Japanese Geotechnical Society Special Publication The 15th Asian Regional Conference on

Soil Mechanics and Geotechnical Engineering

Effect of wall flexibility on the dynamic earth pressure for cantilevered retaining wall

S. B. Jo i), J. G. Ha i) and D. S. Kim ii)

i) Ph.D Student, Department of Civil and Environmental Engineering, KAIST, 291 Daehak-ro, Daejeon 305-701, Korea
ii) Professor, Department of Civil and Environmental Engineering, KAIST, 291 Daehak-ro, Daejeon 305-701, Korea

ABSTRACT

This paper describes an experimental results for effect of wall flexibility on the dynamic earth pressure for
cantilevered retaining wall. Three centrifuge tests were performed on reduced models with different flexibility. The
retaining walls were fixed on the container to clear boundary condition except wall flexibility. This condition was far
from assumption of limit state analysis, which allow sufficient yield displacement in the backfill soil such as
Mononobe-Okabe method, but this paper focused on the change of dynamic earth pressure with wall flexibility. The
phase difference between wall and soil reduced the dynamic earth pressure and dynamic moment has a same phase
with wall inertia for flexible retaining wall. Position of dynamic thrust is H/3 for ordinary retaining wall and upper
bound is 0.5H. Seed and Whitman method overestimate the dynamic thrust when surface PGA was used as a kh, and
using wall base PGA as a kh agrees with Seed and Whitman estimation.

Keywords: retaining wall, earth pressure, dynamic centrifuge, mononobe-okabe, flexibility

1. INTRODUCTION the motions of the retaining wall and backfill, and the
distribution of earth pressure is not triangular distribution
The seismically induced lateral earth pressure on
and also changes with time. Recently Al-Atik and Sitar
retaining structures was initially suggested by Okabe
(2010), Mikola (2012) and Jo et al. (2014) showed that
(1926) and Mononobe and Matsuo (1929) in the 1920s,
M-O method overestimate the dynamic earth pressure for
herein designated the MO method, which was a
non-displacement and yielding retaining walls by dynamic
pseudostatic extension of the classical Coulombs
centrifuge tests. They concluded that the point of dynamic
limit-equilibrium analysis (Coulomb 1776). Inertial force
thrust is at H/3 and M-O overestimate dynamic earth pressure.
of the soil wedge behind wall was considered as an added
However, they used surface peak ground acceleration
static force. For practical purposes, Seed and Whitman
(PGA) of backfill as kh. Surface PGA is amplified
(1970) proposed simplified modification (S-W) that
compared to wall base PGA and dynamic earth pressure
separates the dynamic earth pressure to estimate
can be overestimated due to amplified PGA. They also
coefficient of dynamic earth pressure approximately. Due
proposed triangle shape distribution for dynamic earth
to its simplicity and familiarity of engineers, they are
pressure, not inverted triangle shape suggested by Seed
widely used as global standard in practice.
and Whitman. Even though the MO method is a simple
They proposed the active seismic earth pressure
and powerful tool for evaluating the dynamic earth
coefficient (KAE) and Seed and Whitman suggested
pressure, it only depends on kh and there is no clear
distribution of dynamic earth pressure as inverted triangle
definition for location of kh.
shape. To determine dynamic earth pressure, seismic
In this study, three dynamic centrifuge experiments
coefficient (kh) was considered, however they did not
were performed to evaluate the appropriate location for kh
mention where the reliable location of seismic coefficient
and offer a basic understanding about the dynamic earth
is for evaluating inertial force of backfill soil. Previous
pressure of an inverted T-shape retaining wall with
experimental research using 1g shaking table and
differing stiffness. To clear boundary condition and
centrifuge focused on M-O method whether or not it
eliminate other variables except the stiffness, retaining
provides the adequate total active earth pressure for a
walls were fixed on the container. Results obtained from
yielding retaining wall (Matsuo and Ohara 1960, Ortiz et
this experimental study were compared to those under the
al. 1983, Bolton and Steedman 1982, Steedman and Zeng
hypothetical conditions of the MO and S-W method.
1990). Nakamura (2006) performed a centrifuge test for
This study proposes how to decide the kh for pseudo-static
gravity retaining wall and investigated the discrepancy
analysis and recommendations for distribution of dynamic
between actual seismic behavior and the pseudo-static
earth pressure.
theoretical analysis. He shows a phase difference between

http://doi.org/10.3208/jgssp.KOR-24 911
2 EXPERIMENTAL PROCEDURE 2.2 Model preparation
The sand used in this experiment was silica sand,
Dynamic centrifuge tests were performed at the
which is produced by hammer crusher process. The soil
KOCED Geo-Centrifuge Test Centre in KAIST. Detail
was prepared using dry pluviation. The relative density
of technical specifications for the centrifuge can be
was 85%. Internal friction angle was 45.3 in triaxial
found in Kim et al. (2013a,b) and the testing centrifugal
test and soil-wall inter friction angle was 29.9 in direct
acceleration was 20gc. Earthquake events consisted of
shear test. The sequence of the model construction is as
the motions recorded during Hachinohe and Ofunato
follows. (1) The retaining wall fixed on the container
earthquakes in Japan and the frequency were filtered
base. (2) After placement of each layer, instruments
from 1 to 15 Hz in prototype. The input ground motion
were placed at their designed locations. (3) Sand was
parameters are presented in Table 1 and peak ground
pluviated up to the surface height again. (4) Finally, the
acceleration are increasing up to about 0.25g by stage
front soil of the wall was removed by a vacuum cleaner.
tests. All results are presented with respect to prototype
Plastic sheets were attached using industrial grease to
units according to scaling laws (Schofield 1980) unless
prevent sand from passing through between wall and
otherwise stated.
side walls of the container.
Table 1. Input ground motion parameters.
3 EXPERIMENTAL RESULTS
Earthquake Tm (s) Tp (s) To (s)
Ofunato 0.28 0.16 0.36 3.1 Phase difference
Hachinohe 0.60 0.36 0.51 In the case of flexible retaining wall, dynamic
Note: Tm: mean period; Tp: predominant period; To: smoothed moment has an opposite phase with dynamic thrust (i.e.
spectral predominant period (Rathje et al., 1998). summation of dynamic earth pressure) and similar
phase with wall inertia (Fig. 2). It means that dynamic
2.1 Model Structures and Instrumentation moment was caused by wall inertia, not earth pressure
The inverted T-shape retaining wall models were made during earthquake. When negative moment was
up of aluminum alloy (aluminum 6061; E=68.9 GPa) occurred (i.e. wall towards passive direction) dynamic
considering centrifuge scaling law. The three thickness of earth pressure was increasing, it can be explained by
stem were designed considering Korean standard retaining phase difference between wall and backfill soil in
wall. The detailed dimensions of test models are shown in
deformation.
Fig. 1 and Table 2. Strain gauges were attached over the
height of each wall. Earth pressure transducer and
accelerometer were also attached on the wall to measure
acceleration and dynamic earth pressure. Accelerometer
were installed along the depth in the soil to measure the
amplification and variation with distance from the wall.
The acceleration and the bending moments are positive in
active directions. The earthquake was orthogonally
applied to the retaining wall model.

Table 2. Dimensions of various flexibility walls.

Fig. 2. Phase difference of Flexible wall.

Fig. 3. Phase difference of Rigid wall

912
 As shown in Fig. 3, however, the phase between 3.3 Seismic Coefficient (kh)
moment and thrust were similar for rigid wall case. M-O and S-W method provide a coefficient of
When the negative moment was occurred, dynamic dynamic earth pressure (Kae) and it was calculated by
thrust is close to zero. It means that backfill soil is also only kh. However, they did not clearly mentioned
moving backwards and occurring detachment. The location or appropriate seismic coefficient in the
deformation of flexible wall was occurred by wall backfill, causing a confusion in practice and research
inertia, not dynamic thrust and its phase difference area. Especially, Kae calculated by M-O was rapidly
prevent to generate sufficient dynamic earth pressure increasing if kh is above 0.5g. To evaluate the criteria
during earthquake. for appropriate kh, measured dynamic thrust converted
to Kae and they were plotted with estimation of S-W
3.2 Distribution of Dynamic Earth Pressure
method using wall base PGA and surface PGA for kh.
In this study, the results were compared at three
Fig. 5 show the results of normal wall when Max.
critical states: (1) when the dynamic earth pressure was
PGAff and Max. Pressure were occurred. If the surface
maximum along the depth (Max. Pressure); (2) when
PGA was used for kh, S-W method overestimate the
the maximum moment occurred in the wall (Max.
dynamic thrust at each critical states. Upper bound for
Moment); and (3) when the maximum passive
Kae were 30% and 60% for Max. PGAff and Max.
acceleration occurred at the free-field surface (Max.
Pressure, respectively. If the wall base PGA was used
PGAff).
for kh, the results of Max. PGAff were similar to Fig.
M-O method uses an H/3 for total thrust (i.e.
5(a). Because wall and soil were detached when the
static+dynamic) and S-W method use a 0.6H for
Max. PGAff were occurred for flexible and normal wall
dynamic thrust. Fig. 4 shows the results of thrust height
and those dynamic thrust do not have correlation with
for dynamic increment. Results of rigid wall ranged
PGA intensity.
from H/3 and 0.6H regardless of critical states where as
Fig. 6 show the results of rigid wall and Kae were
those of flexible wall were below H/3, which means
almost same upper bound regardless of critical states.
triangle shape of dynamic earth pressure. In the case of
Surface PGA provide appropriate Kae with 80% of
Max. Pressure, most results were below 0.5H and the
S-W estimation. However, wall base PGA provide the
results of other cases were below H/3. Although the test
well matched Kae up to 0.2g and underestimate for
models were fixed on the container and it can cause a
strong earthquake. Wood (1973) develop solutions for
large and high position of dynamic thrust, test results
an elastic soil stratum on a rigid base with a rigid wall
show that H/3 is appropriate for application point of the
and dynamic thrust of perfectly rigid wall was likely to
dynamic thrust in the normal stiffness wall and 0.5H is
be greater than twice that estimated by M-O method.
for rigid wall.

(a) Estimation using Surface PGA in passive direction

(b) Estimation using Wall Base PGA in passive direction

913
 REFERENCES
1) Al Atik, L. and Sitar, N. (2010): Seismic earth pressures on
cantilever retaining structures, Journal of geotechnical and
geoenvironmental engineering, 136(10), 1324-1333.
2) Bolton, M.D. and Steedman, R.S. (1982): Centrifugal
testing of micro-concrete retaining walls subject to base
shaking, Proceedings of Conference on Soil Dynamics and
Earthquake Engineering, Balkema, Rotterdam, Netherlands,
311329.
3) Coulomb C.A. (1776): Essai sur une application des regles
des maximis et minimis a quelques problemes de statique
relatifs a l'architecture, Memoires de l'Academie Royale pres
(a) Estimation using Surface PGA in passive direction Divers Savants, Paris, France, 7.
4) Jo, S.B., Ha, J.G., Yoo, M., Choo, Y.W. and Kim, D.S.
(2014): Seismic behavior of an inverted T-shape flexible
retaining wall via dynamic centrifuge tests, Bulletin of
Earthquake Engineering, 12(2), 961980.
5) Kim, D.S., Kim, N.R., Choo, Y.W. and Cho, G.C. (2013a): A
newly developed state-of-the-art geotechnical centrifuge in
Korea, KSCE Journal of Civil Engineering, 17(1), 7784.
6) Kim, D.S., Lee, S.H., Choo, Y.W. and Perdriat, J. (2013b):
Self-balanced earthquake simulator on centrifuge and
dynamic performance verification, KSCE Journal of Civil
Engineering, 17(4), 651661.
7) Matsuo, M. and Ohara, S. (1960): Lateral earth pressures
and stability of quay walls during earthquakes, Proceedings of
(b) Estimation using Wall Base PGA in passive direction 2nd World Conference on Earthquake Engineering, Tokyo,
Fig. 6. Coefficient of dynamic earth pressure for rigid wall Japan.
8) Mikola, R.G. (2012): Seismic Earth Pressures on Retaining
Structures and Basement Walls in Cohesionless Soils, Ph.D.
4 CONCLUSIONS Thesis, University of California, Berkeley, CA, USA.
9) Mononobe, N. and Matsuo, H. (1929): On the determination
This study aims to simulate three retaining walls with
of earth pressures during earthquakes, Proceedings of World
different stiffness to qualitatively present the seismic Engineering Congress, 179187.
behaviors of the inverted T-shape cantilever retaining 10) Nakamura, S. (2006): Reexamination of Mononobe-Okabe
walls using the dynamic centrifuge. These retaining theory of gravity retaining walls using centrifuge model tests,
walls were fixed on the container to make a clear Soils and foundations, 46(2), 135-146.
boundary condition. The experimental results lead to 11) Okabe, S. (1926): General theory of earth pressure and
several important conclusions. seismic stability of retaining wall and dam, Journal of the
Japanese Society of Civil Engineering, 12(1), 123-134.
12) Ortiz, L.A., Scott, R.F. and Lee, J. (1983): Dynamic
1) Dynamic moment were large depend on wall centrifuge testing of a cantilever retaining wall, Earthquake
inertia, not dynamic thrust for flexible wall, and Engineering and Structural Dynamics, 11, 251268.
rigid wall has a same phase between wall and 13) Rathje, E.M., Abrahamson, N.A. and Bray, J.D. (1998):
soil. Simplified frequency content estimates of earthquake ground
motions, Journal of geotechnical and geoenvironmental
engineering, 124(2), 150-159
2) Dynamic thrust point were applied at H/3 above 14) Seed, H.B. and Whitman, R.V. (1970): Design of earth
the wall and distribution of dynamic earth retaining structures for dynamic loads, In ASCE Specialty
pressure has a triangle shape not inverted Conference on Lateral Stresses in the Ground and Design of
triangle shape for normal wall. And 0.5H is Earth-Retaining Structures, Reston, VA, 103147.
upper bound for rigid wall. 15) Schofield, A.N. (1980): Cambridge geotechnical centrifuge
operations, 20th Rankine Lecture, Gotechnique, 30(3),
227268.
3) S-W estimation overestimate the dynamic thrust 16) Steedman, R. S., & Zeng, X. (1990): The influence of phase
when surface PGA was used as a kh, and using on the calculation of pseudo-static earth pressure on a
wall base PGA as a kh agree with estimation. retaining wall, Geotechnique, 40(1), 103-112.
17) Wood, JH. (1973): Earthquake induced soil pressures on
ACKNOWLEDGEMENTS structures, PhD Thesis, California Institute of Technology,
Pasadena, CA.
This research was supported by a grant (11
Technology Innovation D02) from Construction
Technology Innovation Program funded by Ministry of
Land, Transport and Maritime Affairs of Korean
government.

914