Long-Term Field Performance of Geosynthetic-Reinforced Soil Retaining Walls
Long-Term Field Performance of Geosynthetic-Reinforced Soil Retaining Walls
Long-Term Field Performance of Geosynthetic-Reinforced Soil Retaining Walls
CDOT-DTD-97-12
Phillip E. Crouse
Jonathan T.H. Wu
Center for Mechanically Stabilized Backfill Research
University of Colorado at Denver
Final Report
May 1996
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
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.
Retaining Wan
17. SECURTm" CLASSIFICATION 18. SECURITY CLASSIFICATION 19. SECURITY CLASSIFICATION 20. LIMITATION OF ABSTRACT
OFREP(,RT OFTmSPAGE OF ABSTRACT
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
2
SURCHARGE
RETAINED SOIL
REINFORCEMENT (TYP.)
FOUNDATION
Figure 1.1
Components of Q GRS Retaining Wall
3
1.1 Background
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.
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.
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.
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:
7
2. Literature Review and Survey of Creep Performance In GRS Retaining Walls
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
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.
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.
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.)
.----1--+....:::...."""""" (M>.)
CONCfIIfTE PANELS
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).
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.
14
the reinforcement (Bright. Collins and Berg, 1994) which is described in
Chapter 3.
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.
15
2.1.4 Japan Railway Test Embankment Project
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".
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.
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.
19
GEOSYNTHETIC REINFORCEMENT
--
-- --
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.
21
Table 2.2
Reinforcement Tensne Strength In Selected Projects
fSeattle Preload Fin Soulheast Wall 41.3 33 1.25 2066-12400 ,10-59 689-/l:n7 689-/l:n7 I
L _ ----- _. - L .. _____ --
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.
25
,
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:
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.
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.
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.
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
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.
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.
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.
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:
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.
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
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
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.
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
(.,)
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
~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.
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
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 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
43
•.
'
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
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.
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
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
Figure 3.6
Reinforcement Creep-nme Curve for the SeaHie Preload Project
3.3 Conservatism Index
51
Table 3.2
Parameters Used to Compute the Consevatl.m Index
Highbury Avenue Highbury Ave. Wall 23.3 9 0.4 :l6 (a) 125 (a) 0 0.27 6.2
~ 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
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
_ _ 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
Pantano Roads
Wan Panel 26-32 16.1 7 1(b) 1.13
~
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.
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.
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.
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.
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 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 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.
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
~
0.00
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)
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.
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
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
ACTUAL DATA
A REGRESSION LINE
w
!<e
£t:
0-
W
W
£t:
U
m = SLOPE
1
TIME (t/t o )
70
Using the plot illustrated on Figure 4.3, the linear relationship can be
represented by the following equations:
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)
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:
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
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.
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.
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
130 IbIft'
$'
Wall Construction
77
Tanque Verde-Wrightstown-Pantano Roads Project
Wall Panels 26-30 and 26-32
Tucson, Arizona, USA
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
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
107'
Project Infonnation
I
IUn~ Weight:
~Sl!!l!ning Soil ProDedies
93.2Ib1ft' (0)
I
...,......,..... - .. -
~O".·, '.:
F." 0;' "",.,. =:
; --
-:.~. ~-J '
!. ~ - =- ,.~.~-r-::
. ~==-~-J
;~-: :§ _ ; -,~ 'r.'
_ ,.. - ; 1t,;W :,
_~ , J
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.)
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)
ITra!IIc loading
Syrchtrqe
81
Federal Highway Administration Research Project
Wall Number 9
Algonquin, Illinois, USA
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%_
~c
SCONFINING SOIL J Design strength: 636lb1ft
'" '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'
.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
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 $TRMII GAUGE
Rejnfon:emen! Strength
t
REINFORCEMENT Layers 1-8 (pqtyester) C. R <..: bOo/",
(TYP.)
32' Short term strength: 12-400 Ibill (b)
Creep Nmited strength: 73161b1ft (e)
Design strength: EIl971b1ft (d)
• t.ayers 9-16 (Polypropylene) <! f!. C. :: 400l G
Short tenn strength: 6133lb1ft
Creep limited strength: 2453lb1ft ,
Design strength: 2D441b111 I
SOUtheast Wal Profile (a) Layers 17-25 (pP/ypropy!enel C .. t . 4 0</..
Short term strength: 41331b1ft
ICreep limited s1r8ngIh: 16531b111
Design strength: 13771b111 I
I:Layers 26,'33 (PoJypropYIene)
Short term strength: 20661b1ft
c. ~ < < <\ 0 ,/.
Noles
a) EloceIpted from the _ _ · P.. 1tH ,,1& o:e 01.12.6 m higI1 geoIextiIewal in _ , Washington", (AIIen,1992).
b) Short tenn strength detennined from the wide width tenSIle strength test (ASTM D-45B5).
e) Long term strength is detennined by muRiplying the short tenn strength by the e.eep reduction coefficient.
d) The design strength is detennined by dividing the long tenn strength by the _ 01 safely.
0) The factor 01 safety accounts lor iniemal stability (Allen, 1992).
I) EstimaIad!nil weight and friction angle used fer design. Actual unit weight was 1341b1ft' and the friction angle varied from If$' 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,
----- -"'-
Project
Interstale Highway 70 Ihrough
Wan
Name
Geotextilo Earth
Height
1ft) n •
Ideg)
T
IbIn'
q
IbIII' Ka CI
I
).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
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:
fI
!
~ ___ F~1l.
('.5'
r---\\~"'E---r~
87
Phil Crouse Conservatism Index 4119/96
calculation Brief
Tult3
C13- CI3 = 0.13
FH3avg
Weighted CI value
Ka= tan ( 45·deg -1f Number of layers and reinforcement each group:
n1 := 5 n2 := 5
Ka =0.28
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
Weighted CI value
CI = 8.72 CI = 8072 for the Tanque Verde - Wrightstown - Pantano Roads Project
Wall Panel 26-30
.
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 =B.ll CI = 8.1 for the Tanque Verde - Wrightstown - Pantano Roads Project
Wall Panel 26-32
cosH _ B)2
~---------~~~~-------
Ka =0.47
CDS(B)2,cOS(B) '[ 1 + ( sinW·sin<+l ).5]2
cos(B) ·cos(B)
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)
FH1avg := FH1
n1
FH1a = 4.36°103 0lbott-1 Average Lateral force
vg for group 1
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 = 0.38 CI = A for the Norwegian Geotechnical Institute Project Wall Section 'N'
, Average lateral forces for each group wj!h same reinforcement strength:
FHavg := FH1
n
-1
FHavg = 545.6·lb·ft
CIValue
92
Phil Crouse Conservatism Index 4119196
Calculation Brief
'-"\~ ....\. i='on...c. y
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
93
Phil Crouse Conservatism Index 4119/96
Calculation Brief
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
]l·ft 9·ft
CI := -·CI1 + - ·CI2
20·ft 20·ft
CI =7.88 CI =7.9 for the Federal Highway Administration Project
94
Phil Crouse Conservatism Index 4119/96
Calculation Brief
Average laterallOrces Tor eacn group WI!!l same relnTorcemem strenam:
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
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
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
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)
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
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)
--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
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)
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.
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.
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.
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
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.
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.
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.
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. (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