Professional Documents
Culture Documents
I 1111i
II lII
111111
II ll -
Contract REPORT
Naval Facilities Engineering Service Center, Port Hueneme, CA 93043-4328
CR 94.003-SHR
June 1994
TRIAXIAL AND TORSIONAL SHEAR
TEST RESULTS FOR SAND
An Investigation Conducted by:
Bruce L. Kutter
Yie-ruey Chen
C. K. Shen
| Department of Civil and Environmental Engineering .
University of California, Davis
Sponsored by
3 1994. ".
20JL
Office of Naval Research .
Arlington, VA
Abstrac( This report presents the results of the labora- also presented. The results from different types are com-
tory tests conducted in triaxial and torsional apparatus. pared. Finally, the conclusions of these laboratory tests
The purposes of this report are not only to support the are also discussed.
calibration and verification of the bounding surface The test results indicate that the shape of phase trans-
hypoplasticity model for granular soil but to provide a formation and failure surfaces were different when viewed
valuable data base for further research in numerical model in the i-plane. It was also found that sampies subject to
simulation and design. rotational shear may be less likely to develop larger strain
Under this contract, two experiments were carried out: during undrained cycling than samples in triaxia] corn-
(1) laboratory samples in the triaxial apparatus and hol- pression/rxtension cyclic tests at similar stress ratios.
low cylinder torsional apparatus, and (2) centrifuge model In the cyclic torsional simple shear test, the maximum
tests including two and three-dimensional structures sub- stress ratio is between values obtained from triaxial con-
jected to static and dynamic loadings. pression and extension tests. In rotational shear tests, the
This report presents the results of triaxial tests includ- stable cycling of effective stress ratios are bounded by the
ing drained and undrained, monotonic and cyclic, and values of triaxial compression and extension failure stress
stress and strain controlled tests. The maximum stress ra- ratios. The maximum stress ratio observed in the rota-
tio achieved in drained triaxial tests was significantly tional shear tests for both compression and extension agree
larger than that in undrained triaxial tests. Results of six reasonably well with the stress ratio in the undrained
hollow cylinder torsional and rotational shear tests are triaxial tests.
Approved for public release; distribution is unlimited.
94 7 25 195
r~l
a- ;Y.-g
116
aI P3-
uZ
4c I
aE
L1 99 ic
M1 41 RSIu -inMm
E M -"'
Ii 0
_ *-%
Si
UI a m -RqR.
8 79 c 4-2 C
E It
ft ft U
cS
t, -c
0 j
REPORT DOCUMENTATION PAGE i I 0WIiLAW,,.
0701-.8
Pulhc pap ngr andf is olection o inform"a dt average 1
s mtimae" I hourperraepo , inuding thetim for revwi"g intructiom, searhing
exftm dmsourc. gthe wd
maimngthe dm needed, and c tingand wing th Cmofinforntn. Sd m omm ming tn
burd estimate or any other epsectof tha lction formlon, inck lngaugtnionforr &jcVisbudn toWa WnHa sevie,
Directoate fornf and Reports, 1215 JefferonDavisHighway, Suite 1204, Arnton, VA 22202-4302,andtotheOfficof MaonLagamat ndB t
Paparwo Reducon Proec (0704-0188),Wa on. DC 20503.
1. AGENCY USE ONLY ILavo biank) 2. REPORT DATE 3. REPORT TYPE AND DATES COVERED
June 1994 Final; October 1990 - June 1994
4. TITLE AND SUBTITLE S. FUNDING NUMBERS
ENo.
LIST OF FIGURES 3
LIST OF TABLES 5
ABSTRACT 6
1. INTRODUCTION 7
2. TRIAXIAL TESTS 23
5. CONCLUSIONS 50
6. ACKNOWLEDGEMENTS 51
7. REFERENCES 52
LIST OF IGUR
Figure 2.2.1
, : Triaxial, Undrained, Strain Controlled, Constant p Test
Figure 2.2.26
Figure 2.3.a : Photograph of the Colored Sand After Triaxial Test N7ODI00A
Figure 2.3.b : Sketch of the Vertical Offset of the Sample After Tfiaxial Test
Figure 2.3.c M u e of the Vertical Offset from Photograph of Sample (N7OD100A)
Figure 2.3.d : Comparisons of Data Measured from Load Cells Mounted on
Both Inside and Outside of Triaxial Chamber
Figure 2.3.1
$ : Traxial, Drained, Strain Controlled, Constant p Test
Figure 2.3.10
Figure 2.4.1
,L : Triaxial, Undrained, Stress Controlled, Cyclic Test
Figure 2.4.35
Fgure 4.1 : Summary Data for Cyclic Stress Ratio Veruw Number of Cycles
to Cause 3% Strain in Traiad, Torsional and Rotational Shear Tests
Figure 4.2 : Relationship Between Cyclic Strain and Number of Cycles in Hollow
Cyfinder Cyclic Torsion Shear Teas
Fore 4.3 : Relationship Between Stress Ratio and Angle in x Plane for
Hollow Cylinder Cyclic Torsion Shear Tests
Figure 4.4. : Summary Plot for Peak Stress Ratio Versus Angle in x Plane in Triaxial
and Cyclic Simple Shear Teats with Stress Paths of the Last Cycle of
Torsional and Rotational Shear Tests
Fiure 4.5 : Summary Plot for Stress Ratio Versus Mean Normal Effective Stress
in Triaxial Compression and Extension Tests
4
IjrOFTABLESq
5
1. ]ONTRODU ['IO
behavior, the effects of the magnitude of the stress ratio are more important than those of the
rotation of principal mess directions. Therefore, triaxial tests are extremely usefiu.
In this study, the results of laboratory testing of granular soil for the calibration and
verification of the bounding surface hypoplasticity model for granular soils (Wang and Dafalias
1990 and Li 1992) are presented. These tests included general tests (e.g particle size analysis,
maximum and minimum dry density and permeability test), triaxial tests and hollow cylinder
torsional tests. All of the laboratory test results reported were performed at The University of
California in Davis (UCD). Several types of triaxial tests including both drained and undrained,
monotonic and cyclic tests were performed. In the hollow cylinder test samples, both cyclic
torsional and rotational shear tests which involved rotation of principal stress directions were
performed.
This report is compiled in two parts. The first part is written in five sections including the
introduction (Section 1), triaxial tests (Section 2), hollow cylinder torsional tests (Section 3),
7
sunmary comparsons (Section 4) and conclusions (Section 5). Afterward, sumnary tables of
test Conditions, plots of A laboratory test results and wmmary plots of comparisons ae presented
in sequence in the appendix of this report. Figures and tales presented in the first pat are
directly related to the contents of the text.
The bacgound of this research, the material used in the lbontory tests and the
laboratory programs are introduced in Section 1.1, Section 1.2 and Section 1.3, respectively.
Finally, the units and the stress and strain definitions used in this report are described in Section
1.4.
8
Table 1.1: ist of Some Earlier Studies in Triaxial Tests
9
Table 1.2: List of Some Earlier Studies in Hollow Cylinder Apparatus
10
1.1. Background and Scope
(Wang and Dafalias 1990). The objective of the present research focused on developing an
experimental data base as follows:
Conduct a variety of triaxial and hollow cylinder torsional tests. These involved
monotonic and cyclic drained and undrained tests. In the hollow cylinder torsional tests, the
effects of rotation of principal stress direction were also to be investigated.
Parallel studies existed as follows:
(1) Determine the parameters of the hypoplasticity model using the data obtained in
triaxial and hollow cylinder torsional tests report by Chen and Kutter (1993).
(2) Conduct variety of centrifuge model tests involving static and dynamic loading of
saturated and dry sand to obtain data from boundary value problem for comparison with dynamic
finite element analysis. The results of these model studies are described in separte reports
Nevada sand was used to perform the laboratory tests. This sand was the same material
used throughout the VELACS (Verification of Liquefaction Analysis by Centrifuge Studies)
project (Arulanandan and Scott, 1993).
Saturated fine grained Nevada sand was used in all of the reported triaxial and torsional
tests. Attempts were made to also study the behavior of a coarse sand, but difficulty arose due to
effects of membrane penetration, therefore, the results of the tests in coarse sand are not
presented.
11
1.3. Testing Program
The laboratory tests in this report are divided into tiree groups: general tests, triaxil tests
and hollow cylinder torsional tests.
1. General tests:
The general tests included maximum and minimum dry density tests, permeability tests and
particle-size analysis. The maximum and minimum dry densities represent the densest and loosest
packing of particles without crushing the grains. The standard test methods ASTM D4253-83
and D4254-83 were used to determine the maximum and minimum dry densities, respectively.
Void ratio is the ratio between the volume of voids and the volume of solid particles in a
mass of soil. The amount of void space within a soil has an important effect on its characteristics.
The specific gravity is the ratio between the mass of dry solids and the mass of distilled water
displaced by the dry soil particles. The summary of specific gravity, dry density and void ratio for
Nevada sand is shown in Table 1.2.1.
The permeability of a soil is a measure of its capacity to allow the flow of a fluid through
it. The fluid concerned in this report is water. Permeability depends on a number of factors (i.e.
particle size, shape, void ratio, degree of saturation, type of flow and temperature). It is,
however, an important parameter for analysis of flow of water in boundary value problems. The
permeability of Nevada sand, measured in the vertical direction, is summarized in Table 1.2.2 for
various relative densities
Particle size analysis is used for classifcation of gramduar soil's particles into a separate
range. Summaries of sieve analysis and grain size distribution for Nevada sand are shown in
Table 1.2.3 and Figure 1.2.1, respectively.
12
Table 1.2.1 : Summary of Specific Graty, DMy Deiity and Void Ratio for Nevada Sand
GS yff, N/m)y(k/3 I . e
2.67 17.33 13.87 0.887 0.551
100
,60-
0
1000 100 10 1 0.1 0.01 0.001
GRAIN SiZE (mm)
13
2. Triaxial tests:
Sevea types of tests were performed in a computer controlled triaxial apparatus They
included both undrained and drained, monotonic and cyclic tests. The triaxial test is a standard
test in geotechnical engineering research. It consists of loading a cylindrical sample of soil with
variable lateral pressure and an axial load. A schematic of the triaxial apparatus is shown in
Figure 1.3.1. Table 1.3.1 shows the nom used for designation of the tiaxial tests, and
Table 1.3.2 indicates the nomenclature used for computer data files. Diskette copies of the test
data will be made available to interested readers. They can be obtained by contacting the first
author of this report. Schematic stress paths for triaxial tests are illustrated in Figure 1.3.2 to
assist the reader in understanding of the nomenclature.
The procedures of triaxial tests can be related to numerous types of practical problems.
The triaxial apparatus can control the magnitude (not the orientation) of the principal stresses, the
drainage and the measurement of pore water pressure. Test results derived from triaxial tests can
provide valuable information for the understanding of soil behavior as well as soil properties for
use in numerical model simulation and practical design.
during earthquake shaking. The triaxial test (explained in previous section) is widely used to
characterize soil in the field and to provide the data base for soil constitutive laws. However, it is
known that triaxial testing can not accotnt for multi-directional shear which has been recognized
14
to represent the real field conditions. Rotational shear, a special case of non-proportional loading,
can only take place under multi-directional loading conditions. Rotational shear is defined in this
report as a stress path for which the second invariant of the deviatoric stress tensor, J, is constant
while the directions of the principal strus rotate. Figure 1.3.4 shows different stress paths used
for torsional tests. The hollow cylinder torsional apparatus has the capability of controlling the
three normal stresses (o, a, and as) independently, and in addition, it can apply a shear stress, q.,
which the triaxial apparatus can not apply. For tests presented in this report, o and o were held
15
0 l0
EIC
1816
es-s
40 C
to
TEST PATH
CIUCE 0 0, C, CC. C,
CIUE/C 0 0,Ea EE.
dIud 0 01 d c C
CIUE 00, EE.
diDd/E (p'=donst.) 0 0, C, Es 01
CIDEIC (p'=donst.) 0 0, E, C3 01
dlDd (p'=CoflsL) 0 01 C, C,
CIDE (p'=Const.) 001 , EE
ClUdylIC 0 0, CE, E~, , E, C6...0
17
Table 1.3.2: Nomenclature Used for Designation of Data Files
NOMNCIATURE DESCRITIION
N6OUxy Normally consolidated, undrained, strain controlled, constant p test.
60: intended relative denity
x: effective confining pressure
y: test mimber
o6OUxy Over consolidated, undrained, strain controlled, constant p test.
60 : intended relative density
x : effective confining pressure
y:-test number
N7ODxy Normally onsolidated, drained, strain contolled, constant p test.
70: intended relative density
x: effective confining pressure
y: test number
CYxNy Normally consolidated, undrained, stress controlled, cyclic test.
x: effective confining pressure
y : test number
CYxOy Over consolidated, undrained, stress controlled, cyclic test.
x: effetive confining pressure
y: test number
NRxCUy Hollow cylinder anisotropically conolidated, undrained, stress
controlled, rotational shear test.
x: (JM)
y_: test number
NKxCUy Hollow cylinder anisotropically consolidated, undrained, stress
controlled, cyclic torsional shear test.
x" (aeA1 )
y_: test number
Table 1.3.3 Nomenclature Used for Designation of Hollow Cylinder Torsional Tests
18
31:
Fj -I-
00
TI;
PC cr4e
12 __
Copison
K~1K=
Extension 3
(a)-
3e
j~d Con.
20
1A&Description o Unit%, Stesses and SWales
a# :Kronecker deht
p :Mean normal Stress
IP, : mea normal deetive sumes
J= 4V~d isotropic invariant of deviatoric stress
R =J/pP : stresratioinvariant
co434Y2
0=Y 00 (JV2)] :Angle in x plane (Lode angle + 30'0)
In the above definition, the summation convention for repeated indices is assumed.
For triaxiatests: (cr 2 cr.)
CF, total vertical stres
CT,' : effective, vertical stress
a2 = 3 : total latieral stresses
C02 3@ effctive lateral strese
U : pore water pressure
C : confining pressure
p'= (aj'+2V3')/3 :mean normal effective stress
21
q = cr,- q3 :deyjatoric su
C, : azialstrain
9 8 :lateral &trin
e = el+2e 3 :.voluietric strain
eq =2(P.,- )/3 :deviatoric strai
For all cyclic torsional and rotational shear tolw in this report, the internal and eternal
cavity pressures were equal, thus, q$-% and various sm and strain quantites are deied
below:
torsional shear stress
P
2q./3
= cr, :mean normal stress
torsional shear strain
e1e.)/
=(e, triaxiai deviatoric strain
K : coefficient of lateral press=r
For fictional materials, the ratio of shear to normal stesis known to be most important
parameter in characteizin- the pro~dmity of the sures state to a filure, condition. R = J/jf is a
parameter which characterizes the ratio of shear to normal stress in principal str ess space. It is
known that the value ofR at filue, Rf, depends on the angle in the x plane, 0. The value of R at
which phase transormaton occurs is also a fimction of 0. This report will use f p, c and e,as
subscripts to indicate Mmiur, phase tas Oin, compession (0 =0) and extension (0= 6)
resPectivel.
22
2. T39AXI" TZSTS
This section provides all triaxial tet information in this projec. Sample preparado is
described in Section 2.1 and tet results a-op from different types of triaxial tests are
presented in Section 2.2 to 2.5. These tests include undrained and drained strain controlled
constant p tests (Section 2.2 and 2.3), undrained stress controlled cyclic tests (Section 2.4) and
isotropic consolidation tests (Section 2.5). In addition, the typical stress paths for different types
of triaxial tests are shown in Figure 1.3.2.
The samples for this study were prepared by pluviation through a 20 inch high Plexiglas
cylinder. The sand was fed by hand using a fimnel through a #16 sieve at the top of the cylinder.
For most of the tests, the relative density was approximately 70%. In a few tests, the sand was
tested at much lower densities. The diameter and height of sample were about 71 m and 150
min, respectively. The top and bottom sides of sand sample were covered by the Plexiglas
pedestals. In order to allow drainage of sample and measurement of pore water pressure, these
Plexiglas pedestals were inlaid with porous stones (diameter = 0.5 inches).
23
2.2. Triadl Undramed Sain Coursd Costat p Tab
(CIUC/Z. CJUEJc, CIUC or C7U)
The rmlts of 26 undrained strain controlled triaxial tes are rqot Thes results (q
vs. p, q vs. e and p! vs. e) are shown in Figure 2.2.1 to Figure 2.2.26. These tests were
performed with various combinations of Overconsolidation Ratio (OCR), confining pressure,
comprne /ension or badc pressure. To check the saturation of the soil specimens, the pore
pressure coeffident (B value) was measured. The B value is the ratio of the increase in pore
pressure to increase in total stress p during isotropic compression while the sample is undrained.
The increase of pore pressure was recorded while the cell pressure was increased by 50 kPa for all
B value mea rements The summary table for these tests is shown on Table 2.2.1.
1. The stress ratios (R = 47;1p) at failure (peak stress ratio) and phase transformation
were obtained from the average of several test results (Figure 2.2.a and 2.2.b) as listed in
the following:
For undrained compression tests:
24
TRIAXIAL UNDRAINED TESTS
am
a-
0-
Y"
_________Mf, = L477hGi.,R.t
01.0-
Mp 0O.963=43li-R,6
* 0.5-
0 MpM=0. 809=41 ..
0 200 0060
Mean Normal Effective Stres (kPa)
Figure 2.2.b Summary Plot for Deviatoric Stress Ratio Versus Mean Normal Effective
Stress in Triaxial Undrained, Tests Under Different Confining Pressure
25
2. The ratio R/Rg=c,-.84 is different fom the ratio at failure R,%/R=Q.61. This is
significant because the hypoplasticity model assumes that cwc-c.
3. Reasonable repeatability of results for tests which had similar testing conditions was
obtained. In this report, for OCR=I, the CIUC/E tests at confining pressure = 100 kPa
were repeated three times (Figure 2.2.2, 2.2.3 and 2.2.4), and repeated two times at both
confining pressures = 250 kPa (Figure 2.2.5 and 2.2.6) and 400 kPa. (Figure 2.2.7 and
2.2.8) In addition, the ClUE tests (OCR=I) were repeated three times at confining
pressure = 100 kPa (Figure 2.2.11, 2.2.12 and 2.2.13).
4. In almost every undrained test, the deviatoric stress increased until cavitation occurred.
After cavitation, the deviatoric stress stabilized, but the stress ratio was observed to
increase. In fact, after cavitation occurs the test essentiady becomes a drained test.
2. In triaxial undrained test, after cavitation the deviatoric stress stabilized but the stress
ratio was observed to increased.
26
2A Trial Draied Strain Controlled Constant p' Tests
(CEDe/, CIDEI CIDC or CIDE)
The results of 10 drained, strain controlled, constant p! triaxial tests are reported. These
results (e vs. e, q vs. e, and q vs. ) are shown in Figure 2.3.1 to Figure 2.3.10. To study shear
band phenomena, 3 samples (N7ODI00A, B and C) were prepared with thin horizontal layers of
dark sand. The thickness of dark sand layer was about 1 mm and the distance between two dark
sand layers was about 25 mm. In addition, to determine the effects of friction in the apparatus on
measurements made by the external load cell, 6 tests (N70DI00A, B, C, P Q and R) were
performed with a second load cell inside the tiaxial chamber. The summary table for testing
conditions of each test is shown on Table 2.3.1.
1. The maximum "failure" stress ratio it observed in drained tests was significantly larger
than that in undrained tests. The averge ste ratios (R = .. ft:pP) on hilure fine were
1.04 and 0.63 in drained compression and extension tests respectively. The ratio (Le.
shape factor in deviatoric plane) at failure cf (c?R,/fR0.6) in drained tests is dosed to
that in the undrained tests. The hypoplasticity model assumes that the shape parameter, c,
is a unique value for both undrained and drained test conditions.
2. Most of the drained test results showed a peak followed by a drop of deviatoric stress.
The drop began when the axial strain reached 6 to 8 % depending on the confining
pressure. The drop was sometimes sudden and sometimes gradual, occurring while the
axial strain increased by an additional 0.2 (Figure 2.3.8) to 3% (Figure 2.3.6). The drop in
stress corresponded to a drop in the dilatancy rate and the formation of a shear band.
27
3. The samples with colored sand we dissected to obaerve any shar bands From tts
interrupted at 3.6% strain, prior to the drop in stress no shear band was apparent. From
tests stopped at 8.5% strain after the drop was complete, it appears that the shear band
was completely developed. A photograph of the colored sand is shown in Figure 2.3.a for
test N70D IOOA.
4. An example sketch which indicates the offset of the sample after triaxial test
(N70D100A) is presented in Figure 2.3.b. From results of three tests N7ODI00A, B and
C, the thickness of the shear band appears to be about 10 to 12 times D5o (mean grain
size). The filure planes were inclined at the angles between 630 and 650. Figure 2.3.c
shows the example measurements of the vertical offset from the photograph of sample
N7ODI00A. The offset at the ends of the sample were measured to be 3.7 mm at the top
and 9.15 mm at the bottom of the sample. From Figure 2.3.4, the axial strain at the end of
test, e*, was 8.4%. Once the shear band is formed, it may be assumed that all of the
deformation occurs on the shear band. Therefore, the axial strain corresponding to the
formation of the shear band, et, can be calculated by
A
e, =e -_T
where, A is the offset across the shear band and to is the initial length of sample. Thus,
the shear band formed at the bottom of the sample at ej = 0.084 - 9.15 I 147.6 =2.2%,
and the shear band formed at the top of the sample when e, = 0.084 - 3.7 I 147.6 = 5.9%.
5. Based on the test results (Figure 2.3.d) measured from load cells mounted on both
inside and outside of taxial chamber, a small offset of the initial reading (about -5 kPa in
compression and +5 kPa in extension) was observed. The maximum offset of test results
was less than 2 kPa in compression test (N7ODI00A, C and R) and reached about 8 kPa in
extension test (N70OD00R). This may be the result of friction induced between the
28
loading rod and the bearing ofthe top cap (se figur 1.3.1). the results forco r
Nggest that fiction is relativy small. For xtension, the friction appears to be mom
important However, thes rsults are only p&minaly invPstaions and more data is
needed to verify this.
1. The max m fa"ilure stress ratio Rf in drained tests was found to be significantly
larger than that in undrained tests.
2. Most of the triaxial drained test results showed a peak followed by a drop of deviatoric
stress. The drop in stress corresponded to a drop in the dilatancy rate and the formation
of a shear band.
3. The thickness of shear band was observed to be about 10 to 12 times mean grain size of
the Nevada sand. The failure planes were incined at the angles between 630 and 650.
29
Zx
miss
NIMF
Ah
Ar
. . Al
xv;,
rwW
7.- NZ .... . k
V;b"
ega
3W
xx
IN, Q
X, n,
It
M
...
Figure 2.3. a Photograph of the Colored Sand After Triaxial Test N70D IOOA
30
Note @W awp c*.w
vm not nheuwvd hits
Figure 2.3.b : Sketch of the Vertical Offset of the Sample After Triaxd Test
9.15 rw
0 2b 4, 80 ab 100 io 14o 10
31
TRIAXAL, DRAINED, STRAIN CONTROLLED, CONSTANT P' TESTS
co4
.040
040
.40
-1 9
Axial Strain (%)
40
10
IO
40
u0 0
0 4
*20
40
Cc40.-Isd
odCl
-11 1 3 7 - 9
Axial Strain (%)
1202
2.4. Trdazial Usidraied Sbres Catrde Cyde Tabt (ClUCydic)
The results of 35 cycki undrained. stres controlled tuiaxial tests, anre ported in Yamur
2.4.1 to Figure 2.435 (q vs. p, q vs. e, and V' vs. et) These tost wme performed with vuyin
cyclic stress rati over-consolidation ratio, confining pressure and testing frequecy For modt of
the tests the testin period were 300 se The sumnary table for these test as Ta" 2.4. 1.
1. The number of cycles to cause 3%/strain for 10t~n stress ratio was obseved and is
discuissed later in connection with Fgur 4.1. The liquefiction of dowe sand. is normally
considered to develop when the pore wate pressure ratio build up to a value of 100%A
and the stains of the order of au 5%(Seed and Idriss, 1982). In this report, hwexver,
3% strain is adopted due to the useffol comparisons to hollow cylinder torsional shear tests
for which the stains only reach aud 3%.
3. Some test results show mean normal effibtive stress, sligtly less than zero. This is the
result of an incorrect zero offset in the pressue transducers.
33
L& Thazld CuMdatls Tfts
Results of 8 isotroic ooulidatkxa tests are uported in Flare 2.5.1 to Veip 2.5.8.
Table 2.5.1 amnmmized the ten dart
1. The oonpresson index I and rebound inde x for critical state soil mechanics
(Schofield and Wroth 1968) were obtained and are prmted in Table 2.5.2. Hee,
A,- Ael/A(p')
/ for virgin o and i= Ae/An(p) for rebound. I and , were
not fund to be onstant. Constant andx values may apply to alimnited range of
coaning presm In any constitutive modeling efforts, thme paramnters ms be
eXperin tay determine over the pressur range of ntrest
and rebound index ie we constants. Here X = de/d(p'/p.)K' for virgin copeso And
e= deld(pllpo) for rebound, in which p. and p' are atmospheric pressure and m
normal effective premm respectively. It is interesting to point out that the values of r
and le (Table 2.5.2) appear to be somewhat better mconstants than ;L and X For example
I varies from 0.0031 to 0.0091, whereas r only varies from 0.0037 to 0.0071.
34
2. Hyppatct modelcmpeo indu v VX =de/dtJ9/p.)X) and rabound Wndx e
35
&lOL JW CYU~M ToRaSLONATIM
Two different types of test (cycPic torsonal showr and rotational sha tests) were
perfomed in the hollow cylinder apparatus at University of Cfo&na, Davis.
The cyclic torsional shear tet was perforned by nimlly applyn the ai loading a on
an isotropically consolidated sample until the desired coefficient of lateral presiare, K (K-
was reached. Shear str s, was than cycically applied.
The stress path of a rotational shear test was performed by ial applying the axial
loading a on an isotropically consolidated mple until the desired value of J
(J = j.2/+ = I( 2 was reached. The sample was then sheared with a constant
value of J. The mean norma stress p was held constant while oz. and (oF-ag) were varied in
The samples used in the hollow cylinder torsional tests were prepared by pluviation
through a 20 inch high Plexiglas cylinder. The sand was fed by hand using a fimnel through a #16
sieve at the top of the cylinder. The inside and outside diameters of samples were about 4 inches
and 6 inches, respectively. The height of samples was about 6 inches. The range of relative
density was 68h to 74%. The sample was covered with top and bottom pedestals and vacwzmed
along the inner and outer vertical surfaces by two separate membranes (thidmess = 0.012 inches).
The inner membrane was fixed by two Plexiglas rng wedges setting on the top and bottom side of
36
pedstal individully Th outer nmnbrne was fixead by o-ing To avoid slippage aWUsk
shear stress from pedestals to the sample, an epoxy coated sand was provided to the pedestals. In
addition 12 lades (6 bades on each pedestal) were placed p ediar to the epoxy coated
sand sunhes, To allow drainage of sample and measurement of pore water pressure, total 6
porous stones (diameter - 0.5 inches) were inlaid on both top and bottom pedestals.
Most of test results showed a slight overshoot in deviatoric stress, (or - a,)143-i, durin
the initial axial loading condition before application of torsional shear stress. This is the result of
friction induced between loading rod and bearing of top cap and backlash of controller. In
preliminary tests, significant binding was observed in the piston, resulting in excessive errors in
the measureme of stresses applied to the sample. This problem was resolved by loosening
certain parts of the axial and torsional load measurement system, but introduced some backlash in
the torsional displ measuring system. The data was "corrected" to eliminate the observed
backlash. Figures 3.2.2, 3.2.17 and 3.3.2 illustrate how the data was corrected. An internal load
cell has been designed and implemented to eliminate these problems, but the results presented in
this report were all obtained before the internal load cell was implemented.
37
&2 Cydlc ToerioalSbear Tests (HCCAUTSCydic)
This report contans the results of four cyclic torsional shear tests performed in the UCD
hollow cylinder torsional apparatus (Figure 1.3.3). These results are shown in Figures 3.2.1 to
3.2.20. The tests were conducted with the following values for coefficient of lateral pressure: K-
o/ z - 0.41, 0.63, 1.0 and 1.38. During a given torsional simple shear test o 0 and a z were held
constant while the shear stress q. (56 kPa) was cycled, q,, = 56 kPa. The summary table for
these tests is shown on Table 3.2.1. The relationship between stress ratio (4J-;/p) and angle 0
in z plane for these four cyclic torsional shear tests will be presented later in Figure 4.2.
1. At the same stress ratio, the number of cycles to cause 3% strain for cyclic torsional
shear tests was larger than that in cyclic triaxial tests. Furthermore, the number of cycles
to cause 3% strain consistently increased as K decreased. K is the initial ratio of oeo r
The relationship between cyclic stress ratio (i.e. (q, - 7.)/(3-.q,') for triaxial test and
1 ,/a' for torsional shear test) and the number of cycles to cause 3% strain will be
2. The mean normal effective stress (shown in Figures 3.2.5 and 3.2.10) exhibited the
phenomenon of stable cycling in stress space between phase transformation and failure
after certain number of cycles in the conditions of K<I. During this cycling, the
magnitude of strains continued to increase. It is interesting to note that the changes in
mean normal effective stress during stable cycles were about 27% and 9% of the cyclic
shear stress, azo, for K-0.63 and K-0.41, respectively. For the cyclic simple shear test
(K=), the mean normal effective stress gradually approached zero at oze=0 (i.e. sample
38
liquefied) after a certain number of cycles (shown in Figure 3.2.15). For the case of K>l
(shown in Figure 3.2.20)1 stable cycling between phase transformation and failure was also
observed, but cyclic change ineffective mean normal stresses during these cycles was
almost equal to the magnitude of shear stress a. and the mean normal effective sess
was much larger than zero at ar,=0.
3. During the stable cycling of stresses the shear strains, e, cycled in the range of 0.25%
and 0.3% for K=0.41 and K=0.63, respectively (shown in Figures 3.2.4 and 3.2.9). The
shear strains e& in cyclic simple shear test (K=1, Le. the major principal stress 450
relative to the vertical in Figure 3.2.14) were 3% in both directions and gradually
increased as the number of stress cycles increased. It is interesting to point out that as K
approaches one, the initial stress approaches an isotropic state (i.e. the initial deviatoric
stress, q,-o, decreases) yet the cyclic shear strains, e., increase. The shear strains
4. For the cyclic simple shear test, the maximum (failure) stress ratio R. = J/p' is 0.737
and phase transformation is observed to occur at RP = J/p'= 0.4. The maximum stress
ratio is in between values obtained from undrained triaxial compression (1 =0.853) and
extension (R=0.524). The stress ratio at phase transformation appears to be less than
that in both triaxial compression (R=0.556) and extension (Rp=0.467) tests. A detailed
explanation of model parnamters for the hypoplasticity model will be discussed in the
report "Calibration and Testing of the Proposed Hypoplasticity Model for Sand" (Chen
5. The maximum stress ratios, R=17;/p', were about 0.71, and 0.56 for the cases of
39
Swua_of kmg observations-
1. At the same stress ratio, the number of cycles to cause 3% strain for cyclic tonional
shear tests was larger than that in cyclic traxial tests. The number of cycles to came 3%
strain consistet increased as the initial ratio of a./G decreased.
2. The effective stres path exhibited the phenomenon of stable cycling between the phase
transformation and failure stress ratios for the cases of K<l and K>I after a number of
cycles. For the cyclic simple shear test (K=I), the mean normal effective stress gradually
approached zero at .9=O (i.e. sample liquefied) after a number of cycles.
3. In the cyclic torsional shear tests, the cyclic shear strain amplitude (e,) seems to
increase as the ratio of qg/ar increases.
4. In the cyclic simple shear test, the maximum stress ratio, P4 is between values obtained
from triaxial compression and extension tests, and the stress ratio at phase tramfomation
was less than that in both triaxial compression and extension tests.
40
3.3. Rotational Sher Tests (R AURS)
The results of two rotational shear tests are reported as follows.. Test results are
presented in Figures 3.3.1 to 3.3.10. Each stress cycle of the rotational shear tests was
accomplished by continuous rotation of principal stress axes at constant values of J. The
direction of the major principal stress rotated from compe I (angle 0=00) to simple shear (0
=300), then continuously rotated to extension (0=600) and finally turned back to con
through another simple shear condition.
In these figures, the filled square, empty square and plus sign symbols represent the
conditions of compression, extension and simple shear, respectively. These symbols were found
to be useful for mapping and cross referencing from one plot to another. The two tests were
conducted using different values of J (Figures 3.3.1 and 3.3.6). The chosen values of J (56 kPa
and 40 kPa) corresponded to stress ratios that were 33% and 24% of the stress ratio at failure in
triaxial compression tests, respectively. The summary table for these tests is shown on Table
3.3.1.
1. The relationship between cyclic stress ratio, J-/p,, and the number of cycles to
cause 3% strain will be explained in later section and is shown in Figure 4.1. The
inclination of curve in the rotational tests appear to be higher than that in the triaxial tests.
More data is required to define the relationship.
2. For both rotational tests, the stress ratios, (a. - o,)/( r3--p-)
and J"Ip',stabilized
after reaching a large strain (shown in Figures 3.3.4 and 3.3.9). Although the
phenomenon of stabilization of stress ratio was observed, the mean normal effective stress,
p', continued to change cyclically. Figures 3.3.3 and 3.3.8, show the change of mean
41
normal effecti stress, p', with the shear strain, Z.*. Intially, the mean normal effectie
stress gradually decreased resulting in an increase in stress ratio until the "stabilization" of
stress ratio occurred. During the "stable" cycles, p' tended to increase as the direction of
stress moved from compression (0=00) to simple shear (0=300), to extenson (9=600) and
p' tended to decrease as direction of loading rotated from extension to simple shear, to
compression. The minimum value of p occurred at ,-00 (triaxial compression). The
maximum value of p' occurred at 0=600 (triaxial extension). The value of p' was nearly
constant as 0 reduced from 600 to 300.
3. The relationship between shear stress (, and shear strain, e, for both rotational shear
tests are shown in Figures 3.3.2 and 3.3.7. For 9 increasing from 00 to 600 and 0
decreasing from 600 to 00, the relationship between aF and eA appears to be symmetrical.
The maximum shear strain ( .) occurs after the peak value of a.q when Fze is about
half of the peak a,,. The deviatoric strains 2(ez-E)/3 were not symmetrical, extension
strains were three or four times greater than compression strains in both tests. The
deviatoric strain consistently returned to near zero when the deviatoric stress, was a
maximum (0=00). Similar to the relationship between qzq and .9, the maximum deviatoric
strains occurred after the maximum deviatoric stress when the magnitude of the deviatoric
stress was about half of its maximum value.
4. Although the shear stress and deviatoric stress ratios stabilized near their phase
transformation and failure surfaces, the magnitude of shear strain, eg, and deviatoric
strain, 2(;-/3, increased as the numbers of stress cycles increased. The stabilization of
stress ratio is clear in the repeatable loop in the plot of (a0 - qe)/(43p') vs.
(Figures 3.3.4 and 3.3.9). The increasing rate of strain is also clear in the spiral shape in
42
the plot of 2(%-eW vs. e.,(FIgures 3.3.4 and 3.3.9) In teat NR4 CUSO (J-40 kPa),
while the shear strain reached about 3%, a sudden increase of sher strain was observed.
5. The cycling of stress ratios, R= 4J-/p' and (C,-,,)/(..p') (Fires 3.3.5 and
3.3.10), from compression (e=00) to extension (0=600) through the simple shear state (0
-300) are shown for tests NR40CU5O (J=40 kpa) and NR56CUSO (J - 50 kpa),
respectively. These results revealed that the stress ratios stabilized after a certain number
of cycles. The maidmum stress ratios, R are about 0.85 in compression and 0.6 in
extension for both tests. It is interesting to point out that the ratio, R is
equal to 0.7 which is between the value c, and cf determined from undrained triaxial tests.
In the triaxial tests, the ratio cp was 0.84 (Cp=Rp/ on phase transformation line) and cr
was 0.61 (ci"Rg/Rh on failure line). The values of R (0.853), Rf, (0.524), Rw (0.556)
and R (0.467) determined from undrained triaxial tests as indicated on Figure 3.3.5 and
3.3.10 for reference, the stable cycling of stress ratio are bounded by the values of Rh and
Rh.
1. The stress ratios, (a, - a)/(-F3"p') and 47;1p'%traced a stable loop after reaching a
larger strain. Although the phenomenon of stabilization of stress ratio was observed, the
mean normal effective stress, p', continued to change cyclically. The maximum value of p'
occurred at 0=600 (triaxial extension).
2. Although the shear stress and deviatoric stress ratios stabilized near their phase
transformation and failure surfaces, the magnitude of shear strain and deviatoric strain
increased as the numbers of stress cycles increased.
43
3. in rotational she test, the relationship between o, and e was found to be
symmetrical but the deviatoric strains, 2(ci-e in exension were tiree or four tims
gremer than those in compesion.
4. In rotational shear testsu the stable cycling of sress ratios are bounded by the values of
44
4. SU MMARY'COMPARISONS OF "IALI TOIMM MOAL AND ROTATIONAl.
SEmeAR TES RESU LTS
Figure 4.1 shows the summary data for cyclic stress ratio versus number of cycles to cause
3% strain in triakial, torsional and rotational shear tes. It should be noted tha
TJ;;= 4-c-: c) 2 /3+ a.2 for rotational shiear tests and 47 -a~or/3 fobr triaxidaltests,
(i.e. a,,-O for triaxial tests). For torsional shear tests, or,, was cycled with constant values for
(q,- c.), which depended on the values for coefficient of lateral pressure K (K=q/). Thus,
these cyclic stress ratios were calculated by (q, - qo)/(4 .. ., /o' and J,/p, for
triaxial, torsional and rotational shear tests, respectively. These results indicate that the soil
samples which have similar relative density are more resistant to cyclic loading for the rotational
tests than for symmetric cyclic compression/extension tiaxial tests. Note that triaxial tests with
unsymmetrical loading with a larger magnitude of compression than extension never developed
large strains (for example, CY250NI, see Table 2.4.1). At the same stres ratio, the number of
cycles to cause 3%strain in the torsional shear tests was larger than that in the cyclic triaxial tests.
Based on these results, liquefaction potential appears to be overestimated by conventional
symmetrical triaxial cyclic test results. This appears to be inconsistent with observations of
Towhata and Ishihara (1985). Further investigation is required.
The consistent picture which emerges is that the occurrence of triaxial extension states of
stress is damaging. In the torsional shear tests, ca was cycled while ai-ae was held constant
corresponding to four different initial values of K=o/a1 . For K>I, the initial state is one of
triaxial extension, and for K<I the initial state is triaxial compression, and for K=1 the initial state
is isotropic. For K=I the cycling of cre corresponds to simple shear. A summary plot for the
relationship between cyclic strain, e, and number of cycles is shown in Figure 4.2. For K=1.38
(extension), 3% strain developed in 4 cycles, for K=I, 3% strain developed in 7 cycles but for
K=0.41, only 0.25% strain and for K-0.63, only 0.3% strain developed when the tests were
terminated after 15 cycles.
45
Fiure 4.3 shows the relationship betweenMsrm ratio, 4J_. /p', and the Mle 0 in z plane
for cyclic torsional shear tests performed with different values of coefficient of lateralpressure K.
For the cyclic simple shear test (K-1) (Gzu."-." and o. was cycled) the anle 0 was
approximately constant (0=300). For K=oa/o<1 (iLe. ajza-o* and oY. was cycled) the
angle 0
was started at 00 and cycled with the cycling of shear mess a The madmum m s rios for
both K-I and K<l were about 0.71. For the case of K> 1, the pattern is similar to that for K<I
but the angle 0 was started at 600, and the maximum stress ratio was observed to be 0.56.
Figure 4.4 shows a summary plot of stress ratio versus angle 0 in the z plane for undrained
and drained triadal, torsional shear and rotational shear tests. The data points plotted in this
figure represent the failure points in varying conditions of triaxial tests, the maxmun point in
torsional shear test with K-1 (cyclic simple shear). The lines in Figure 4.4 show the last cycle of
cyclic torsional and rotational shear tests. This plot night be thought of as a map of the failure
surface. Note that the dta points at drained and undraed triaxial tests represent average values
from more than one test. The angles of 08, 300 and 600 represent the triwdal comn (aY
r=-c, %9=0), simple shear (G.-.9 0 ) and triaxial extension (;<o,-oa,
0d, a..0) tests.
respectively. The path for rotational shear tests is a loop between 00 and 600. The torsional
shear tests for the cases of K=oda/o >1 begin from 0=00 and the angle 0 increases as Iao
increases. For K<I, the test path starts on 0=600 and 0 decreases as the magnitude of (YA
increases.
These results (Figure 4.4) show that the maximum stress raio,,/p, in triaxial drained
compression tests is significantly larger than that in undrained compression tests. This finding
also can be concluded from the data plotted in Figure 4.5. In Figure 4.5 the filled square and
empty circle symbols represent the maximum and ultimate stress ratios, q/p', in drained triaxial
tests respectively, and the solid lines represent the undrained trixial stress paths under difrent
confining pressures. Note that the error bar for the drained tests represents the scatter in stress
ratios in drained triaxial compression and extension tests. The maximum stress ratio, q/p', in
46
drained triaxial tests is larger than that in undraied tests in both compresion and eatmsion. The
difference in maximum stress ratio for triaxial undrained and drained tests appea to be
decreasing as the mean normal effective stress increases. The larger stres atios at faimr in
drained tests are also apparent in undrained teats after cavitation occurs. The upward hook at the
end of the undrained stress path corresponds to cavitation.
Based on the results plotted in Figure 4.4, the maximum stress ratios in the rotational
shear tests for both compression (&=0*) and extension (0.-60) are also found to agree reasonably
well with the stress ratios in the undrained triaxial tests. It was observed that the maximum sess
ratio in the torsional shear test (K= 1) was larger than that in rotational shear tests in simple shear
direction (0=300).
Several types of laboratory tests involving general tests, triaxial shear teats and hollow
cylinder torsional and rotational shear tests were conducted on samples of Nevada sand using
derent facilities and apparatus at the University of California, Davis. Regarding the results
obtained from these tests, the following enumerated observations were made.
2. The maximum "failure" stress ratio Rf observed in drained triaxial tests was significantly
larger than that in undrained trivdal tests. The difference appears to be decreasing as the
mean normal effective stress increases (Figure 4.4). The larger stress ratios at failure in
drained triaxial tests are also apparent in undrained tests after cavitation occurs (Figure
4.5).
3. By measuring the deformed shape of colored sand layers in some drained triaxial
samples, the thickness of shear bands was observed to be about 10 to 12 times mean grain
47
sze of the Nevaa sand (F'Qur 2.3.a and 2.3.b) or about 1.7 to 2.0 nmL The iwe
planes were obsevd to be inclined at anses betwem 63 and 65. A drop in deviatoic
stress (strain softening) was observed to occur a the hilure pla=e davoped (eg., Fip
2.3.8).
4. Over a limited range of confining pressure, values for the compresion index ;V and
rebound index ie in the hypoplasticity model appear to be more closely a constant than
values for ) and K for the critical state soil mechanics model (Table 2.5.2). However, the
hypoplasticity model assumes that r and ie are constants.
5. At the same stress ratio, the number of cycles to cause 3% strain for cyclic torsional
shear tests was larger than for cyclic triaxial tests (F%ure 4.1). Also, sod samples tested in
rotational shear appear to be more resistant to cyclic loading tham samples sulject to
symmetrical cyclic trinxial tem. Thu the lkqecion potential appears to be
overestimated by conventional symmetrical triaxial cyclic tes results. This conclusion
needs to be firther verified since it contradicts the well-know observatio by Towhata
and Isbihara (1985).
6. For the cyclic simple shear test (K=Oa/-1l), the mean normal effective stress gradually
approached zero at a,=O (Figure 3.2.15). The effective stress path exhbed the
phenomenon of stable cycling between the phase trnfrmto and failure stress ratios
for the cases of K<1 and K>I after a number of cycles (Figure 3.2.5, 3.2.10 and 3.2.20).
7. In the cyclic simple shear test, the maximum stress ratio is between values obtained
from triaxial compression and extension tests (Figure 4.4).
48
S. 1 the wrist of cyic torioal shea tesrs, the rae of chane of cycl ohr suri
ampEitude seeM to iwm as the ratio of af; increases (iur 4.2).
9. In rotational shew tets, the stress ras, (q - ,)/(,.r. p') and Pf/p, traced a
Stae loop af-er a mMber of c (Fe 3.3.4 and 3.3.9). Trhe sm re a red
to oscillate between the phase trans on nd failiresurfrac.
10. In rotational shear tests, lthm the shear sand deviatoric su ratios stabilized
near their phase trans-omation and failure surfces, the magitl of shear strain, e and
deviatoric strain, 2(%-e Y3, increased as the mumbers of stress cycles increased (igure
11. In rotational shear tests, 0 increasig from 00 to 600 and 0 decreasing from 600 to 00,
the relationship between q. and e., appears to be symmetrical but the deviatoric strains
2(;-e/3 were not symmetrical (Figure 3.3.2 and 3.3.7). The deviatoric strains in
12. In rotational shear tests, the stable cycling of stress ratios are bounded by the values of
triaxial compression failure stress ratio R. and extension failure stress ratio It (Figure
3.3.5 and 3.3. 10).
13. The madmum stress ratio observed in the rotational shear tests for both compression
and extension are found to agree reasonably well with the stress ratios in the undrained
49
e maor of this research has bee the generation of a wide divmsy of
laboratory tet data. Thee tgt data not only can be used to support the calibration and
veMiication of the bounding suface hypoplasticity model for granular soil but also provide a
valuable data base for futher research in constitutive model studies. Summay compaisa are
also made between triaxial, cylic torsional and rotational shear tets. The tet results indicate
that the rotations of principal stress directions have very important effects on the soil deformtion
and strength .
Regarding the results obtained from these tests, soil samples tested in rotational shear
were found to be more resistant to cyclic loading than samples in symmetrical uiaxial cyclic tests.
The maximmum "filue" stress ratio in drained triaxial tests was significantly larger than that in
undrained triaxial tests. Also, the shape of phase transformation and failure surfaces were
different when viewed in the z-plane.
In the cyclic imple shear test, the sres ratio on the failure line is between values obtaned
from tirxial compression and extension tests. In rotational shear tests, although the shear stress
and deviatoric stress ratios stabilized near their phase transformation and failure surfces, the
magnitude of shear and deviatoric strains still increased as the numbers of stress cycles increased.
Also, the stable cycling of stress ratios are found to be bounded by the values of triaxial
compression and extension failure stress ratios. In addition, The maximum stress ratio observed
in the rotational shear tests for both compression and extension are found to agree reasonably well
with the maximum stress ratio in the undrained triaxial tests.
50
yx
6. ACXX'OW rl'M'T
This research reported herin was conducted under Naval Civil Engineering Laboratoy
contract number N47408-89-C-1058. This support is gratefily acknowledge. Dr. Ted Shuar
provided valuable advice and review comments at several stages of this research.
In this research, forty-five uiaxial tests were conducted by James Wickitrom, fifteen
triaxial tests were conducted by Shiyo Chen and six torsional tests were accomplished by Chi-
Wen Lin.
The technical assistance and computer programs from Dr. Xang-Song Li in all aspects of
Many thanks to Mr. Bill Siuis and Mr. Tom Kohnke, the development tecmicians, for
their valuable suggestions and excellent contrbutions to improve the test apparatus.
Thanks are also due to Dr.Han-Wei Yang and Dr. S. Jafroudi for their previous
developments of the hollow cylinder torsional apparatus at U. C. Davis.
51
Alarcon, A., Chameau, J. L. and Leonards, G. A., "A New Apparatus for Investigting the Strs-
Strain c of Sands, Geoec a Tst Jounal, Vol. 9, No. 4,1986,
PP204-212
Arulamnandan, K. and Scott, IL F., "Project VELACS - Control Tes Result", Journal of
Geotcical Engineering ASCE, Vol. 119, No. 8, August 1993, PP1276-1292.
ranandan K. and Scott, R.F., Werification of Numerical Procedures for the Analysis of Soil
Liquefaction Problems', Proceedings of the International Conference, Davis, Califora
Vol. 1, October, 1993.
Castro, G., "Liquefaction and Cyclic Mobility of Saturated Sands", Journal of Geotedhcal
Engineering Division, ASCE 101(GT6), 1975, PP551-569
Chen, Y.-R. and Kutter, B. L., 'Calibration and Testing of the Proposed Hypoplasticity Model for
Sand', report to Naval Civil Engineering Laboratory, Department of Civil and
Environmental Engineering. University of California, Davis, 1994.
Chu, J., Lo, S.-C. K and Lee, L K., 'Strain-Softening Behavior of Granular Soil in Strain-Path
Testing", Journal of Geotechnical Engineering, ASCE, Vol. 118, No. 2, 1992, PP191-208
Dasseault, M. B., "A Versatile Hollow Cylinder Triaxial Device', Can. Geoteck J. 18, 1981,
PP1-7
Desai, C. S. and Siiwardae,. I ., "Constitutive Laws for Engineering Material With Emphasis
on Geologic Materials", Prentice Hall, 1984.
Gilbert, P. A- and Marcuson, W. F., "Density variation in Specimen Subjected to Cyclic and
Monotonic Load', Journal of Geotechnical E ASCE, VoL 114, No. 1., 1988,
PP1-20.
Gutierrez, M, Ishihara, K. and Towhata, 1."Flow Theory for Sand During Rotation of Principal
Stress Direction", Soils and Foundations, Vol. 31, No. 4, Dec. 1991, PP121-132.
52
Gutierrez, Ishihara, K. and Towhata, "Model for the Deformation of Sand During Rotation of
Principal Stress Directions, Soils and Fouidations, Vol. 33, No. 3, Sept. 1993, PPI05-
117.
Hight, D. W., Gens, A. and Symes, M. J., "The Development of a New Hollow Cylinder
Apparatus for Instigatmg the Effects of Principal Stress Rotation in Soils",
Geotechnique, 33, No. 4, 1983, PP355-383.
Head, K., "Manual of Soil Laboratory Test', Volume 1, 2 and 3, ELE Internationa Limited,
1985.
Hettler, A. W., and Vardoulakis, I., "Behavior of Dry Sand Tested in a Larger Triaxial
Apparatus', Geotechnique, Vol. 34, No. 2, 1984, PP183-198.
Ishihara, K, Tatsuoka, F. sad Yasuda, S., "Undrained Deformation and Liquefaction of Sand
Under Cyclic Stresses', Soils and Foundations, Vol. 15, No. 1, March, 1975, PP29-44.
Lain, W.-KI and Tatsuoka, F., "Effects of Initial Anisotropic Fabric and q2 on Strength and
Deformation Characteristics of Sand", Soils and Foundations, Vol. 28, No. 1, 1988,
PP89-106.
Li, X. S., "Free Field Soil Response Under Multi-directional Earthquake Loading", Ph.D.
dissertation, University of California, Davis, 1990.
Li, X S., Chan, C. K. and Shen, C. K, "An Automatic Triaxial Testing System", Advanced
Triaxial Testing of Soil and Rock, ASTM, STP77, 1988, PP95-106.
Miura, K, Miura, S. and Told, S., "Deformation Behavior of Anisotropic Dense Sand Under
Principal Stress Axes Rotation", Soils and Foundations, Vol. 26, No. 1, Mar. 1986,
PP36-52.
Pradel, D., Ishihara, K and Gutierrez, M., "Yielding and Flow of Sand Under Principal Stress
Axes Rotation', Soils and Foundations, Vol. 30, No. 1, Mar. 1990, PP87-99.
Riemer, M. F., Seed, R. B., Nicholson, P. G. and Jong, H. L., "Steady State Testing of Loose
Sands: Limiting Minimum Density", Journal of Geotechnical Engineering, ASCE, Vol.
116, No. 2, 1990, PP332-337.
Saada, A. S. and Townsend, F. C., "State of the Art: Laboratory Strength Testing of Soils",
Laboratory Shear Strength of Soil, ASTM STP 740, 1981, PP7-77.
Saada, A. S., "State of the Art Paper - Hollow Cylinder Torsional Devices: Their Advantages and
Limitations", Advanved Triaxial Testing of Soil and Rock, ASTM STP 977, 1988, PP766-
795.
53
Schofield, A. N. and Wroth, C. P.) 'Critical State Soil Mechanics, Mcgrw-Hill Pulishing
Company, Great Britain, 1968.
Seed, R B. -Soil Liquefiction and Cyclic Mobility Evaluation for Level Ground During
Earthquake', Journal of Geotechnical Engineering Division, ASCE, Vol. 105, No. MT2,
1979, PP2OI-255.
Seed, H B. and Idriss, L Kl, "Ground Motions and Soil Liquefaction During Earthquke,
Earthquake Engineering Research Institufte 1982.
Tatsuoka, F., -Some Recent Developments in Triaxial Testin System for Cohesionless Soils ,
State-of-the-Art Paper, Advanced Triaxial Testing of Soil and Rock, AST?4 STP77,
1988, PP7-67.
Tatsuoka, F., Muramatsu, M. and Sasaki, T., -Cyclic Undrained Stress-Strain Behavior of Dense
Sands by Torsional Simple Shear Test", Soils and Foundations, Vol. 22, No. 2, June 1982,
PP55-70.
Townsend, F. C., "A review of Factors Affecting Cyclic Triaxial Tests", Dynamic Geotechnical
Testing, ASTMK STh 654,1978, PP356-383,
Vaid, Y. P. and Chem, L. C., "Cyclic and Monotonic Undrained Response of Saturated Sands",
Advances in the Art of Testing Soils Under Cyclic Conditions, Proceedings, ASCE 1985,
PP120-147.
Vaid, Y. P., Sayao, A., Hou, E. and Negussey, D., "Generalized Stress-Path-Dependent Soil
Behavior with a New Hollow Cylinder Torsional Apparatus", Can. Geotech. J. 27, 1990,
PP601-616.
Wang, Z. L., "Bounding Surfatce Hypoplasticity Model for Granular Soils and its Application",
Ph.D. dissertation, University of California, Davis, 1990.
Wang, Z. L., Dafalias, Y. F. and Shen, C.K, "Bounding Surface Hypoplasticity Model for Sand",
Journal of Engineering Mechanics, ASCE, Vol. 116, No. 5, May 1990, PP983-1001.
54
Wilson, D. and Kutter, B. L., "Cewrig Tests on Cicular and Reaw Model Foundtion,
report to Naval Civil En eering Laboratory, Department of Civil and Environmental
EUniversy of Calfornia, Davi 1994.
Yamada, Y. and Ishihara, X, "Undrained Deformation Characteristics of Sand in Multi-
Directional Shear", Soils and Foundations, Vol. 23, No. 1, Mar. 1983, PP61-79.
Yamashita, S. and Tol, S. Effects of Fabric Anisotropy of Sand on Cyclic Undrained Triaxial
and Torsional Strength", Soils and Foundations, Vol. 33, No. 3, Sept. 1993, PP92-104.
55
SUMMARY TABLES AND SOME RESULTS
ON
TRIAXIAL, TORSIONAL AND ROTATIONAL SHEAR TESTS
Table 2.2.1 Summary of Triaxial, Undrained, Strain Controlled, Constant p Test Data
for Nevada Sand
Table 2.3.1 Summary of Triaxial, Drained, Strain Controlled, Constant p' Test Data
for Nevada Sand
Table 2.4.1 Summary of Triaxial, Undrained, Stress Controlled, Cyclic Test Data
for Nevada Sand
Table 2.5.1 Summary of Triaxial, Consolidation and Rebound Test Data
for Nevada Sand
Table 2.5.2 Summary of Compression Index Xand Rebound index K for Nevada Sand
Table 3.2.1 Summary of Hollow Cylinder Undrained, Stress Controlled, Cyclic
Torsional Shear Test Data for Nevada Sand
Table 3.3.1 Summary of Hollow Cylinder Undrained, Stress Controlled, Rotational
Shear Test Data for Nevada Sand
C4C
a (4 la
IaaC
0% 42 lowlaw
A.. 00
C 0 0o 0D4
C4 m4 C4 rc
%0Go' f ~ ~ '0 (
C, Co
C-O
o s.
.1 3a
o4 eq
C cv
CC
z0 %0
oI5 111le Iis
hAw InCAI0
gm0
o o C
"Ill
AA
ZOg
9 em
e4 eq t- ~
Go0
Co Co '0 i'CC
i--l l Co
z z z
a
Co C2 ClS %.d
,,~~~ ,..+1,
%..-
- on-, ,.
+z
+j o0C
410-$ 8. _
00- , ,.
1) .2 Co soc
00 S 0 0 0 (E '0
t
00 000 al VI
00 r-0 0 0 00
-_$ -- , . , . 0 ,.
1 C C> _ > C
I-.
o Q- ,i
4 -0" 60r- ~ +C, C
-" .
>C vn
- :-N &
Cp
0%~
0 ~~ %0%0 0% ~ % 0
C
Co C) C> C
57 C
In= =
j~~~n. 0~- ~ '-8
ai pm 4^@i -i4
A n.
2 0 C C) 00 001
en en m
In .0
eel W% on 0
C)(4 In 0 I
'0 - 00 t- ' % 00
In In= In In 1n
C-4 @0
o 0 0 0
C) 0 C0 0
0~C00 0 00 0C 00o
W) W%W1 W1 n
W% n 4n $6n An %^
In(% 0 (4 en C
04 q
-q C 4
-q
0'00% 0 % % 0% 0 % % 0 0% LA
1
_ON
_ en 0 - - - -
t00 too e.
N0 00.0 NO %0 % 0
0
No 0 0
4~ 0 0
NO
Ini n
aq '* an an Cn '
00
%n4 $0% W1 W1 W1 W1 W010
--
C0 04 04 C4 C4 0 -
%%0
F r U
'0 4_
00 o0 G 0
10~00 C', 00
02
C CC Co C
cc00 00 S
i 0 0 '0 '
I"'It~I i
00
('
CCC C CI
CC
CD00
Ca _ _ _ _
C4 SKc
~ii
ill
ii;
ro- so - -0c
1q 88888nn
C..2)
GM4-8
4.
C4 m4
C~Cod cc cc ccC c;
0
o 00 %C 00 g00 t-%C00 o%
00el~
'0 o0-
c co o c c= c o c c c c 4
CURC C
CP Sa o C ,.
0 C l 10C .C l4 DC 4 D' ,C oC
cncc cco c ;zt-cc "6~
olC ) oC 4 DD 0C
-~g =o C 0 8 8 c oo g CCl cc Opcc
C?4en 84 C4
06 .0 c.
0 00 -4
40 Cl. C
U U u U ) .0 0
.41
000
00 0 0G
00
C4 Ci
.2
W 00 COO
C0 C
4)-
Z4l
E- z
TEST RESULTS
ON
TRIAXIAL UNDRAINED, STRAIN CONTROLLED, CONSTANT P TESTS
o 100
0-
-100-
0 100 200 400
700-
no-
30 -
0
100 -
.100
-1 1 :
w
I
AXIAL STRAIN (%)
400
~350
Wac
wU 250
~2
U-
UA-
UJ 150
-J
100
Z 50
z
AXIAL STRAN(%
~100
40
-M
m
No
cc 100
0
.100
400
AXIAL STRAIN()
450 -
400200
350
6 150
0 5
z 10
2 2 0 2 4 6S10.
AXIAL STRAIN(%
Figure 2.2.2 Triaxial, Undrained, Strain Controiled, Constant p Test (N6OU1001)
N80UI 002: TRUAXAL UNDRAINED STRAIN CONTROLLED
S-
we
S 0-
-10
-'U
~400
p100
0-
-100
0 4
AXIAL STRAIN(%
450-
400
~350
300
LL. 200
U-
z
0 22 ; 0.
AXIAL STRAIN(%
]Fig=r 2.2.3 TriaxiaL, Undrained, Strain Controlled, Constant p Test (N6OUI 002)
N6OU1003: TRIAXIL UNDRAINED STRAIN CONTROLLED
550
4W -
a00
0*
41
50 150 250 350 450
no
400
01 50
I-
-11S 5 7 9 1
00
W AXLAL STIN (%)
4 00
50
1450-
3 50
AXIAL STRAIN (%)
1
250
4W 2
ic
z
505
AXIAL STRAIN(%
Figure 2.2.4 TriaxiaL, Undrained, Strain Controlied, Consti Lest (N6OU 1003)
N60U2501: TRIAXIAL UNDRAINED STRAIN CONTROLLED
4.3
41
-42
S 0.9
42
4-06 __ _______________________________________
-0.435
Lii 450
3 400
z 250
300
z
.1 i 7
AXIAL STRAIN(%
Figure 2.2.5 Triaxial, Undrained Strain Controlled, Constant p Test (N60U2501)
N60U2502: TRAXIAL UNDRAINED STRAIN CONTROLLED
1.1 -
Iu
0.7
,A
0.3
0.4
0.1
0-
1.1
0.7
GA
0G
0.1
02-
0.1 -I
650
500
Vsm
400
~j 350
-1 1 3
0.48
02
0.1
1 1
t 0.7
0.6
c 0.4
0 22 a
S0.2
0.1
.0.1-
42
-1
700-
w 400
9 200
10
&.3
02
0a1
0-
no 4;0 No No0 m0
mu 0.8
0.5
0.2
0.1
A)UAL STRMJ %
Nuo
U. 5oo0
M -
cc 450
0
z
300-
2-0.5 2.1.5
5 3.54.
10.7
021
.41
420 4W 500 546 iSo 80 700 740
1.2
1.1
0.8
cc 0.5
0.2
0-
700
w w-
U. 00
U. 50
4250-
-1 10 3
40-
-70
40
40
.100
-110
10 30 56 76 SO 110
10 -
470-
-W
-~
40
040
-7o
-100
-110 -
-120 *
110
~go
w 70
U-
*W 80
Z 20o
.4 -- 2-1 0
-10
-14
-22
-24,
0 20 46 0 80 10
2-
0-
C1 -i
-16
-24
-7 -5 4 -
110
100 -
090
0 -
w7
WU 40
30 -
0 20
z
z10
S7-5 - -1
AXIAL STRAIN %)
Figure 2.2.11 Triaxial, Undrained, Strain Controlled, Constant p Test (N60UI006)
N60U1008: TRIAXIAL UNDRAINED STRAIN CONTROLLED
-100 -
S-150
50-
4W0
-w0
50 150 2i0 350 4W0
50 -
0-
~.1o0
0200
0 m
-450
-9 -7 -3
4W-
IL
20
3W
15 150
0
W 200
z
50
.9 -7 5 -3 -
AXIAL STRAIN(%
Figure 2.2.12 TriwaL Undrained, Strain Controlled, Constant p Test (N6OU1 008)
N6OU1009: TRIAXIAL UNDRAINED STRAIN CONTROLLED
50
-100
,
150
400
-450
vo-
0 100 N00 300
MEAN NORMAL EFFETIVE STRESS (PA)
0~
-0
00
L-1
-450
~4W
00
4W
E 5o -
2 50-
U-
U. 200 -
-10 -
0
zz 50
-10 -4-4-2 0
Figure 2.2.13 Triaodal, Undrained, Strain Controiled, Constant p Test (N60U 1009)
N601.2506: TRIAXIAL UNDRAINED STRAIN CONTROLLED
0.S
as-
S 0.6
0.4
0.
0.2
S 0.1
0-
-0.2
-100 100 300 500
0.9
CD W
0.A
S 0.2
0.1
0-
.0.1
-0.2 -
4 -4 -202468
AXIAL STRAIN(%
CD500
cc 400
> 300
U- 200
-
< 100
AXIAL STRAIN(%
Figure 2.2.14 Triaxial, Undrained, Strain Controlied, Constant p Test (N60U2506)
N60U4006: TRIAXIAL UNDRAINED STRAIN CONTROLLED
10-
.100 :o:; ; ;
100
-m
100
4200
MEN XORAL
SETAIN STES%) CA
~-m0
-100
-W
-7 -53
!z
F550
ws
lii 300
1 250
0 200
z
100 -3 -1
AXIAL. STRAIN()
Figure 2.2.15 Triaxial, Undrained, Strain Controiled, Constant p Test (N60U4006)
1ON400Ul: TRLAXL'AL UNDRAINED STRAIN CONTROLLED
4W -
200
250
10 -
0 100 -40
350
~250
200
i-o
100
50
AXIAL STRAIN(%
w
I- 3
w 250
a:100
0
zz 5
AXIAL STRAIN(%
Figure 2.2.16 Triaial, Undrained, Strain Controlled, Constant p Test (IloN400U1, Dr=-18/,O)
40N250U1: TRIAXIAL UNDRAINED STRAIN CONTROLLED
wo
100
Go-
0O
13
00
0-
W. 3o50
U-
w 3
zZ 200
-1 5 79
No
0
4w
100
0 -
40 860 120 l0 200 240 US US 260 400
700-
400
50 ,
~200
100
0 -
-1 15
400
260
140
cc 120
w 00
60
20-
113
CCIA STR20 (
Fiur 212.1 Trail nrieSri otold osatpTs 00
060U2001: TRIAXIAL UNDRAINED STRAIN CONTROLLED
OO
I"~
400
No
40
N-
1u
00
r
ao
-100
AXE STRAN (%
)
3o
-10 1 4
w340
-140
-100
-14
0
-a-
'(40-
~40-
S-120
-140
~-100
400
-W"
35
2
-74
w 20
W 200
AXIAL STRAIN I%)
U.
U-
cc100
45 -5 -3
I-400
100 -
0 -
g0
~400
0
-4W
4 4 -4 -2 0
g58O-
5w:
w4 W
w 4w00
CC200
..
z
I-
100
70
3W
400
2
00-
1 3 7 9
AXIAL STRAIN (%
U00
W 250
L
IL.
c2100
s5o
AXIAL STRAIN(%
Figure 2.2.22 TriaL, Undrained, Strain Controlied, Constant p Test (050U1)
060U1001: TRIAXIAL UNDRAINED STRAIN CONTROLLED
7o
no o
00-
0-
50 150 250 3i0 M50
so0-
~~400
No "
200
10
0 2 4 6 a 10
. i450
~4w
fu)
~300
LU250
L
u2 0 0
ir150
0
z
5i0
0 10
so-
70-
-100-
No
.00
-1 5
(g 50 AXAL ST'AN I%
w
CC 50
W 250
U-
CC150
0
z
~50L
13 5 71
AXIAL STRAIN (%)
Figure 2.2.24 Triaxial, Undrained, Strain Controlled, Constant p Test (060U 1002)
050U4: TRIAXIAL UNDRAINED STRAIN CONTROLLED
0
40-
.100
.140
S.100
MEAN NORMAL EFFECTIVE STRESS (KPA)
.100
.140-
-40
30 -
40
3400
AXIAL STRAIN (%)
L 40 -
1320-
4L
ISO -2
ITSOi AXLA (
> 220
20
w
<020-
_j 140
CC100
6 140
0020so - -
-60 _ _ _ _ _ _ 2__ _ 0 _ !t
1(40
-100
.380
-no
-300
-0 -W
0150
4W
-10 -8 - -4 -2 0
cc -m~100
00
do0
400
U) 280
U)360
OW 240
AXIAL STRAIN (%)
> 200-
b 60
LU 40
U 20
cc 10 _ _ _ _ _ _ _ _ _ _ _ _ _ _
-10 4 -
-4
0 2 a 10
110 -
740
90 M
~so
20
10
0 24 6810
VO AL STRAIN (%)
110 -
100
9g0
jr 70
rh so
20
30
0-0
VOLUMETRIC STRAIN(%
Figure 2.3.1 Triaxial, Drained, Strain Controlied, Constant p'Test (N7ODSO1)
N7OD1 001: TRLAXLJ DRAINED STRAIN CONTROLLED
0-1--
-4
Q12
405
0
04
45
-1 1 3 7,, 9
AXIAL STRAIN(%
W160
02
.0
0.40-
40
-100
t44
-2 0 2 4 6 - 10
AXIAL STRAIN %)
i400
.00
200
~-100
-1.2
>.1.4
-1 1 3 S 7 S
I-
20
00
40
0- -1 1 3 5
AXAL STRAN (%
Z140
2w 12*
~100
so
140
VOLMETRIC STRAN ()
Figure 2.3.4 Triaxial, Drained, Strain Controlled, Constant p' Test (N70D100A)
N7OD100B: TRIAXIAL DRAINED STRAIN CONTROLLED
0P
-04
-,
-12
-1.4
190
140
100
Q4 0
20
0~
A05 0.U. . .
AXIAL STRAIN(%)
so
140
~120
w
~100
-40
240
20
A4
AG4 -
0 2
-1.
AXIAL STRAIN(%
140
~120
100
20
S0
ISO
240
21- 1. 12 84
100TRCSTAN %
soue236 Tail rind tanCnrleCosatp et(7DOC
N711005: TRIAXIAL DRAINED STRAIN CONTROLLED
&.4
-a0-
IA
-IA0
.120
*40
O o
-2
AXlAL STRAWN(%
120
0
so
~i40
0 0
40-
-2
VOLUMETRIC STRAN(%
Figur 2.3.7 TraiaL, Drained Strain Controiled Constantp' Test (N7OD 1005)
N70D2505: TRIAXIL DRAINED STRAIN CONTROLLED
0.4
f4
Q4 4
-1.4
-1.6
-l.4
AXIAL STRAIN(%
500-
7400
3oo
100
o-100
4200
-11 -9
AXIAL STRAIN(%
500
00.
'300
~100
0 0
S-100
410
0.4
-742
44
0
40-
-0.1 -9 -7 0 4
2-
-100
0-
io
-20
-4
cc 40
-100
VOLUIMETRIC STRIN(%
Figure 2.3.9 :Triaxial, Drained, Strain Controlled, Constant p'Test (N7OD lOOP)
N70D100R: TRIAXIAL DRAINED STRAIN CONTROLLED
Z 84-
~3
10 -
1 -
100 -9- -
140
010-
4a0
o~ so
-40
-100
-120'1
ISO -
IGO
140
120
0
o
20
0
~-20
~-40
o-60
40
-100
-120-
-1 1 35
VOLUMETRIC STRAWN(%
Figure 2.3.10 Triaxia, Drained, Strain Controlled, Constant pTest (N7OD lOOR)
TEST RESULTS
ON
TRIAXIAL UNDRAINED, STRESS CONTROLLED, CYCLIC TESTS
14
.12
02
-2
-4
.10
-12
E0.
-.
14
Q o-
10
.,o
-10
.40
0-0
iao 4 --
r -
10
.--
IL40
~10 4 - -2 0 -, 2
90
710
taoOZ0
-10
t
AXIAL STRAIN Cy
Figure 2.4.2 Triaa, Undraind, Stress Controlied, Cyclic Test (CYlOONI)
CY10ON2: TRIAXIAL UNDRAINED STRESS CONTROLLED
310
,o0 fi I
L
-.
10
3030 70 90
30
10
:
0
-10
1 4.14 .9 . 4.02 .2
AXAL STRAIN )
_ ,-,
100
70
AXLAL STRAWN (%)
uj
450
AXLAL STRAIN(%
Figure 2.4.3 Triaxial, Undrained, Stress Controlled, Cyclic Test (CY100N2)
CY100N3: TRIAXIAL UNDRAINED STRESS CONTROLLED
170-
140
lo-
W110
~100
970
70
s0
50
410-
10-
10
~10
310200sos i1214
~10
40
10
U 40 -
30
30
20120
~10
w 0
13-
Z10
2
-0.9 02061141. . .
soAI AXIA (%
Fi5
70 r 4. Trail-nrieSrs otoid ylcTs CIO 3
CY1 00N4: TRIAXLAL UNDRAINED STRESS CONTROLLED
~140
S100
20
0
40 606 100 120 140 IGO Ig0
240-
6
WI
~140
20
1l
cc100
0
40
20
.170
150
~140
~130
20
wJ 110
IL
cc70
AXIAL STRAIN(%
Figure 2.4.5 Thiaxial, Undrained, Stress Controlled, Cyclic Test (CY100N4)
CY100N5: TRIAXIAL UNDRAINED STRESS CONTROLLED
40-
10
[0
40-
40-
S10
AXIAL MARDN(%
~100
70 -
a408 4.01 0.01 0.08 0.06 0.07 0.06 01Z.
AXIAL $TRMN %
Figur 2.4.6 TridAi Undrained, Stres Controlled Cyclic Test (CY100NS)
CY100ON6: TRIAXIAL UNDRAINED STRESS CONTROLLED
40
-40 -161 s6 10 v
-40
-101
AXIAL STPAIN(%
M 0
*so
43
0 -
~20
~10-
AXIAL STRAIN()
Figure 2.4.7 Triaxal, Undrained, Stress Controlled, Cyclic Test (CY100N46)
CY25ONI: TRIAXLAL UNDRAINED STRESS CONTROLLED
40
-40
40
AXIAL STRAIM%
~240
~120
.. 100
AXIAL STRAIN %
Fgure 2.4.8 TriaxiaL, Undrained, Stress Controffed, Cyclic Test
(CY2SONI)
CY250N2: TRIAXIAL UNDRAINED STRESS CONTROLLED
70
40
21)
-10
-40
S40
30
-010
*10
~240
10
460
140-
460
520
~-40
S-60
480
-100
-120
-140
AXIAL STRAIN(%
240
IISO
~140
LU120
ILO
w
CC40
AXIAL STRAIN(%
Figure 2.4. 10 Tiaxial, Undrained Stres Controlled Cyclic Tesn (CY250N3)
CY2SON4: TRLAXIAL UNDRAINED STRESS CONTROLLED
a0--
70
S20
.10
~-70
-50
s
70
as
-50
420
10
-40
-0
AXIAL STRAIN(%
?20D-
~240
22-
200100
0t 40
wz2 0
-10
AXA TRI %
Fiur 2..1TixaUdand tes otoid ylcTs C204
CY250N5: TRLAXIAL UNDRAINED STRESS CONTROLLED
10
~a
-W
-00L
Q-
10
0-
o n-1
I-7
-3 -11
20
~~21
AXLSTAN %
20
-10
-60
-7h
04.0
730
-40
-60
-70- 77
.7 -5 31
AXIAL STRAIN(%
280 -
~240
zU 20
0-
AXIA STR00 (
Fiue241 rail ndand tesCnUled ylcTs C206
CY250N7: TRLAL. UNDRAINED STRES CONTROLLED
70-
-40
-50
-70
150 1;0 220 260
~40
.30
~2
~1:
-10
0 40
-430
.40
4012 -0.1 406 -406 404 402 0 0.06 0.04 0.06 0.06
AXIAL STRAIN(%
~240
W~ 210
200
U-
~170
z
140-
-no
.40
-40
-140
-200 'lO0G40IO i
140-
If.100
-40
240
0-
~-40
0 0
-80
-100
-140
AXIAL STRAIN(%
~260
40
2 0
-7 200 .5
AXIA STRSNO
Fiur 2..5IGOaUnrieSrs CnrieCcicTs C20
CY250N9: TRIAXIAL UNDRAINED STRESS CONTROLLED
140
so-
40
~g0.20
~40
140
w40 "
.100
240
240 S-42
S2 o
-0 40-
o
9 40
44-4 -2 02
AXIAL
AXIAL STR, iN (%)
STRAIN()
s4o
o.
20
40
.0
140-
-
AXIL STRAIN(%
2ft
Fiur 2..7TixaUdand tesCnrleCci et(YSNO
CY25ON41 1: TRIAXIAL UNDRAINED STRESS CONTROLLED
so
G-1
40
30
270
70
30
40-
Ii-
U0-
ALSTRAIN(%)
40
~so
30
01
Qo -
5-10
I-0
~40
-00
70
~20
~10
240
o-
40-
2GO
-50
-70
AXIAL STRAI N%
~240
ISO
S140
U100
sao
"60
40
Z20_
-20-
4 4 -20 2 4
AXItAL STPAWN%
Figure 2.4.19 Tia-xial, Undrained, Stress Controiled, Cyclic Test (CY2SON 12)
CY250N13: TRIAXIAL UNDRAINED STRESS CONTROLLED
I0
o iso a go
MEAN NORMAL EWECTIE STRIESS WN
140-
-n
.0
-m
~240
U-
140
so
so so -
"S0-
40-
I0:
J40~
10
M ANXC~IAL
SFETAIN~ s QP
20
am-
1*1
100
-12
4410 --- 0;
UU
-12
-12
48 4 -4 -202
~40
~10
-10- XA
AILSTRAWN (%)
Figure 2.4.22 TriaxiAl Undrained, Stress Controlled, Cyclic Test (CY5001)
CY1 0001: TRIAXIAL UNDRAINED STRESS CONTROLLED
40
~10
.0
-10
- 10 10 36 50 70 90 110
40-
30
210
Q 0 -o
.10
S110
,100
~so
w
~10-
20 -
40-
30 5070q
410-
10163
. 0
w--
-03-
~10 -AS
20
0 -
~10 -
40
0~
l.o
~.10
-10 ;0 30 50 70 9O
40
~10
0
~-10
-40
AXIAL STRAIN ()
0 -
so-1
so
ZO100
40
U.
U.
0 10
z
40
70-
0 10
10 10- 0 7 0 10 ,0 10 1!0 1
0r 40
-1090-
-40
-70
-10 10 -6
i0 70 -2 110 1" 150 170 190
70
40
20
10
0
O -10 0
c-40
-60
-70 - ___________
-20 2
U)
~140
WLU20
U. 8
0: 40
0
Z 20
z
AMUL STRAIN(%
Figure 2.4.26 TriaxiaL, Undrained, Stress Controlied, Cyclic Test (CY2000 IX)
CY200021: TRIAXIAL UNDRAINED STRESS CONTROLLED
20
40
40-
.40
-
40
60 -
cc 40 -
~.0
0 -
40
-80
-202
O~200
~140
w
S120
00
U1
U-
U.
w 80
CC 40AA
0Z 20 -
Z0
4 4 .4 -2 o= =
3 -1
40-
-10 - ,
0
10 -
al
0-10
F01 .
w
Ir 97
LL-
20 \
02
40-
01
-40
-10 ;0 i0 50 70 O
40-
30.
a, 0
.10
0 0-
-40
,100 ..
0-
cr 70_-
C-50,o
W,40
3 0
0-
O 10 -
z
z 0-
10
420
~10
I:
00
-40
-50-
~40
~2D
~10
0-1
-3 -
~-*D
0 40
003 401 0.01 0.03 &;05
AXIAL STRAIN(%
~101
96
93-
-9J
8S7
2 as
~81.
007-.0 403 401 0.01 0.0 0.05-,
-40
2 0-
30-
~-30
-40
-10 ;0 -7 -3-11 76
So-
O
- -10-
4.0
0 50
"100-
S2 0
O 10
Z40
Qo
I-10
-40-
-0 11
~40
~20
~10
40
-40
,100o
cc so
IL
LL 40
30
~20
4 4
I -10 1 09 o1 1
~40
-
-40
10110
50
0 10-
-5o-
9040
so -7- -
110
370
2D-
10
-40
-50
-10 10 360 5o 7o so 11o
so -
10
10
-30 -
-40
-7 -5 3
w o-
AXIAL STRAIN(%
,o
110
710
40 4
10
-10
41 3O 41 43 4 47 40 O 1
f5 MEAN NORMAL U'CTW STN S "(A)
Is
-5
-15
- .
-W
IS
-I-w
Fu 2 S
.140
47 -,
w6_
45-1 -O0 ,.O ,05 .0 .0 0.1 .0 0. .0 . 01
S NS
Lo
0Mlb
SOTP
GAm T T I
GLOW
Sm~~O PiiI
______ IOSUDTO TEST
- ---
----
!I:II~~Loi *
I -Pil
_____
TIAL COSUDTO
iliji ) TESTii
0_ _ __ _ 1 4 .r
P~ ls
1011.
1'4
I ~- 44 ~ . - -4 .. 4)
am$-'1 1t
am4 4.1 * -- ''
GAM________________I
;
an&
ago.
a am t 9LOGI
U1 14~3
ama I 1 h 4 I)
Figure~~
: rail 2.. osIdtinTs (Y0C9
TRIAXIAL CONSOUDATION TEST
(oWCIooI)
ii I 2 22
0, ,-
o \:I
___"_____ ji
! " i ______ :
urn ____. . . .
Q622!
)' urns
0= i : "6
0114 l____.. .
0112 OA
i tl s t
1? 4 ,a
101 1" 6 IS
SQilRT(P)i i I
TRIXIA COSOUDTIONTES
aa j
DAW1
OI-M
__II i i
LOGOP
am______
am82-
2 ______
______ __
0 CO25
SQRT(P)
_ 1J "-[iI(001006) i -
_______
I
imI!_i_ i,!,l ,
I
:
I 1-11
!
___
iiI-
.: i-] l
!!I
! i i i i ,! \ ', i I I !!
_____i-iii
!IE1i'-",I iZ !Ii-i
__ _ i i i
'ii Ii
I i
__!!___
l"
i .'
!X ii
-
am
I __,___ : :
0.
LOOP
o*
2 - TRIAXIAL CONSOUDATION TEST
(06001o006)
ame
SORT(P)
140-
Ioo
bI
o
p o-
0 40
t0.
40 -40 J0 0 2 40 l 0
Fig
40-
10-
00
0C02 04 o461 0i 610161k1 61i 1 61 W20,
MenNraU8 fetveSrsP
6
-4W
Amo
a.
4I 475 45 .0 0 065 is 6
.
30.
-101
01
0 012 44 is is 1.
020
1w
a040
z
cU
000
I.4
0,4
Z Uhft0.1%b
dhrt ot m(uw pv
- 65 45 sos ds 65
Shear Strain, e. (%)
Q;
S
2
6o
0)
40-
0-
4*0 ioCi)
i oG0 i D101; 0I_4 04 i 8,'
Q;
10.
4 46 k 4 40 0
U
Shear Stress, . (KP()
s
I.2
o.2
0
E
ca 30
W.-
10
40-
C'10
040
015
-C1
-I 7S 5 -is -5 05 0-2
Q;
a--
0
040
0
C)
a 0m
z
lowtW0.1%If EPILkz
W
S 4-1 47 ss
s.
a
0
__ 1.t1
0
1.
o30.41
40
0.-
*10
-20-
040
40
5&
d- 40-
20"
-0.
(U
b 0 . - , I
40-0 -40 -30 JO -10 0 1o a 4D 40 50 0
100-
- Rf=0.737
7O.-O4
30-
Cn
i-
0,
0 0 20 30 40
6 50 gO 70 4 0 100 110 10 16 40 150 160 170 160 190 200 210 220
Mean Normal Effective Stress, p' (KPa)
00-
00
- -
CO
2
cc,
-40
(U -W
G !4 4 0- 1 2 3
40
10-
30
.20
40-
0
0
2 -15 -1 4s To i1 1.5 2
no-
a-
.b
1. 12O-
10-
0
4) 100-
co
0
z
-M:-2 -.
5 :l .4s 10 as 1'.5s
200
100
000
W
00
E
0 0
z
00
0.
2.
1.5
0L5
0 -2
46 4 -3 -2 1 0 2 3 4 5
Figure 3.2.14 Hollow Cylinder Cyclic Torsional Shear Test (NK IOCU5O)
NK10CU50: UNDRAINED STRESS CONTROLLED TORSIONAL SHEAR TEST
60-
0 -
40
b. -30
Li?
I-
1i
Y 0-
20
o.
30
104
40-
CO)
0- 4
0-
Qi
!01
Deviatoric Strain, 2e ./ %
w 1W
E
zC
S
S
4 4 4~4 .
8* .3%Wsa ft EPSILONzo
4 4 0 2 3 4
Shear Strain, e. (%)
'I)
a..60.
Cu
U)
9)
-40
"60 ! i-! !
t; 1.
rn 40
-10-
--
C1
40-
40-
0\
Co -I "
(0 0*,W - -u 0
(0
Shear Strain, e..:,%
S .10.
- 40-
, 0-
co
4W 4i i ip~~ 5
I Sb Shsw
140
0140
0
z 0
20,
22-0
Shift0.1%left for upper part
1W Exbmdon
0) 12D
01W
0
w
00
I| I
0
-5 -4 04
Shear Strain, e, M%
0.2F
1.2&
0
Shift 0. 1%left for upper part
U)U
-1.5
Q .7
(0
U)
0 Cmnression 0 E~dauion
0, so 0;
'0 0 * ot 0 0o so 0
0.*
0.21
0v 4
4 *~ Conwr"sio 0 Emteiion
0 00 0 Go 0 so 0 so 0 60 0 so 0 90 0 00 0
10.
0 -1
0i40
U)
10)
0-
40-
020
00
-6 1
CL
Y 50
00-
00
070
Deitrc tan (c ./
ISO- +
120,
W W
E
0
z
C 40,
W
0-
5
040
4)
E
0
z
a,,
00
U; .L I SWS Sh
1.
425-
a:a
0 4
4.5-
ft Rh
II, G
2 Rio
0 .4-
02
OL9" or "f
0U Oanimproaion 0 8"ioin
0 0 0 60 0 60 60 0
dt &w-
-42 f
04
4-
0 0 0 0 0 ;0 so
ZC05
C,,
C
A
0<
00
Go
Z ~E A
U U z
a o
U) c L- oL
co0N
2-
o *
0 1 0 00 0
0
-2-
A
n K=1.38 KI0K=.3
0 0 =04
-3 1 1 1 1
0.1 1 1010
Number of Cycles
Figure 4.2 Relationship Between Cyclic Strain and Number of Cycles in Hollow
Cylinder Cyclic Torsion Shear Tests
CYCLIC TORSIONAL SHEAR TEST
0.8 2 1
0.6 3 4
p
0.4-
0.2-
0 I ! I I
-10 0 10 20 30 40 50 60 70
Figure 4.3 Relationship Between Stress Ratio and Angle in x Plane for
Hollow Cylinder Cyclic Torsion Shear Tests
TRIAXIAL AND TORSIONAL TESTS
0.6- 422
0.4-
0.2-
0-
--30 0 30 60 90
(File: COMBI)
Figure 4.4 :Summary Plot for Peak Stress Ratio Versus Angle in ic Plane in Triaxial
and Cyclic Simple Shear Tests with Stress Paths of the Last Cyci e of
Torsional and Rotational Shear Tests
TRIAXIAL TEST RESULTS
(N. C. NEVADA SAND; CONSTANT P; STS-P-CE)
2.6-
2.4 MAXIMUM STRESS RATIO FOR DRAINED TEST
- ....
2.2- .... ULTIMATE STRESS RATIO FOR DRAINED TEST
.2- DRAINED FAILURE ENVELOP
16....... ........
....
......
...................
... ..................
. ..
.....
........
.....
........
14
4... . .......... .
IL1.2-
q:
0.8
0.6 - 1 2 34
OZ0. COMPRESSION
0.2-
0
O) 0.4-
N CRTCAL STATE
Co 0.8- . ............
-
............
S . ........................... ...
1 . ..
....." .... ........ .- .................
1.2 - - ./ FAILURE EiNEL(3P
0
1.4 ,_,__
0 200 400 600 800
Figure 4.5 Summary Plot for Stress Ratio Versus Mean Normal Effective Stress
in Triaxial Compression and Extension Tests
DISTRiMBrTON LIST