Missouri University of Science and Technology

Scholars' Mine
International Conferences on Recent Advances 1981 - First International Conference on Recent
in Geotechnical Earthquake Engineering and Advances in Geotechnical Earthquake
Soil Dynamics Engineering & Soil Dynamics

27 Apr 1981, 2:00 pm - 5:00 pm

Soil Behavior Under Dynamic Loadings

T. Goltom
Michigan State University, East Lansing, Michigan

G. Baladi
Michigan State University, East Lansing, Michigan

Follow this and additional works at: https://scholarsmine.mst.edu/icrageesd

Part of the Geotechnical Engineering Commons

Recommended Citation
Goltom, T. and Baladi, G., "Soil Behavior Under Dynamic Loadings" (1981). International Conferences on
Recent Advances in Geotechnical Earthquake Engineering and Soil Dynamics. 14.
https://scholarsmine.mst.edu/icrageesd/01icrageesd/session01b/14

This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.

This Article - Conference proceedings is brought to you for free and open access by Scholars' Mine. It has been
accepted for inclusion in International Conferences on Recent Advances in Geotechnical Earthquake Engineering
and Soil Dynamics by an authorized administrator of Scholars' Mine. This work is protected by U. S. Copyright Law.
Unauthorized use including reproduction for redistribution requires the permission of the copyright holder. For more
information, please contact scholarsmine@mst.edu.
Soil Behavior Under Dynamic Loadings
T. Goltom
Research Assistant, Dept. of Civil and Sanitary Engineering, Michigan State University, East
Lansing, Michigan

G. Baladi
Assistant Professor, Dept. of Civil and Sanitary Engineering, Michigan State University, East
Lansing, Michigan

SYNOPSIS Stress-controlled cyclic triaxial tests were performed on clay samples obtained from the subgrades of existing
highway pavements in the State of Michigan. These samples were unconsolidated or isotropically consolidated and then
tested under the following conditions:
1) confining pressures of 34.5 psi and 172.4 kPa (5 and 25 psi),
2) a range of stress ratio from 2 to 4,
3) a constant frequency of one cycle per second, and
4) at the natural water contents.

Test equipment included an MTS electrohydraulic closed-1oop system which applies the sustained and cyclic loads to the
sample in a stress-controlled mode, a minicomputer which controls the actuator of the MTS system to yield a loading
function similar to that applied, by tandem axle truck, to the pavement section, and measuring devices including two
vertical and two radial LVDT(s), a load cell, a pore water pressure transducer, recording devices and an ascilloscope.

The test results indicate that, for all samples, the plastic strain rate (in the logarithmic mode) assumes two values
per test. The first of these values is constant from cycle number one to cycle number 100 while the second value is
also constant and it controls the plastic strain beyond cycle number 100. This abrupt change in the strain rate values
was observed in the unconsolidated and consolidated samples. Also, it was found that the ultimate stress ratio that it
can be applied on the sample during cyclic loading is a function of the confining pressures and other sample parameters.

INTRODUCTION are proposed each satisfying one or more aspects of the

fatigue phenomenons and yet none being adequate for all
In recent years the studies of soil behavior under re- cases (17,15,8).
peated loading conditions were given considerable atten-
tions particularly in the field of pavement design. This Generally speaking, silt and clay subgrade materials
is so because researchers realized that the application (subjected to cyclic loading) will exhibit a stiffening
of stresses to pavement materials by moving wheel loads behavior as the number of load applications increases
is a transient one (4, 10, 2, 9, 15 and 6). To study (16,15,12). Each load application produces elastic
the effect of cyclic loading on subgrade cohesive soils, (resilient) and plastic (permanent) deformations. The
stress-controlled cyclic triaxial tests were performed increment of plastic deformation of cohesive soils was
on clay samples obtained from the subgrade of existing found to be large during the first few load cycles. This
highway pavements throughout the State of Michigan. increment decreases in subsequent load applications.
These samples were unconsolidated or isotropically con- After large number of load repetitions the change in
solidated in a triaxial chamber and then they were permanent deformation per cycle becomes very small. The
subjected to undrained cyclic loadings under different total permanent deformation of the test samples, however,
testing conditions. These conditions included: increases with increasing number of cycles. Several
a) a range of cyclic stress ratio from 2 to 4, and models were used by different investigators to present
b) confining pressures of 5 and 25 psi. the cumulative permanent strain (Ep) and the number of
In this paper the characteristics of the measured axial load repetitions (N) data. The most popular of these,
permanent deformations are discussed. however, is the log log model (19) and the semi-log
model (16,8,1,11). In this paper the former model will
be used to study Ep and N.

BACKGROUND INFORMATIONS

Timoshenko (17) credits Poncelet as being the first to TESTING PROCEDURE AND EQUIPMENT
consider the strength of materials under repeated load-
ings and to introduce the term "fatigue" to describe the In this research program, all samples were trimmed to
resulting strength deterioration. Timoshenko (17) also 5.4 em (2.126 inches) in diameter and 13.6 ern (5.3543
credits Wohten for conducting the earliest and most inches) in length. The sample axial and radial deforma-
extensive repeated load tests. Wohten found that the tions and the pore water pressure were measured using
number of load cycles to failure increased as the cyclic linear variable differential tranducers (LVDT). Two LVDT
stress intensity decreased. Other investigators (5,13, were mounted verically 180° apart and two horizontally,
14) studied the fundamental aspects of fatigue and one at the middle of the sample and the other at one-
developed hypotheses to explain the experimental data. third of the sample length from the base. The loading
These studies are still continuing and several theories system consisted of a closed loop electro-hydraulic

129
 130

actuator. Most tests were conducted up to 30,000 cycles TEST RESULTS

using a minicomputer modified sinusoidal wave form
(shown in Figure 1) and a constant frequency of one The grain size distribution curves of the test materials
hertz. are shown in Figure 2. Typical cyclic triaxial test
results are plotted in Figures 3 through 5. The perma-
cu nent strains (£pl were calculated by dividing the
u
<:: average axial deformation (average of both longitudinal
cu
s.. LVDT) by the original sample length. The samples were
....cu consolidated under a confining pressure of 34.5 and
........
Q
172.4 kPa (5 and 25 psi) until the secondary compression
.,., curve was defined after which the repeated load test was
cu commenced. The data in the figures are plotted in terms
.....s.. of the logarithm of £p against the logarithm of N. The
til straight lines in the figures were obtained using least
.... square curve fitting technique. The two straight lines
...."'0.u in each figure intersect at approximately cycle number
100. The angle of intersection is called the angle B
....k<:: in the figures •
p..
Time
FIGURE 1. MINICOMPUTER MODIFIED SINUSOIDAL WAVE FORM DISCUSSION

Examination of Figures 3,4,5 and 6 indicates that the

relationship between the cumulative permanent strain (£pl
and the number of load repetitions (N) could be modeled
using the following equation:

log Ep log ai + si log N OR

(1)
ai Nsi
~
,... 60 1---J------__.:tr-___J\---+-----+----+--;
cu where: log ai is the intercept in the figures,
.....<:: si is the slope of segment i of the straight lines,
....
i ~ 1 or 2 relative to the proper segment of the
~ 40 1---+-----+.......l~t-.-'!\---1-----+-----t--1 straight line in the question.
cu
~ It is apparent from the figures that the data points follow
~ two distinctive straiqht lines. The first of these lines
20r----+------t-----~---~~~~~-+-------~--~ spans cycle& number 1 through number 100, while the second
line is for all cycles beyond cycle number 100. Similar
data ,concerning the second straight line, were reported
by several investigators (3, 19,16). They recommended
that the first few cycles ,up to cycle number 100, be used
10 1.0 0.1 .01 .001
for sample conditioning to seat the cap on the sample and
Grain Diameter (mm) to eliminate possible end imperfections.
FIGURE 2 GRAIN SIZE DISTRIBUTION CURVES Further studies of Figures 3,4,5 and 6 indicate that the
angle of intersection (6) between the two straight lines

SYMBOL <\ Log a s od w G

s
0'3

• 2 .00034 178 121 20.4'1; 2. 7:2

• 3 .00108 171 121 21.2'1; 2.72

A. 4 .00598 170 121 20.0'1; 2. 7:2

N
FIGURE 3. AXIAL PERMANENT STRAIN (Ep) VERSUS NUMBER OF LOAD
REPETITIONS (N) FOR SAMPLES CONSOLIDATED UNDER
CONFINING PRESSURE OF 5 PSI, SITE #3.
 131

100~------~--------~--------~--------r--------,

SYMBOLS o1 B
- !Log a od w G
s
03

• 2 .0015 175 121 19.8% 2.72

A 2.5 . 0135 172 120 20.6% 2. 72

• 3 .0162 170.9 122 20.0% 2.72

N
FIGURE 4. AXIAL PERMANENT STRAIN ~ p) VERSUS NUMBER OF LOAD
REPETIRIONS FOR SAMPLES CONSOLIDATED UNDER A CONFINING
PRESSURE OF 25 PSI, SITE #3.
2.~--------~------~--------~--------T---------
SYMBOLS o1 Log a
- B oct w G
s
03

• 2 . 0002t 168 118 22.4% 2.70

• 3 . 0012. 165 118 21.8% 2.70

• 4 .0173 164 118 21.8% 2.70

FIGURE 5. AXIAL PERMANENT STRAIN (£p) VERSUS NUMBER OF LOAD

REPETIONS (N) FOR SAMPLES CONSOLIDATED UNDER CONFINING
PRESSURE OF 5 PSI SITE #2

o1
SYMBOLS - Log a B oct w G
0'3 s

• 2 .00047 166 119 20.0% 2.70

• 2.5 .0017 179 118 19.8% 2.70

A 3 .0108 180 118 19.8% 2.70

.01
lO
~--------~--------~--------~------~~------~
10
1
10 lO 10 10
N
FIGURE 6. AXIAL PERMANENT STRAIN (£p) VERSUS NUMBER OF LOAD
REPETITIONS (N) FOR SAMPLES CONSOLIDATED UNDER
CONFINING PRESSURE OF 25 PSI, SITE #2.
 \32

changes as the confining pressure and the cyclic stress 100

ratio change. This may suggest that the sample behavior
and the characteristics of permanent strain beyond cycle
number 100 are influenced by the first 100 cycles and
perhaps the first few cycles. This finding was expected
because as the number of load repetitions increases the

. ... . •. .--
sample stiffness increases and thus the rate of plastic 10

-. --
strain decreases. Consequently, if it is desired to study
and compare the axial permanent strain of different
samples or different soils, all specimens should be con-
ditioned at the same stress level and to the same number
of load cycles. Once the sample conditioning is reached
the desired cyclic stress could be applied and the test
w

1.0
~ - 03 25 psi

.
~ --
could be commenced. The data through the conditioning
cycles ,however, should be collected and reported regard-
less of its irregularities. Further, sample conditioning 03 5 psi
is a part of the sample stress history which may have
substantial effects on the characteristics of the sample
plastic strain. 0.1 ~
As it was expected, the higher the cyclic stress diffe- 10°
rence the higher the permanent strain and for the same N
stress ratio the higher the confining pressure the higher FIGURE 9. AXIAL PERMANENT STRAIN (q) VERSUS NUMBER OF LOAD
the permanent strain. These results are shawn in Figures REPETITIONS (N) FOR od/03 = 2.0, SITE #3.
7 and 8 and 9 respectively. It should be noted at this
time that each data point in Figure 7 represents an in-
dependent sample tested to 30,000 load repetitions.
10~---------.---------r---------,--------~---------,
Further examinations of Figures 3,4,5,6,8 and 9 indicate
(not very conclusively) that the angle of intersection
(B) increases as the stress ratio decreases. The ultimate
value of this angle is 180° for a stress ratio of l. For
this case (if the sample is relatively undisturbed) the 25
slope of the single straight line will be related to the
compression index of the sample.

CONCLUSIONS

On the basis of the test results, the following conclu-

sions were drawn:
O.ll------1----:;;;:...J..~:::~~~t::~==-+------l 5

a) all soil samples prepared for cyclic loading tests

should not be subjected to any sample conditioning
prior to testing. If sample conditioning is desired
however, then all the test data (starting from
cycle number one) should be collected, analysed
and reported.
b) The cumulative plastic strain of clay samples is N
related to the number of load repetitions by an
FIGURE 8. AXIAL PERMANENT STRAIN (Ep) VERSUS NUMBER OF
exponential function of the form of equation l.
The parameters of the equation are related to the LCAD REPETITIONS (N) FOR 0d/a 3 = 1.0, SITE 3.
sample and test variables.

-80 REFERENCES
·rl
Ul
8 l) Barksdale, R.D. (1972), "Laboratory Evaluation of Rutt-
2') ing in Base Course Materials", Proceedings of the Third
<lJ
u International Conference on Structural Design of As-
60
.,c
<lJ phalt Pavements, London, England, p. 161-174.
Ill
4-1
4-1
·rl
2) Campen, W.C., and Smith, J.R.,"Use of Load Test in the
Q
40 dasign of Flexible Pavements", Special Technical Pub-
lication No. 79, ASTM.

3) Cannon, R.H. Jr. (l976),"Dynamic of Physical System",

McGraw Hill Book Company, New York.

4) Casagrande,A. and Shannon, W.L., (1948), "Research on

Stress-Deformation and Strength Characteristics of Soils
and Soft Rocks under Transient Loading", Publication
8 10 of Harvard Graduate School, Engineering Soil Mechanics,
Series 31.
Axial Permanent Strain (%)
FIGURE 7. PRINCIPAL STRESS DIFFERENCE VERSUS 5) Thiers, G.R.,and Seed, H.B.,"Cyclic Stress-Strain Cha-
AXIAL PERMANENT STRAIN. racteristics of Clay", Journal of the Soil Mechanics
and Foundation Division, SM2, p. 555-569.
 133

6) Hittle, J.E., and Goetz, W.H.,(l947) ,"A Cyclic Load

Test Procedure", Special Technical Publication No. 79,
ASTM.

7) Hufferd, W.L., and Lai, J.S.,(l978),"Analysis of N-

layered Viscoelastic Pavement Systems", Final Report
to the Fderal Highway Administration, FHWA-RD-78.22.

8) Lentz, R.W., and Baladi, G.Y.,(l980)," Simplified pro-

cedure to Characterize Permanent Strain in Sand sub-
jected to Cyclic Loading", Proceedings of the Inter-
national Symposium on Soils under Cyclic and Transient
Loadings, Swansea, England.

9) Mcleod, N.W., (1947) ,"A Canadian Investigation of Load

Testing Applied to Pavement Design", Special Technical
Publication No. 79, ASTM.

lO)Mitry, F.G.,(l964),"Determination of the Modulus of

Resilient Deformation of Untreated Base Course Mate-
rials", Ph.D. Thesis, University of california,
Berkely.

ll)Annaki, M., and Lee, K.L.,(l977) ,"Equivalent Uniform

Cycle Concept for Soil Dynamics", Journal of the Geo-
technical Engineering Division, GT6, Vol. 103.

l2)Morgan, J.R.,(l966)," The Response of Granular Mate-

rials to Repeated Loading", Proceedings of the Third
International Conference on Structural Design of As-
phalt Pavements, London, England.

l3)Parmelee, R.A., Penzien, J., Schaeffey,C., Seed, H.B.

and Theirs, G.R.,(l964) ,"Seismic Effect on Structures
Supported on Piles Extending Through Deep Sensitive
Clays", Report to California State Division of H.W.
University of California, Berkeley, California.

l4)Penzien,J., Schaeffey,C., and Parmelee,R.,(l964),

"Seismic Analysis of Bridges on Long Piles", Journal
of the Engineering Mechanics Division, ASCE, Vol. 90,
EM3.

lS)Seed, H.B.,and McNeill, R.L. (l956),"A Comparative

Study of Soil Deformations in Normal Compression and
Repeated Loading Tests", Highway Research Board,
Bulletin 141.

l6)Seed, H.B., and McNeill ,R.L., (l957),"Soil Deformation

under Repeated Stress Applications", ASTM Special Tech-
nical Publication 232.

l7)Timoshenko,L.P., (1953) ,"History of Strength of Mate-

rials", McGraw Hill Book Company, New York.

l8)Luo, w.K.,(l973) ,"The Characteristics of Soils sub-

jected to Repeated Loads and their Applications to
Engineering Practice", Japanese Society of Soil Me-
chanics and Foundation Engineering, Vol. 13, No,l.

l9)Yoder, E.J., and Witczak, M.W.,(l975) ,"Principles of

Pavements Design", Second Edition, John Wiley and Son
Publishing Company, New York.