Long-Term Field Performance of Geosynthetic-Reinforced Soil Retaining Walls

Download as pdf or txt
Download as pdf or txt
You are on page 1of 121

Report No.

CDOT-DTD-97-12

LONG-TERM FIELD PERFORMANCE OF


GEOSYNTHETIC-REINFORCED SOIL
RETAINING WALLS

Phillip E. Crouse
Jonathan T.H. Wu
Center for Mechanically Stabilized Backfill Research
University of Colorado at Denver

Final Report
May 1996

Prepared in cooperation with the


U.S. Department of Trausportation
Federal Highway Administration
The contents of this report reflect the views of

the author who is responsible for the· facts and

the accuracy of the data presented herein. The

contents do not necessarily reflect the official

views of the COlorado Department of Transportation

or the Federal Highway Administration. This report

does not constitute a standard, specification, or

regulation.

i
REPORT DOCUMENTATION PAGE FORM APPROVED
OMB NO. 0704-0188
Public reporting bu.'1ien for this collection of infonnation is estimated to average 1 hour per resporu;c,including the time for reviewing instructions, searching existing data sources,
gathering and Ir.aintaining the data needed I and completing and reviewing the collection of infonnation. Send comments regarding this burden estimate or any other aspect of this
collection of infonnation. including suggestions for reducing this burden, to Washington Headquarters Services, Directorate for Infonnation Operations and Reports, 1215 Jefferson
Davis Highway, Suite 1204, Arlington, VA 22202-4302, and to the Office of Management and Budget, Paperwork Reduction Project (0704-0188), Washington, DC 20503.
1. AGENCY USE ONLY (Lea~" Blank) 2. REPORT DATE 3. REPORT TYPE AND DATES COVERED

May 1996
4. TI1LE AND SLlJ1TI'LE S. FUNDING NUMBERS

Long-Term Field Performance of Geosynthetic-Reinforced Soil Retaining Walls


6. AumORS(S)

Phillip E. Crouse and Jonathan T. H. Wu


7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(S) 8. PERFORMING ORGANIZATION
REPORT NUMBER
Center for Mechanically Stabilized Backfill Research CDOT-DTD-97-12
University of Colorado at Denver
9. SPONSORINGIMONITORING AGENCY NAME(S) AND ADDRESS(S) 10.SPONSORING~ON~RING

Colorado Department of Transportation AGENCY REPORT NUMBER


4201 E. Arkansas Ave. CDOT-DTD-97-12
Denyer, Colorado 80222
11. SUPPLEMENTARY NOTFS

Prepared in Cooperation with the U.S. Department of Transportation, Federal


Highway Administration
12a. DISTRm1:TION/AVAILABILITY STATEMENT 12b. DISTRIBUfION CODE

No Restrictions: This report is available to the public through the


National Technical Infonnation Service. Springfield, VA 22161
13. ABSTRACT (Maximum 200 wOrds)

A study was undertaken to synthesize field long-term performance data of full-scale geosynthetic-reinforced soil (GRS) retaining walls. Upon
conducting an extensive literature review and survey, seven GRS retaining walls of which the performance had been monitored for extended periods
of time were selected for this study. To assess the wall performance, a conseratism index (CI) was dermed tu quantify the relative degree of
conservativeness of the walls. In addition, a simple analytical equation was developed for predicting creep deformation of a GRS wall beyond
the measurement period. The analytical equation was derived based on the syuthesized behavior of the GRS walls that the logarithmic creep rate
decreased linearly with logarithmic time.
This study conclusively indicated that all the GRS retaining walls with granular backfill deformed very little due to creep and were stabilizing
with time. The current design methodology to account for creep is overly conservative when well-compacted granular backftll is employed.
Using results of a soil-geosynthetic composite performance test in conjWlction with the analytical equation, long-term creep deformation of a
GRS wall under project specific conditions can be predicted in a rational manner throughout its design life.

14. SUBJECT TERMS 15. NUMBER OF PAGES

Creep deformation 112


Geosynthetic 16. PRICE CODE

Retaining Wan
17. SECURTm" CLASSIFICATION 18. SECURITY CLASSIFICATION 19. SECURITY CLASSIFICATION 20. LIMITATION OF ABSTRACT
OFREP(,RT OFTmSPAGE OF ABSTRACT

Unclassified Unclassified Unclassified


CONTENTS

Chapter
1. Introduction ......................................................................................................... 1
1.1 Background .......................................................................................... 4
1.2 Research Need .................................................................................... 5
1.3 Research Objectives ............................................................................ 6
1.4 Report Organization ............................................................................. 7
2. Uterature Review and Survey of Creep
Performance in GRS Retaining Walls ................................................................ 8
2.1 Project Descriptions ............................................................................ 11
2.1.1 Interstate Highway 70 through Glenwood Canyon Project ........... 12
2.1.2 Tanque Verde - Wrightstown - Pantano Roads Project .................. 14
2.1.3 Norwegian Geotechnical Institute Project ....................................•. 15
2.1.4 Japan Railway Test Embankment Project ........................................ 16
2.1.5 Highbury Avenue Project ................................................................... 16
2.1.6 Federal Highway Administration Research Project ......................... 17
2.1 .7 Seattle Preload Fill Project .................................................................. 1.7
2.2 Design ApproQch Evaluation ............................................................ 18
2.2.1 External and Internal Stability ............................................................ 18
2.2.2 Lateral Forces .......................................................................... ........... 19
2.2.3 Reinforcement Tensile Strength ........................................................ 21
2.2.4 Partial Factors of Safety ..................................................................... 29
2.2.5 Facing Rigidity ...................................................... ... ............................. 1
3. Project Long-Term Performance ...................................................................... 34
3.1 Instrumentation and Measured Parameters .................................... 34
3.2 Reinforcement Strains and Wall Movemenl.. ..................................35
3.2.1 Interstate Highway 70 through Glenwood Canyon Project ........... 35
3.2.2 Tanque Verde - Wrightstown - Pantano Roads Project .................. 38
3.2.3 Norwegian Geotechnical Institute Project ..................................... .41
3.2.4 Japan Railway Test Embankment Project....................................... .43
3.2.5 Highbury Avenue Project .......................................................... .........45
3.2.6 Federal Highway Administration Research Project ........................ .45
3.2.7 Seattle Preload Fill Project ................................................................ .46
3.3 Conservatism Index............................................................................ 51
3.3.1 Creep-Rate and the Creep Modulus ............................................... 53
4. An Approach to Estimating Creep Using a Laboratory Test .......................... 58
4.1 Creep in Labaratory Tests .................................................................. 58
4.1.1 Laboratory Creep Test Descriptions ................ .................................. 59
4.1.2 Laboratory Test Creep Rate .............................................................. 64
4.2 Laboratory and Full-Scale Creep Rate Comparison ...................... 67
4.3 An Analytical Solution for Estimating Creep Strain ............... .......... .70
5. Summary and Conclusions ............ .................................................................. 34
5.1 Summary ............................................................................................ 73
5.2 Conclusions .......................................................... ............................... 73
5.3 Recommendations for Future Study ....................................... .......... 74
Appendix
A. Project Description Sheets ......................................................................... 76
B. Conservatism Index Calculation Brief ....................................................... 84
C. Plots used to compute the creep modulus ....................................•........ 97

References ................................................................................................. 108


TABLES
Table
2.1. Selected Full-Scale Field GSR Retaining Wall Projects ...................... 9
2.2. Reinforcement Tensile Strength in Selected Projects ..................... 22
3.1 . CRC, Reintorcement Strain and Wall Movement for the
Selected Projects ............................................................................... 37
3.2. Parameters Used to Compute the Conservatism Index ................. 52
3.3. Creep Modulus for the Selected Projects ........................................ 57
4.1. Creep Modulus for the FUll-Scale Walls and Laboratory Tests ....... 69
FIGURES
Figure
1.1. Components of a GRS Retaining Wall ............................................... 3
2.1. GRS Retaining Wall Project Location Map ....................................... 10
2.2. Wall Profiles for Selected GRS Retaining Wall Projects ................... 13
2.3. Forces Using the Tie-Back Wedge Method ...................................... 20
2.4. Parameters of a Geosynthetic Stress-Strain Curve .......................... 24
3.1. Reinforcement Creep-Time Curves for the Tanque - Verde-
Tanque - Wrightstown - Pantano Roads Project ............................. .40
3.2. Reinforcement Creep-Time Curves for the Norwegian
Geotechnical Institute Project .......................................................... 42
3.3. Tensile Force in Reinforcement and Displacement for the Japan
Railway Test Embankment Project .................................................... 44
3.4. Reinforcement Creep-Time Curve for the
Highbury Avenue Project ................................................................... 48
3.5. Reinforcement Creep-Time Curve for the FHWA Project .............. .49
3.6. Reinforcement Creep-Time Curve for the Seattle
Preload RII Project .............................................................................. 50
3.7. Creep-rate-Time Curve Illustrating the Creep Modulus .................. 54
3.B. Creep-Rate-Time Curve for the Selected Projects .......................... 55
4.1 . Schematic of the Long-Term Soil/Geosynthetic Performance Test
Device (Helwany and Wu, 1996) ..................................................... 61
4.2. Schematic of the Modified Soil/Geosynthetic Performance Test
Device (Ketchar! and Wu 1996) ...................................................... 62
4.3. Creep-Rate-Time Ratio Plot ............................................................... 70
1. Introduction
Retaining walls have become an increasingly popular method for
retaining earth to accommodate worldwide development of transportation
and other structural systems. Conventional gravity and cantilever retaining
walls that extemally resist lateral earth pressure can be costly and difficult to
build because of their large rigid mass. However. a new type of retaining
wall is available that derives its stability from within the backfill (i.e .• is
internally stabilized) and is demonstrating distinct advantages over
conventional retaining walls.

In France. H. Vidal introduced modem applications of soil-reinforced


retaining walls in the 1960s (Vidal. 1966) using metal strips for reinforcement.
The idea of intemally stabilizing soil is to strengthen the soil mass by the
inclusion of planar reinforcement whose function it is to restrain the
development of tensile strain in the direction of the reinforcement.
Reinforcement can be inextensible (e.g .• metals) or extensible (e.g ..
geosynthetics) . Since 1980. geosynthetics have been used for reinforcement
due to their flexibility and low cost. Soil reinforced with geosynthetics is
referred to as geosynthetic reinforced soil (GRS). Some of the advantages of
GRS retaining walls over conventional retaining structures include:

• Their flexibility allows greater tolerance to foundation


settlement;
• Construction of GRS walls is rapid and requires only
"ordinary" construction equipment; and
• GRS retaining walls are generally more economical than
conventional retaining walls.

The primary components in a GRS retaining wall include the


reinforcement. wall facing. reinforced soil backfill. retained soil. and
foundation soil. Figure 1.1 illustrates these components in a typical GRS
retaining wall.

Since the development of GRS technology, researchers have


identified three characteristics that are not well understood when reinforcing
soil with geosynthetic material. These include:

• Lateral earth pressure distribution;


• Failure surface; and
• Creep.

This study focuses on creep in a GRS retaining wall. Lateral earth


pressure distribution and the failure surface have been addressed by several
other researchers and is ongoing.

2
SURCHARGE

RETAINED SOIL

REINFORCEMENT (TYP.)

FOUNDATION

Figure 1.1
Components of Q GRS Retaining Wall

3
1.1 Background

Since the mid-1980s researchers have attempted to characterize the


long-term behavior of GRS retaining walls. The overall research objective has
been to understand their long-term behavior to guide the development of
rational methods of analysis and design. Although this has been the overall
objective, researchers have approached the problem from three different
aspects:

• Instrumenting full-scale GRS retaining walls;


• Soil/geosynthetic composite laboratory creep tests; and
• Element laboratory creep tests of geosynthetics.

Since the 1960s numerous full-scale GRS retaining walls have been
built and instrumented to quantify their performance. However, these walls
typically were monitored for relatively short periods of time due to financial
constraints and/or instrumentation damage. Since the late 1980s researchers
have built a few full-scale GRS retaining walls that have been monitored for
extended periods of time to quantity their long-term performance. The
results from these instrumented walls have been individually documented,
but have never been investigated in a unified manner.

In 1994 a soil/geosynthetic composite laboratory creep test was


developed by Wu (1994a) and Wu and Helwany (1996) to characterize the
complex behavior of the soil/geosynthetic composite. The test simulates the
composite by transferring stresses applied to the soil in a manner similar to
the typical load transfer mechanism in a GRS retaining wall. Ketchar! and
Wu (1996) continued the research by developing a simple test procedure to
assess the long-term behavior of GRS walls and tested various soils and
reinforcement materials under different conditions.

4
The current state of practice is to account for creep by performing a
creep test on the reinforcing element. Laboratory tests such as the
procedure contained in the American Society for Testing and Materials
(ASTM) D5262 Test Method entitled "Tension Creep Testing of Geotextiles" is
used to determine a creep-limited strength of the reinforcement elements.
The test consists of applying a constant load for a minimum duration of
10.000 hours to an eight-inch-wide specimen. Because of the obvious time-
constraint of the test. estimated creep-limited strengths are typically used for
GRS retaining wall designs instead of performing the actual test. The creep-
limited strength is computed by applying a creep reduction coefficient
(CRC) or partial factor of safety to the geosynthetics' short-term strength.
Current American Association of State Highway and Transportation Officials
(AASHTO) design methods recommend reducing the short-term strength by
as much as 20 to 80 percent to account for creep.

The fundamental assumption in using results from geosynthetic creep


tests is that the soil/geosynthetic composite wall will behave the same as the
reinforcement element. However. results from full-scale and laboratory tests
have reveoled that the geosynthetics perform significantly better when
confined in GRS walls than predicted by the element creep tests due to
stress redistribution in the soil/geosynthetic composite. Because of this
discrepancy. current design methods are overconservative and are inhibiting
the development of GRS technology.

1.2 Researc:h Need

Since geosynthetics are creep-sensitive materials. designers are


concemed about providing adequate margins of safety to account for
creep in permanent GRS retaining wall applications. This. along with the lack

5
of quantitative long-term performance data has led to the misunderstanding
of the complex behavior of the soil/geosynthetic composite resulting in
overconservative designs. Therefore, the first research need is to compile
existing, quantitative, long-term performance data, from full-scale, well-
instrumented GRS retaining walls. The second research need is to develop a
rational method for estimating creep for the design life of the structure based
on the creep behavior of the soil/geasynthetic composite instead of the
geosynthetic element alone.

1.3 Research Objectives

The three main research objectives include:


1. Compile long-term performance data from field projects involving
well-instrumented GRS retaining walls;
2. Develope a means to quantify the conservativeness of the designs;
and
3. Develop a rational method to estimate creep based on laboratory
creep test of the soil/geosynthetic composite deformation

To meet the first objective, the following tasks were performed:

• An extensive literature search was performed to determine what


projects could be used for the study of long-term performance;
• A request for information was sent to experts in GRS technology;
• Specific projects were selected for the study; and
• Specific design and performance data from the selected projects
were compiled and summarized.

To meet the second objective, the following tasks were performed:

• The actual or design creep reduction to the reinforcements' tensile


strength is compared to reductions recommended by AASHTO;
• A conservatism index (CI) was developed to quantify the
conservativeness of the design and;

6
• A simple procedure was developed to predict creep using a simple
laboratory test and analytical equation can be used to predict
creep
To meet the third objective, the following tasks were performed:

• The laboratory test procedure used to model the creep behavior of


the soil/geosynthetic composite was described;
• The laboratory creep tests were validated using the performance of
the selected projects; and
• A rational procedure was developed using the laboratory test and
analytical equation to estimate creep for the design life of a GRS
retaining wall.

1.4 Report Organization

Chapter 1 presents the introduction, background. research needs and


research objectives. Chapter 2 describes the projects selected from the
literature survey. Chapter 3 describes the design and long-term
performance of the selected projects. Chapter 4 describes the method to
estimate creep using the laboratory soil/geosynthetic model creep tests.
Chapter 5 describes the conclusions and recommended future research.
Appendix A contains the selected project descriptions. Appendix B contains
the conservatism index computation and Appendix C contains the graphs
used to compute the creep modulus.

7
2. Literature Review and Survey of Creep Performance In GRS Retaining Walls

Since the 1960s researchers have built and instrumented numerous


full-scale soil-reinforced retaining walls to quantify their performance.
However. due to financial constraints and/or instrumentation damage.
researchers could monitor the wall performance for only relatively short
periods of time. In the 1980s. transportation officials began using GRS
retaining walls for highway and railway renovation projects and sponsoring
research in GRS technology. With support from the transportation resources.
researchers installed instruments in some of these walls to monitor their long-
term performance under actual service and field conditions.

In this study. an extensive literature review and survey was conducted


to collect information on the projects that used GRS retaining walls that had
been monitored for extended periods of time (i.e .. greater than six months) .
A survey was developed and sent to 10 internationally renowned experts to
obtain information on GRS projects under their direction. From the literature
review and survey. seven GRS retaining wall projects were selected. These
projects typically had well-documented. long-term reinforcement strain
data. wall deformation data. and design data. The projects selected are
listed in Table 2.1. The locations of the projects are illustrated on Figure 2.1 .

8
Table 2.1
Selected Full-Scale Field GRS Retaining Wall Projects

MonHoring Principal
Proled Construded Duration location Researcher
Interstate Highway 70
Glenwood Springs.
through Glenwood 1982 7 months R. Barrett
Colorado. USA
Canyon
Tanque Verde-
Tucson. Arizona.
Wrightstown-Pantano 1985 7 years J. Collin
USA
Roads

Norwegian Geotechnical
1987 4 years Oslo. Norway R. Fannin
Institute

Japan Railway Test


1987 2 years Tolcyo. Japan F. Tatsuoka
Embankment

Highbury Avenue London. Ontario.


2 years
I 1989
Canada
R. Bathhurst

Federal Highway Algonquin. Illinois.


1989 1.3 years M.Simac
Administration USA

Seattle.
Seattle Preload Fill 1989 1 year l. Allen
Washington. USA

9
Hlahbury Avenue Pro/oct 'orwoalan Geotechnlcol Inltltut. Prolec'
~attl. Preload Fill Pro .ct
l.ooatIofI Il..ondorI. OrMrto, Conodo '-0'''''' 0.10, tto."woy

_""
LoccrUon SeottM. WQ.hI~on. U"
Rt!nlorcem.n\ Geottltt..
R.w~ c.ovrw RWoI", .... ,MI ...
HMght 12J..l FNt
~ 15.7 fMl
HtIgM 141.3 r..t
.....
Built
0010
I Way.
II Year
'181
o.o.mIw, '181
.... Jvly.
4 Yeo,..
,.87

o q
~~VSo\
rid \1"
~d~
20 FNt
July. 1989
1.3 Y«Jf1I
-=~
f)
Tonque Verde-Wrfghtalown-Pantono
Roadl Project - ----- II:J
locGtlon !ruteofl, Mzano, USA
Inte",tat. Highway 70 Ja.an Railroad Teol Embanlcme-:
R,WOfument GfIoOttd
Hlli9ht I II r...
through Glenwood Conyon '--"" ~•

.........
Locatlon ~ ~ o.r... ~ R.lnfo~mtnt c.o.,1d
~ber, 1985
~lnforcelMnl Qeolextlle Hei9h1 18.4 feet
7 Yea,.
Hel'11t 18 FoNt Buill Januo!),. 1088
Bunt. lt42. Data 2 YMn
DolG 7 WonlM

figure 2,'
GRS RetaInIng Wall project LocatIon Mop
The walls built for each project represent a variety of GRS retaining
walls. The walls range from 15 feet to over 40 feet in height and typically
include surcharge loads comprised of earth fills or highway loads.
Reinforcement materials consist of polypropylene or polyester geogrids and
geotextiles ranging in short-term strength from 400 to over 12.000 Ib per foot
width. The faGing used on the walls consists of concrete modular blocks and
panels or exposed surfaces. Some of the walls are constructed on poor
foundations while others are constructed on competent foundation
materials. The environmental conditions vary from freezing temperatures in
Ontario, Canada, to temperatures up to 11 10 Fahrenheit for walls built in the
state of Arizona, USA.

Although the selected projects consist of a variety of GRS retaining


wall types, all the walls performed exceptionally well. The maximum strains
measured in the reinforcement in all cases were less than five percent. In
some cases, the designs predicted strains of 40 to 60 percent. In other cases,
the walls were designed to fail, yef failure could nof be achieved. The
following secfions provide a brief descripfion of fhe selected projects and
design approach. Chapfer 3 provides fhe performance evaluation.

2.1 Project Descriptions

The following secfions provide a brief overview of fhe projecfs


selecfed from fhe literafure review and survey. The GRS refaining walls builf
for each project are illusfrated on Figure 2.2. Selected project information is
provided on project descripfion sheets in Appendix A. The projecf
description sheefs include informafion such as the wall componenfs (Le.,
confining soil, facing, and reinforcement type). reinforcemenf strength.
surcharge, and schedule showing dates of milesfone events such as the

11
beginning of conslruction, surcharge loading, and moniloring period. A
schemalic of Ihe relaining wall(s} and project pholographs are also included
on Ihe projecl descriplion sheels.

2.1.1 Interstate Highway 70 through Glenwood Canyon Project

In April of 1982, Ihe Colorado Departmenl of Highways designed and


conslructed a series of inlernally reinforced walls for Ihe Inlerslale Highway
70 project through Glenwood Canyon. The Glenwood Canyon follows Ihe
Colorado River Ihrough Ihe scenic Roclcy Mounlains of Colorado, USA, near
Ihe city of Glenwood Springs. The relaining walls were buill over highly
compressible sills and cloys allhe bose of Ihe canyon. Because of
archilectural and environmenlal conslrainls, transportalion officials lesled a
series of inlemally reinforced relaining walls including a reinforced earth wall.
relained earth wall (VSL), a wire-mesh reinforced wall. and a geolexlile-
reinforced wall. The geolexlile reinforced wall was one of Ihe first full-scale
GRS walls conslructed in Ihe USA.

The performance of Ihe GRS relaining wall was observed for several
yeors; however, quanlilalive performance dala was documenled for only
Ihe first seven monlhs of service. The wall was designed 10 delermine Ihe
lower slability limils of a GRS relaining wall. Iherefore geolexliles having
relalively low lensile strenglhs (Le., 400 10 900 Ib/H) were used for Ihe
reinforcement. In June, 1983, a 15 fool high surcharge was applied 10 Ihe
top of Ihe wall in an altempllo collapse Ihe wall. However, failure never
occurred.

12
L SURCtWlGE

TOP OF
TEI.lPORARY ROAD -V-" "NCI.ND
' ' ' / 34' SURCHMGE

--
Wall Panel 26-30
RETAINED SOIL
1IWFIO . . . . . .

,....,'
I• 16.4'
• federal Highwgy Adm;ni:strot;QO project
Wall Na . 9
Interstgte Hjghway 70 through Glenwgod Canyon Hjgbbyrv Ayenye project AlgoOQuin. Illinois. USA
. Geotextjle Egrth Retgjnjng Wall London OntgriQ. Cgngdg
GlenwgQd Sorings. CQIQradQ. USA '2'
ImN"OftCDIENT (TYP.)

Wall Panel 26-32


Tongue Verde-Wrightstown-pgntQno Roads PrQject
TycsQn. Arizong. USA

.----1--+....:::...."""""" (M>.)

CONCfIIfTE PANELS

'-'' ' ' 'IED SOlI.

Jgpan Rgilway Teat Embgnkment proj@ct


Norwegjgn Geotechnjcgl Instjtute
JR Embgnkment No. 1
RElNf'ORCajENT Skedsmo. Norway Experiment stgtion Qf Japgn Rgilwgy Technicgl Resegrch Instjtyte
(T'/P.)
32' TQkyQ. Japgn

Segttle PrelQad All Project


Southeast Wgil
Scottie, WgshingtQn. USA

figure 2.2
Wall Profiles for Selected GRS Retaining Wall prolects

13
In 1983 and 1993. samples of the reinforcement were exhumed to
determine the survivability and durability of the reinforcement (Bell and
Barrett. 1994). The strength of the exhumed reinforcement was compared
with that of archive samples. The results of the test are described in Chapter
3. Additional project information can be found in Federal Highway
Administration (FHWA) Report No. CHOH-DTP-R-86-16 entitled "Evaluation of
Fabric Reinforced Earth Wall" (Derakhsandeh and Barrett. 1986).

2.1.2 Tanque Verd. - Wrtghtstown - Pantano Roads Project

In 1984 and 1985. 46 GRS retaining walls were constructed in the city
of Tucson as part of the Tanque Verde Grade Separation Project. In
September of 1985 two of the walls were instrumented (Wall Panels 26-30
and 26-32) to monitor their performance during and after construction.
Approximately seven years of performance data have been published for
the two instrumented walls (Collin. Bright. and Berg. 1994). The original
design and instrumentation information is contained in an Federal Highway
Administration (FHWA) report entitled "Tensor Geogrid-Reinforced Soil Wall"
(FHWA. 1989). Other papers have been written by Berg. Bonaparte. Anerson.
and Chouery (1986) and FIShman. Desai. and Sogge (1993) describing the
construction and performance of the walls.

The city of Tucson is located in the southem part of the state of


Arizona. USA. in the Sonora desert where summer temperatures can reach as
high as 111 0 Fahrenheit. Soil temperatures within the wall reached as high as
970 Fahrenheit. Elevated temperature environments for geosynthetics were a
potential design concem since the high temperatures may accelerate
mechanisms of degradation. Similar to the Colorado project. reinforcement
samples were exhumed after 11 years of service to examine the durability of

14
the reinforcement (Bright. Collins and Berg, 1994) which is described in
Chapter 3.

2.1.3 Norwegian Geotechnical Institute Project

In 1987, the Norwegian Geotechnical Institute (NGI) built a full-scale


GRS retaining test wall in Skedsmo, Norway. The purpose of the wall was to
establish characteristics of creep in the reinforcement. Skedsmo is located
near the city of Oslo, Norway, in northern Europe. The climate at Oslo is
moderate with temperatures ranging from 380 Fahrenheit in the winter to 64 0
Fahrenheit in the summer. Rainfall can be heavy at times with approximately
40 inches of rainfall annually.

The wall was instrumented in two sections, 'J' and 'N', each with a
different arrangement and spacing of the reinforcement. Approximately
four years of performance data have been published for the two
instrumented sections (Fannin and Herman, 1992). Following construction, the
wall was monitored for approximately four weeks under self-weight loading.
Thereafter, the top of the wall was cyclically loaded by using water tanks
that applied a maximum contact pressure of 6,000 Ib/ft2. After
approximately two months of cyclic loading, the tanks were removed and a
permanent 1O-foot-high surcharge was placed on top of the wall applying a
uniform and sustained pressure of 10,000 Ib/ft2.

The original design and instrumentation information are contained in


the paper entitled "Geosynthetic Strength - Ultimate and Serviceability Limit
State Design" by Fannin and Hermann (1992). An additional paper
describing the project Fannin and Hermann, (1990) has also been published.

15
2.1.4 Japan Railway Test Embankment Project

Two test embankments-were constructed at the Experiment Station of


Japan Railway Technical Research Institute near Tokyo, Japan. The test
embankments were part of a series of embankments constructed with sand
and Tokyo's sensitive clays in the 1980s to develop an intemal reinforcing
system that could withstand its heavy precipitation events (Tatsuoka,
Tateyama, Tamura, and Yamauchi). The first test embankment (JR Number
I) was backfilled with sand while the second embankment (JR Number 2)
was backfilled with clay. JR Number 1 was selected for this study.

JR Number 1 was constructed in 1988 to evaluate the stability of GRS


embankments with rigid facing. Instruments were installed during
construction and monitored for approximately two years until 1990, when it
was loaded to failure. The facing consisted of rigid cast-in-place concrete
panels installed in five wall segments. One wall segment consisted of
discrete panel squares for comparison with the rigid panels. The overall
project infonmation can be found in a paper written by Tatsuoka, Murata,
and Tateyama (1992) entitled "Permanent Geosynthetic-Reinforced Soil
Retaining Walls used for Railway Embankments in Japan".

2.1.5 Highbury Avenue ProjKt

The Royal Military College of Canada has published several papers


documenting the long-tenm performance of a GRS retaining wall used in
reconstructing and widening Highbury Avenue in London, Ontario, Canada.
The wall was instrumented during construction in late 1989. Approximately 2
years of performance data have been published through August of 1991
(Bathurst. 1992). The research objective for the project was to collect
performance data from a well-instrumented in-service GRS retaining wall to

16
evaluate its long-term performance. Additional information can be found in
the paper by Bathurst (1992) entitled " Case Study of a Monitored Propped
Panel Wall".

2.1.6 Federal Highway Administration Research Project

From 1984 to 1989. the FHWA sponsored several soil reinforcement


research projects at its stone quarry in Algonquin, Illinois, USA. One project
consisted of building a wall referred to as 'Wall 9". The wall was built to
quantify the long-term behavior of continuous filament polyester geogrid
reinforcement and dry-stacked, soil filled facing units (Simac. Christopher
and Bonczkiewicz. 1990). The test wall was constructed with a very low
factor of safety to evaluate the applicability of existing design methods. The
internal stresses were monitored for three months. then an inclined surcharge
approximately seven feet high was place and monitored for approximately
1.3 years.

2. 1. 7 Seattle Preload RII Project

In March of 1989. the Washington State Department of Transportation


designed and supervised the construction of a series of GRS retaining walls to
provide a preload fill in an area of limited right-of-way located in Seattle.
Washington. USA. The tallest wall (southeast wall) constructed for the project
had a height of 41.3 feet and supported 17.4 feet of surcharge fill. Since this
wall was significantly higher than any previously constructed wall.
instrumentation was installed to monitor its performance. The wall was
monitored for approximately one year after which it was demolished.
Specific design information can be found in the paper entitled "Performance
of a 12.6 m High Geotextile Wall in Seattle. Washington" (Allen. Christopher.
and Holtz. 1992).

17
2.2 Delign Approach Evaluation
This section summarizes the approach used to design the GRS
retaining walls selected for the study previously described. The purpose for
evaluating the design approach is to illustrate how the current
methodologies address design considerations such as extemal and intemal
stability. creep. construction damage. and biological degregration of the
reinforcement. Each of these considerations add conservatism to the
design. When the conservatism from each of these design considerations is
combined. the GRS retaining wall design can be grossly overconservative.

2.2.1 Extemal and Intemal Stability

The design consideration for external stability is satisfied when there is


an adequate safety margin for failure due to sliding. foundation bearing and
overall slope failure. Similar to the design approach for conventional
retaining walls. extemal stability is based on limit equilibrium analysis where
destabilizing forces (e.g .• lateral earth pressure) against the reinforced soil
mass are resisted by stabilizing forces (e.g •• reinforced soil mass weight and
external forces) with adequate margins for safety. Internal stability is satisfied
when the wall is sufficiently stable against failure within the reinforced soil
mass. External stability design methods are well understood and are
therefore not addressed in this study. However. internal stability design
methods for GRS retaining walls have not been well-established and can
vary from one design to another.

The retaining walls selected for this study were designed using a
commonly used design approach. In general. the internal stability of the
selected walls was satisfied using an ultimate-strength approach based on
the method of limit equilibrium. The ultimate-strength approach applies

18
factors of safely to the ultimate strength of the materials (i.e •• soil.
reinforcement and facing) or to the computed quantities (i.e .• forces and
moments) or to both the ultimate strength and calculated quantities (Wu.
1994b). The specific quantities and strength parameters include:

• Lateral forces from the surcharge. reinforced soil mass and retained
soil;
• Reinforcement tensile strength; and
• Facing rigidily.

Due to the lack of reliable empirical data. somewhat arbitrary factors


of safely are used. which have resulted in overconservative designs. The
following subsections describe how the quantities. strength parameters and
associated factors of safely were determined for each project.

2.2.2 Lateral Forces

Lateral forces on a GRS retaining wall can be described by two


important characteristics. The first characteristic is the location of the failure
surface. The second is the lateral earth pressure distribution providing the
driving forces. As mentioned previously. these two characteristics are being
studied by others.

In general. the retaining wall designs in the selected projects assumed


a Rankine planar failure surface through the reinforced mass. The part of the
reinforcement that extends beyond the assumed failure wedge is
considered to be tension-resistant tiebacks (frequently referred to as the tied-
back wedge method) as illustrated on Figure 2.3. The tie-back wedge
method of analysis assumes that the shear strength of the reinforced soil
mass behind the wall is fully mobilized and thus active lateral earth pressures
are developed.

19
GEOSYNTHETIC REINFORCEMENT

RANKINE FAILURE SURFACE

--
-- --
FAILURE WEDGE-

- -- -- --
---
RESISTING
TIE-BACK FORCE
-- -- -- --
- --
-- --- --- -- --
- --- - - -- -
-- -- --- - -- -- -- ---
- -- -- -- -- -
-- - -- -- -- -- -- -- -- --
figure 2,3
Forces Using the TIe-Back Wedge Method

20
The second characteristic is the assumed lateral earth pressure
distribution. Typical lateral earth pressure distributions such as the linear
Rankine surface typically overestimate the lateral force on the reinforced soil
mass adding conservatism to the designs. Clayboum and Wu (1993)
compared six design methods and revealed that there are very significant
discrepancies in the factors of safety for various design methods due to
varying earth pressure distributions. In a typical wall examined in that study,
the combined factors of safety ranged from 3 to 23, depending on the earth
pressure distribution used. Typically, a linear Rankine lateral earth pressure
distribution was assumed for the selected projects. In most cases an active
condition was assumed. However. the Interstate Highway 70 through
Glenwood Canyon project design assumed "at resf' conditions.

2.2.3 Reinforcement Tensile Strength

In the tie-back wedge method of analysis. the lateral earth pressures


are resisted by the tensile strength of the reinforcement. This is the design
component that is adjusted to account for creep since geosynthetics are
comprised of creep-sensitive polymers. The adjustments include reducing
the short-term tensile strength to account for creep and then further
reductions to account for construction damage and biological degradation.
The strength. adjusted for creep, is referred to as the creep-limited strength.
The creep-limited strength adjusted for construction damage and biological
degradation is referred to as the design-strength. The short-term, creep-
limited, and design tensile strengths for the types of reinforcement used in the
selected projects are summarized in Table 2.2. Each type of reinforcement
strength is described in the following sUbsections

21
Table 2.2
Reinforcement Tensne Strength In Selected Projects

Average Cr••p Cr. .p·


Numb.ral R....lorc ...... nI Shorl· T.rm R.,ducllon UmIIod DOIlgn-
Wall Height Relnforc:ement Spacing Strength cc>.mcl.nt str.ngth Str.ngth
ProJoct Name (It) Layers (tt) (Ib/tt) (%) (Iblll) (Ib/II)
Interstate Highway 70 Ihrough Geotextlle Earth
16 17 0.9 400-1150 401055 220 to 3.40 220 to 340
Glenwood Canyon Relaining Wan

Wai Panel 26-30 15.6 10 1.6 5400 37 1933 1327


anque Yerde-Wrlghlstown-
Pantano Roads
Wall Panel 26-32 16.1 10 1.6 5400 37 1933 1327

~ WaII5ectlon J 15.7 4 2.2 833-3600 NA NA 833-3600


Norwegian Geotechnical
insmule
Wall Section N 15.7 8 2 833-3600 NA NA 833-3600

Uapan Railway Test JR Embankment


16.4 17 I 1880 NA NA 1880
Embankment No, I

HIghbury Avenue HIghbury Ave. Wall 23.3 9 0.4 2000-3450 NA NA 2000-3450


!federal Highway
Wall No,9 20 8 2.5 2604 60 1560 1032
Adminlstrallon

fSeattle Preload Fin Soulheast Wall 41.3 33 1.25 2066-12400 ,10-59 689-/l:n7 689-/l:n7 I
L _ ----- _. - L .. _____ --

NA ~ Not available In the Hteralure


2.2.3.1 Short·Term Strength

The short-term tensile strength of the geosynthetic reinforcement is


determined by applying a tensile load to an unconfined or confined test
sample at a constant strain-rate until failure occurs. During the loading
process, both load and displacement are measured to obtain a stress-strain
curve as illustrated on Figure 2.4.
The maximum tensile stress is typically referred to as the ultimate stress
or short-tenm stress. The strain at failure is typically referred to as the
maximum strain. Stress is typically measured in load per unit width and the
strain is computed by dividing the elongation by the original specimen
length. These values are illustrated on a typical stress-strain curve on Figure
2.4.

23
SHORT-TERM,..'
.,.-- ....... ......
b
en
STREN~H..JTuLT L./_ __ ____ . . . " RUPTURE
en
w // I
~ /
/ I
I
I ~I
I ~I
I ~
I ~I
I ~I
I ~I
I ~I
/ I
STRAIN, E:

figure 2.4
Pgrameters of g Geosynthetjc Stress-Strain Curve

24
The American Society for Testing and Materials (ASTM) recently
standardized the procedure for determining the unconfined short-term
strength and maximum elongation for geosynthetics which is described in
ASTM Test Method D 4595, "Tensile Properties of Geotextiles by the Wide
Width Strip Method". The ASTM D 4595 wide-width test uses a geosynthetic
sample that is 8 inches in width and 4 inches in gage length. The sample is
stressed uniaxially at a constant strain rate of 10 percent per minute until
failure occurs. The short-term strengths for the reinforcement used for the
Interstate Highway 70 through Glenwood Canyon, Norwegian Geotechnial
Institute, FHWA and Seattle Preload Fill Projects were determined by this
method.

The short-term strength for the Tanque Verde - Wrightstown - Pantano


Roads Project was determined using a four-inch-wide sample stressed
uniaxially at a constant rate of 2 percent per minute. The test method for the
short-term strength of the reinforcement used in the remaining two projects
(the Highbury Avenue and Japan Railway Test Embankment projects) were
not available in the literature. The smaller width sample used for the Tanque
Verde - Wrightstown - Pantano Roads project most likely produced a weaker
load-displacement response of the sample due to the Poisson effect (Wu
and Tatsuka, 1992) therefore adding conservatism to the design.

25
,

2.2.3.2 Creep·limlted strength

The creep-limited strength values reported in the literature for the


selected projects are listed in Table 2.2. The CRC for the projects that
reported it in the literature are also listed. The CRC is computed using the
creep-limited strength and short-term strength as illustrated in Equation 2.1 .

CRC = TcreepiTulf Equation 2.1

Where: CRC = Creep reduction coefficient


Tcreep = Tensile strength accounting for creep
Tulf = Short·term strength

As shown in Table 2.2, the CRC values used for the selected projects
range from 40 to 65 percent. For comparison, The AASHTO-AGC-ARTBA Joint
Committee Task Force 27 (AASHTO, 1990) recommends the following CRC
values for different polymer-type materials:

Polymer Type Creep Reduction Coefficient


Polyester 40%
Polypropylene 20%
Polyamide 35%
Polyethylene 20%

For example, the creep-limited strength for a reinforcement with a


short-term strength of 1,000 Ibltt would be 200 Ibltt using a CRC of 20
percent. The reinforcement materials used for the selected projects were
manufactured from polypropylene and polyester polymers. Although the
CRC values used in the selected projects where higher than the
recommended values (i.e .. less conservative) the reinforcements exhibited
very small strains over extended periods of time as will be discussed in
Chapter 3.

26
The creep-limited strength for the Tanque Verde - Wrightstown -
Pantano Roads project was determined by McGown (1984). Rapid creep
tests were performed to determine the creep-limited strength for the geogrid
reinforcement used in the project. These tests consisted of developing
isochronous load-strain curves at varying temperatures, strain rates and loads
to determine a load below which rupture by a ductile yield was not likely to
occur. Isochronous curves can be used to determine the load in a
geosynthetic for a certain strain at a given time. The other projects arbitrarily
selected various creep reduction coefficients to account for creep instead
of performing actual element tests.

The current AASHTO design procedure recommends determining the


creep-limited strength by the following method. Controlled laboratory
creep tests are performed for a minimum duration of 10,000 hours for a
range of load levels on reinforcement samples. The samples are then tested
in the expected loading direction, in either a confined or unconfined mode,
and at an assumed in-ground temperature of 700 Fahrenheit. The test results
are then extrapolated to the required design life using the procedure
outlined in ASTM 02837. From the creep test, two tensile loads should be
determined: the limit state tensile load [TOmi!), and the serviceability state
tensile load (T.orvice). The limit state tensile load is defined as the highest load
level at which the log time creep-strain rate continues to decrease with time
within the design lifetime without inducing either brittle or ductile failure. The
serviceability state tensile load is defined as the load level at which total
strain will not exceed 5 percent within the design lifetime. The design lifetime
is typically 75 years. AASHTO recommends that critical walls be designed for
a 1OO-year lifespan (MSHTO, 1990).

27
Since these creep tests take an extended amount of time, the majority
of designers used the recommended default values listed above in Section
2.2.3.2. Using default CRC value results in using only 20 to 40 percent of the
reinforcment's short-term strength. Moreover, partial factors of safety for
construction damage, durability, and overall intemal stability further reduce
the creep-limited strength to obtain the design-strength as described below.

2.2.3.3 Design Strength

The design strengths reported in the literature for the selected projects
are listed in Table 2.2. The design strength is the tensile strength of the
reinforcement used for design purposes. Most design methods use a partial
factor of safety approach to compute the design strength where the creep-
limited strength (i.e., Tfimit and/or T,ervicej is adjusted to account for site-specific
conditions. The MSHTO-AGC-ARTBA Joint Committee Task Force 27 currently
recommends the following procedure using the portia I factors of safety to
compute the design strength.

1. Compute the allowable long-term reinforcement tension based on


a limit state criterion given by:

Tal =Tfimit!FO*FC*FS
Where: Tal = Allowable long-term tension based on a limit state
criterion
Tfimit = Creep-limited strength based on a limit state
FO = Partial factor of safety for polymer durability
FC = Partial factor of safety for construction damage
FS = Overall factor of safety to account for uncertainties
in structure geometry, fill properties, reinforcement
properties and externally applied loads

28
2. Compute the allowable long-term reinforcement tension based on
serviceability state criterion given by:

=Tservice/FC*FD
Tas

Where: Tel =Allowable long-term tension based on


serviceability criterion
T.orvieo = The allowable long-term tension based on a
serviceability state

3. The design strength should be the lesser of Tor and Tes.

The partial factor of safety for durability accounts for the degradation
of the geosynthetic reinforcement due to chemical and biological exposure.
In the absence of product-specific durability information, AASHTO
recommends that the FD should be between 1.10 and 2.0. The partial factor
of safety for construction damage accounts for damage (i.e.. rips, punctures)
to the reinforcement during wall construction. In the absence of full-scale
construction damage tests, AASHTO recommends that the FC should
between 1.25 and 3.0. For penmanent, vertically faced GRS retaining walls
the minimum overall factor of safety should be no less than 1.5 (AASHTO,
1990). The partial factors of safety used in the selected projects are
described below.

2.2.4 Partial Factors of Safety

2.2.4.1 Factor of Safety for Durability

None of the selected projects directly used a factor of safety for


durability. However, reinforcement samples were exhumed from the walls
built for the Interstate Highway 70 through Glenwood Canyon and the
Tanque Verde - Wrightstown - Pantano Roads projects located in Colorado
and Arizona respectively. Reinforcement samples were exhumed

29
approximately 11 years and 8 years after construction for the Colorado and
Arizona projects. respectively .. After the samples were exhumed. they were
tested to determine their tensile strength and compared with the tensile
strength of archived samples cut from the same reinforcement material lots
used in construction. The Colorado project used a non-wooven geotextile
reinforcement manufactured from polypropylene and polyester polymers.
while the Arizona project used a geogrid reinforcement manufactured from
a polypropylene polymer.

The results from the durability testing indicate that the geosynthetic
material degrades very little over time in normal soil conditions. In both
projects. no significant decrease in tensile strength was observed in the
exhumed samples (Bright et al.. 1994; and Bell and Barrett. 1994) . For
comparison. the current factor of safety recommended by the Task Force 27
report (e.g .. 1.10 to 2.0) reduces the creep-limited tensile strength of the
reinforcement by 10 to 50 percent.

2.2.4.2 Factor of Safety for Construction Damage

Similar to the factor ot safety for durability (FD). the factor of safety for
construction damage (FC) was left out of the design computations for the
selected projects. The reinforcement samples exhumed from the Colorado
project exhibited an average 27 percent loss of strength based on element
tensile strength due to construction damage (Bell and Barrett. 1994) even
though the wall performed very well.

Similar to element tests for creep. the reduction in the element


strength due to construction damage represents only the behavior of the
reinforcement alone without accounting for the confinement of the
reinforced soil and soil/reinforcement interaction. Recently. San and Matsui

30
(San and Matsui L-J) performed a test on a 2O-foot-high wall where the
reinforcement embedded in the wall was cut using electrical wiring. The
reinforcement was cut at varying lengths starting from a distance furthest
from the face and progressing to the face of the wall. Each time the
reinforcement was cut. lateral and vertical displacements and
reinforcement strains were measured. After all the reinforcement layers had
been cut within approximately 1.5 feet behind the face. the total lateral
displacement was only approximately 1.5 inches. Based on the tie-back
wedge design concept. the wall should have collapsed once the
reinforcement was cut inside the Rankine failure surface. This test provides an
excellent illustration of the fact that neither construction damage or
degradation of geosynthetics will hinder its reinforcing function. Cutting the
geosynthetic reinforcement into small segments following construction can
be considered an extreme form of construction damage and
biological/chemical degradation. Apparently. whether the reinforcement is
continuous or not has little effect on the function of the reinforcement to
restrain lateral deformation of the soil.

The test performed by San and Matsui can provide reasons for the
good performance of GRS retaining walls even with construction damage
like in the Colorado project. From the test results and performance of the
selected case studies. two conclusions regarding the factor of safety for
construction damage can be made:

• Element tensile strength tests on exhumed reinforcement does not


characterize the impact to a GRS retaining wall due to construction
damage; and
• The recommended construction damage factors of safety (Le .. 1.25
to 3.0) are overconservative.

31
2.2.4.3 Overall Factor of Safety

The Seattle Preload Fill located in Washington, USA and the Tanque
Verde - Wrightstown - Pantano Roads Project located in Arizona, USA. used
overall factors of safety of 1.2 and 1.5 respectively in their designs. In both
cases, the walls performed very well. Since soil properties can vary, a
recommended overall factor of safety of 1.5 may be reasonable in GRS
retaining wall designs. By using a factor of safety of 1.5, the reinforcement
design strength is computed by reducing the short-term tensile strength by 33
percent.

2.2.5 facing Rigidify

By placing geasynthetic reinforcing in the soil, the strength of the soil is


improved such that the vertical face of the soil/geosynthetic composite is
self-supporting; therefore, most designs ignore the resistance of the facing .
However, most GRS walls use facing for atheistic purposes and to prevent
raveling between the reinforcing elements. Most types of facing include
concrete modular blocks that are dry stacked in front of the wall. Other
types of facing materials include rigid concrete panels and wrapped
geosynthetics. The Seattle Preload Fill, Interstate Highway 70 through
Glenwood Canyon and Norwegian Geotechnical Institute projects used a
wrapped geotextile face as illustrated on Rgure 2.2. and in the project
photographs in Appendix A. Shotcrete was placed on the Glenwood
Canyon project wall to prevent ultraviolet degradation of the geotextile.
Modular block type facing was used for the FHWA research project illustrated
on Figure 2.2 and in the project photographs in Appendix A.
The Japan Railway Embankment. Tanque Verde - Wrightstown -
Pantano Roads and Highbury Avenue projects used rigid concrete panels. In
the latter two projects, the facing was mechanically attached to the

32
reinforcement. For these two projects, the reinforcement strains were highest
at the face than at other locations along the reinforcement. This is due to
larger settlement of the reinforced fill relative to the rigid facing (Bright, 1994:
and Bathurst, 1992). The Japan Railway project used a flexible concrete
panel on the middle section of the wall to compare the wall's performance
using rigid and flexible facing material. The portion with the flexible facing
exhibited much larger deformation than the rigid facing (Tatsuoka, 1992).

33
3. Project Long-Term Performance
This chapter summarizes the performance of the GRS retaining walls
selected from the literature review and survey. The following section
describes the instrumentation and measured parameters used to quantify
the long-term performance of the walls. Section 3.2 provides the overall
performance of the walls including the reinforcement strains and wall
movements. Section 3.3 describes a conservatism index (CI) that was
developed to quantify the conservativeness of the designs used in the
selected projects. Section 3.3.1 describes the creep modulus developed to
quantify the rate of creep.
3.1 Instrumentation and Measured Parameters

For each of the selected projects, instruments were installed during


construction to quantify the behavior of GRS retaining walls in field
conditions. The long-term performance was quantified by recording
instrument readings periodically over an extended period of time and
documenting the results in published papers. Specific behavior parameters
were monitored for each project depending on the project's objectives as
described in Chapter 2. In general. the behavior parameters listed below
were measured:

• Horizontal and vertical displacements of the reinforced soil mass;


• Reinforcement strains in selected layers and locations; and
• Extemal and internal soil temperatures.

Strain gauges were installed on selected layers of reinforcement at


varying distances from the face of the wall. The primary objective in most of
the projects was to determine the location of the maximum strain in the
reinforcement. This would confirm the theoretical location of the failure

34
surface assumed for design. The second objective was to measure the
magnitude of strain in the reinforcement during and after-construction. The
location and type of instrumentation used for each project are illustrated on
the project description sheets provided in Appendix A.
3.2 Reinforcement Strains and Wall Movement

The maximum reinforcement creep strain and wall movements for


each project are listed in Table 3.1. If the creep strain was unavailable in the
literature for a particular project, it was computed based on the incremental
change in total strain. Note, that the creep strain listed in Table 3.1 refers to
the deformation of the wall due to creep occuring after construction. The
movement listed in Table 3.1 refers to the total displacement of wall since
the beginning of construction. In some cases, the majority of the movement
was during construction. The CRC used for the design and recommended
by AASHTO for each project is also listed.
3.2.1 Intentale Highway 70 through Glenwood Canyon Project

The Interstate Highway 70 through Glenwood Canyon project was


purposely designed to determine the lower stability limits by designing at or
near the equilibrium factor of safety. It was anticipated that the
reinforcement would exhibit excessive strains on the order of SS percent, yet
little movement within the reinforced soil mass was observed. Approximately
one year after the wall was constructed, a surcharge load was applied to
the top in an attempt to create failure conditions. The surcharge consisted of
a 1S foat high soil embankment applying a pressure of approximately 1,950
Ib/ft2. However, failure never occurred.
The wall was constructed on a weak foundation soil and experienced
significant movement. The retaining wall experienced over two feet of
differential settlement from one end of the wall to the other due to

3S
consolidation of underlying clays. Despite the large differential settlements,
only small strains occurred in the reinforcement (Derakhashandeh and
Barrett, 1986).

The eRe values used in the design of the wall ranged from 40 to 55
percent for reinforcement layers manufactured from polypropylene type
polymers and 65 percent for the reinforcement layers manufactured from
polyester type polymers. AASHTO recommends eRe values of 20 and 40
percent for polypropylene and polyester respectively (AASHTO, 1990). The
eRe values used for the design are over two and one and half times less
conservative for the polypropylene and polyester reinforcement layers
respectively, yet the wall performed very well.

Since the wall performed better than anticipated, the researchers


concluded that the mechanisms of geosynthetic reinforcement soil are not
well understood and the ability to select allowable loads is limited (Bell,
1983). They also concluded that more full-scale walls should be
instrumented and monitored to better understand the behavior of the
soil/geosynthetic interaction.

36
Table 3.1
CRC. Reinforcement Strain and Wall Movement for the Selected Projects

eRe eRe
ProJect
Wall
Name
H
(II)
t
(years)
....
(II)
De.lgn AASHTO
(%) (%)
Ee","
(%)
Ymu
Movement

(In)
Xm..
(In)
Interstate Highway 70 through Geotextlle Earth
16 0.8 0.9 40-55 20-40 NA 3.5 5.15
Glenwood Canyon Retc:ining Wall

Wall Panel 26-30 15.6 7 1.6 37 20 <1 NA 3.7


Tanque Verde-Wrightstown-
Pantano Roads
Wall Panel 26-32 16.1 7 1.6 37 20 <I NA 3.7

(.,)
Norwegian Geotechnical Wall Section J 15.7 4 2.2 NA 20 0.5 NA NA
'-J
Instllute
Wall Section N 15.7 4 2 NA 20 0.6 NA NA
apan Railway Test Embankment JR Embankment
16.4 2 I NA 40 NA I -0.4 I
No. 1

Highbury Avenue Highbury Ave. Wall 23.3 2 0.4 NA 20 1.5 NA 1.7


Federal Highway Admlnlslration
Wall No. 9 20 1.3 2.5 60 40 0.7 NA -2

~atlle Preload AI Soulheast Wai 41.3 1 1.25 40-60 20-.10 0.7 1.4 to 1.6 5.5-0.3

~
NA ~ Not available in Ihe literature Sc::mw: = Maximum creep strain in the reinforcement
H ~ Height Ym", ~ Total vertical movement of the wall
t ~ Moniloring dUl'ation x.- ~ Total horizontal movemenl
Average reinforcement spacing
Sc.-vg =
CRC Design ~ Creep reduction coefficient used In the design
CRC AASHTO ~ Creep reduction coeffiecient recommended by AASHTO
3.2.2 Tanque Verde - Wrightstown - Pantano Roads Project

The performance of Wall Panels 26-30 and 26-32 was monitored for
approximately seven years after construction. Geogrid reinforcement strains
were measured in the bottom. middle and top layers of the two wall panels
using resistance strain gages and inductance coils. Strain readings from the
inductance coils had a large variance due to low strains in the
reinforcement. therefore the results were believed to be unreliable (FHWA.
1989) so the readings from the strain gauges are reported in this study.

Reinforcement strains were measured during construction. two weeks


after construction and thereafter on an annual basis. The post-construction
strain measurements were adjusted to account pretensioning and
compaction during construction so that strains measured after construction
would be the result of creep.

The lateral movement of the wall was measured by surveying points


at the top of the wall. The points were surveyed during construction and up
to one month after construction. During construction. the top of the both
walls moved laterally approximatley three inches while the bottom of the
wall remained stationary. Uttle movement was observed after construction.

The mean total creep strain in the reinforcement after construction is


illustrated on Figure 3.1. As illustrated on Figure 3.1. the strain increased in the
reinforcement during the first year of service indicating that creep was
occurring. Thereafter. however. the creep strain remained generally
constant indicating that the wall had stabilized with time. The maximum
creep strain recorded was less than 1.0 percent. Based on isochronous load-
strain curves developed by McGown (1984). the load induced in the

38
reinforcement at 1.0 percent strain was approximately 265 Ib/tt. This is
approximately only5 percent of the short-term strength (5,400Ib/tt).

39
1.0

g 0.8
I! 0 .6

liE 0.4 -+-Wall Panel 26-30


~
• _ _ Wan Panel 26-32

i

'" 0.2 ·

0.0 ••---+----+---+----if----+---+----+---+--_f----l
o 200 400 Enl Enl 1000 1200 1400 lEnl 1800 2CXlO
Tim. Alter Conmuctlon (day)

Figure 3.1
Reinforcement Creep-TIme Curve. for the Tanque Verde - Wrightstown - Pantano Road. Project
3.2.3 Norwegian Geotechnical Institute Project

The performance of the Noregian Geotechnical Institute project wall


sections 'j' and 'N' was monitored for approximately four years since its
construction. Both the force and strain was measured in the reinforcement.
Section 'N" had twice as many layers of reinforcement than Section 'j'.
Following construction, the wall was monitored for approximately four weeks
under self-weight loading. Thereafter, the top of the wall was cyclically
loaded for two months followed by a permanent surcharge.

The mean total creep strain in the reinforcement for the two sections
following application of the permanent surcharge loading is illustrated on
Rgure 3.2. The creep strain was determined from the incremental increase in
the total strain begining 10 days after the surcharge was placed. The
maximum strain over the four years was approximately 0.5 and 0.6 percent in
section 'j' and 'N' respectively. The maximum tensile force in the
reinforcement after the permanent surcharge reported in the literature was
approximately 200 Iblft for both of the sections. This is approximately 6
percent of the short-term strength (3,600 Ib/ft). The eRe value used in the
design for this project was unavailable in the literature. The eRe value
recommended by AASHTO for the reinforcement type used in the two
sections is 20 percent (AASHTO, 1990).

41
1.4

&t
_ 1.2
.5
i'" 1.0
t
! 0.8
U
1:
•~ 0.6
!!
.....
N
..'"
C0 0.4

0.2
-+-Wall Section N
- - - Wall Section J

0.0
0 200 400 Em Em 1000 1200 1400
Time Alter Surcharge (day]

FIgure 3.2
ReInforcement Creep-TIme Curve. for the NorwegIan Geotechnlcallnatltute Project
3.2.4 Japan RaHway Test Embankment Project

The performance of the Japan Railway Test Embankment JR Number


1 was monitored approximately two years since its construction. The vertical
and lateral displacement and tensile force in the reinforcement was
measured in three wall sections (cross sections D-D, F-F, and H-H) illustrated
on the project description sheets in Appendix A. Figure 3.3 illustrates the
monitoring results.

As illustrated on Rgure 3.3, the tensile force in the reinforcement


increased during the first eight months reaching a nearly asymptotic state
similar to the performance of the other projects. The maximum tensile force
in the reinforcement was approximately 131 Ib/ft. This is approximately only 7
percent of the short-term strength (1 ,880 Ib/ft). The eRe value used in the
design for this project was unavailable in the literature. The eRe value
recommended by AASHTO for the reinforcement type used in the two
sections is 40 percent (AASHTO, 1990).

43
•.
'

{' ~_~f., ~~~:;:;:::~:. ;~.;:;,:::::::.::::'.:::::.:::::~ ~ ~ ~:::::::.::::~:"": ~. ". _.~.::: ~".::::.:::


~ ..

Note: See Appendix A for Section locations

Figure 3.3
Tensile Force in Reinforcement and Displacement of the
Japan Railway Test Embankment Project
(after Ta1suoka F.• Murato O. and Tateyama Moo 1992)

44
3.2.5 Highbury Avenue Project

The Highbury Avenue Wall was monitored for approximately two


years. Reinforcement strain was measured after the props holding the
concrete panels were removed. Reinforcement strain was measured
thereafter in December 1990; then in March 1990; July and August 1990; and
a year latter in August 1991. The creep strain was based on the incremental
change in the strain since December 1990. The maximum reinforcement
creep strain was approximately 1.5 percent based on the mean creep strain.
The mean creep strain over time is illustrated on Figure 3.4. Similar to the
previous projects, the wall exhibited creep over the monitoring period,
however, had begun to stabilize with time. The CRC value used for the
project was unavailable in the literature.

3.2.6 Federal Highway Administration R.search Project

Wall nine built for the FHWA project was monitored for approximately
one year. Reinforcement strain and total wall movement was recorded
more frequent then the previous projects. Instrument readings were
recorded on an almost daily basis during construction and during placement
of the surcharge load. The surcharge was completed November 10, 1989.
Thereafter instrument readings were recorded nine times up through
November 11, 1990.

The maximum creep strain computed after the surcharge load was
placed is illustrated on Figure 3.5. The creep strain was based on the
increament increase in total strain. As illustrated on Figure 3.5, the creep
strain shows that the wall was becoming stable with time. The maximum
creep strain was approximately .7 percent over the one year monitoring
period. The total lateral movement after the props were released was

45
approximately 3.6 inches. The measurement was based on the vertical
inclinometer directly behind the face of the wall. Most of the movement is
most lilcely due to the tensioning of the reinforcement.

The eRe value used for the design was 60 percent. AASHTO
recommends a more conservative eRe of 40 percent for the type of
reinforcement used in wall nine. Although the eRe value used in the design
was one and half times higher (e.g., less conservative), the reinforcement
strains were very small.

3.2.7 Seattle Preload Fill Project

The southeast wall for the Seattle Preload Fill project was monitored for
approximately one year after its construction. Similar to the FHWA wall,
instrument readings were recorded on a frequent basis. The maximum
reinforcement creep strain in the reinforcement over time is illustrated on
Figure 3.6. Creep strain was determined immediately after the surcharge
was placed on the wall. As illustrated on Figure 3.6, creep was occurring
and beginning to stabilize. The maximum creep strain recorded in the
reinforcement was less than O.S percent.

The eRe values used for the design were 40 and 60 percent for
polypropylene and polyester type reinforcement respectively. AASHTO
recommends eRe values of 20 and 40 percent for polypropylene and
polyester respectively. The eRe values used were two and one and half
times less conservative than the recommended values, yet very little strain
was observed in the reinforcement.

The researchers concluded that the low strain were the result of lower
than expected load level in the reinforcement or due to poorly understood
interaction between the reinforcement and the confining soil. Additionally,

46
the reinforcement was damaged during construction damage with no
apparent impact to the performance.

47
2.0

1.8

_ 1~

~
~ lA

~ 12
I
U 10

~
1E M

~
M
i- M

0.2

0.0 i " - - - - + - - - - f - - - - - - , > - - - - - < - - - - < - - - - + - - - - - - <


o 100 2CXl 3lO 4CXl &Xl eoo 700
Time After Prop Relecue (day)

Figure 3.4
Reinforcement Creep·TIme Curve for the Highbury Avenue Project
-•...
u
0'
A.

~
...
%

--•...
..r:
0

.., •
.
~

~ i- ~
II! ;)
.0
j ...
&E
ii:1=
~ ~ •
Q.

•0...•
."


~

~
~
•E
0:
- c
•E

2
§ -0
c
ii
110

fi!

r--+--+-~--4---r--+--+-~--~~ 0
~~ o
O o
,",:~LI)"If:O"lN
0 0 dod
(%) ulDlJS d ....:l fuawa"Jolu!all

49
-
~ 0.8
0.9

j 0.7
'"t 0.6
!
U 0.5
C
~ 0.4
..
{ 0.3

gJ . 0.11
• 0.2

.. ........: ....
o .... I : -I
a 50 100 150 200 250 300 350

Time Aller Surchar". (day)

Figure 3.6
Reinforcement Creep-nme Curve for the SeaHie Preload Project
3.3 Conservatism Index

The selected GRS retaining walls vary from conservative to less


conservative designs. For example, the Interstate Highway 70 through
Glenwood Canyon and Section 'J' of the Norwegian Geotechnical Institute
projects were purposely designed to determine the lower stability limits by
designing at, near equilibrium or even below factors of safety. Conversely,
the Highbury Avenue and Tanque Verde - Wrightstown - Pantano Roads
projects were designed using more conservative assumptions.
Due to the variability in retaining wall designs, direct comparison of
the selected projects would be misleading. Therefore, a conservatism index
(CI) was developed so that the design of the walls could be evaluated. In
general, the CI value is computed using a limit equilibrium analysis where
resisting lateral force provided by tensile strength of the reinforcement is
divided by the driving lateral force of the earth. The CI value is based on the
same principles of limit equilibrium used in the current design methods where
the resisting tensile force is entirely provided by the reinforcement and
redistribution of stresses due to the soil/geosynthetic interaction are ignored.
The CI value takes into consideration the reinforcement strength, number of
reinforcement layers, and active lateral earth pressure caused by the
retained soil and surcharge. These parameters and the resulting CI for each
project is listed in Table 3.2. Detailed computations are provided in
Appendix B.

51
Table 3.2
Parameters Used to Compute the Consevatl.m Index

Wal Height I ... ~. Y q


Project Name (tI) n (tI) (deg) IbIft' IbllI' Ka CI
Inlerslate Highway 70 through Geolexflle Earth
16 17 0.9 :l6 130 1960 0.27 0.44 I
!Glenwood Canyon Retaining Wall

Wall Panel 26-30 15.6 10 1.6 34 122.5 369 0.26 8.7 I


anque Verde-Wrlghlstown-
Pantano Room
Wall Panel 26-32 16.1 10 1.6 34 122.5 369 0.26 8.1
R)
Wall Sedlon J 15.7 4 2.2 3B 108.8 1044 0.47 0.4
NO!Wegian Geotechnical
Institute
WaR Sedlon N 15.7 7 2 3B 108.6 1044 0.47 1.6

apen Railway Test Embankrnenl


JR Embankment No. I 16.4 16 1 :l6 (a) 93.2 0 0.27 3.5

Highbury Avenue Highbury Ave. Wall 23.3 9 0.4 :l6 (a) 125 (a) 0 0.27 6.2

Federal Highway Administration


Wall No.9 20 8 2.5 :l9 125.6 728.6 0.23 7.9

ealtle Preload RI Southeast Wall 41.3 33 1.25 38 130 2210 0.26 26

~ Wt;
n • Number of rmfOlcem8ut lIyers q • Surcharge a) Estlmatod val\Je
.... = Average spacing between reinforcement fayers Ka • Active lateral earth pressure ooofflclent
• = Design Internal frtctlon onglo Ct • Conservatism Index
1 = DesIgn un. weight
The CI is an index value to indicate the relative conservativeness of a design.
Similar to a factor of safety concept, a CI value close or less than one is
considered a less conservative design. A design with a greater CI value is
more conservative relative to a design with a smaller CI value. As an
example, if project A has a 01 value of 3 and project B as a DI value of 5,
theoretically, project B should perform better (i.e., smaller displacements and
strains) than project A.
The CI for the selected projects ranged from 0.4 to 8.7. The less
conservative designs have a CI of 0.4 and include the Interstate Highway 70
through Glenwood Canyon project and wall section 'J' of the Norwegian
Geotechnical Institute project. Both these walls were purposely designed
using less conservative assumptions, however still performed very well. Since
none of the selected projects exhibited large strains, it is difficult to correlate
the CI value with an under designed GRS retaining wall to determine the
lower bound CI. However, a CI value greater than 0.4 would indicate a
more conservative design since the walls with the lower CI values
demonstrated good long-term performance.
The Tanque Verde - Wrightstown - Pantano Roads project wall panels,
Highbury Avenue and Federal Highway Administration projects have CI
values ranging from 7.9 to 8.7. Each of these projects used high tensile
strength reinforcement ranging from 2,000 Ibltt to 5,400 Ibltt (short-term
strength). For comparison, the Interstate Highway 70 through Glenwood
Canyon wall used reinforcement layers with a short-term tensile strength of
220 Ib/ft.
3.3.1 Creep-Rate and the Creep Modulus

Creep-rate is the time-rate at which a GSR retaining wall deforms


under a sustained load. The change in the creep-rate can be used to

53
quantify the stabilization of a GRS retaining wall due to creep. A constant
creep-rate would indicate that the wall is deforming at a constant rate
which would be considered secondary creep. An increasing creep-rate
would indicate that the wall is deforming at an increasing rate which would
be considered tertiary creep .. In either cases, the wall could conceivably
reach a creep failure condition. Conversely, a decreasing creep-rate would
indicate that the wall was stabilizing with time reaching an equilibrium
condition.

Figure 3.8 illustrates the creep rates computed for the selected
projects. As illustrated on Figure 3.8, there is a decreasing trend in the creep-
rates indicating the GRS walls were stabilizing over time. This behavior has
also been observed in laboratory soil/geosynthetic composite creep tests
conducted by Ketchart and Wu (1996). Moreover, the decreasing trends
were close to linear when plotted on logarithmic scale. The slope of the
linear relation is referred to as the creep modulus (CM). The CM is illustrated
in the example below on Figure 3.7.

LINE
w
~
0:::
D..
W
W
0:::
(..)

CM = CREEP MODULUS

TIME
Figure 3.7: Creep-rate-TIme Curve Illustrating the Creep Modulus

54
1E+<Xl

I E-Ol

_ _ _ Pr_ FII Projec1


>; lE.Q2
ftI
:!;!
e:. lE.Q3
-O--Federal Highway A_lion
Project

I _ _ Highbury AVtIfU> Prcject


111
111
..
II:
Q.
lE.Q4

-+- Norwegian GeoIechnicallnstitute


15 lE.Q5
Project - Section N

_ _ Norwegian Geotechnlcallmltltule
Projec1- SectIon J
lE.Q5
- 0 - Tanque Verde - Wrightstown -
Pontano Roods Projee1- Waa Pane
lE.Q7 2&3:1
-.-Tanque Verde - Wrlghtstown-
1 10 lime (day) 100 1000
Pantano Roods Project - Wall Pane
2632

Figure 3.8
Creep·Rate·nme Curve for the Selected Projects
The CM value provides a means to characterize the long-tenm perfonmance
by quantifying the slope of the creep-rate time curve The CM computed for
each project are listed in Table 3.3. .The regression lines used to compute the
CM are illustrated on the Figures C.1 through C.5 in Appendix C. The CM for
the selected projects range from 0.57 to 1.13 %/day2. This is a fairly narrow
range given the wide variety of retaining wall types in the study. The
decreasing slope in the creep-rate and similar slopes were also observed in
the laboratory tests perfonmed by Ketchart and Wu (1996). The CM may be a
good parameter to characterize the creep behavior of the soil/geosynthetic
interaction which will be discussed in Chapter 5

Based on the CM, the creep-rate for the selected projects are decreasing at
a rapid rate indicating that the walls are stabilizing with time. Moreover, if
demonstrates the arbitrary nature of reducing the reinforcements short-term
strength by up to 80 percent using element creep tests, CRCs and/or partial
factors of safety.

56
Table 3.3
Creep Modulus for the Selected Projects

Wall Height t II" eM


Project Name (II) (ye<m) (") ("/day')
nterstate Highway 70 through Geolextlle Earth
16 0.8 NA(a) NA
Glenwood Canyon Retaining Wall

anque Verde-Wrlghtstown- Wall Panel 26-30 15.6 7 lIb) 0.92

Pantano Roads
Wan Panel 26-32 16.1 7 1(b) 1.13

Norwegian Geotechnical Wall Secflon J 15.7 4 98 1.1


nstiMe
Wan Section N 15.7 4 89 1.08

apanese Railway Test


JR Embankment No. 1 16.4 2 NA NA
Embankment

~ighbury Averue Highbury Ave. Wall 23.3 2 98 1.5

Federal Highway Administration Wall No.9 20 1 89 0.57


,.
~eatlle Preload All Southeast Wall 41.3 1 68 0.41

~
R' = Regression confidence coefficienl .
eM = Creep modulus. Positive indicates decreasing slope.
t = Monitoring duration

NQIln
a) NA indicates that data was unavailable for computafton.
b} R2 = 1 since regression Hne developed from two data points.

57
4. An Approach to EstImating Creep Using a Laboratory Test
Chapter 3 demonstrated that the current design methods significantly
over-estimate the magnitude of strain in the geosynthetic reinforcement and
movement of the wall face caused by creep. However, the longest period
of performance data for any of GRS retaining wall is less than 10 years. Most
applications require that permanent retaining walls be designed for a
minimum service life of 75 to 100 years (AASHTO, 1992). Thus, a rational
means for estimating creep based on the soil/geosynthetic interaction is
needed. This Chapter describes an approach for estimating creep using a
simple laboratory test and analytical solution.

4. 1 Creep In Laboratory Tests

Development of the method begins with evaluating recent laboratory


tests conducted at the University of Colorado at Denver to determine creep
behavior of soil/geosynthetic composites. Wu (1994a) and Wu and Helwany
(1996) developed a laboratory test procedure to characterize the
soil/geosynthetic composite behavior. The apparatus used in for the
procedure allows the stresses applied to the soil to be transferred to the
geosynthetic reinforcement in a manner similar to typical GSR walls. Using
the device, they conducted two long-term performance tests, one using a
clay backfill and the other using a sand backfill. A second study was
performed In 1995 by Ketchart and Wu (1996). They simplified the testing
apparatus device and performed tests on various soils and geosynthetics
under different conditions, including accelerated creep tests at elevated
temperatures.

58
In both studies, it was observed that the long-term deformation
behavior of the soil/geosynthetic composite was significantly affected by the
time-dependent behavior of the soil and the geosynthetic reinforcement. In
general, if the confining soil has a tendency to creep faster than the
geosynthetic reinforcement, the geosynthetic will impose a restraining effect
on the deformation of the soil through friction and/or adhesion between the
two materials. Ketchar! and Wu (1996) observed that in eoch case with
granular soil, the creep rate decreased over time. This behavior was also
observed in the reinforcement strains for the full-scale walls described in
Chapter 3. The following subsections describe the laboratory test, the test
results and the procedure developed to estimate creep deformation over
the design life of a GRS wall.

4.1.1 Laboratory Creep Test Descriptions

The test apparatus developed by Wu and Helwany (1996) consists of a


Plexiglas box with thin sheet-metal sides approximately 1.5 feet by 3 feet in
size. A layer of reinforcement is sandwiched between two soil blocks placed
inside the box using techniques similar to field construction procedure. Then
the composite is loaded with a sustained surcharge load. The side-wall
adhesion between the Plexiglas and the soil was minimized by creating a
lubrication layer at the interface of the two materials to create plain strain
conditions. The test apparatus is illustrated on Figure 4.1

One of Wu and Helwany's tests consisted of placing an Ottawa sand


into the testing apparatus using a air-pulviation method. Once half the sand
was placed. a layer of geotextile was placed and securely attached to the
two sheet metal plates, followed by the remaining sand. Another layer of
geotextile was then placed at the top of the sand. The soil/geosynthetic

59
composite was loaded with a sustained vertical load of approximately 16
Ib/in2 for 30 days. The stress-strain behavior of the geotextile was determined
by performing a series of element geotextile creep tests to compare its
behavior with the behavior of the soil/geosynthetic composite. Lateral and
vertical deformation and reinforcement strain were measured over the
testing period. The test results indicate that the element creep test over-
estimated the strain in the reinforcement by a factor of four consistent with
the performance of the full-scale retaining walls described in Chapter 3.

Ketchart and Wu modified the apparatus developed by Wu and


Helewany so that the lateral supports could be released to model "worst"
case conditions. This would be similar to removing the modular blocks or
other type of facing from the front of a GRS retaining wall, exposing the soil
and the reinforcement. Similar to previous test procedures, soil was placed
in the test apporatus and compacted to mid-height. A layer of geotextile
reinforcement was then placed (without attaching to the side walls)
followed by compocted soil to the top of the apparatus. The sample was
then subjected to a sustained surcharge for a period of 30 days. In some
cases, the apparatus was placed in a room with elevated temperatures to
accelerate creep of the geasynthetic. The test apporatus is illustrated on
Figure 4.2.

60
T;
4S em c

,
I
I d
I
:T·'b . . ..
90 em

'. ,.

Dimensions: l&11ends:
Sand-Backfill Test: a Geosynthetic Reinforcement
L = 81.3 elil b Soil
H = 30.5 em e Steel Plate
d Rigid Container wi~h
Lubricated Side Walls
Clay-Backfill Test: c Sustained Load
L ,. 45.7 cm f Rigid Plate .
H .. 25.4 em

Rgure4.1
Schematic of the Long-Term SoIl/Geosynthetic Perfomance
Test Device (Helwany and Wu.1996)

61
sus~~~ea p~.S5U~.

LOQ.cllng pla. te

pa.nel

LVDT 5UppO,..'tlno
'tubR

6 In.

Cal Before Releasing Lateral Supporting Panels

r-" 'o,....'V ~l~ te


\1l1:.~1ll S~PD~'tlng p4r.L

L VlJl sUPpc,.'tiI'lQ
'tlolile

Figure 4.2
Schematic of the Modified Long-Term Soll/Geosynthetfc Perfomance
Test Device (WU and Ketchart, 1996)

62
The types of tests performed during Ketchart and Wu's study are
described below.

A total of 11 tests were performed during the study. From the testing
program, six of the tests conducted using a granular backfill is of interest for
this report. These include the tests described below (Ketchart and Wu, 1996).

• Test 0-1 : Test 0 -1 was performed using a heat-bonded nonwoven


polypropylene low-strength geotextile having a short-term tensile
strength of 420 Ib/tt. and an average vertical pressure of 15 Ib/in2
at a temperature of 700 F. The test was performed to determine
the creep behavior of the soil/geosynthetic composite using a low-
strength reinforcement. Reinforcement strain was measured in
addition to lateral and vertical displacement.

• Test H-1: Test H-1 was performed using a woven geoteX/ile having
a short-term tensile strength of 4800 Ibltt and an average pressure
of 30 Ib/in2. at a temperature of 125o F. The test was performed to
determine the creep behavior of the soil/geosynthetic composite
using a large load at an elevated temperature.

• Test R-1: Test R-1 was performed using a woven geoteX/ile having a
short-term strength of 4800 Ibltt and an average vertical pressure of
15 Ib/in2 at a temperature of 70 of. The test was performed to
determine temperature effects on the creep behavior of the
soil/geosynthetic composite by comparing the results with test R-2.

• Test R-2: Test R-2 was performed using the same material and
loading as R-1 except at an elevated temperature of 125o F. The
test was performed to determine temperature effects on creep
behavior of the soil/geosynthetic composite by comparing the
results with test R-1.

• Test R-3: Test R-3 is a duplicate test of R-2 to determine the


repeatability of the test method.

• Test W-1: Test W-1 was performed using a woven geotexille having
a short-term tensile strength of 1440 Ibltt and an average pressure
of 151b/ in 2 at an elevated temperature of 125 o F. The test was
performed to determine the temperature impacts to the creep

63
behavior of the soil/geosynthetic composite using a low strength
reinforcement.

The granular bacKfill consisted of a road base comprising of a silty


sandy gravel. The soil was prepared 2 percent wet of optimum moisture
content and compacted to 95 percent of the relative density or
approximately 126 Ib/ft3 having an intemal friction angle of 34°.

4.1.2 Laboratory Test Creep Rate

Ketchart and Wu measured lateral and vertical displacements and in


one test, strain in the reinforcement due to creep. Lateral displacements
were measured using linear voltage deformation transducers installed at the
mid-height of the testing apparatus where the reinforcement was located.
Strain was measured in the reinforcement for test 0-1 only. The lateral
displacement over the time period for each of the above tests are plotted
on Figure 4.3. From the lateral displacement data, the lateral creep rate was
computed and plotted on Figure 4.4. The creep rate based on the
measured maximum strain in the reinforcement for test D-1 is also shown.

As illustrated on the Figure 4.3, the effects of geosynthetic strength,


temperature, and loading all impact the time-dependent behavior of the
soil/geosynthetic compasite as to be expected. However, there is a linear
decreasing trend in the creep rates of all the tests when plotted on a
logarithmic scale as illustrated on Figure 4.4. After performing a linear
regression on each of the data sets, the confidence (R-squared) coefficient is
on the order of 94 percent demonstrating a good linear fit. Moreover, the
slopes of the linear relation or eM are approximately the same for all the
tests.

64
0.16
/
0.14 /
:cu
c
;.
0.12 /
C ..,..-.......
• 0.10

iu 0.06
~
J! ......... TestD-l

I 0.06
......... TestH-l
-<>- Test w- r
I
~

0.04 rf ___ Te,t R-3


g; _ _ TestR-2
0.02

0.00
o 5 10 15 20 25 :lO
Tim. (day)

FIgure 4.3
Lateral Dlsplacement·Tlme Curves for Soll/Geosynthetlc Laboratory Teat
IE+(!;
lE+04
lE+03
lE+02
1E+{)1
lE+oo

ii lE.Q1
e-~
!~
lE.Q2
U lE-03
0-
0-
lE-04
lE'(!; _ _ lab Test D-I Irelnforcement strain
lE'(!; rate I
_ _ lab Test D-llcisplacement rate I
lE.Q7
-+-lab Test R-2ldisplacement rate)
lE-OO
0.01 0.1 1 10 100 I -o-lab Test W-I Idlsplacement rate)
Time (day)
_ _ lab Test H-J Idisplacement role)

--+-lab Test R-J Idisplacement rate)

Figure 4_4
Creep-Rote-TIme Curve. from Soil/GeolyntheHe Laboratory T••t
4.2 Laboratory and Fun·Scale Creep Rate Comparison

Rgure 4.5 illustrates the creep rates computed for the selected full-
scale projects and the creep rates computed from the lateral displacement
of the laboratory tests. It is observed that the full-scale data fits well with the
laboratory data and shows a continuing decreasing trend in the creep rate.
Moreover the CM for the selected projects and the laboratory tests are
nearly the same. The CM values are listed in Table 4.1 . The plots used to
compute the CM values are provided in Appendix C.

The full-scale creep performance also demonstrates the validity of the


testing procedure developed by Ketchart and Wu by accurately modeling
the soil/geosynthetic integration of a full scale GRS retaining wall with
granular soil. The full-scale creep performance also demonstrates that the
laboratory procedure can determine the creep behavior of a
soil/geosynthetic composite material consisting of granular soil in a relatively
short amount of time unlilee the 10,000 hour element creep tests currently
required.

67
IE+(!;
lE+04
lE-t03
lE+02
- lE+01
! lE-OO
~ lE-Ol
-II
II<
lE-02
It lE.Q3
! lE-04
U
?» lE.(l;
lE.()6
lE-07
1E.()6
0.01 0 .1 10 100 1000
Tim. (day) SolllGeoaynthe1lc Laboratory Creep
Tests
-Selected Full-Scali Wall ProJects
Figure 4.5
Combined Creep.Rate·nme Curve. from Selected Projects and Laboratory Test,
Table 4.1
Creep Modulus for the Full-Scale Walls and Laboratory Tests

Wall HeIght I It' CM


Project Name (II) (years) (%) (%'day)
Intemate Highway 70 lhrough Geotextlle Earth
16 0.8 NA(a) NA
Glenwood Canyon RetainInQ Wall

anque Verde-Wrightstown- Wall Panel 26-30 15.6 7 I (b) 0.92


Pantano Roads
Wall Panel 26-32 16.1 7 I (b) 1.13

Norwegian Geotechnical Wan Sec1ion J 15.7 4 98 1.1


nstitute
wan Section N 15.7 4 89 1.06

apanese Railway Test JR Embankment


16.4 2 NA NA
Embankment No. I
Highbury Avenue Highbury Ave. Wall 23.3 2 96 1.5

!Federal Hlghway Admlnistraffon Wail No. 9 20 I 89 0.57

~eatffe Preload RII Southeast Wall 41.3 I 66 0.41

aboratory Test 0-1 I 0.06 99 1.41

Laboratory Test H-I I 0.08 96 1.17

Laboratory Test R-I I 0.08 I (b) 4.06

Laboratory Test R-2 I 0.08 94 1.36

Laboratory Test W-I I 0.08 98 1.35

WmnQ
~ =Regression confidence coefficient.
CM = Creep mocMus. Po~tiv" Indicates decreasing slope.
t = MonitOring period

tJ!m
0) NA indicates thai data was unavailable for computation.
b) R' = 1 since regression line developed from two data points.

69
4.3 An Analytical Solution for EstImating Creep Strain

The previous sections demostrated that the creep-rate for the


laboratory soil/geosynthetic composite tests and full-scale walls could be
represented as a straight line when plotted on a lograthimic scale. To
determine the creep strain at any given time, the strain-rate can be plotted
with time as illustrated in Figure 4.3.

ACTUAL DATA
A REGRESSION LINE
w
!<e
£t:
0-
W
W
£t:
U

m = SLOPE

1
TIME (t/t o )

Figure 4.3: Creep-Rate-Time Ratio Plot

70
Using the plot illustrated on Figure 4.3, the linear relationship can be
represented by the following equations:

log (:c)= -m'log(I:) + log(A)


Equation 4.1

or,

dt c
dl :=
()-m
A· I:
Equation 4.2

else
Where: dt = Creep-rate (%/day)
=
m Slope of the log (t/to) vs. log (d&c/dt) curve
A = Creep-rate coefficient (%/day)
=
t TIme (day)
to = Reference time (day)

The creep-rate coefficient (A) is the creep-rate corrosponding to a tit 0


value of one. The reference time (to) is the time at which the creep-strain
begins. Typically this would be at the end of construction of a wall. For
example, if it took 30 days to complete the construction of the wall, to would
be 30 days.

Integrating Equation 4.2 gives the creep-strain expressed in Equation


4.3:

tc:= klom .(!..)1-m+ C


1- Io
Equation 4.3

Where: C = Integration constant

71
Equation 4.3 can be solved by two unique solutions. Knowing that the
-A-to

creep strain (te) is zero at t =to, C will be equal to 1 - m when the slope (m)

is not equal to 1_ When the slope (m) is equal to 1, C equals zero. Thus, the
analytical solution for determining creep strain at a given time can be
expressed by Equations 4.4 and 4_5:

te := A-to _ (..!..)1 - m __A_-_to_ When: m"l


1- m to 1- m
Equation 4.4

When:m= 1
Equation 4.5

Note that Equations 4.4 and 4.5 are only valid for a soil/geosynthetic
composite that exhibits a constant value of m. The smaller the m value, the
larger the creep-strain.

72
s. Summary and Conclusions
S.l Summary

The three main research objectives of this study included:

1. Compile long-term performance data from field projects involving


well-instrumented GRS retaining walls;
2. Develope a means to quantify the conservativeness of the designs;
and
3. Develop a rational method to estimate creep based on laboratory
creep test of the soil/geosynthetic composite deformation.

The fIrSt objective was accomplished by surveying experts in GRS


technology and pertorming an extensive literature search. From the survey
and literature search, seven well-documented GRS retaining wall projects
around the world were described and analyzed.

The second objective was achieved by showing that walls designed


using a CRC that was greater than the CRCs recommended by AASHTO (Le.,
less conservative) performed exceptionally well under a variety of
conditions. The CI was develop to provide a measure of conservativeness in
the designs. Even with a low CI for some of the projects, the walls performed
exceptionally well.

The third objective was achieved by developing a simple procedure


for estimating creep based on the observed decreasing creeJrl"Ote of the
soil/geosynthetic composite. By using the simple testing procedure
developed by Ketchart and Wu and the analytical solution, the creep can
be predicted for any given time after construction for project specific soil
and reinforcement types.

73
5.2 Conclusions
From the study, the following conclusions can be made:
1. GRS retaining walls with granular backfill deform very little due to creep:
The GRS retaining walls selected for the study represent a variefy of
wall fypes using granular backfill and field conditions. The maximum creep-
strain in the reinforcement were less than 1.5 percent.

2. The actual reinforcement load is over-estimated:


In some of the selected walls, the tensile load in the reinforcement
could be estimated. In all those cases the tensile load was less than 10
percent of the reinforcements short-term strength. This suggests that the
design strength required for the reinforcement is too large
(overconservative). The design strength is the result of overconservative
creep reduction coefficients (CRCs) and partial factors of safefy required by
the AASHTO design method. This results in limiting the fype of reinforcement
in GRS walls to only higher-cost, high-strength geosynthetics.

3. The GRS retaining walls were stabilizing with time:


In all of the selected walls and laboratory tests, the cree~te was
decreasing with time indicating that the walls were stabilizing with time. The
tensile forces in the reinforcement are likely to decrease with time as the
creep strain-rate becomes very small (known as "stress relaxation").

4. A simple laboratory test and analytical equation can be used to predict


creep:
It was observed that the logarithmic creep-rate for the full-scale walls
and laboratory soil/geosynthetic composite tests decrease in a linear
relationship with logarithmic time. From this observation, an analytical
equation can be used to predict creep during the design-life of a GSR wall.
In full-scale applications, the simple test developed by Ketchart and Wu may
be performed to determine the creep modulus of a project specific
geosynthetic/soil composite. Long-term creep deformation of the wall can
then be determined in a rational manner by the analytical solution.

5.3 Recommendations for Future Study

A wide variefy of GRS retaining walls are being studied based on the
literature search and survey. However, the focus of the research seems to be
in several directions. The overall direction of the projects selected for this

74
study was mainly to demonstrate the functionality of geosynthetic soil
reinforcement and that it basically "works". The projects selected can be
considered the first full-scale GRS retaining walls in field conditions that have
been monitored for extended period of times. Future research in monitoring
the performance of full-scale walls should be focused in the areas of lateral
earth pressure distribution. localion of the failure surface and creep so that
specific data is collected to better understand the complex behavior of the
soil/geosynthetic composite.

Future research is required to determine the impact of material types


and the environment on the creep-rate relationship used to predict creep-
strain. Eventually. a database of creep-rate-time curves for specific
soil/geosynthetic composites could be established so that the magnitude of
creep ca n be estimated using analytical solutions.

75
Appendix A
Project Description Sheets

76
Interstate Highway 70 through Glenwood Canyon Project
Geotextile Earth Relaining Wall
Glenwood Springs, Colorado, USA

OP Of GEOTEXTILE TYPE
,EIAPORARY ROAD
• lC:.:k.T, o~ or) 17 • •• •• 2 ,,
7 5 3

.s:;.u"~.. c~l...of
16
15 •• ••
2
2
• ,,
7
7
5 3
3

,,,
14 2 7 5 3
13
12
•• •• 2
2
• 7
7 5
3
3

0:
11
10 •• •• 2
2
• 7
7,, 5
3
3
Wg
~ 8 •• •• •• ••
2
2
3
3 2
7
6 •• •• •• •• 2
2
3
3
Z
Z
5
4
•• •• •• ••
2
2
3
3
Z
Z

••• •• •• •• •• •• • ••
3
2 5 5 5 5 5
REINFORCEIAENT (TYP,) 5 5 5 5 5 5 5
1
2 3 4 5 6 7 8 9 10
IReIloining Wa. S c - . : (0) SEGMENT

ICor1finillll Soil: Well graded clean


Foundation Soil:
Reinforcement:
Facing: oh_
Campi assible sitts and
Geotextile (nonwooven polypropylene and polyester)
Wrapped face with

Short Term Cnoep Limited Creep Reduction


IlO!~!I!il!!@ strength (e) stranqthm Coefficientlg'l
400 IbIII @ 14l'!b strain 220 Ibill 55%
680 IbIII @ 145% stnoin 3751b111 55%
680 IbIII @ 66% _in 345 Ibill 4l'!b
18661b111 @ EIl'I6 _in 670 Ibill 4l'!b
4551b111 @ EIl'I6 sInoin 295 Ibill 66%
11551b1ft@75%strain 700 IbIft 66%
5251b1ft @ IIO'J(, strain 210 IbIII 4l'!b
a&llblft@ 55%_ 34J IbIft 4l'!b

130 IbIft'
$'

IPelmalnent Fill - 15 ft high

Wall Construction

MoniloMg instruments "'" not shown for clarity,


Excerpted f r o m " _ Eaf1I>,Reinforced Retaining Wall Tests: Glenwood Canyon,C_" (Bell, 1983).
Informatioll ...... jAed from the F-..t Highway Ad",; lislndioll" report entitled " E " -
Fabric R _ Earth Waif'. (1lenIIashandeIl and - . 1986).
PP = poI,,,opyIei'" type polymer. PE = polyester type polymer (DerakIlshonleh and - . 1986).
The short term strength was determined by the wide width tensile_ ata ~ strain rat. 0/ 10%. The specimens were
in _ prior to the test (DeraI<hshandeh and BareIt, 1986).
..--limiled streng1h was _ned by muftiplying 1I1e short term strength by the creep nodUction coeffICient.
creep reduction coefficients ...... dolennlned by reinforcement element tests performed by Dr. Richard Bell at Oregon State
(Derakhshandeh and Barrett, 1986).
samples were exhumed in 1984 to Investigate surfflilbility (Bell and Barrett, 1994).

77
Tanque Verde-Wrightstown-Pantano Roads Project
Wall Panels 26-30 and 26-32
Tucson, Arizona, USA

Wal DImg COnsb'uction (a)


9ARRIER~~ ...
TRAFFIC CONCRETE PAVEMENT Prolect Information
................. . ," ..... . ,...
'P~E"\"
WaUC.., _ _
2c.-1o ConfirVng Soil: W~~sand
Foundation: Wefi1Iraded grlMIy sand
Reinforcement: Geogrid
Facing: Concrete Panels
FTNAL GRADE
El.EVATlON

t o,_ Beinl2l~!IlMW StrenGth


Short term strength: 5400 IbIIt (b)
l.£.VEl.ING PAD 12' REiNFORCEMENT Creep limited strength: 19:33 IbIII (c)
Performance Ilml strain: 10'1!> (c)
TRAFFIC BARRIER
CONCRETE PAVEMENT Factor of safety: 1.5 (d)
Design strength: 13271b111
.,," . " ,' C r . p R _ Coef.: 37%

eonftnina son Properties


Un~Weighl: 122.51b1ft' (a)
Friction Angle: 34'(a)

.' . IDes~ SUrcha:=:n.


LEVEUNG PAD 12' REINFORCEMEN'
I
""'"

E>o:erpled from the Federa) Highway Administration Report No. FHWA-EP-OOOO1.al5.


Short term strength determined from unconfined tensile lest using a 4 inch wide sample tested for all layers at a constant 2%
rate at standard test cancfdians of 2JJ' C and 65% relatiVe humidity by McGown. et. al. (1985).
~ed from isochronous sliffneas curves _oped by McGown, et. aI. (1985).
Acoounts for ~ in design (FHWA, 1989).
A geogid sample was _ from. seporalesection oflhewall for ~anaIysis (Bright et. al., 1994).

78
Norwegian Geotechnical Institute Project
Wall Sections 'J' and 'N'
Skedsmo, Norway

Project Information

Will Coml!!!nentoll!)
Confining Soil: Welli/faded medium to fine
..nd
Foundation Soil: Gravely sand
R.i"f,"ce"'et~: Geogrid (polypropylene)
Facing: Exposed

ReiI.f9ls*,1Eld strength lei


7.2' Prima[X Beicforcement
REINFORCEMENT (TYP.)
Short term shngth: 3ED) IbIIt @ 13%
strain
Intermediate ReIDf2~ment
Short term slrength: 833lb11t @14% strain
(longitudinal)

Confinl~g ~II Pro_


Unft Weight: 10s.81b111' (b)
Friction Angle: :38' (b)

SW!:harge Ibl
9.8' . f!/:EiT IbIft' cyclic load
1044 IbIIt' unWorm surcharge from fl1l9.8 ft
Soction 'J' and 'N' PToftle
high
~
a) Excerpted from the paper entitled "Performance data ,--_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _,
for • sloped reitforced soU w.r (F..nn, 1990).
b) Excerpted from the paper entitled "Geosynthetic
STrength - Uftimate and Servfcoability Limft Sbde
Design" (Fannin, 1992)
c) Personal correspondance (Fannin, 1995). Long
term strength values used in the deaign were not .
provided.
d) Non-unifonn spacing of primary reinforcement
(Fannin, 1992).
e) Uniform spacing of primary reinforcement.
Intermediate reinforcement was used when spacing of
the primary layers moceeded 3 feet (Famin, 1992).

79
Japanese Railway Test Embankment Project
JR Embankment No.1
Experiment Station of Japan Railway Technical
Research Institute

JR Embanlo'nenl No.1 at Completion (Background)

107'
Project Infonnation

Will Coml!!nenta (I)


Confining So~: Sand with 16% fines
R_cemet~. Geogrld (Polyester)
Facing: Caat-ln-ploce UM!inl0r0ad
concntIe and cflSClele penoIs(b)

IShort Term Strength:


Relnt'on:oment S!rJIJqth
1880 JbIft (a)

I
IUn~ Weight:
~Sl!!l!ning Soil ProDedies
93.2Ib1ft' (0)
I

...,......,..... - .. -
~O".·, '.:
F." 0;' "",.,. =:
; --
-:.~. ~-J '

!. ~ - =- ,.~.~-r-::
. ~==-~-J
;~-: :§ _ ; -,~ 'r.'
_ ,.. - ; 1t,;W :,
_~ , J

Plan and Cross Sections (b)

Excerat,od from the paper entitled "Permanent geosynthotio-relnforoad soil retaiinll walls used for railway embankments In Japan"
1992).
sectiano, oflhe_ wall had contn>us rigid facing of ~ cast.fl-place ... ainIOIced concntIe will ighUy relnfolced
IcorllUUC1IOIl joinIs. Facing for the middle - . ccnsisIed of discn!Ie paneI-. (TaIaJoIoI. 1992).

80
Highbury Avenue Project
London, Ontario, Canada

r (;ON<:RE'IE PANElS
REINFORCEMENT (TYP.)

Wall Under Construction (a)


Project Infonnation

Components (bl
Confining Soil: Coarse sandi
Foundation Soil: Dense sandy till
lEVELING PAD ReinfOi oeme"t: Geogrid
(Polypropylene)
sc.ou: Facing: Concrete Panels
~--
~- ,b
Reinforcement Strenath
Wall Profile (aJ Not Available (c)

I Confin!nq Sol P!llpor!!os Ibl


Friclicn Angle: "!I:f - C

ITra!IIc loading
Syrchtrqe

_ " Cese sWdy or. mon_


0) Elccerpted from the ""per entitled "Review of Three Insirumented Geogrid Reinforc:ed Soil Retaining WtI/Its" (Bathurst, 19511 ).
b) E>cerptecI from the _ propped panel waII", (BathunIt. 1992).
c) Specific reiilfoccetll8ilt strength datil used In the design was not available. Hovvever. the reinforcement used was 8 Tensar
UX1aJO (Bathurst, 1992). c..p limited strength for UX1aJOranges from 2,0001<> 3,4501b'ft ec:c:cJI"Qng I<> manuf_
fiIeraIure (T....., 1996).

81
Federal Highway Administration Research Project
Wall Number 9
Algonquin, Illinois, USA

WaU Completioli (a)

Pro'eet Information
Wall CionmooetM Itl
Confining Soil: Well graded sand and gravel
Foundation Soil: Medium dense gravely sand
INCLINOMmR r SURCHARGE Reinforcement: Geogrid (Polyester)
...l. -_ Facing: Modular Blocks

1[====~====]Ran~~ko~gI~!~"~~~~~~~======l
V/
MODU!AR ,\--/ .
,
INCLINED
FAC ING UNIT
v""'" 34'
SURCHARG E
] cO MachinePirection
"''' '''''' .' 'i'': =T~:~=~: ~:~/5%strain(b)
f:::
l- J:{:~~t~~~:~:~~~;~~~~~ti:~?!j} stre~:
DesIgn 1002 IbIft (d)
t:
,- rR~_~FORCEMENT (TYP.),/ Cross Mac!!ne 0irection
Short Term strength: 1572lb1ft@ 19%_

f~ Creep Um~ed strength: 9241b1ft

~c
SCONFINING SOIL J Design strength: 636lb1ft

o 20 :::: 1",",6'. Performance Umft stJain:


Factor of Safety: 1.5 (e)
10% (a)

'" '5'
(>
~;
C""", Reduction Coef; &0% (0)

t: !e
'j Confining Soli Properliu m
= t:i'
~.¥{
UM Weight: 12S.6Ib1ft'
F= Friction Angle: "!II'

-
~,

Surcharge/al
Uncompecled fiU 6.9 feet ft above the top of the waU
2.7.j Untt Weight: lC6.61b1113
, 8.9
14.1'

Wall Number 9 Profile

.Hll!!!!i..
a) El<cerpIed from the paper _
b) The short term sIrengII1 is excerpted from Simac (1990). Assumed to be
(ASTM 0-4596).
_ned _
"1nstrurnerlt2d field performanoe of a 6 m geogrid soil war. (Simac, 1990).
on the wide width tensile strength test

c) The long tenn strength 18 .'' ' ' J>Ied


from Simac (1990). Assumed to be Determined using 10,000 hour creep tests described in the
FedonJ Highway Administration'. _ to"", 27 report (FHWA, 1969).

e) The factor of safety ac:c:cuu tor long tenn _1Iy and construcIion
Adminislrolion's _ foroe 27!epor1 (FHWA. 1969).
site.-.-
d) The design strength is detennined by dividing the long tenn strength by the factor of safety.
described in the F _ Highway

82
Seattle Preload Fill Project
Southeast Wall
Seattle, Washington, USA

SURCHARGE ~
• BONDED RESISTANCE STRAIN . GA.:.;::::
o MECHANiCAl EXliNSOMET£R
INCLINOMETER - £NnH PRESSURE CDJ..
CASING a TtlERMOCOUPl[
6 RDtOTE SETTl.DIENT GAUGE
II IIDIJCTANCE COIL $TRMII GAUGE

...... Wall alter Construo\lOn a)


Pro -eet Infonnation
_Cw ............ lal
Confll1ing Soil: Gravely sand
Foundation: Granular soil overlying
compressible aoft clay
Reinforcement: G_le
Facing: Wrapped face

Rejnfon:emen! Strength

t
REINFORCEMENT Layers 1-8 (pqtyester) C. R <..: bOo/",
(TYP.)
32' Short term strength: 12-400 Ibill (b)
Creep Nmited strength: 73161b1ft (e)
Design strength: EIl971b1ft (d)
• t.ayers 9-16 (Polypropylene) <! f!. C. :: 400l G
Short tenn strength: 6133lb1ft
Creep limited strength: 2453lb1ft ,
Design strength: 2D441b111 I
SOUtheast Wal Profile (a) Layers 17-25 (pP/ypropy!enel C .. t . 4 0</..
Short term strength: 41331b1ft
ICreep limited s1r8ngIh: 16531b111
Design strength: 13771b111 I
I:Layers 26,'33 (PoJypropYIene)
Short term strength: 20661b1ft
c. ~ < < <\ 0 ,/.

Creep ImiIecI strwr9h: 8271b1ft


Design SlrengIIl: 13771b111

Confining Soli Properties m


Unit Weight: 13llblft'
Friction Angle: 3if'
SUn:hlrge!al '
Sloped (2:1) fill 17 ft aboYe the top 01 the wall I

Noles
a) EloceIpted from the _ _ · P.. 1tH ,,1& o:e 01.12.6 m higI1 geoIextiIewal in _ , Washington", (AIIen,1992).
b) Short tenn strength detennined from the wide width tenSIle strength test (ASTM D-45B5).
e) Long term strength is detennined by muRiplying the short tenn strength by the e.eep reduction coefficient.
d) The design strength is detennined by dividing the long tenn strength by the _ 01 safely.
0) The factor 01 safety accounts lor iniemal stability (Allen, 1992).
I) EstimaIad!nil weight and friction angle used fer design. Actual unit weight was 1341b1ft' and the friction angle varied from If$' to
1fT' (Allen, 1992).

83
AppendixB
Conservatism Index
Calculation Brief

84
Phil Crouse Conservatism Index 4119/96
Calculation Brief
Purpose:
The purpose of this calculation brief is to determine the conservatism index (CI) for the
selected projects.

Methodology:
The CI value is computed using a limit equilibrium analysis where resisting lateral force
provided by tensile strength of the reinforcement is divided by the driving lateral force of
the earth. The CI value is based on the same principles of limit equilibrium used in the
current design methods where the resisting tensile force is entirely provided by the
reinforcement and redistribution of stresses due to the soillgeosynthetic interaction are
ignored.
The CI value is based on the average lateral force (F) acting on the reinforcement layers
assuming a linear Rankine active pressure distribution. The CI value is computed by
dividing the short-term tensile strength of the reinforcement by the average lateral earth
pressure acting on the wall. If the wall has different reinforcement spacings or
strengths, a weighted CI value is computed. The computation is illustrated below
followed by a summary of the results and detailed computation for each selected
project.

T
H,
----- -"'-

eI. = "T1l1e"'y wH~1l1i' T..... lT;: $HDf!.T-TIf>lM.


..s rLt::'l'J.c"i"H-.
~,-::. fl.rr/'tlF.~(f;:;"Me-""i
85
Results

Project
Interstale Highway 70 Ihrough
Wan
Name
Geotextilo Earth
Height
1ft) n •
Ideg)
T
IbIn'
q
IbIII' Ka CI
I

16 17 35 1:Jl 1950 0.27 0.44


Glenwood Canyon Retaining Wall
anque Verd...Wrlghtslcwn-Pantano Wall Pane! 263)
15.6 10 34 122.5 S 0.28 8.7
Road.
Wall Panel 263)
16.1 10 34 122.5 S 0.28 8.1
Norwegian Geotechnical Wall Section J 15.7 4 38 108.8 1044 0.47 0.4
ex
0- Inst~ute Wall Section N 15.7 7 38 108.8 1044 0.47 1.8
iJapanese Railway Teat Embankment JR Embankment
16.4 16 35 (a) 93.2 0 0.27 3.5
No. I
Highbury Avenue Hlghbury Ave.
23.3 9 35(0) 125 (a) 0 0.27 8.2
Wall
Federal Highway Administration Wall No. 9 20 8 :l9 125.6 728.6 0.23 7.9
~.alIlePreload FIR SOUltHIaslWall 4f .3 33 36 1:Jl 2210 0.26 2.6

).gmt !!1m
n • Number of relnforcemerrt layers III) Value estimated since not available In the literature.
•• Oaslgn Internal friction Ingle
y = Oaslgn unll weight
q = Surcharge
Ka • Acllve laleral earth pre.sure colfficlent
CI • Conservatism Index
PhU Crouse Conservatism Index 4119/96
Calculation Brief

Interstate Highway 70 through Glenwood Canyon Project


Properties:
H := 16·ft Hq := lS·ft Height of Surcharge
Ib
~ := 3S·deg yq := 130·- Assumed unit weight of surcharge
ff3
y := 130·-
Ib
q := Hq·yq q =1.9S0103 °lbott""2
ft3
Ayerage lateral forces for each group with same reinforcement strength:

Ka= tan ( 45·deg - ;f Number of layers and reinforcement each group:


n1 := 9 n2 := S n3 := 3
Ka =0.27

Group 1: Group 2:
FH1= Ka.[ ,S·Y· (9.S.ft)2 + q.(9.S.ft) ] FH2= FH1 + Ka·[o.s.y. (4.3·ft)2 + q.(4.3.ft) ]
FH1avg= FH1 FH2avg := FH2
n1 n2
-1 3 -1
FH1avg = 734.420Ibon FH2 = 9.21°10 °lbon
Average Lateral force Average Lateral force
for group 1 for group 2
Group 3:

FH3 := FH2 + Ka.[0,S·Y·(2.2.ft)2 + q. (2.2.ft) ]


FH3avg := FH3
n3
3 -1
FH3avg = 3.49 010 °lbott
Average Lateral force
for group 3

fI
!
~ ___ F~1l.

('.5'

r---\\~"'E---r~
87
Phil Crouse Conservatism Index 4119/96
calculation Brief

Conservatism index for each group


Reinforcement strength for each group:
_ Ib Ib
Tult1 .- 400· - Tull2= 680·-
ft ft

CI1 := Tult1 CI1 = 0.54


FH1avg
. Ib
Tult3 = 455·-
CI2 := Tull2 CI2 = 0.37 ft
FH2avg

Tult3
C13- CI3 = 0.13
FH3avg

Weighted CI value

CI := 9.5.ft. CI1 + 4.3·ft. CI2 + 2.2·ft. CI3


16·ft 16·ft 16·ft

CI = 0.44 ct- OM for Intersta1e Highway 70 through Glenwood canyon Project

Tangue Verde - Wrightstown. Pantano - Roads


Project Wall Panel 2HO
Properties:
Ib
• := 34·deg Tult := 5400· -
ft
y := 122.5·-
Ib
ft3 q = 359.;
Ayerage lateral forces for each group with same reinforcement strength:

Ka= tan ( 45·deg -1f Number of layers and reinforcement each group:
n1 := 5 n2 := 5
Ka =0.28

FH1 := Ka.[.5 ·y· ( 10.1 ·ft)2 + q.( 10.1·ft) 1


FH1avg := FH1
H~ ,
,
! \
\o.-l F.1'1
n1
' i \
-1 ~ _ l '"
r--,
FH1avg = 558.31 o lb o ft
Average Lateral force
for group 1
.
: \ .
I.

88
Phil Crouse Conservatism Index 4119/96
Calculation Brief
GrouD2:
FH2 := FH1 + Ka.[0.5.y.(5.5.ft)2+ q.(5.5.ft)]

FH2avg= FH2
n2
3 -1
FH2 = 3.87'10 ·Ib·ft
Average Lateral force
for group 2
Conservatism index for each group
Tult
CI1 := - " " - CI1 = 9.67
FH1avg

CI2 := Tult CI2 = 6.97


FH2avg

Weighted CI value

CI := 10.1·ft. CI1 + 5.5·ft .C12


15.6·ft 15.6·ft

CI = 8.72 CI = 8072 for the Tanque Verde - Wrightstown - Pantano Roads Project
Wall Panel 26-30

Tanque Verde - Wrightstown - Pantano - Roads


Project Wall Panel 26-32 .

FH1 := Ka.[ .5'1'( 10.6.ft)2 + q.( 10.6.ft) ]


FH1avg := FH1
n1
-1
FH1 avg = 604.3 Ib o ft
0
(, ,"0.,," 1- 1-
Average Lateral force I~I' ;
S LA'IEts
T.l.L~: S40Ci(n-
:,
;...\
for group 1 i"
~

FH2= FH1 + Ka.[ 0.5·y·(5.5·ft)2 + q.(5.5.ft) ] ~


__
~""4f--

.
FH2avg := FH2
n2
Average Lateral force
for group 2

8'1
PhU Crouse Conservatism Index 4119/96
Calculation Brief
Tult
CI1 .- CI1 =8.94
FH1avg

CI2 := Tult
FH2avg

Weighted CI value

CI := 10.I ·ft. CI1 + 5.5·ft .CI2


15.6·ft 15.6·ft

CI =B.ll CI = 8.1 for the Tanque Verde - Wrightstown - Pantano Roads Project
Wall Panel 26-32

Norwegian Geotechnical Institute Project Section 'N'


Properties:

.:= 38·deg Tult := 3600. Ib


ft
B := 26.6·deg Sloped face
lb
1 := 10B.B. Ib
3 q := 1044·- n := 7 Reinforced layers spaced evenly
11 f(
Average lateral forces for each group with same reinforcement slrength:

cosH _ B)2
~---------~~~~-------
Ka =0.47
CDS(B)2,cOS(B) '[ 1 + ( sinW·sin<+l ).5]2
cos(B) ·cos(B)

FH1 := Ka. [ .5·Y·(15.7.ft)2 + q.(15.7.ft)]

FH1avg := FH1
n
3 1
FH1avg = 2.01'10 ·lb·tt-
Average Lateral force

Conservatism index
Tult CI = 1.8 for the Norwegian Geotechnical
CI- CI = 1.79
FH1avg Institute Project Section 'N'

90
Phil Crouse Conservatism Index 4119/96
Calculation Brief
Norwegian l:ieo1ecnmcal mSDlU1e ..rolect ~ectlon ..1.

Properties:
Ib
+:= 38·deg Tult := 3600·- B := 26.6·deg Sloped face
ft
Ib
1:= 108.8·- Ib
Ie q:= 1044·-
ft2
Average lateral forces for each group with same reinforcement strength:

cos(+ _ B)2
Ka- ----------==~~~---------
Ka = 0.47
COS(B)2.COS(B).[ 1 + ( sinW·sin(!/l) ).5J2
cos(B)·cos(B)

n1 := 2 2 reinforcement layers in group 1

n2 := 2 'I reinforcement layers in group 2

FH1avg := FH1
n1
FH1a = 4.36°103 0lbott-1 Average Lateral force
vg for group 1

FH2 := FH1 + Ka· [.5·1· (4.S·ft/ + q.( 4.S·ft) ]

FH2avg := FH2
n2
3 -1 Average Lateral force
FH1avg = 4.36°10 °lbon
for group 2

TI
is:'

I :
14.5',

i f

91
Phil Crouse Conservatism Index 4119/96
Calculation Brief

Weighted CI value
CI1 '- Tult CI1 = 0.41 CI2 ~ Tult CI2 = 0.31
FH1 FH2

CI:= 11.2·ft. CI1 + 4.5·ft .C12


15.7·ft 15.7·ft

CI = 0.38 CI = A for the Norwegian Geotechnical Institute Project Wall Section 'N'

Japanese Railway Test Embankment Project


Properties:
Ib
• :"', 35·deg Assumed Tult := 1880·-
ft
Ib
T := 93.2·- n := 16 Reinforced laye/'S spaced evenly
ft3

, Average lateral forces for each group wj!h same reinforcement strength:

Ka= tan ( 45·deg -1f


Ka =0.27 T
FH := Ka.[ .5·y·( 16.4.ft)2] 1~.4 '

FHavg := FH1
n
-1
FHavg = 545.6·lb·ft

CIValue

CI := Tult CI = 3.45 CI = 3.5 for the Japanese Railway Test Embankment


FHavg Project

Highbury Avenue Project


Properties:
Ib
, := 35·deg Tult := 2000·-
ft
Ib
T := 125·- Assumed
ft3

92
Phil Crouse Conservatism Index 4119196
Calculation Brief
'-"\~ ....\. i='on...c. y

Ka := tan ( 45·deg - ~r Number of layers and reinforcement each group:


n1 := 5 Number of reinforcement layers in group 1
Ka =0.27 n2 := 4 Number of reinforcement layers in group 2

FH1= Ka· [.5'1'( 16.3·ft)2 )

FH1avg := FH1
n1

,
-1
FH1avg = 899.99·lb·ft Average Lateral force
for group 1

Group 2:
,,
I
FH2 := FH1 + Ka. [ 0.5·7·(7.ft)2]

FH2avg := FH2
, II
~~.?
n2 !
3 -1
FH2avg = 1.33'10 ·Ib·ft

Weighted CI value

CI := 16.3.ft. CI1 + ~.CI2


23.3·ft 23.3·ft

CI =8.23 CI = 8.2 for the Highway Avenue Project

Federal Hjghway Administration Research Project


Properties:
Ib Ib
+= 39·deg Tult: = 2604·- yq := 105.6·- Unit weight ofsurcharge
ft ft3
Ib
T: = 125.6·-
ft3 Hq := 6.9·ft Height of surcharge

q= yq·Hq q = 728.64·lb·ft--2 Surcharge

93
Phil Crouse Conservatism Index 4119/96
Calculation Brief

Ka = tan ( 45·deg - ~r Number of layers and reinforcement each group:


n1 := 5
Ka =0.23 n2 := 3
Group 1:

FH1= Ka.[.5.y.( 1Ht)2 + q.]l.ft]

FH1avg := FH1
n1
-1
FH1avg = 710.45-lb-ft Average Lateral force
for group 1 'l-: 7"z B ~ Ib/Ff'"

Group 2: -,---. 1 1 I L
,,
~
6QO.,~ ~

!: I.""I;;'~
, ....~~ ~ l £04 IblFT
!
FH2= FH1 + Ka. [ 0.5·Y· (9.ft)2 + q.9.ft] I
,,
FH2avg := FH2
n2

FH2avg =2.07-103 -Ib-ft-1


Weighted CI value

]l·ft 9·ft
CI := -·CI1 + - ·CI2
20·ft 20·ft
CI =7.88 CI =7.9 for the Federal Highway Administration Project

Seattle Preload Fill Project


Properties:
+:= 36·deg Hq := 17·ft Height of Surcharge
Ib
yq := 130· - Assumed unit weight of surcharge
Ib fl3
1 := 130·-
~ q := Hq.yq
3
'q = 2.21 '10 -Ib·ft
-2

94
Phil Crouse Conservatism Index 4119/96
Calculation Brief
Average laterallOrces Tor eacn group WI!!l same relnTorcemem strenam:

Ka := tan (45.deg -1) 2


Number of layers and reinforcement each group:
n1 := 8 n2 := 9 n3 := 8 n4 := 8
Ka =0.26
Group 1:

i
'6RO\.);:' '1.
FH1= Ka· [.5.y-( I I 03.ft)2 + q.( 1I .3.ft) ] 9 U:.".£'\)
~'T,"Z0('6IbJF'"
FH1
FH1avg .- - 6Ro:i p - " - -
n1 '" \.. A-ot E:t\.>
3
FH1avg = 1.08 010 °lbott-
1 4\.3' ''''c"t ~ ~IH II,F,-
I
Average Lateral force ~l 'b' l
:>
6~CNQ?
g "."'6',,-,).
-
for group 1 i l -r~~.;\= (':)tlllJl/Fi
; ! a' 611""4 - -
Group 2: ) '1>' E? /"'A"'.r"~
2 ] \2.400 1'tJ/FT
FH2= FH1 + Ka· [ 0.5·y·(10·ft) + q·(10·ft) ~--"---'I--~~_ _ _ _ _-!~~...l4_--l.

FH2avg := FH2
n2
FH2 =1.61 0104 0lbott-1
Average Lateral force
for group 2

Group 3:
FH3= FH2 + Ka.[ 0.5·Y' ( 10.ft)2 + q.( 10.ft) ]
FH3avg := FH3
n3
3 1
FH3avg = 2.94°10 °lb°ft-
Average Lateral force
for group 3
Group 4:
FH4 = FH2+ Ka.[0.5.y.(10.ft)2 + q:(10.ft)]

FH4avg := FH4
n3
3 1
FH4avg = 2.94°10 0Ib°ft-
Average Lateral force
for group 4

95
Phil Crouse Conservatism Index 4119/96
Calculation Brief
Reinforcement strength for each group:

fu fu fu Ib
Tult1 := 2066·- Tult2 = 4133·- Tult3 := 6133·- Tult4 := 12400·-
ft ft ft ft

Tult1
CI1 .- CI1 = 1.91
FH1avg

Tult2
CI2 .- CI2 =2.32
FH2avg

Tult3
CI3 .- CI3 = 2.09
FH3avg

Tult4
CI4 .- CI4 = 4.22
FH4avg

Weighted CI value

CI - C11· 1l.3·ft + CI2. 10·ft + C13. 10·ft + C14. lO·ft


41.3·ft 41.3·ft 41.3·ft 41.3·ft

CI = 2.61 CI = 2.6 for the Seattle Preload Fill Project

I\lc~~·. ScI> ~"P<t_'" A t-c>1l.. ~~<>:S6<-,," ~€"<"\I'"\\'''4.


""- l2.t!i=e"7l.c:", ,,-'

96
Appendix C
Conservatism Index

97
lE+OO

_ 1E·01

I lE.Q2
i
e-
li!
u 1E-03

~ lE-04
10 100 1000
Tlma Mia, Conslrucllon (day)
_ _ WaU Panel2&3J
_ _ Wan Panel 2632

Figure C.l
Creep-Rate-Tlme Curve for the Tanque Verde - Wrightstown· Pantano Roads Project
lE+OO

lE.()1

! lE.Q2

~ lE.(13

i
D.
lE-04
II
!!
U lE.Q5

lE.Q5
-.0
-.0
lE.o7
100 lCXXl
nme Allel SUlcharge (day)
-6- Section N
___ Section J

-Regression line for section N


RSQR= .89
- - - Regression line for section J -
RSQR~ .00
Figure C .2
Creep-Rate-Tlme Curve for the Noweglan Geotechnical Institute Project
lE+OO

1E-01
~

t
~
~ lE-02
'"Il. ~
. ... "I
a 1E-03
I
,. .:.'"

8
lE-04
100 lCXXl
Time Aller Prop Relea•• (day)
-+- Highbury Avenue Wall
-Regression line - RSQR = .96

FIgure C.3
Creep·Rate·nme Curve for the Hlghbury Avenue Project
lE+OO .. --- ...

lE-Ol

ii lE-02
!; -
~!!.
u

I-... Cit p.or


.......
-o- lE-03 r---....

lE-04
10 100 1000
TIme After Surcharge (day)
--Regression line RSQR = .89
-....W.119

FigureC.4
Creep-Rate-Time Curve for the Federal Highway Administration Project
1E+00 -_ .. - --. .. _A. .-.--. ,-

1E.{)1
~

1E-02

.; ~ 1E.re
~ ~§~.~I
"'i
.a. ..... ~
1---- - .
~

~
!~ - I I I II : ...
lJ
~~
1E-04

0 1E.(5
'" -

1E.oo

1E-07
10 100 1CXXJ
Time Aller Surcharge (day)

_ _ Sealile Preload An Project


-Regres~an ins RSQR = .68
Figure C.S
Creep-Rate-Tlme Curve Sea"le Preload Fill Project
::: r Fi if
..... 1E+01
~
~ 1E+OO
~
..... 1E.Q1
! 1E.Q2

i
D
1E-03
li!!i!;iiiiiliiiillliiii!!ii!!iil!i!iillliii!!ii!!Ii1i1!iiiili!!i!ii!!iii!iiiilli
8 U 1E.Q4~_I111==-="
1E.a;~
1E.Q6
1E.Q7 I I I III I I III I I III I I III
0.01 0.1 10 100
nme(day)
_ _ Lab Test 0-1

- 0 - Reges~on ine RSQR =


0.99

Figure C.6
Laboratory Creep Test 0·1
lE<01

lE+OO

1E'()1

i lE.Q2
l
! 1E.()3

t
U
lE-04

lE.(l;
~
lE.()6

1E'()7
0.01 0.1 1 10 100
Time (day)

_ _ Lab Test H·l

--Regr...lon Una
RSQR-.96
Figure C.7
Laboratory Creep Test H-1
lill In
III ,
i

, ,

fJ.
~

" V
i
Z
\J

II ";".
II:

11111 :
.. I-
_ <II
" a.
1/ !
::JU
e
II .~~
II.. 0

i
j

q
~ o

-
UJ

105
§

~
"
It:
a
II>
.,
It:

:;1 "
:J
"
~e
51
1il
"
t-
.co
!l
..
.:!

+ I

....
-j N

III:

.-••..
~,.

0 ....
.. 0
::1-
m.,
~

.... 0
..a
.!:!

q
~ o
w

(ADp/%) aJDI daal:)


106
lE+02

lE+01

lE+OO

1 E-01

i
'>;
lE-02

! lE-03

!
<J
1E-04

lE.(l;
0
'-I
1E-OO

lE-07

lE-OO
0.01 0.1 1 10 100
TIme (<hIy)

-6--Leb Test W-1

Figure C.10 _ _ Regression Lin. RSQR = .\lEI


Laboratory Creep Test W-1
References

Allen T.M., Vinson T.S., and Bell, J.R. (1982). ''Tensile Strength and Creep
Behavior of Geotextiles in Cold Regions Applications", Proceedings of
the 2nd International Conference on Geotextiles. Las Vegas, Vol. 3,
August 1982, pp 775-780.

Allen T.M., Christopher B.R., and Holtz R.D. (1992). "Performance of a 12.6 m
High Goetextile Wall in Seattle, Washington", Geosvnthetic-Reinforced
Soil Retaining Walls. Ed. Jonathan T.H. Wu. AA Balkema Publishers,
Rotterdam, Netherlands, pp. 81-100.

AASHTO (1990). "In Situ Soil Improvement Techniques", Report on Task Force
27, Joint Subcommittee Report of MSHTO-AGC-ARTBA American
Association of State Highway and Transportation Officials, Washington,
D.C.

AASHTO (1992). ''Standard Specifications for Highway Bridges", 15th edition,


AASHTO Subcomittee on Bridges and Structure.

ASTM D 4595. "Tensile Properties of Geotextiles by the Wide Width Strip


Method", American Society for Testing and Materials, Philadelphia, PA.

108
Bathurst, R.J •• and Simac M.R. (1991) ."Review ofThree Instrumented Geogrid
Reinforced Soil Retaining Walls". Geosvnthetjcs: Design and
Performance. Vancouver Geotechnical Society - 6' th Annual. One
Day Symposium. May.

Bathurst. R.J. (1992). "Case Study of a Monitored Propped Panel Wall",


Geosvnthetic-Reinforced Soil Retgining Walls. Ed. Jonathan T.H. Wu.
A.A. Ballcema Publishers, Rotterdam, Netherlands, pp. 159-166

Bell. J.R., Barrett. R.K. and Ruclcman, A.C. (1983), "Geotexfile Earth-Reinforced
Retaining Wall Test: Glenwood Canyon, Colorado", Transportation
Research Record 916, National Academy Press, Washington, D.C .• pp.
59-69.

Bell, J.R. and Barrett. R.K. (1994) "Survivability and Durability of Geotextiles
Buried in Glenwood Canyon Wall", Transportation Research Board
Meeting Paper, Washington, D.C. January. 1994

Berg. R.R .• Bonaparte. R.• Anerson. R.P, and Chouery. V.E. (1986).
"Design, Construction and Performance of Two Geogrid Reinforced Soil
Retaining Walls". Proceedings from the Third Intemational Conference on
Geotexfiles. Vienng. Austria . pp. 401-406.

109
Bright. D.G., Collins, C.G .. and Berg, R.R. (1994). "Durability of
Geosynthetic Soil Reinforcement Elements in Tanque Verde Retaining Wall
Structures", Transportation Research Record No. 1439, National Academy
Press, Washington, D.C., pp 46-54.

Christopher B.R.. Bonczkiewicz C, and Holtz R.D. (1992). "Design, Construction


and Monitoring of Full Scale Test of Reinforced Soil Walls and Slopes",
Recent Case Histories of Permanent Geosynthetic-Reinforced Soil
Retaining Walls. Ed. Fumio Tatsuoka and Dov leshchinsky, Balkema
Publishers, Rotterdam, Netherlands, pp. 45-60.

Clayboum A.F. and Wu, J.T.H. (1993). "Geosynthetic-Reinforced Soil Wall


Design," Geotextiles and Geomembranes, Vol . 12, pp. 707-724.

Collin C.G .. Bright, D.G., and Berg, R.R. (1994). "Perfonmance Summary of the
Tanque Verde Project-Geogrid Reinforced Soil Retaining Walls",
Proceedings. Earth Retaining Session, ASCE National Convention,
Atlanta, GA

Derakhashandeh, M. and Barrett, R.K. (1986). "Evaluation of Fabric


Reinforced Earth Wall", Federql Highway Administration Report No.
CDOH-DTP-R-86-16, lakewood, CO, September.

11 0
Fannin, R.J. and Hermann, S. (1990). "Performance Data fora Sloped
Reinforced Soil Wall", Canadian Geotechnical Journal, Volume 27,
Number 5, pp. 676-686.

Fannin, R.J. and Hermann, S. (1992). "Geosynthetic Strength - Ultimate and


Serviceability Umit State Design", Proceedings of the ASCE Specialty
Conference on Stability and Performance of Slopes & Embankments II,
University of California, Berkeley, Ca., pp. 141 1-1426.

FHWA (1989), "Tensar Geogrid-Reinforced Soil Wall", Federal Highway


Administration Report No. FHWA-EP- 9QOO1"()()5, Washington, D.C.,
January.

Fishman, K.L., Desai, C.S., and Sogge, R.L. (1993). "Field Behavior of
Instrumented Geogrid Soil Reinforced Walr' Joumal of Geotechnjcql
Engineering, Vol 119, No.3, August 1993, pp 1293-1307.

Ketchart K. and Wu, J.T.H. (1996). "Long-Term Performance Tests of Soil-


Geosynthetic Composites", CDOT-CIT-96-1, Colorado Department of
Transportation, Denver, CO.

McGown A., Andrawes K.Z., Yeo K.C., and DuBois D. (1984). "The Load-strain-
.time behaviour ofTensar Geogrids", Proceedings, SYmposium on
..
Polymer Grid Reinforcement in Civil Engineering, London, England, pp
11-1 7.

111
Simac, M.R., Christopher, B.R., and Bonczkiewicz. C. (1990). "Instrumented
Field Performance of a 6 m Geogrld Soil Wall", Geotextiles.
Geomembranes and Related Products. Ed. by Den Hoedt, Balkema
Publishers, Rotterdam, Netherlands, pp. 53-59.

San, K. and Matsui, T (....._-1). "1 G Self Weight Failure Test and Simulation of a
Geosynthetic Reinforced Retaining Wall" submitted to ASCE
Geotechnical Journal for publication review.

Tatsuoka F., Tateyama M., Tamura Y., and Yamauchi, H. (--1, "Lessons
Learned from the Failure of a Serles of Full-Scale Geosynthetic-
Reinforced Soil Retaining Walls", Nonpublished.

Tatsuoka F., Murata 0., and iateyama M. (1992). "Permanent geosynthetic-


reinforced soil retaining walls used for railway embankments in
Japan", Geosynthetic-Reinforced Soil Retaining Walls. Ed. Jonathan
i.H. Wu. A.A. Balkema Publishers, Brookfield, VT, pp. 101-130.

Vidal H. (1966). "La terre armee', Annis L'ints. Techn. de Batiment et des
Travaux Publics, Serle Materlaux 30, Supplement no. 223-4, July-August

112
Wu J.T.H. and Tatsuka F (1992) . '"Laboratory Model Study on Geosynthetic
Reinforced Soil Retaining Walls", Discussion, Journal of Geotechnical
Engineering. American Society of Civil Engineers. Vol. 118, No. 3, pp.
496-498.

Wu, J.T.H. (1994a). "Long-Term Creep Behavior," Recent Case Histories of


Permanent Geosvnthetic-Reinforced Soil Retaining Walls. Ed. Fumio
Tatsuoka and Dov Leshchinsky, Balkema Publishers, Brookfield, VT, pp.
343-344.

Wu, J.T.H. (1994b). "Design and Construction of Low Cost Retaining Walls: the
Next Generation in Technology", CTI-UCD-1-94, Colorado
Transportation Institute, Denver, CO.

Wu, J.T.H. and Helwany, S.M.B. (1996). "A Performance Test for Assessment of
Long-Term Creep Behavior of Soil-Geosynthetic Composites,"
Geosvnthetic intemational. Journal of International Geotextile Society,
Vol. 3, No. 1, pp 107-124.

113

You might also like