INFORMATION TO USERS

This manuscript has been reproduced from the microfilm master. UMI films
the text directly from the original or copy submitted. Thus, some thesis and
dissertation copies are in typewriter face, while others may be from any type of
computer printer.

The quality of this reproduction is dependent upon the quality of the

copy submitted. Broken or indistinct print, colored or poor quality illustrations
and photographs, print bleedthrough, substandard margins, and improper

W
alignment can adversely affect reproduction.

In the unlikely event that the author did not send UMI a complete manuscript
IE
and there are missing pages, these will be noted. Also, if unauthorized
copyright material had to be removed, a note will indicate the deletion.
EV
Oversize materials (e.g., maps, drawings, charts) are reproduced by
sectioning the original, beginning at the upper left-hand comer and continuing
from left to right in equal sections with small overlaps.
PR

ProQuest Information and Learning

300 North Zeeb Road, Ann Arbor, Ml 48106-1346 USA
800-521-0600

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
 W
IE
EV
PR

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
 University of Nevada, Reno

Behavior of Underground Piping Joints Due to

Static and Dynamic Loading

W
IE
EV
A dissertation submitted in partial fulfillment of the
requirements for the degree of Doctor of Philosophy in
Civil Engineering
PR

by

Ronald D. Meis

Dr. M. Maragakis/ Dissertation Co-advisor

Dr. R. Siddharthan/ Dissertation Co-advisor

May 2003

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
 UMI Number: 3090863

W
IE
EV

__ ___ <B>
PR

UMI
UMI Microform 3090863
Copyright 2003 by ProQuest Information and Learning Company.
All rights reserved. This microform edition is protected against
unauthorized copying under Title 17, United States Code.

ProQuest Information and Learning Company

300 North Zeeb Road
P.O. Box 1346
Ann Arbor, Ml 48106-1346

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
 I \ I\ I ksl I ^
CM \l V \l )A THE GRADUATE SCHOOL
We recommend that the thesis
prepared under our supervision by

RONALD D. MEIS
entitled

Behavior of Underground Pipeline Joints Due to

Static and Dynamic Loading

W
be accepted in partial fulfillment of the
requirements for the degree of
IE Doctor of Philosophy
EV
—Manos Maragakis, Ph.D., Co-Advisor

Raj Siddharthan, Ph.D., Co-Advisor

PR

Itani, Ph.D., Committee Member

Yanyao Jiang, Ph.D., Committee Member

Ik ___________________________
John Anderson, Ph.D., At-Large Member

Marsha H. Read, Ph.D., Associate Dean, Graduate School

May, 2003

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
 ABSTRACT

This report describes the procedures and results of an empirical data research program

designed to determine the static and dynamic behavior of some typical restrained and

unrestrained underground pipe joints, such as their axial and rotational stiffness, axial

force capacity, and moment bending capacity. Pipelines have suffered damage and

failure from past earthquakes and have been shown to be vulnerable to seismic motions.

It has been well documented that a majority of pipeline failures have occurred at

unrestrained pipe joints while restrained joints have a capacity to resist pull-out, and

W
therefore, both unrestrained and restrained pipe joints need to be examined and their axial

and rotational stiffness and their strength characteristics need to be investigated in order
IE
to help mitigate potential damage and failure. Five different material types with eight

different joint types and several different pipe diameters were used in this testing
EV
program. The test results are given as load-displacement plots, moment-rotation plots,

and tables listing the axial and rotational stiffness, force capacities, and bending moment
PR

capacities. A comparison is made between static results and dynamic results to determine

if static testing is sufficient to characterize the dynamic behavior of pipe joints. This

report also suggests methods to use the test results for a finite element pipeline system

analysis and for risk assessment evaluation.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
 ACKNOWLEDGMENTS
The project described in this report was funded by the Multidisciplinary Center for

Earthquake Engineering Research (MCEER) located at the State University of New York

at Buffalo under a grant from the National Science Foundation (NSF). The authors are

grateful for this funding and support. However, it must be noted that the opinions

expressed in this report are those of the authors and do not necessarily reflect the views of

MCEER.

W
IE
EV
PR

Reproduced with permission o f the copyright owner. Further reproduction prohibited without permission.
 TABLE OF CONTENTS

SECTION TITLE PAGE

1 INTRODUCTION I
1.1 Background I
1.2 Past Performance of Pipelines 4
1.3 Past Research 5
1.4 Test Specimen Description 11

2 AXIAL STATIC EXPERIMENTS 21

2.1 Description 21
2.2 Test Assembly Configuration and Instrumentation 22
2.3 Test Methodology and Loading 25
2.4 Test Results 25

W
3 AXIAL DYNAMIC EXPERIMENTS 39
3.1 Description 39
3.2 Test Assembly Configuration and Instrumentation 40
3.3 Test Methodology and Loading
IE 43
3.4 Seismic Motion Records 44
3.5 Test Results 49
3.6 Combined Load-displacement Plots 61
Comparison Between Dynamic Loading and Static Loading Results 66
EV
3.7

4 STATIC AND DYNAMIC BENDING EXPERIMENTS 71

4.1 Description 71
4.2 Test Assembly Configuration and Instrumentation 72
4.3 Test Methodology and Loading 75
PR

4.4 Test Results 76

4.5 Combined Moment-Theta Plots 90

5 APPLICATION OF TEST RESULTS 95

5.1 Description 95
5.2 Risk Assessment Evaluation 95
5.3 Analytic Finite Element Analysis 101
5.4 Example: Computer Analysis of a Pipeline System 103

6 OBSERVATIONS AND CONCLUSIONS 140

7 FUTURE RESEARCH and INVESTIGATION 144

8 REFERENCES 147

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
 Appendix A AXIAL STATIC EXPERIMENTS
TEST REPORTS and LOAD-DISPLACEMENT PLOTS

Appendix B AXIAL DYNAMIC EXPERIMENTS

TEST REPORTS and LOAD-DISPLACEMENT PLOTS

Appendix C STATIC AND DYNAMIC BENDING EXPERIMENTS

TEST REPORTS and MOMENT-THETA PLOTS

Appendix D RISK ASSESSMENT EVALUATION DEVELOPMENT

Appendix E DEVELOPMENT OF SOIL STRAIN -

SEISMIC VELOCITY RELATIONSHIP

W
IE
EV
PR

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
 LIST OF ILLUSTRATIONS
FIGURE TITLE PAGE
1-1 Ruptured ISO mm dia. cast iron bell from the Northridge earthquake 9
1-2 Cracked bell on 200 mm dia. cast iron pipe from the Northridge earthquake 9
1-3 Slip-out of joint on 450 mm dia. cast iron pipe from the Kobe earthquake 9
1-4 Slip-out of joint on 200 mm dia. ductile iron pipe from the Kobe earthquake 9
1-5 Slip-out of joint on 450 mm dia. cast iron pipe from the Kobe earthquake 10
1-6 Shear failure on 200 mm dia. cast iron pipe from the Kobe earthquake 10
1-7 Slip-out of joint on 300 mm dia. ductile iron pipe from the Kobe earthquake 10
1-8 Failure o f300 mm (12”) dia. steel main from the Northridge earthquake 10
1-9 Joint types for ductile iron pipe 12
1-10 Cast iron pipe with lead-caulked joint 13
1-11 Ductile iron pipe segments 14
1-12 Ductile iron pipe with retaining ring 15
1-13 Ductile iron pipe with a gripper gasket joint 16
1-14 Ductile iron pipe with bolted collar joint 17

W
1-15 Steel pipe with lap-welded joint 18
1-16 PVC pipe with push-on rubber gasket joint 19
1-17 Polyethylene pipe with butt-fused joint 20
2-1 Load frame and actuator configuration for static load testing
IE 23
2-2 Exploded view of actuator, test specimen, and loading frame 24
2-3 Location of external instrumentation for static load testing 24
2-4 Typical smoothed load-displacement plot showing key zones and points 29
EV
2-5 Example of load-displacement plot with approximated bi-linear curve 29
2-6 Load-displacement for ductile iron pipe with push-on rubber gasket joints 30
2-7 Cut section of ductile iron pipe with push-on rubber gasket joint 30
2-8 Load-displacement for cast iron pipe 31
2-9 Load-displacement for ductile iron pipe with gripper gasket joints 31
PR

2-10 Load-displacement for ductile iron pipe with retaining ring joints 32
2-11 Load-displacement for ductile iron pipe with bolted collar joints 32
2-12 Load-displacement for steel pipe with lap-welded joints 33
2-13 Load-displacement for PVC pipe with push-on rubber gasket joints 33
2-14 Load-displacement for PE pipe 34
2-15 Maximum load capacity for different pipe diameters of restrained joints 37
2-16 Elastic stiffness for different pipe diameters of restrained joints 37
3-1 Dynamic test assembly and shake-table 41
3-2 Plan view of shake-table, specimen, restraint frame and loading arm 42
3-3 Location of external instrumentation for dynamic load testing 42
3-4 Northridge Arleta station normalized velocity time-history 45
3-5 Northridge Arleta station response spectra 46
3-6 Northridge Sylmar station normalized velocity time-history 46
3-7 Northridge Sylmar station response spectra 47
3-8 Northridge Laholl station normalized velocity time-history 47
3-9 Northridge Laholl station response spectra 48

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
 vi

3-10 Typical elastic stiffness curves for restrained joints 51

3-11 Load-displacement curves for 200 mm cast iron pipe 52
3-12 Load-displacement curves for 150 mm DIP with push-on joint 52
3-13 Load-displacement curves for 200 mm DIP with push-on joint 53
3-14 Load-displacement curves for 150 mm DIP with gripper gasket joint 53
3-15 Load-displacement curves for 200 mm DIP with gripper gasket joint 54
3-16 Load-displacement curves for 150 mm DIP with retaining ring joint 54
3-17 Load-displacement curves for 200 mm DIP with retaining ring joint 55
3-18 Load-displacement curves for 150 mm DIP with bolted collar joint 55
3-19 Load-displacement curves for 200 mm DIP with bolted collar joint 56
3-20 Load-displacement curves for 150 mm steel pipe 56
3-21 Load-displacement curves for 200 mm steel pipe 57
3-22 Load-displacement curves for 150 mm PVC pipe 57
3-23 Load-displacement curves for 200 mm PVC pipe 58
3-24 Load-displacement curves for 150 mm PE pipe 58
3-25 Load-displacement curves for 200 mm PE pipe 59
3-26 Load-displacement curves for DIP with push-on joints 62
3-27 Load-displacement curves for DIP with gripper gasket joints 62

W
3-28 Load-displacement curves for DIP with retaining ring joints 63
3-29 Load-displacement curves for DIP with bolted collar joints 63
3-30 Load-displacement curves for steel pipe with lap-welded joints
IE 64
3-31 Load-displacement curves for PVC pipe with push-on joints 64
3-32 Load-displacement curves for PE pipe with butt-welded joints 65
3-33 Restrained joint axial stiffness 65
3-34 Restrained joint ultimate load 66
EV
3-35 Static-dynamic ultimate load comparison for 150 mm diameter pipe 69
3-36 Static-dynamic ultimate load comparison for 200 mm diameter pipe 69
3-37 Static-dynamic elastic stiffness comparison for 150 mm diameter
restrained joints 70
3-38 Static-dynamic elastic stiffness comparison for 200 mm diameter
PR

restrained joints 70
4-1 Test specimen and actuator configuration for bending testing 74
4-2 Bending test assembly elevation 74
4-3 Location of external instrumentation for bending testing 75
4-4 Typical moment-theta plot with approximated straight-line curves 81
4-5 Moment-theta plot for 200 mm dia.. cast iron pipe 81
4-6 Moment-theta plot for 150 mm dia. ductile iron pipe with push-on joint 82
4-7 Moment-theta plot for 200 mm dia. ductile iron pipe with push-on joint 82
4-8 Moment-theta plot for 150 mm dia. ductile iron pipe with gripper gasket joint 83
4-9 Moment-theta plot for 200 mm dia. ductile iron pipe with gripper gasket joint 83
4-10 Moment-theta plot for 150 mm dia. ductile iron pipe with retaining ring joint 84
4-11 Moment-theta plot for 200 mm dia. ductile iron pipe with retaining ring joint 84
4-12 Moment-theta plot for 150 mm dia. ductile iron pipe with bolted collar joint 85
4-13 Moment-theta plot for 200 mm dia. ductile iron pipe with bolted collar joint 85
4-14 Moment-theta plot for 150 mm dia. steel pipe 86

Reproduced with permission o f the copyright owner. Further reproduction prohibited without permission.
 vii

4-15 Moment-theta plot for 200 mm dia. steel pipe 86

4-16 Moment-theta plot for 150 mm dia. PVC pipe 87
4-17 Moment-theta plot for 200 mm dia. PVC pipe 87
4-18 Moment-theta plot for 150 mm dia. PE pipe 88
4-19 Moment-theta plot for 200 mm dia. PE pipe 88
4-20 Static moment-theta plot for ductile iron pipe with push-on joints 91
4-21 Static moment -theta plot for ductile iron pipe with gripper gasket joints 91
4-22 Static moment -theta plot for ductile iron pipe with retaining ring joints 92
4-23 Static moment -theta plot for ductile iron pipe with bolted collar joints 92
4-24 Static moment -theta plot for steel pipe with lap-welded joints 93
4-25 Static moment -theta plot for PVC pipe with push-on joints 93
4-26 Static moment -theta plot for PE pipe with butt-fused joints 94
4-27 Comparison of static rotational stiffness 94
5-1 Pipe joint capacity chart 100
5-2 Pipe-soil friction transfer chart 100
5-3 Plan of piping system geometry 104
5-4 Diagram of lateral spread displacement distribution 106
5-5 Piping system elements 106

W
5-6 Load pattern distribution 110
5-7 Straight piping system model with soil springs 114
5-8 Joint configuration IE 115
5-9 Laboratory measured load-displacement plots for DIP joints 118
5-10 Laboratory measured typical joint moment-rotation plot 118
5-11 Load-displacement plot for axial soil spring input data 120
5-12 Load-displacement plot for transverse soil spring input data 120
EV
5-13 Applied displacement amplitude pattern on main branch 121
5-14 Applied displacement amplitude pattern on tee branch 122
5-15 Computed main branch nodal displacements along pipe axis
from applied displacements in the 9=0 direction 123
5-16 Computed tee branch nodal displacements along pipe axis
PR

from applied displacements in the 9=0 direction 124

5-17 Computed main branch nodal displacements along pipe axis
from applied displacements in the 9=90 direction 124
5-18 Computed tee branch nodal displacements along pipe axis
from applied displacements in the 9=90 direction 125
5-19 Joint separation for unrestrained joints on the main branch
loaded in the 9=0 direction 128
5-20 Joint separation for unrestrained joints on the tee branch
loaded in the 9=0 direction 129
5-21 Joint separation for unrestrained joints on the main branch
loaded in the 9=90 direction 129
5-22 Joint separation for unrestrained joints on the tee branch
loaded in the 0=90 direction 130
5-23 Joint separation for retaining ring joints on the main branch
loaded in the 9=0 direction 130

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
 5-24 Joint separation for retaining ring joints on the tee branch
loaded in the 0=0 direction 131
5-25 Joint separation for retaining ring joints on the main branch
loaded in the 6=90 direction 131
5-26 Joint separation for retaining ring joints on the tee branch
loaded in the 6=90 direction 132
5-27 Joint separation for gripper gasket on the main branch
loaded in the 6=0 direction 132
5-28 Joint separation for gripper gasket joints on the tee branch
loaded in the 6=0 direction 133
5-29 Joint separation for gripper gasket joints on the main branch
loaded in the 6=90 direction 133
5-30 Joint separation for gripper gasket joints on the tee branch
loaded in the 6=90 direction 134
5-31 Joint separation for bolted collar on the main branch
loaded in the 6=0 direction 134
5-32 Joint separation for bolted collar joints on the tee branch

W
loaded in the 6=0 direction 135
5-33 Joint separation for bolted collar joints on the main branch
loaded in the 6=90 direction 135

loaded in the 6=90 direction

IE
5-34 Joint separation for bolted collar joints on the tee branch

5-35 Number of joints and corresponding separation distance for main branch
136

loaded in the 6=0 direction 136

EV
5-36 Number of joints and corresponding separation distance for tee branch
loaded in the 6=0 direction 137
5-37 Number of joints and corresponding separation distance for main branch
loaded in the 6=90 direction 137
5-38 Number of joints and corresponding separation distance for tee branch
PR

loaded in the 6=90 direction 138

A-l Load-displacement plot for 200 mm cast iron pipe 150
A-2 Load-displacement for 100 mm DIP with push-on rubber gasket joint 151
A-3 Load-displacement for 150 mm DIP with push-on rubber gasket joint 152
A-4 Load-displacement for 200 mm DIP with push-on rubber gasket joint 153
A-5 Load-displacement for 250 mm DIP with push-on rubber gasket joint 154
A-6 Load-displacement for 150 mm DIP with gripper gasket joint 155
A-7 Load-displacement for 200 mm DIP with gripper gasket joint 156
A-8 Load-displacement for 300 mm DIP with gripper gasket joint 157
A-9 Load-displacement for 150 mm DIP with retaining ring joint 158
A-10 Load-displacement for 200 mm DIP with retaining ring joint 159
A -ll Load-displacement for 300 mm DIP with retaining ring joint 160
A-12 Load-displacement for 150 mm DIP with bolted collar joint 161
A-13 Load-displacement for 200 mm DIP with bolted collar joint 162
A-14 Load-displacement for 100 mm steel lap-welded pipe 163

Reproduced with permission o f the copyright owner. Further reproduction prohibited without permission.
 ix

A-15 Load-displacement for 150 mm steel lap-welded pipe 164

A-16 Load-displacement for 200 mm steel lap-welded pipe 165
A-17 Load-displacement for 250 mm steel lap-welded pipe 166
A-l 8 Load-displacement for 150 mm PVC pipe 167
A-19 Load-displacement for 200 mm PVC pipe 168
A-20 Load-displacement for 300 mm PVC pipe 169
A-21 Load-displacement for 150 mm PE pipe 170
A-22 Load-displacement for 200 mm PE pipe 171
B-l Load-displacement Plot for 200 mm cast iron pipe 173
B-2 Load-displacement for 150 mm DIP with push-on rubber gasket joint 174
B-3 Load-displacement for 200 mm DIP with push-on rubber gasket joint 175
B-4 Load-displacement for 150 mm DIP with gripper gasket joint 176
B-5 Load-displacement for 200 mm DIP with gripper gasket joint 177
B-6 Load-displacement for 150 mm DIP with retaining ring joint 178
B-7 Load-displacement for 200 mm DIP with retaining ring joint 179
B-8 Load-displacement for 150 mm DIP with bolted collar joint 180
B-9 Load-displacement for 200 mm DIP with bolted collar joint 181
B-10 Load-displacement for 150 mm steel lap-welded pipe 182

W
B-l 1 Load-displacement for 200 mm steel lap-welded pipe 183
B -l2 Load-displacement for 150 mm PVC pipe 184
B-13 Load-displacement for 200 mm PVC pipe
IE 185
B-14 Load-displacement for 150 mm PE pipe 186
B -l5 Load-displacement for 200 mm PE pipe 187
C-l Dynamic moment-theta for 200 mm cast iron pipe 190
C-2 Static moment-theta for 200 mm cast iron pipe 190
EV
C-3 Dynamic moment-theta for 150 mm DIP with push-on rubber gasket joint 192
C-4 Static moment-theta for 150 mm DIP with push-on rubber gasket joint 192
C-5 Dynamic moment-theta for 200 mm DIP with push-on rubber gasket joint 194
C-6 Static moment-theta for 200 mm DIP with push-on rubber gasket joint 194
C-7 Dynamic moment-theta for 150 mm DIP with gripper gasket joint 196
PR

C-8 Static moment-theta for 150 mm DIP with gripper gasket joint 196
C-9 Dynamic moment-theta for 200 mm DIP with gripper gasket joint 198
C-10 Static moment-theta for 200 mm DIP with gripper gasket joint 198
C-l 1 Dynamic moment-theta for 150 mm DIP with retaining ring joint 200
C-12 Static moment-theta for 150 mm DIP with retaining ring joint 200
C-l 3 Dynamic moment-theta for 200 mm DIP with retaining ring joint 202
C-14 Static moment-theta for 200 mm DIP with retaining ring joint 202
C-l 5 Dynamic moment-theta for 150 mm DIP with bolted collar joint 204
C-l 6 Static moment-theta for 150 mm DIP with bolted collar joint 204
C-l 7 Dynamic moment-theta for 200 mm DIP with bolted collar joint 206
C-18 Static moment-theta for 200 mm DIP with bolted collar joint 206
C-19 Dynamic moment-theta for 150 mm steel lap-welded pipe 208
C-20 Static moment-theta for 150 mm steel lap-welded pipe 208
C-21 Dynamic moment-theta for 200 mm steel lap-welded pipe 210
C-22 Static moment-theta for 200 mm steel lap-welded pipe 210

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
 X

C-23 Dynamic moment-theta for ISO mm PVC pipe 212

C-24 Static moment-theta for 150 mm PVC pipe 212
C-25 Dynamic moment-theta for 200 mm PVC pipe 214
C-26 Static moment-theta for 200 mm PVC pipe 214
C-27 Dynamic moment-theta for ISO mm PE pipe 216
C-28 Static moment-theta for ISO mm PE pipe 216
C-29 Dynamic moment-theta for 200 mm PE pipe 218
C-30 Static moment-theta for 200 mm PE pipe 218

W
IE
EV
PR

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
 LIST OF TABLES
TABLE TITLE PAGE

1-1 Earthquake damage data summary 5

2-1 Test results summary for static axial loading 34
2-2 Joint static axial stiffness values and force levels for restrained joints 35
2-3 Joint static axial stiffness values and force levels for unrestrained joints 36
3-1 Northridge earthquake station record data 48
3-2 Joint dynamic axial stiffness values and force levels for restrained joints 60
3-3 Joint dynamic axial stiffness values and force levels for unrestrained joints 61
3-4 Comparison of dynamic and static axial yield force and elastic stiffness 68
4-1 Joint static rotational stiffness values 89
5-1 Example analysis member types and material behavior 116
5-2 Joint and member axial properties 119

W
5-3 Joint rotational properties fro laboratory results 119
5-4 Loading configurations considered 121
5-5 Resulting maximum nodal displacements along pipe axis 125
IE
EV
PR

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
 1

SECTION 1
INTRODUCTION

1.1 Background

Pipelines transporting water, gas, or volatile fuels are classified as part of the

infrastructure "lifeline" system and are critical to the viability and safety of communities.

Disruption to these lifelines can have disastrous results due to the threat they pose in the

release of natural gas and flammable fuels, or in the restriction of needed water supply

required to fight fires and for domestic use. M. O’Rourke (1996), Iwamoto (1995),

W
Kitura and Miyajima (1996), T. O’Rourke (1996) and other authors have documented

pipeline damage and failures caused by wave propagation of seismic motions, surface
IE
faulting, and by permanent ground deformations resulting from liquefaction and

landslides. Figures 1-1 to 1-8 show examples of joint failures during the Northridge and
EV
Kobe earthquakes. A large number of pipeline failures have occurred at joints due to

pull-out of unrestrained bell and spigot type joints and the fracture and buckling of
PR

welded joints on steel pipes. Singhal (1984) performed testing on 100 mm, 150 mm, 200

mm and 250 mm diameter ductile iron pipe with push-on rubber gasket joints to

determine their structural and stiffness characteristics when subjected to axial pull-out

loads. He showed that the resistance to pull-out of unrestrained push-on joints is quite

low, less than 2 kN (500 lbs) in magnitude, which suggests that the cyclic nature o f the

forces induced by earthquakes and by the resulting ground deformations is an important

design concern for pipelines with unrestrained joints. The use of commercially available

joint restraining devices such as retaining rings, gripper gaskets, and bolted collars can

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
 greatly increase a joint’s capacity to resist pull-out, and therefore, decrease the

probability of joint failure.

The resulting interruption in service and the economic consequences of repair and

replacement of damaged pipe can be severe for communities as well as for pipeline

owners. Some preliminary strategies have been implemented to address the problem of

service disruption. Some pipeline owners are willing to let the inevitable damage occur

and to by-pass the damaged area with temporary flexible hosing until repair to the

pipeline can be made. This strategy is based on two assumptions: 1) the time of

W
disruption until the by-pass can be installed is tolerable, and 2) the redundant lines will

have the capacity to provide vital services. Another strategy employed for seismic
IE
damage mitigation is to develop a long-term program of pipeline upgrade to a more

seismic resistant design. If this is in conjunction with regular replacement of older and
EV
corroded pipes, it may be part of a normal maintenance program and the cost can be

incorporated into an annual maintenance expense. Other pipeline owners may select to
PR

develop a seismic upgrade program for pipelines that still have remaining economic life,

with the cost budgeted in a special seismic upgrade account. In either case, there is a

possibility that the time-span to complete the upgrade may be excessive and the

probability of a major earthquake occurring during this time-span may be high.

However, if pipeline owners were able to assess the damage potential of zones within

their service area, certain portions of their system and corresponding upgrade plans could

be prioritized according to the damage potential which would reduce the probability of

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
 3

major earthquake damage occurring within that zone. A comprehensive program of this

type can help in mitigating potential damage and the consequence of failures.

This report discusses an empirical research project designed to determine the static and

dynamic axial and rotational stiffness and the strength characteristics of a number of

common types of pipe joints, both restrained and unrestrained. It must be recognized that

a pipe joint, especially one with a restraining device, is an assembly of structural

mechanisms, each with highly non-linear properties such as friction sliding, compressive

behavior, tensile restraint, and surface gouging and extrusion. As such, the examination

W
of the behavior of pipe joints requires empirical testing of the joint assembly as a whole.

The results of this testing can help in assessing the response of pipelines to seismic
IE
motion and ground deformation and identify areas of potential damage. The data from

this research can be used in a computer based finite element pipeline system analysis or
EV

in a risk assessment evaluation to determine probable joint failure (see Section 5). A

complete evaluation of the effects of seismic motions on pipelines must also include the
PR

evaluation of the soil-pipe interaction and how strains in the soil are transferred to the

pipe (see Appendix E).

This experimental project included testing of different types of pipe joints and materials

and was divided into three phases: 1) static axial loading, 2) dynamic axial loading, and

3) static and dynamic bending loading. Static axial loading was initially done, not only to

obtain static axial behavior characteristics, but also to get a benchmark o f the maximum

force level capacities of the individual pipe joints so that the dynamic axial testing phase

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
 4

of the project could be properly planned and designed. Since static actuators are able to

deliver a greater level of loading to a specimen than dynamic actuators, static axial

loading was performed on a larger number of pipe joints and diameters, while the

diameters of pipe for axial dynamic loading and bending loading were limited due to the

load capacity limitations of the loading assemblies. The results of the experiments

produced extensive empirical data on the static and dynamic axial and bending stiffness

and failure levels of the specimens tested. They also allowed comparisons between

restrained and unrestrained joints, between different pipe diameters, and between static

and dynamic loadings. However, the characterization of the behavior of joints are limited

W
to the specimens tested and should not be extrapolated to other joint types or pipe

diameters. IE
1.2 Past Performance of Pipelines
EV

Pipeline damage that occurred during recent earthquakes has been well documented. T.

O'Rourke (1996) reviewed the performance and damage of pipelines for various
PR

earthquakes and its effects on different lifeline systems. Table 1-1 summarizes the

amount of damage that occurred to pipelines in some recent earthquakes. In the 1989

Loma Prieta earthquake, the major damage was concentrated in areas of soft soils, such

as in the Mission district in San Francisco. In the San Francisco, Oakland, Berkeley, and

the Santa Cruz areas, there were almost 600 water distribution pipeline failures. In the

1994 Northridge earthquake, over 1400 failures were reported including 100 failures to

critical large diameter lines. In the 1995 Kobe earthquake, 1610 failures to distribution

water mains were reported along with 5190 failures to distribution gas mains.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
 5

Figures 1-1 to 1-8 are field photographs of some typical types o f failures that occurred in

the Northridge and the Kobe earthquakes.

TABLE 1-1 Earthquake Damage Data Summary (O’Rourke 1996)

1989 Loma Prieto Earthquake

San Francisco, Oakland, Berkeley 3SO repairs to water lines
Santa Cruz 240 repairs to water lines
Overall area >1000 repairs to gas lines
1994 Northridge
Los Angeles area 1400 repairs to water lines
107 repairs to gas lines
1995 Kobe Earthquake
Kobe City 1610 repairs to water lines
5190 repairs to gas lines

W
The evidence and documentation shows that earthquakes will cause damage and failure
IE
of pipelines due to transient wave motion and ground deformation, resulting in disruption
EV
to communities and utility services, and risking the life-safety of citizens. Research into

the behavior of pipelines and in particular, pipe joints, both restrained and unrestrained,

must be done in order to understand how and where piping systems fail, and to develop
PR

mitigation methods to reduce the damaging results of earthquakes.

13 Past Research

Extensive research studies have been performed in the past, investigating the effects of

seismic motions and ground deformations on buried pipelines, focusing on the extent and

causes of failures, and the determination of their structural properties. Current testing

conducted by manufacturers has been limited to determining the pressure capacity and

pressure rating of pipes and pipe fittings, and is essentially a proof-testing procedure to a

Reproduced with permission o f the copyright owner. Further reproduction prohibited without permission.
 6

pre-specified level. Past earthquakes have shown that pipelines will fail during seismic

events and ground movements, and that research into pipeline behavior is essential. Past

research can be divided into three areas: 1) review and extent of pipeline damage, 2)

theoretical and analytical evaluation of pipelines and pipe joint behavior, and 3) empirical

testing of pipe joints to determine their structural properties.

Iwamatu et al. (1998) document failures and the failure rate per km in the 1995 Kobe

earthquake. They provide a comprehensive summary on pipeline damage in terms of

pipe material, joint type, and the failure mechanisms that were observed. They also

W
report that the majority of pipeline failures were at the joints, and the predominant modes

of failure were slip-out o f the joints and the intrusion of the spigot into the bell end. They
IE
observed that in steel pipes, failure occurred in the welded joints.
EV
Kitura and Miyajima (1996) document failures in the 1995 Kobe earthquake. They report

that the majority of pipeline failures were at the joints and the predominate modes of
PR

failure were slip-out of the joints and the intrusion of the spigot into the bell end,

especially in small diameter cast iron pipes. These researchers provide a comprehensive

summary on pipeline damage in terms of pipe material type, joint type, and the failure

mechanisms that were observed.

Wang and Cheng (1979) state that “ most literature on pipeline failure due to earthquakes

indicated joints being pulled out and crushed are the most common modes of failures”.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
 7

Trifunac and Todorovska (1997) have a detailed investigation for the amount and types

of pipe breaks occurring during the 1994 Northridge earthquake. They report that the

"occurrence of pipe breaks in those areas during earthquakes can be correlated with the

recorded amplitudes of strong ground motion— In their paper, they note the

distribution of pipe breaks and present empirical equations which relate the average

number of water pipe breaks per km of pipe length with the peak strain in the soil or

intensity of shaking at the site.

T. O'Rouke and Palmer (1996) review the performance of gas pipelines in Southern

W
California over a 61 year period. Statistics are provided for 11 major earthquakes starting

from the 1933 Long Beach earthquake up to the 1994 Northridge earthquake. The paper
IE
states "an evaluation is made o f the most vulnerable types of piping, failure mechanisms,

break statistics, and the threshold of seismic intensity to cause failure, and damage
EV
induced by permanent ground displacements".
PR

Newmark (1967), in a seminal paper on wave propagation in soil, develops the

relationship between seismic motions and the resulting soil strains and curvatures, and

shows that the strains induced in the soil are related to the velocity of the seismic motion

and the shear wave velocity of the soil. This paper is cited by almost all subsequent

research publications that focus on the evaluation of pipeline behavior and earthquakes.

Wang (1979) summarizes the seismic motion and soil strain relationships. Using these

relationships as a basis, Wang develops a simplified quasi-static approach to determine

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
 8

the relative pipeline displacements and rotations, and proposes design criteria and a

methodology to resist seismic wave propagation effects.

Singhal (1984) performed a number o f experiments on rubber gasketed ductile iron pipe

joints to determine their structural and stiffness characteristics. The joints were subjected

to axial and bending static loading for pipes that were encased in a "sand box" that

allowed the soil-pipe interaction and overburden pressures to be included. The author

gives failure criteria in terms of deformations for various sizes of pipes and suggests a

modified joint detail to provide greater deformation capacity. His results showed that the

W
resistance to pull-out of the spigot end from the bell end is low.

IE
Wang and Li (1994) conducted studies on the damping and stiffness characteristics of

flexible pipe joints with rubber gaskets, both axial and lateral, and subjected to dynamic
EV

cyclic loading. They provide expressions for energy dissipation and for equivalent axial

and lateral stiffness.

PR

Reproduced with permission o f the copyright owner. Further reproduction prohibited without permission.