least partially backfilled before the shoring is removed. Under those conditions,
the U-shaped shields cannot be used as visualized above. If the backfill material
requires tamping, they cannot be used at all. However, some utilities are using
a thermal backfill that is in fluid or slurry when paced in the trench. The
U-shaped shield can then be used up-side-down and later pulled up. The standard
shoring that is pounded deep enough into the bottom of the trench so that it can

3-38
hold its position without being crossbraced presents none of these problems, but
is very expensive.

Under the worst of soil conditions, a trench cannot be excavated without installing
some type of shoring in advance. In beach sand, or when trenching below the water
line, soil shifting will take place at shallow trench depths. One solution is to
drive the shoring in from the surface and remove the soil from the channel that is
formed. Unless there are few obstructions, or the location of obstructions is
known with high accuracy, this technique has the disadvantage of possibly driving
shoring into other utility lines. Under such soil conditions, the shoring cannot
be removed until after backfilling has been completed.

The best solution for shoring under very poor conditions appears to be a type of
drop-in shoring that can be automatically put in place as the digging progresses.
One utility has described a technique that appears, with modification, to be capa­
ble of meeting this requirement. As presently practiced, a box is laid in the
partially excavated trench and tilted at an angle. The box is essentially two
wooden walls connected by cross-bracing placed so as to avoid interfering with the
backhoe. As the backhoe removes the soil, the box drops lower into the trench until
it is finally at the desired depth. The side of the box is then used to hold the
soil back while standard shoring is driven into position beside it. After the
shoring is in place, the box is moved on and the process repeated. When obstacles
are encountered, the box is lifted up, like the trench shield, and moved to the
other side of the obstruction; the box can be maneuvered very close to an obstruc­
tion.

A modification of this technique appears promising. The inefficiency of the present

system is that: the box must be cross braced in such a way as to interfere as
little as possible with the action of the backhoe; the position of the box is
changed by the crew in the trench; and the box is lowered into the trench by its
own weight, which is necessarily low. All three of these problems could be alle­
viated if the box panels were mounted on arms attached to and moved by the machine
doing the digging, either a backhoe or trencher. If the box sides could be mounted
rigidly like large paddles on arms extending from the excavating machine, no cross
would be required, and they could be mechanically lifted and moved to reduce the
dependence on manual labor.

3-39
 I 11

FASTENER

Figure 3.7 U-Shaped Trench Shields

1 1
I SMS I

Figure 3.8 Possible Shield Connectors

Figure 3.9 Box Excavator Shield

3-40
Manhole Design and Installation

In HPOF pipe-type cable and ducted, extruded dielectric cable manholes serve two
purposes during construction: (1) they provide physical strength when the cable
is pulled into the pipe, (2) they provide large spaces which can be sealed and
dehumidified when the cable sections are spliced. Since the manholes may have to
remain in position for an indefinite period of time, they must also be capable of
supporting the weight of traffic above them. However, after the circuit served by
the manholes is energized and tested, the manholes often represent just a tax
liability and under these conditions may be filled with thermal sand and abandoned.
For this reason, they represent an opportunity for cost reductions.

Prefabricated Manholes. Generally, the present practice is to pour concrete man­

holes at the site. The actual size will vary with location according to the prac­
tices and preferences of the local utility, but the dimensions are roughly 6' by
15' x 8' high with 9" walls. There are several methods by which these manholes
could be prefabricated with possible cost reductions.

There are two reasons usually given for not prefabricating manholes: (1) most man­
holes must be modified to fit the specific site conditions, and (2) transmission
manholes are so large that providing a crane capable of lifting the precast section
at the site would eliminate any savings. One utility recently (1972) started in­
stalling small prefabricated manholes for their ducted distribution lines. Prior
to that time, it was felt that a standardized manhole could not be used often
enough to warrant the initial investment. A search of the records for the previous
five years indicated that only 25% of those manholes had required modification be­
cause of specific site conditions. Since starting the prefabrication program,
only two out of almost 200 manholes installed have been poured on site. It appears
that system designers started choosing manhole locations with the standardized
shape in mind. The savings reported were $2000 to $3000 per manhole, roughly half
the cost of poured-in-place manholes. While local conditions will differ, it may
be very valuable for each utility to perform a detailed review of the actual man­
hole construction details for their area.

The manhole of the preceding paragraph was typically a 6' x 12' x 7' box with 6"
walls. It is cast in two pieces, transported to the site on trucks fitted with
an A-frame crane, and lowered into place and joined. Each half weighs roughly
nine tons. The largest manhole installed in this manner was 9' x 14' x 8' with
7" walls. The two halves each weighed roughly 16 tons and, in that case, required

3-41
a flat-bed trailer and a special crane for installation. This is about the same
size as the transmission system units. Hence by using larger transportation equip­
ment and, if necessary, by casting in perhaps three sections (two identical end sec­
tions and a middle section), it should be possible to realize savings on the con­
struction of the larger transmission line manholes comparable to those obtained
with manholes used with distribution systems.

If the cost of a large crane-truck were so high as to preclude the precasting and
installation of an entire manhole, savings might still be obtained if individual
manhole walls were precast and assembled on site. If a manhole is to be abandoned,
the top could be a girder-reinforced steel plate or a reinforced concrete section
set at ground level, to be recovered later for further use. At least one utility
is presently considering this type of system, but the benefits appear limited.

New Manhole Design. The previous section on prefabrication assumed that the man­
hole design would remain static and discussed less expensive methods of construc­
tion. Improved designs could load to further savings.

All manholes should be moisture tight since there will be cable splices at each,
but not all manholes need to be equally strong. Clearly all manholes must be
capable of withstanding overhead pressure and soil pressures on the walls, so this
will be the primary design consideration in most situations. When these loadings
are high, the following considerations are unnecessary.

For the HPOF pipe-type cable, the maximum tension in the pulling cable is on the
order of 40,000 lbs. Much of this stress is borne by the pipe and transferred to
the walls and roof of the manhole by temporary bracing. The tension when the
electrical cable first enters the pipe is about 4000 lbs. Much of this is caused
by frictional forces between the pulling cable and the inside of the pipe. This
implies that the manhole where a winch is located must be capable of withstanding
and distributing much higher loads during the pulling operation than the manholes
where cable reels are located.

If after a cable section is pulled into the pipe it is possible to move either the
winches or the cable reels (not both) to the next unused manhole, it will be pos­
sible to install manholes that are designed for a specific function. Although it
will not always be possible to place the cable reels or winch to pull in both
directions from the same hole, installation of any manhole which is not over-
designed for its function could be a savings.

3-42
However, destandardizing manholes from one design to two, as described above,
may eliminate some of the savings available from a prefabricated construction
technique.

The extra strength required at the pulling end is needed for only about four days
during the actual pull. Hence the type of manhole needed at the cable-reel end
can be installed at all locations with, when required, an additional internal,
steel load-bearing structure that can be quickly assembled and disassembled inside
the manhole. Larger manhole entrances may be required to install this structure.
The winch crews that were observed during this study spend much of their time
waiting for cables to be prepared for pulling; this time could be used for in­
stalling the temporary steel structure.

It is possible to apply the concept of temporarily strengthening manholes under

more adverse soil conditions. The basic manhole design could be much weaker than
the soil pressures would demand if each were to be strengthened by some temporary
internal structure for the time between construction, backfilling, and energizing
the circuit. What is envisioned is that all manholes would be very lightweight
(suitable for total prefabrication) and strengthened by an internal, reusable
structure to allow it to safely support traffic and retain the soil in place. Man­
holes where the winch would be located would be further strengthened with a reus­
able structure possibly installed by a winch crew. All internal structures would
be removed before the manholes were abandoned. This approach would not reduce the
number of units required for the project, but merely allow the most expensive ele­
ments of the units to be used for several projects. Obivously, if the manholes are
needed for future installations of other systems along the same route they should
not be abandoned, and the most traditional precast unit should be used.

Pipe and Cable Handling

Rigid systems like the HPOF pipe-type cable and the gas-insulated (SFg) cable re­
quired extensive handling and fabrication when being installed. The individual
pipe sections can be either laid in the trench and then joined together, or joined
outside the trench and whole lengths pulled into the trench at one time. The
latter approach is very costly for the large pipes used with the gas insulated
cable. Reductions in the costs associated with installing the pipes could lead
to significant savings.

In both the gas-insulated SFg cables and the HPOF pipe-type cables, pipe sections
are joined by welding. The welds can be (and are) radiographed before the trench

3-43
is backfilled to locate flaws that are likely to become leaks. It is difficult to
use radiographs to locate weld defects in the SF systems, because of the shadow
6
from the center conductor. However, personnel engaged in interpreting x-ray
films claim that it is still a useful technique.

0-Ring Joints. It has been suggested that switching to a soft-metal O-ring joint
or automatic welding systems might lead to significant savings; the possible savings
would be expected to accrue from reduced specialization of the crew making the
joint (lower cost/man-hour), reduced time for making the joint, and reduced joint­
testing costs. The benefits of an O-ring joint appear to be very small or non­
existent. However, automatic welding equipment appears to hold considerable pro­
mise .

O-ring joints requiring the addition of flanges to the steel pipe sections for HPOF
pipe-type cable have been rejected for the following reasons:

• Reductions in field costs would probably be offset by the increased

costs of the pipe;
• Welded joints are as strong as (or stronger than) the pipe being
welded, and are fully capable of transferring and distributing all
stresses. On the other hand, flange joints, represent a real dis­
continuity in the pipe; and
• Welding would still be required around cable joints because a pipe
sleeve must slide back over the pipe end during the cable-jointing
process.

The costs of joining two pipes in the gas-insulated SF are significantly higher
than those for the HPOF system. Although this fact is shown in the tables in
Section 2, the difference is even more impressive when one considers the time
involved. It takes approximately 20 minutes to make weld in a steel pipe, one steel
pipe per circuit. While cleaning the pipe is important, that can be performed by
other members of the crew costing less per hour than the welder. For the multiple-
finger, tulip-type connector, it takes over three hours per weld for the SF system,
and requires three welds per circuit. About one third of the time is required for
the extremely important cleaning procedure.

Because of the large diameter of the pipe, and the large space between two pipe
ends, a great deal of time is required to fill that space. The space between the
two pipes provides the flexibility necessary to align and join the inner connector.
Hence, an O-ring flange system seemed promising. When the SF^-system pipes use a
slip-fit connector, an O-ring would be useless, because a welder would still be
required to weld the pipe sleeve to the pipe.

3-44
Automatic Welding Systems. Several automatic welding systems are available on the
market and modification of one of these could possibly be used to reduce welding
costs. Most systems use a stationary torch and rotate the piece being welded. These
would obviously have limited applicability for HPOF systems. One interesting ap­
proach is presently being used for welding sections of submarine oil pipelines
(^3 feet i.d.). The joint is welded simultaneously from the inside and outside:
the inside weld is made by a welding head carried by a "crawler" (battery-powered
vehicle) inside the pipe, and the outside welding head moves on a metal track
strapped to the pipe near the joint. The crawler also carries the x-ray source for
testing. Using this total approach on an 8 inch i.d. pipe may not be feasible, but
development of the outside portion of the system for use on the smaller pipes is
promising. Both approaches should be investigated in greater detail recognizing
that the weld specifications for oil pipes and transmission cable pipes differ in
important respects.

UNDERGROUND CABLE INSTALLATION IN RURAL AREAS

Prior sections of this report have concentrated on the problems that are encountered
in urban and suburban installations of transmission lines, i.e., lines installed
principally under roads where man-made obstacles are likely to be encountered. This
emphasis has been dictated by the very small number of rural transmission lines
previously installed. The few short sections of underground transmission lines
that have been installed in the countryside, have been constructed to meet special
local conditions, such as airports or river crossings. Recently, however, there
have been indications that, in the future, longer sections of rural transmission
may be required to be placed underground. While the installation techniques used
in urban installations can be used in the rural environment, it is possible that
other techniques, not usable in the city, might be used to reduce installation
costs. The purpose of this section is to identify those areas where R&D might
lead to cost savings, when and if transmission lines are installed underground in
the rural environment.

There are several important differences between the rural and urban environments.
The urban installation site is generally characterized by the following conditions:

• It is frequently covered by hard surface that must be removed and

replaced
• The surface grade across the trench is quite small and well defined
• Access to the area for equipment and supplies is not difficult
• The impact of the construction activity on local traffic must be
kept to a minimum

3-45
 There are restrictions on the length of trench that may be open at
any one time
The load carrying capacity of the surface along the route is known
The locations of bed rock and other natural obstacles are reasona­
bly well known; also general soil conditions and the height of the
water table
Water sources and drainage systems are available
Man-made obstacles, e.g., other utility lines, are frequently en­
countered but their locations are known with reasonable accuracy
Many bends of the route are likely
Right-of-way maintenance considerations are minimal
The right-of-way is usually publicly owned
The ground must be returned to its original conditions as soon as
possible.

A rural installation site, on the other hand, is characterized in general by the

following, different conditions:

• There is no paved surface to be removed or replaced

• The surface grade across the trench can vary sharply along the route
• Access to the area for equipment and supplies may be generally poor;
frequently the main access will be along the cable right-of-way
• The impact of the construction activity on the environment may be
significant
• It is unlikely that there will be restrictions on the length of
trench that can be open at any one time
• The load carrying capacity of the surface along the route can vary
sharply and may not be well known before route surveying
• The location of bed rock, buried boulders and other natural obstacles
may not be known; also soil conditions and height of the water table
• Natural water sources and drainage are not available at all points
along the route
• Man-made obstacles, e.g., roads and other utility lines, are en­
countered infrequently but may present locally difficult problems
• The land along the right-of-way may be privately owned and acquisi­
tion and legal access may pose problems
• The surface of the ROW is not normally returned to its original
condition, and subsequent maintenance of ROW and cable may be
difficult.

Because of the differences in the environments, many of the options that were not
usable in the city may now be considered. These differences will effect

• The selection of the cable

• The designed cable environment, and
• The installation method.

3-46
In order to avoid obstructions and minimize the impact on traffic in the urban
environment, ducted cables (HPOF pipe-type cables or ducted extruded dielectric
cables) using short lengths of duct assembled on site are used almost exclusively.
The number of curves in the duct must be kept to a minimum to minimize pulling
tension. Thus, in order to go under obstructions, and reduce the number of bends,
cables are normally installed deeper than would be required by thermal considera­
tions. Thermal bottlenecks encountered en route (e.g., steam lines) might require
rerouting of the circuit or the installation of special cooling equipment. Perma­
nent or semi-permanent manholes are used. Installation is done using small, flexi­
ble equipment. Spoils are removed from the site and new material brought in. All
of these general approaches may be changed when the line is installed in the rural
areas where R&D can provide opportunities for reducing the cost of installing cables
in the rural environment by examining the different options that are available.

Cable Section

HPOF pipe type cable is the most common type of underground power transmission
cable used in the United States today. However, directly buried LPOF paper cables
are commonly used in European cities.* The primary reason for this difference
involves the types of restrictions placed on the installation activities; other
conditions are roughly comparable. Pipe type cable is chosen in the United States
because, for roughly the same or slightly lower costs than the LPOF paper cable,
the pipe will provide protection for the cable and meet the limitations on the maxi-
(13)
mum length of open trench. Once the latter are removed, directly-buried
cables can be reconsidered.

In the rural environment, ducting will be required only for forced cooled systems;
ducts assembled from short sections of rigid materal to go under obstacles would
be needed only in unusual circumstances. When ducting is used, long lengths of
flexible ducts delivered to the site on spools may be most suitable.

It is difficult and expensive to install gas-insulated cable in the urban environ­

ment because of the large size of the present, isolated phase cables and the time
required to assemble each joint. The attractiveness of the system in the urban
environment stems from its high power transmission capacity per unit of trench
cross-sectional area. Thus, despite the problems, the installation of a gas-
insulated cable in the city can be more attractive at high power levels than the
HPOF pipe-type cables of equivalent capacity.

techniques which eliminate much of the labor, once required to install these
cables, are now being used.

3-47
In the rural environment, the problems are different, the large trenches that
are required for the gas-insulated cables do not present as severe a problem as
they do in the city, but the high capacity per unit of trench cross-sectional area
is no longer as important either. The long critical distance of the gas-insulated
cable, however, will be significant advantage in long rural installations.*

In summary, for long distance rural installation, the factors that appear to be
most important in the selection of an underground transmission system are:

• Suitability for installation over long distance in presumably

variable terrain with minimum interruption for joints, etc.
• Maximum thermal independence of the local environment
• Suitability for long distance transmission with minimum thermal
and reactive compensation.

Cable Environment

Route Surveying. While there are fewer man-made obstacles along a rural route than
along an urban route and less precise information about the location of these ob­
stacles is required, there are many more natural obstacles along the route and
significantly less information is available. For this reason, route planning and
surveying can be as important as in the city. For example, the wide variety of
soil and water conditions likely to be encountered along the route can have a
significant cost impact. The soil must be capable of supporting heavy construction
equipment without major modification, but if there are too many large boulders,
the cost of excavation may be significantly increased. Too sharp a grade across
the proposed trench will require costly grading of the ROW in order to allow the
use of some heavy equipment. The location of the local water table will have a
significant impact on some cable types, and if it is necessary for workmen to
enter the trench, determining where and when shoring is required will also be
important. In certain parts of the country, off-the-road ROWs will have a very
irregular topography because of land costs and pre-emption of the best (level,
ledge-free and dry) land for building, etc. There must be enough information so
as to choose a route that will minimize the installation problems and their as­
sociated costs.

Some of the urban surveying methods described in Section 3 (p. 3-2) can be of use
during the rural survey, but there are some differences. For example, sonic remote
sensing systems were found to be less useful than electromagnetic systems for the

*The distance where the required capacitor charging current equals the cable
ampacity.

3-48
urban environment because of the lack of accuracy and problems associated with
coupling the sonic energy through the road surface to the soil underneath. The
absence of a road surface in the rural environment may allow the economical use of
an acoustic sensor to locate natural obstacles and the water table. The use of a
thermal probe to determine when the trench might be cut narrower than usual (i.e.
when the native, undistrubed soil can be used as the thermal backfill) may prove
helpful.

Thermal Considerations. The thermal environment for a cable installed in rural

areas must be the same as that for cable in the city (for the same ampacity) but
the techniques that can be used to obtain that environment can be different. For
example, there will be few, if any, extraneous thermal sources along a rural trans
mission route so that thermal bottlenecks are unlikely to be a problem. But drain
age conditions along the route can lead to severe problems if they create a vari­
able thermal conductivity of the backfill material.

Natural Cooling. In the urban installation, the ROW is normally along the street
and no ROW maintenance is required or allowed. The surface is normally covered
with macadam and can become quite hot when the sun is high; this can seriously
affect the shape of the optimum trench configuration and the circuit ampacity. In
the rural installation, the surface is covered with vegetation and does not become
excessively hot. The present theoretical approach with calculations of ampacity
and trench configuration may be more accurate in the rural than in the urban envir
onment..

When a cable is naturally cooled and surrounded by granular material, it is impor­

tant to restrict the growth of deep-rooted vegetation close to the cable. Such
plants would compete with the cable system for whatever water is available and
thereby degrade the thermal environment during the dry season. But the surface
must be covered with some vegetation to reduce erosion and soil drying by evapora­
tion. The cost of maintaining the proper amounts and types of vegetation along
the ROW is not excessive, and may be small compared to the cost of maintaining the
access road along the ROW. A more important problem may be environmental opposi­
tion to the installation because of the continuing need to control vegetation.

Forced Cooling. The use of a forced cooled system or a backfill that does not
depend on the moisture content to maintain the thermal conductivity simplifies the
thermal problem (as discussed in Section 3 (p. 3-7). These modifications confine

3-49
the need to remove the natural vegetation to the path that the equipment will fol­
low. After installation is completed, the access road along the ROW must still be
maintained to ensure timely access to the cable for repairs; however, the road may
be narrower than the ROW required for naturally cooled cables and thus be more
acceptable to the public.

Forced cooling of LPOF self-contained cables by circulating and cooling the oil in
the core is not new and may be used to reduce the demands on the external thermal
environment. However, its usefulness over long distances is limited by power
losses. For the LPOF forced cooled systems considered in Reference 13, about 8%
of the input power is lost in a circuit 100 miles long. (Periodic placement of
reactive compensation is assumed.) This power loss is the sum of the transmission
losses and the power used to run the refrigeration and oil pumping equipment. Be­
cause this loss rate is quite high, this approach will effectively exchange the
reduction in installation costs for an increase in operating costs. Forced cooling
of the oil is more appropriate for use in the city where the distances are short
and the cost of obtaining additional capacity by adding circuits might be prohibi­
tive.

The losses associated with forced cooling HPOF pipe type cables by circulating and
cooling the oil are also quite high. The losses in the cables considered in Ref­
erence 13 over 100 miles vary between 5% and 11% of the circuit capacity so that
forced cooling of oil in long underground HPOF pipe type cables is unlikely to be
satisfactory.

Forced cooling of gas insulated cables seems more appropriate to the rural environ­
ment than to the city. The energy losses are relatively low, the power density is
high, and the critical distances are quite long. Their attractiveness is likely
to be increased if and when a current program to develop a flexible gas insulated
cable is successful.

Another method of forced cooling a cable is to cool the cable exterior by circulat­
ing cooled water in pipes near or around the cable. The water would be cooled
periodically, perhaps by evaporation as in a cooling tower. The increased energy
cost (besides resistive and dielectric losses) is that energy required to pump
the water through the pipes and run the cooling tower.

The simplest water-cooled system involves water circulating through plastic pipes
laid between the cables. But, if there are voids in the soil surrounding the

3-50
cables, hot spots can develop. This cooling approach is successfully used with
conventional burial techniques, when the backfill is carefully compacted around
the cable and the voids are eliminated. Neither of the two techniques described
earlier, for compacting backfill materials, tamping in lifts and puddling is attrac­
tive in the rural environment. Tamping in lifts is expensive and tends to slow
down the installation process. On the other hand, puddling (mixing the soil with
water and pouring into place) requires large quantities of water, that may not be
easily obtained. A low thermal conductivity backfill between the cable and the
cooling pipes may also be obtained from materials that are installed as a liquid
and solidify in place (e.g., the limestone slurry described (p.3-12). But their
usefulness may be limited by the requirement for water.

A second method of water cooling would have water passing in direct contact with
the cable exterior. The only reliable method of effecting this contact is by in­
stalling the cable in an oversized duct through which water passes. Two methods
of placing the cable in the duct can be considered: a) the duct (flexible or
rigid) is buried and the cable is pulled into place (as in urban installations),
or b) a flexible duct can be extruded around the cable and the combination of duct
and cable buried as a unit. Each of these systems requires further consideration.

Installation Methods

As previously mentioned, installation methods used in the city can be used in the
rural environment. The only major change might be in methods of crossing bodies
of water. However, there appear to be opportunities to develop new techniques
which can reduce installation costs in rural areas; it is these opportunities which
are of interest in this section.

Excavation Techniques. The excavation of uniform trenches across open flat country
through loose soil can best be accomplished by automated trenching equipment, i.e.,
trenchers. It is unlikely, however, that ideal conditions will be found along the
entire excavation route so that a machine that is more flexible than present tren­
chers, as described (p.3-29), would be.very useful for installing cables in the
rural environment.

A transmission route which requires an extensive amount of trenching through rock

will seldom be chosen. A typical route is likely to contain short section through
rock but most of the route will involve soils more easily excavated. The use of
the water-cooled cables will allow the trench through the rock to be reduced in width

3-51
significantly (^40%) and the cost of excavation will be reduced proportionally.
Additional cost reductions that might be obtained by using improved, automated hard
rock surface mining techniques require further investigation. But since the length
through rock is expected to represent only a small percentage of a typical installa­
tion, the advantages of developing automated hard rock mining techniques will be
relatively small to the electric power industry, compared to other industries. For
this reason, the development of these improved techniques should not be supported
by the electric power industry at this time. However, as new hard rock breaking
techniques become available, they should be used for power installations whenever
required.

As mentioned, earlier, the presently available boring systems are either extremely
expensive or unsuitable for use in the urban environment over long distances. They
are used only when alternatives (e.g., detours) are even more expensive. This sit­
uation generally prevails in the rural environment, but they can be profitably used
with water-cooled cables in the rural environment because the horizontal hole(s)
may not need to be backfilled. The accuracy requirements are not quite so stringent
in the countryside as they are in the city.

The size and type of hole that can be provided by the "Subterrene," provided it can
be developed to a commercially acceptable state, may be suitable for use with water-
cooled systems. The characteristics of the hole lining would make any other duct
unnecessary. Other boring techniques, not suitable for naturally cooled cable,
could also yield cost-effective systems for use in the rural environment with
forced-cooled cable, for instance, pipe pushing may be suitable in some soils. A
report of the various options has been produced by Ingersoll-Rand, under EPRI/ERDA
sponsorship.

Automated Installation Techniques. The least expensive method of installing under­

ground cables, when soil conditions are favorable, is by plowing the cables directly
into the soil. The low cost of this method is derived from performing the three
parts of the process (opening the earth, planting the cable behind the plow share
and closing the earth) with just one piece of equipment and eliminating the need
to move large amounts of soil. However, the technique (with present equipment) is
unsuitable for use with the high voltage transmission cables considered in this
study because of the need to control the thermal environment.

If the cable is to be naturally cooled, the usual requirement of installing special

backfill materials makes the plow system completely unacceptable. Even if the

3-52
thermal characteristics of the undisturbed soil are ideal for use as backfill, the
lack of control over the degree of compaction of the soil around the cable, and
therefore its thermal conductivity, argues against the use of present plow systems.
If the equipment could be modified as a result of an R&D program so that soil near
the cable were predictably compacted, the use of plowing systems could significantly
reduce installation costs.

Plows may be more useful when the cables are water-cooled. If the water is passed
through flexible ducts between the phases, the problem of controlling the level of
soil compaction remains but this approach would reduce the need for special back­
fill materials and make the plow useful with a larger variety of soil types. If
desirable, the cable and the duct reels could be mounted on one large machine and
all five elements installed in one pass. The elimination of the need to shore the
trench walls would further reduce the costs of this system.

If ducting were used to circulate water in direct contact with the cable exterior,
the present plow's lack of compaction would not matter. Two methods of providing
this contact have been mentioned. The first method involves flexible ducts buried
in one pass of the plow into which the cable is later pulled; in the second method
the cable is installed in the duct at the factory and the three duct/cable combina­
tions would be plowed in, in one pass. The problem of protecting the duct/cable
during the plowing process, however, may prove difficult.

The plow cannot be used where rock is present along the route. For this reason,
the development of equipment that can successfully excavate through soil and soft
rock with economies similar to those of a plow, would be more desirable.

The vertical element trencher described earlier was made up of two distinct elements.
The first element was capable of cutting and loosening relatively soft rock. The
second element was the mucker which followed to remove the loosened soil and cut
rock from the trench. Normally, the soil removed from the trench must be disposed
of or stored alongside the trench until after the cable is installed. If the cable
or duct were laid in the trench as soon as the mucker passed, the soil could be
restored to the trench immediately after that. A possible arrangement is to have
cable reels mounted on the vehicle carrying the mucker and have the soil pass from
the mucker, across the machine on a conveyer belt and returned to the trench
immediately after the cable is laid. When the mucker is used, many soil conditions
(e.g., beach sand) might require shoring to maintain the soil integrity, even for
the very small time that the trench is open. In such cases, the vehicle carrying

3-53
the mucker must provide moving shoring in the form of trench shields on the vehicle.
The usefulness of the automated mucker technique will depend largely on the ability
of properly designed shields to eliminate the requirement for standard shoring
systems.

Devising a method to install a liquid backfill (such as limestone slurry) around a

cable being placed in the soil appears more difficult than installing ducts for
water cooling. The two problems are: 1) containing the slurry around the cable
while the slurry sets to a solid, and 2) making sure conditions underground allow
the slurry to set properly. Presumably, the slurry could be contained by some kind
of duct around the cable, but then the system has begun to resemble the water-cooled
systems described earlier. The problem deserves further investigation.

Installation of Rigid Cable Systems. The installation of rigid pipe type systems
over long distances through the rural environment appears to be similar to the in­
stallation of long distance water and gas pipelines. There are, however some sig­
nificant differences which will affect the cost savings that may be expected by
the use of pipeline installation techniques. These differences are:

• There is no need to control the thermal environment of gas and water

pipelines. The pipeline trench can be narrower than for cables and
the native soil can be used for backfill.
• Only one pipe is usually installed at a time. Electric cable systems
usually require multiple pipes.
• The cleaning and welding process at each pipe joint is relatively
simple and can be performed rapidly. The equivalent process for
gas insulated power cables can take as much as three hours.
• The effect of a small installation defect (e.g., water in the pipe)
on a gas or water pipe may be small and repairs can be scheduled at
the convenience of the pipeline company. A defect on a gas insulated
electric cable can cause arc overs and repairs must be performed as
rapidly as possible.

The net effect is that each problem associated with installing a pipe line is more
severe when the pipe is to be part of an electric transmission line than when it is
to be used for any other purpose.

While the pipeline installation systems have been developed to a high level of
efficiency, it is not clear how many of these efficiencies can be transferred to
the installation of rigid electric cable systems. The process of installing rigid
electric systems using automated techniques is cumbersome and therefore expensive,
compared to the installation of gas lines or flexible, directly buried electrical

3-54
cable. For this reason, we recommend that the development of automated installa­
tion techniques suitable for use with flexible electric cables (e.g., extruded di­
electric cables) be given priority.

STANDARDIZATION

The particular items and areas related to underground high voltage cable installa­
tions where standardization may result in economic benefits are as follows: 1) HPOF
cable system hardware, 2) inventories, and 3) testing procedures. These areas are
primarily related to HPOF systems, since the two other types of cable—extruded di­
electric and compressed-gas insulated—generally do not have as widespread a choice
of joints, terminations and hardware. There are only two U. S. manufacturers of
Compressed Gas Insulated systems at present, and it seems likely that a utility
would continue to work with its initial choice (assuming satisfactory performance).
Six manufacturers supplied 138 kV extruded dielectric cables to the Waltz Mill
test facility. Two domestic suppliers presently produce terminations for all of
these cables. Splices differ in conductor connector choice, type of wrapping tape,
and physical dimensions. Although this area may benefit from standardization in
the future, we turn now to HPOF systems where economic benefits to public utilities
are a near-term possibility.

HPOF Cable Systems

For a given size of steel pipe in an HPOF cable system, there appear to be a variety
of flanges, reducers and trifurcators available. While this presents a choice among
items for a utility, it leads to delays in field installation. As a specific example,
consider the flanges attached to the pipe ends for pressure testing. Variations in
the number of bolt holes, the O-ring diameter, and the diameter of the O-ring groove
preclude the use of a common flange for a given diameter (e.g., 10 inch) steel pipe.
The same general argument applies to reducers and night caps.

If cable manufacturers and utilities would agree to establish a standardized set

of HPOF components for a given pipe size (in terms of physical specification), it
would reduce the amount of time that the contractors must spend custom machining
adapters and obtaining material from assorted suppliers. Consequently, this portion
of the installation would be done faster, implying a reduced cost for the total
job.

Pothead manufacturers have attempted to standardize by using a range of adapaters

and inserts on a given size (voltage rating) of porcelain insulator. They can thus

3-55
accept a range of cable conductor sizes and insulation thickness for a given voltage
rating, thereby reducing the number of different porcelain insulators required.

HPOF conductor joining is another area where a variety of components are marketed.
Although this is advantageous from a competitive bidding viewpoint, delays may
develop in the field installation of these joints because the presses, dies, etc.,
of one manufacturer are not generally compatible with those of another. However,
some efforts have been made in this area, with several manufacturers designing joint
kits to accept tooling of the two major suppliers. Limited die interchangeability
does exist.

The net result of this situation is either additional delay in the field or an in­
creased inventory of joining tools carried by the installation division or contrac­
tor; in either case, there are increased costs. It is doubtful that a standardized
joint for a given cable conductor size (kcmil) and material could be agreed upon by
all manufacturers. In particular, the small producers would have difficulty com­
peting economically with the high-volume standardized production of the major man­
ufacturers. Nevertheless, for a given utility, standardization on a certain
manufacturer or cable joint system may result in benefits comparable to or greater
than those achieved through the price savings gained through competitive bidding.

Inventories

Standardization of spare parts inventories is tenuously connected to underground

cable installations through the capital investment required in spare cables and
the storage costs accruing through time. If neighboring utilities can agree on a
set of standardized cables for various current ratings at a given voltage, it may
be possible to distribute the pared-down inventory load through the group. For
example, suppose Utility A decided upon a 2250 kcmil 115 kV cable for a particular
installation, based on initial economic considerations. Suppose further that a
neighboring Utility B had 2500 kcmil cable, identical in other respects. Assuming
a mutually satisfactory arrangement could be established, it is possible that
Utility A would benefit economically in the long run by specifying 2500 kcmil cable
for its installation, and foregoing the purchase of spare cable. A's reliance on
B would presumably be balanced by other spare cable that A had which B would "use"
as a spare.

Even after cables have been purchased for a particular job, it is possible that
neighboring utilities may be able to pare their cable inventory through substitution

3-56
of similar cables. One utility in the Northeast has proposed such an arrangement,
and specified the variations it would be able to accept in terms of conductor size
and insulation thickness. Past experience has shown for this utility that spare
cable lengths are very rarely put into service. Thus, it would seem that a rela­
tively expensive (interms of initial cost and carrying costs) and inactive inventory
may be safely reduced through coordination of system needs between utilities.

This concept of shared, reduced inventory may be applied also to pumping systems,
terminations, joints and other HPOF hardware. The specific benefits in each case
depend on the cost of the item, the warehousing costs of a utility, and price differ­
ential between a "desired" item and the nearest acceptable "standard" item. The
concept may be implemented immediately solely through inter-utility negotiations,
and obviously does not require additional R&D effort or funding.

Regarding the quantitative benefits that may accrue to utilities which arrange to
share inventories, we can provide rough estimates with the following approach. The
cost of storing inventory has three basic components: 1) the annual fixed charges
that apply to the total dollar value, base on the cost of capital; 2) the direct
storage costs (warehouse space, taxes on warehouse, etc.); and 3) inventory over­

head (administrative) costs. If we assume a 9% cost of capital for item 1, direct

storage costs on the order of 2% for item 2, and overhead charges of 2%, our estim-
mate of annual carrying cost per unit of inventory value is 13%. Recognizing that
additional coordination work will be required among all utilities which participate,
we will assume a coordinating cost equal to 1% of inventory value. The result of
these assumptions is an estimated carrying cost of 14% per annum.

The next item to consider is how much reduction in total inventory value may be
achieved using a coordinated approach. It seems fairly conservative to postulate
a 15% reduction in total value. Optimistic projections may range from 20% to 30%.
We feel that a compromise figure of 20% is reasonable and sufficient for the pur­
pose of illustrating estimated benefits.

Table 3.1, which covers a range of inventory values from $250,000 to $10 million,
has been drawn up to show the potential benefits of coordinated inventories. If
the approximate value of a utility's inventory is known and falls within this range,
the estimated annual benefits may be derived. The middle range ($500,000 to
$5,000,000) should cover most large utilities, and the calculations are straight­
forward. However, there is a possibility that the largest utilities may not be

3-57
 TABLE 3.1

ESTIMATED ANNUAL BENEFITS OF COORDINATED INVENTORIES

Individual Utility
$250,000 $500,000 $1,000,000 $5,000,000 $10,000,000
Inventory Size

Annual Carrying
35,000 70,000 140,000 700,000 1 ,400,000
Charge at 14%
i-58

Reduced Inventory
200,000 400,000 800,000 4,000,000 8,000,000
via Coordination

Annual Carrying
28,000 56,000 112,000 560,000 1,120,000
Charge at 14%

Estimated Benefits
7,000 14,000 28,000 140,000 280,000
due to Coordination
able to find similarly sized coordinating partners such that a 20% reduction in
size may be unattainable. In any case, by substituting its own estimates of carry­
ing charge and potential inventory reduction, any utility can estimate annual
savings using this approach.

Testing Procedures

Certain testing procedures, when applied cumulatively, may be somewhat redundant.

HPOF system tests include visual weld inspection, x-ray weld photographs, vacuum
tests, and pressure tests. The pipe is shipped with protective caps, filled with
dry nitrogen at night during the pipe laying/welding phase, and sealed for pressure
testing prior to cable pulling. The tests are performed using pressurized dry nitro­
gen, at 300 to 600 psi for a duration of 1 to 3 days (depending on a given utility's
preference). After these pressure measurements have been taken, vacuum testing is
performed.

To summarize this general area, it may be economically advantageous if a standard

procedure can be accepted by all utilities. If, for example, a 24-hour test at
400 psi on an 8 inch I.D. pipe system has resulted in satisfactory performance for
a given utility, then proper proof of this performance may convince a utility em­
ploying a 500 psi/48-hour test that reliability would not suffer with the shorter
test.

From a redundancy viewpoint, once a pipe system has passed pressure and vacuum
tests, the value of maintaining a vacuum prior to pulling in cable is not clearly
evident. Assuming the pipe is filled with dry nitrogen at a nominal pressure of
25 psi, moisture-laden air would not enter. The previous vacuum test would also
have removed entrapped air. It thus appears that maintaining a vacuum just prior
to pulling is of questionable value, and that a nominal overfill of dry nitrogen
would be sufficient.

Standardization in the various items and fields noted previously may be carried
out through cooperative activity among neighboring utilities or through the efforts
of initiating and coordinating groups like the Insulated Power Cable Engineers
Association (OPCEA), IEEE, Association of Edison Illuminating Companies (ABIC),
EEI, American National Standards Institute (ANSI), and National Electrical Manu­
facturers Association (NEMA). ANSI is primarily a coordinating body that reviews
and seeks comments on proposed standards submitted to it. It is difficult to
assign specific cost benefits to each area of standardization discussed, but some

3-59
areas are more amenable to calculation than others (such as the distributed
shared inventory). It should be emphasized again that such areas of potential
cost reduction can be explored immediately, and that the benefits do not hinge on
the commitment of substanial research funding.
 Section 4

RECOMMENDATIONS FOR RESEARCH & DEVELOPMENT

In this study of the reduction of the cost of installation of underground power

cables we have examined almost all aspects of the process of installation which
might be changed to reduce the overall cost of underground transmission of elec­
trical energy. The principal omissions have been the design of underground cables
and the technologies of underground transmission.

There appears to be no dramatic discovery in sight which will reduce the cost of
installation to a small fraction of its present value. On the other hand, there
are several opportunities for reducing the costs of individual parts of the in­
stallation process. The cumulative effects of these will justify the efforts to
capture them. Although the technical aspects of installation received principal
attention, it was clear that savings could be made in non-technical areas, for
example:

• Communication. Different techniques of performing the same activity

are presently used by different electric utilities for reasons that
are not obvious. For example, custom appears frequently to decide
against the replacement of "thermal sand" by a solid thermal environ­
ment. Much could be gained by encouraging discussion between util­
ities to stimulate critical examination of local procedures.
m Management and Planning. The installation of an underground trans­
mission cable is complex and expensive because of the many differ­
ent activities, trades and skills involved, several of which do not
involve technical knowledge. While it is recognized that an installa­
tion will seldom run according to plan, primarily because of the
absence of prior knowledge of every detail of the route, we believe
that more careful and detailed organizational planning would be
helpful. This subject has not been examined in depth, but it is
worthy of more attention than it has received.
• Cooperation Between Utilities. The discussion of standardization
of components in Section 3 introduced the subject of underground
cable inventories. Because of the high cost and the high relia­
bility of underground transmission cables, spare cables may remain
in stock for long periods of time. If inventories could be re­
duced by neighboring utilities agreeing to use common types of
cable and spare parts, substantial savings will be made.

4-1
Improvements in these non-technical areas cannot be brought about by research and
development programs nor can they be mandated by any central authority. The con­
ditions under which each utility operates can vary significantly from service area
to service area within a state and from state to state; many of the differences are
important and will affect the degree of improvement in these areas that can be ex­
pected. If operating conditions of two utilities are too dissimilar, there will
be few attempts to communicate or cooperate. ERDA and EPRI could encourage improve­
ments by sponsoring conferences and/or newsletters to communicate specific successes.
It is recognized that many utilities are willing to try new methods and techniques
if the benefits to be gained are properly demonstrated elsewhere and then properly
communicated to the industry. Each utility will estimate for themselves the bene­
fit that might accrue from the proper use of the new technique in their own opera­
ting environment.

The analysis of methods and costs of underground cable installation presented in

Sections 2 and 3, has indicated that there are several areas in which research and
development is likely to produce useful benefits. Within urban areas, the benefits
are likely to appear in the form of greater speed and predictability of installation
and thus of lower costs. Precise cost benefits are hard to identify since they
vary so greatly from one installation to another according to local geology,
regulations and labor costs. In rural areas, the benefits will appear in the
ability to use techniques which are not usable in the city but which are inherently
more rapid and capable of continuous operation.

The program represented by the following list of projects contains research topics
that fall into three categories:

Program A: Route Planning and Design

Program B: Pavement and Soil Removal Techniques
Program C: Installation Methods

These programs are relatively independent so that, to a large extent, they can be
performed simultaneously. The total effort required by the project lies between
forty-five and fifty-five man years but this estimate is tentative since the efforts
required by some parts of the program will not be known until the conclusion of
initial exploratory phases. The total program should be capable of completion
within eight years.

4-2
PROGRAM A: ROUTE PLANNING AND DESIGN

Project Al: Computerized Route Planning

R&D in this area has the potential of significantly reducing the amount of manual
effort expended in determining a satisfactory cable routing through complex urban
areas. The manpower savings would result from computerized analysis of blueprints
and similar sketches of existing plans or layout drawings. Ultimately, a system
could be devised which would calculate the optimal cable (pipe) routing through an
intersection, plot a new map of the area, specify which foreign service lines must
be relocated (if any) and estimate the cost of performing all construction work.

For such a system to be realized, R&D is required in several areas. These include:

(a) Further development of three-dimensional piping layout computer

programs. In addition to route clearance calculations, the program
should perform route calculations based on minimum bending radii
considerations, standard pipe lengths (if applicable), optimal
spacing of splices and maximum cable lengths.
(b) Development of apparatus capable of readily translating printed
layout sketches and blueprints into data compatible with the com­
puter system specified above. Conventional digitizers and optical-
electronic systems offer promise as initial components of prototype
apparatus.
(c) Research on methods of computer processing data from active survey­
ing systems, such as those employing radar or magnetic field gradient
sensors, into forms which can be input directly to the planning
computer.

Expected Effort

Task (a) - Layout Program Development 2.5 man-years

Task (b) - Graphic Data Processing Apparatus 2.0 man-years
Task (c) - Active System Data Analysis 1.5 man-years

Project A2: Subsurface Surveying System

The use of underground survey systems employing either low frequency EM radiation,
radar or magnetic field sensing can potentially reduce the costs of both explora­
tory excavation (test pits) and actual route excavation. It is unfortunate that
the areas of maximum possible benefits, crowded urban intersections, also provide
the most difficult operating environment. Nevertheless, as more cable installa­
tions are planned for suburban areas the near-ideal operating environment for radar
systems should result in improved knowledge of soil, rock ledge and water table
along the proposed cable route. If these potential advantages are to be realized,
R&D in the following areas is recommended.

4-3
 (a) Research into optimal output waveshape, frequency, return signal
analysis and output data for radar systems. Specific tasks would
include:

• Better understanding of information derived from the polariza­

tion of reflected signals.
• Continued development of the synthetic aperture approach for
general purpose radar systems.
• Continued development of computer processing techniques to
retrieve information from dispersion data.
• Possible development of multi-antenna interferometric systems
to enhance positioning accuracy and produce addtional subsur­
face propagation data.
• Continued development of the pulsed radar system developed
solely for pipe detection.
(b) Development of prototype hardware (at least two types of systems).

(c) Development of output data processing techniques to produce survey

maps.

Expected Effort

Task (a) - Radar system research 4 man-years

Task (b) - Prototype Hardware (at least two
different systems) 8 man-years

Task (c) - Output data processing 2 man-years

Improved Cable Thermal Environment

R&D in this area has the potential to reduce the cost of installing new cables and
to increase the power rating of both old and new cables. The cost reductions would
result from a reduced trench width or the use of the excavated native soils or new,
less expensive materials for the backfill instead of expensive importing thermal
sand. The increased power rating of individual circuits could reduce the need for
additional capacity. R&D in this area should concentrate on devising a general
method of determining just where and when significant improvements can be made and
what site and routing data will be required to make this determination. The speci­
fic areas that seem to hold significant promise for overall cost reductions are
described in the following projects.

Project A3: Exploration of Cable Thermal Environment

The thermal conductivity of the native soils may be high enough to allow the width
of the trench to be reduced and/or to allow the spoil to be used as the thermal

4-4
backfill. Methods of determining in advance when these fortuitous circumstances
exist should be explored. The R&D program should include:

(a) The development and testing of a thermal probe suitable for survey­
ing the thermal conductivity of the undisturbed native soil along
the proposed cable route.
(b) The development of a methodology for conducting thermal surveys to
take into account peak cable usage and the effect of localized
events (e.g., recent rains and poor drainage).

Expected Effort

Task (a) - Thermal Probe 2 man-years

Task (b) - Methodology of Thermal
Surveying 0.5 man-years

Project A4: New Thermal Materials

The development of new low-cost materials with high thermal conductivities which
do not depend strongly on compaction or water content would lower costs by removing
the need to import and install thermal sand. A program to develop, test and eval­
uate new thermal materials should also include consideration of how the material
would be transported and installed, conditions under which it can and cannot be
used, effects on the cable materials, etc. Specialized equipment, if required,
for transportation and installation should not be developed until after several
materials have been investigated and it is clear which have the most promise of
being used.

Expected Effort

Task (a) - Search for Promising Materials 0.5 man-years

Task (b) - Laboratory Development and Test
of Candidate Systems (two programs) 3 man-years
Task (c) - Field Test and Evaluation 2 man-years

Project A5: Thermal Analysis of Cable Environment

The standard methods of computing the proper trench configuration and cable ratings
assume that the primary heat sink is the surface of the ground and that as the sur­
face temperature varies, the cable rating will vary. This may not be the situation
during the whole year but it is a good approximation if peak usage occurs during
the summer, the maximum cable capacity may be underestimated. How much it is under­
estimated will depend on the soil surface, the local climatology, and recent cable
usage. The optimum shape of the trench will vary similarly with local conditions.

4-5
Research into how the local and seasonal thermal conditions may effect the cable
rating will lead to significant cost savings if it proves possible to defer or re­
duce the cost of new cables. This program should include:

(a) Development of a method of thermal analysis that includes the effects

of local climate, soil temperatures and diffusivity into the determin­
ation of the maximum cable ampacity, and hence of calculating optimum
trench configuration.
(b) Validation of the above analysis with an experimental program.

Expected Effort

Task (a) - Improved methods of thermal analysis 1.5 man-years

Task (b) - Experimental validation 2 man-years

Project A6: The Thermal Interaction of Cables Arranged One Above the Other

Dual circuits are now normally installed side by side. When it proves difficult
to route these circuits through complicated intersections, an alternative path
may sometimes be found if the two circuits could be placed one above the other.
The acceptability of the vertical arrangement depends on its effect on the thermal
rating of the cable. Research into the thermal acceptability of this arrangement
will lead to cost reductions if it proves possible to reduce the need for detours.
The research should include:

(a) Determination of the relative ampacities of the two cables in the

vertical configuration compared to those in the horizontal, assuming
that:
(i) the surface is the heat sink
(ii) the method which accounts for local climatology correctly
predicts the temperature and location of the heat sink.
(b) Determination of the optimum trench configuration for the vertical
configuration.
(c) Validation of the model by experiment.

Expected Effort

Task (a) - Ampacity of circuits in vertical configuration1man-year

Task (b) - Trench configuration 0.5 man-years
Task (c) - Validation 2 man-years

PROGRAM B: PAVEMENT AND SOIL REMOVAL TECHNIQUES

A large portion of the cost of installing underground lines is comprised of the

cost of excavating, disposing and replacing the soil in the trench and shoring

4-6
the trench walls. The most economical excavation technique can be determined only
by considering the specific conditions at the site. However, there are some im­
provements of standard techniques that can effect cost reductions. These improve­
ments will not be useful along every portion of every cable route, but when they
can be used, the cost reductions appear to be significant. The following R&D pro­
jects are intended to realize these possible cost reductions. The projects are
not listed in any order of priority.

Project Bl: High-Power Microwave Pavement Breaking Systems

In any situation where removal of concrete paving is hampered by the close proxi­
mity of objects subject to damage from impact breakers, pavement cracking by a
high-power microwave heating system should possess the lowest potential for damage.
As noted in Section 3, depending on the nature of the paving, additional apparatus
may be required to separate the interlocking fragments produced by the microwave
system. The optimum pavement breaking system would use microwave heating to weaken
and crack concrete, followed by bursts from a small water cannon to separate the
pieces.

It is anticipated that development rather than research is the primary area worthy
of attention in creating a prototype microwave heating system. The program should
emphasize the importance of developing a system that is compatible with the con­
struction environment at the job site. Requirements for a prototype system should
include the following:

(a) Higher power than existing systems

(b) Provision for shielding personnel from reflected microwaves
(c) Ease of maintenance and operation
(d) Operating costs comparable to impact breakers

Expected Effort 4.5 man-years

Project B2: Plasma Torch Pavement Breaking System

The plasma torch system is another alternative to impact-type breakers. It is

the only alternative which can break solidly reinforced concrete without additional
cutting apparatus to sever the reinforcing bars. Current prototypes have experi­
enced problems in the erosion of torch materials adjacent to the workface, exces­
sive noise and high power comsumption. R&D is thus necessary in the following
areas to produce a system suitable for field trials.

4-7
 (a) Design R&D (materials, performance, efficiency)
(b) Development and test of a prototype field unit

Expected Effort

Task (a) - Design R&D 2 man-years

Task (b) - Prototype field unit 4 man-years

Project B3: Water Cannons for Surface Removal

It is anticipated that devices utilizing ultra-high-pressure water may be success­

fully employed in two areas related to surface removal. These are (a) slicing of
hard (concrete) paving surfaces, and (b) demolition of concrete of rock-like ob­
structions. In the first application, the normal operating mode would employ a
thin, continuous ultra-high-velocity stream of water as a cutting tool. Advantages
over a concrete slitting saw include less noise, minimal dust production, and the
fact that there are no cutting discs to sharpen or replace.

The impact devices used to fracture obstructions rely upon minute "slugs" of water
which are forced from chambers under ultra-high pressure. By creating and exploit­
ing cracks in the object, they cause it to fail in tension; since tensile strength
is much less than compressive strength for concrete and rocks, breakup is achieved
more easily.

Several prototypes of slitting machines and fracture machines have been constructed;
testing programs are underway at the individual laboratories of origin. Current
R&D efforts should be directed toward eliminating nozzle material fatigue problems
in the impact-extrusion type of device and raising the pulse rate of those systems
employing annular cumulation of pressure. Pavement slitting systems may need addi­
tional effort in designing the cutting tool so that splash and minute objects flying
out of the cut will not be problems to the operator. The hybrid technique cited
in Section 3 also deserves additional attention, since it can produce pulses faster
than cumulation systems with more pressure than impact systems. Specific areas
of R&D include the following:

(a) Development of ultra-high pressure systems designed specifically

for pavement slitting. Requirements for successful application
are:
• Designed for machine mounting
• High cutting rates in concrete
• Ease of maintenance and operation
• Economics comparable to slitting saws

4-8
 (b) Development of ultra-high pressure systems designed for breaking
up concrete paving, rocks,old foundations and other obstructions
of a similar nature. A successful system should:
• Operate with minimal spray from the workface
• Operate efficiently independent of physical orientation
• Break up material at rates comparable to current machine-
mounted hydraulic/pneumatic breakers
• Demonstrate ease of maintenance and operation
• Demonstrate favorable operating economics

Expected Effort

Task (a) - Hydraulic cutting system development 3 man-years

Task (b) - Hydraulic breaker development 3 man-years

Project B4: Improved Trenching Machines

Present trenching machines under favorable conditions can excavate soil very quickly
and economically. However, they are highly specialized machines that cannot turn
corners easily or operate when the surface is on a significant grade. The develop­
ment of more flexible trenching equipment for use in some suburban areas and in the
rural environment can effect significant savings. While such a piece of equipment
would be useful to a variety of industries, its development should be supported by
the electric utility industry because, in general, its requirements are the most
stringent, and in many respects, its need are the greatest.

A program to develop a more flexible trenching machine should include the design,
construction and test of a trencher which has a digging element capable of cutting
or breaking up frost or soft rock as well as loosening the soil. The spoil should
be removed from the trench either by the digger or by a soil removal system which
can compensate for a surface grade. The soil removal unit and the digger should
both be able to operate close to an obstruction and to move around curves easily.

Expected Effort

Task (a) - System analysis, preliminary design,

and design of development program 2 man-years
Task (b) - Final design, construction and test Effort to be
developed in
Task (a)

Project B5: Water Jet Excavation Techniques

Water jet and air and water jet systems for moving soil are well established tech­
niques in the construction industry. But use of these systems for clearing soil

4-9
away from obstructions in a trench represents a new application. Their use would
effect savings by substituting a mechanized approach for the manual approach being
used now. However, management of the water and the soil moved in this manner re­
presents a new problem. A program to develop methods of solving the problems
should include:

(a) The assembly of a water jet soil removal and water management system
capable of meeting the requirements of water management and disposal
in the urban environment using commercially available equipment.
(b) Test and evaluation of the system in the field under a variety of
soil conditions.
(c) Design, construction and test of an integrated system if successful.

Expected Effort
Task (a) and (b) 1 man-year
Task (c) 2 man-years

Project B6: Improved Boring Systems

Present methods of boring a horizontal hole into which an electric transmission

cable can be installed are limited by their high cost and poor accuracy. Of these
two, the poor accuracy is considerably more serious; the high cost can be justi-
field if their use can reduce the total circuit length significantly. But, the
poor accuracy will severely limit the usefulness of the approach in the suburban
and urban environments, the most probable environments for new underground lines.
Thus, any program to develop new boring equipment should concentrate first on de­
veloping new methods of guiding and steering the borers and secondarily on reduc­
ing the costs. We suggest that the development of new boring equipment be ap­
proached in the following steps:

(a) Guidance and steering equipment.

(1) Design new equipment capable of being demonstrated on presently

available boring equipment.
(-2) Prototype construction, demonstration and evaluation.
(b) System analyses and design.
(1) Perform system and cost benefit analyses to specify and to
provide justification for developing boring equipment not
now being developed and that is intended to meet the
needs of the electric utility industry.
(2) Design a force cooled, electrical cable system capable of
being installed in the horizontal holes produced by new
boring equipment designed for other applications, e.g.,
obtaining soil samples.
(c) Design low cost new equipment for boring and lining long, large
diameter horizontal holes for use with electric transmission lines.

4-10
 (d) Prototype construction, demonstration and evaluation.

Expected Effort

Task (a) - Guidance and Steering Techniques 1. 3 man-years

2. Determined by
Task 1
Task (b) - Systems Analysis 2 man-years
Task (c) and To be determined
(d) by Task (b)

Project B7: Automated Soil Removal System

The extreme variability of conditions in the urban environment is most economically

met by excavating the trench with a backhoe. However the backhoe spends much of
the time lifting the excavated soil out of the trench and placing it on a truck or
on the side of the trench. Using an automatic soil remover like a conveyer system
to lift the soil would increase the time the backhoe can spend digging and increase
the overall excavation speed. A program to develop an automated soil removal
system should consist of the following tasks:

(a) Design an automatic soil removal system which can be used with back-
hoes for installing cables in the city streets. Since obstructions
in the trench will be common, the equipment must be capable of being
moved around obstacles without excessive interruption of the excava­
tion process.
(b) Build, test and evaluate a prototype of the automatic soil removal
system.

Expected Effort

Task (a) - Preliminary design of equipment 1 man-year

Task (b) - Build, test and evaluate prototype Effort deter-
mined by
Task (a)

Project B8: Improved Shoring Techniques

The cost of using the most common shoring techniques is defined almost entirely
by the cost of the labor. The development of a reusable shoring system with
perhaps higher materials costs but lower labor costs could reduce the average
cost per foot of trench. The design of such a shoring system should allow it to
be used with a large variety of soil conditions and some obstructions in the
trench. This design and development effort should include the following tasks:

4-11
 (a) Design of a reusable, drop-in shoring system which will meet the
requirements of the OSHA and the utilities under a large variety
of soil and water conditions.
(b) Construction, field testing, and evaluation of the shoring system
under a variety of conditions of soil, trench depth, obstacles in
the trench, etc.

Expected Effort

Task (a) - Preliminary design 1 man-year

Task (b) - Construction and test Determined by

Task (a)

Project B9: Temporary Shoring System

Some soils are so unstable that it is necessary to provide shoring before or during
excavation of the trench. At present, shoring is driven into position from the
surface before excavation. The development of a machine which can provide shoring
to be used briefly during excavation and replaced later, will reduce costs and also
the danger of driving shoring into existing underground equipment. A program for
development of a system to provide temporary shoring should include the following
tasks:

(a) A definition of the characteristics of equipment to be mounted on

an existing excavator to provide temporary shoring under a variety
of soil conditions and trench depths; preliminary design and devel­
opment of program plan.
(b) Construction, field test and evaluation of temporary shoring machine.

Expected Effort

Task (a) - Preliminary design and program plan 1 man-year

Task (b) - Construction, field test and evaluation Determined by

Task (a)

PROGRAM C: INSTALLATION METHODS

Project Cl; Automatic Welding and Testing of Pipe Joints

Most of the automatic welding equipment presently available is not suitable for
use in the field. However, automatic welding equipment is now being developed and
marketed for use on large diameter submarine oil pipes. Some of this equipment
also includes a provision for automatic testing of the joint. If some of this
equipment Could be modified for use on the smaller pipes used for HPOF pipe type
cable, significant cost reductions could be obtained through reduced labor costs.

4-12
increased reliability and reduction in the time required to weld and test the
pipe.

The most promising automatic welding equipment is quite large and in one case in­
dues a "vehicle" that moves under its own power inside the pipe. It is designed
to be used on a pipe laying barge. Whether this equipment can be modified for
satisfactory use on small pipes lying in a trench in unclear. A program to modify
existing automatic welding and testing equipment should include:

(a) Detailed investigation of present automatic pipe welding and joint

testing equipment designs and concepts to determine which equipment
shows the most promise of being able to meet the requirements of the
electric utilities; estimate of final equipment costs and possible
system benefits.
(b) Modification of promising designs and construction of units for use
on smaller pipes.
(c) Field test and evaluation of the equipment performance.

Expected Effort

Task (a) Detailed investigation of present

automatic field welding techniques 0.5 man-years
Task (b) Modification and construction of Based on results
promising designs of Task (a);
probably 5-10
Task (c) Field test and evaluation
man-years

Project C2: Reduction of Manhole Costs

The development of methods to reduce the costs of manholes shows promise. The
use of concrete manholes, precast in two sections, is well established in many
areas of the country and the cost advantages are significant compared with on-site
construction. The next promising step appears to be development of temporary re­
usable manholes that are suitable for use in the urban environment. The steps
necessary to develop temporary manholes are as follows:

(a) Design reusable temporary manholes capable of meeting the stress

requirements of soil and traffic loading, and which can be used
with existing installation and transportation equipment; analyze
methods of field use including the use of temporary reinforcement
during cable installation if necessary; estimate costs and benefits
to the user.
(b) Construct, field test and evaluate a prototype temporary reusable
manhole.

4-13
Expected Effort

Task (a) - Preliminary design and analysis 1 man-year

Task (b) - Construction field test and evaluation Determined
by Task (a)
probably 1-
2 man-years

Project C3: Integrated Transmission Cable System for Underground Rural Use

The successful development of an automated method of installing direct-burial

cable in the rural environment under favorable conditions promises to reduce in­
stallation costs provided the design of the cable, the provision of the thermal
environment and the methods of cable installation are approached as elements in
one integrated underground transmission system. Because of the variability of the
thermal characteristics of soil, means to provide rapidly a predictable thermal
environment for the (long distance) rural underground cable becomes a major aspect
of the problem. The preferred approaches appear to be either (1) the automated
installation of a water duct into which a single phase cable can later be pulled
or, (2) the automated installation of the combination of a cable with an oversized
water duct extruded around it. The former approach seems to have the advantages
of a higher intrinsic reliability and the ability to use presently available equip­
ment and cable, although the installation is no longer a continuous process. We
suggest that the development of an integrated underground cable transmission sys­
tem be approached in the following steps:

(a) Systems Analysis

1. Establish the need for an integrated rural underground cable

system in terms of the following characteristics:
• Number of installations and typical length expected
during the period 1980-1990.
• Range of voltage and transmission capacities required
• Expected locations, soil and topological conditions
2. Develop functional specifications for the cable system and
its installation methods as a result of (1).
3. Consider alternative approaches; select the most promising
conceptual design, prepare preliminary system plans in suf­
ficient detail for initial cost estimates to be developed.
4. Perform a detailed cost benefit analysis to provide justifi­
cation for further work.
5. Develop a program plan.
(b) Prototype system construction, test and evaluation.
(c) System demonstration.

4-14
Expected Effort

Task (a) - System analysis and preliminary design 3 man-years

Task (b) and (c) Determined
by Task (a)

Project C4: Testing Procedures

Testing procedures vary considerably between one utility and another. An examina­
tion of present practices in order to remove redundant tests and decide upon accep­
table standards would lead to substantial benefits.

It is suggested that this type of study be undertaken by a utility workshop con­

vened for the purpose, followed by a Task Force consisting of utility engineers.

Project C5: Standardization

Although "Standardization" may not be regarded, strictly speaking, as a subject

for technical R&D, it has become apparent in the course of this investigation that
greater use of standarized components, construction techniques, test and installa­
tion methods could lead to substantial savings. These can occur through reduced
inventories, a large pool of experienced installation personnel, and greater utili­
zation of specialized equipment. It is suggested this topic is suitable for in­
vestigation by a Task Force comprised of electric utility engineers.

4-15
 REFERENCES

1. The Use of Advanced Technologies for Locating Underground Obstacles, EPRI/ERDA

Final Report 78-20-0, Palo Alto, California, August 1974.
2. "Radar/Computer System Profiles Underground," Electrical World, August 1, 1974,
p. 82.
3. Van Dyke, J., "A Portable Plastic Pipe Locator," AGA Distribution Operating
Section, Annual Conference, May 7, 1974.
4. "Pipe and Cable Location," Sub-Search Corporation, 70 Mt. Hope Avenue,
Rochester, New York.
5. Handbook of Chemistry and Physics, 40th Edition, Chemical Rubber Publishing
Company, 1958.
6. Dingwell, I., et al., The ADLPIPE Design Information System, Arthur D. Little,
Inc., Cambridge, Massachusetts, June, 1973.
7. Long, George M., "IGT Program Report on Quiet Pavement Breakers," Pipe Line
Industry, June, 1972.
8. Personal conversation with M. Thorpe, President, TAFA Division, lonarc Cor­
poration, Bow, New Hampshire.
9. Long, George M., "Quiet Pavement Breaking Research and Development at the
Institute of Gas Technology," AGA Operating Section Proceedings, 1972.
10. Clipp, L. L., "Water Cannon Shows Promise for Gas Utility Trenching,", Pipe
Line Industry, July, 1973.
11. "Soil Thermal Characteristics in Relation to Underground Power Cables," AIEE
Committee Report, Paper No. 60-785, December 1960.
12. Fink, L. H. and Smerke, J. J., "Control of the Thermal Environment of Buried
Cable Systems—Part II," AIEE Transactions P.A.S., June, 1958.
13. "Underground Power Transmission," Arthur D. Little, Inc., Cambridge, Massachu­
setts, 1971.
14. "System Requirements and Analysis of an Experimental Guide Tunneler, ERDA/
EPRI Final Report," Ingersoll-Rand Research, Inc., Princeton, New Jersey,
August, 1975.

R-l