ASIAN JOURNAL OF CIVIL ENGINEERING (BUILDING AND HOUSING) VOL. 11, NO.

5 (2010)
PAGES 575-583

EXPERIMENTAL ANALYSIS OF THE DYNAMIC STRESS

DISTRIBUTION AT THE SOIL FOUNDATION INTERFACE

A. Boumekik, S. Belhadj-Mostefa*, and F. Meribout

Department of Civil Engineering, University Mentouri of Constantine, 25000, Algeria

ABSTRACT

This paper presents the dynamic stresses within the soil-foundation interface zone measured
by an experimental process. The cubic foundation prototype is subjected to a cyclic loading
directly generated by the testing machine at the foundation center. The stress enhancement
induced by the dynamic loading may lead to a partial confinement with an eventuality of
instability risks or resistance loss; showed by a particle rearrangement. The tests have been
conducted on two soil samples namely dense and medium dense sand. Experimental
displacements, measured at the soil foundation interface are represented, as a function of the
number of cycles, for a constant stress level. The dynamic behaviors of used sample soils
are analyzed and some concluded remarks are presented.

Keywords: Interaction soil-structure; dynamic loading; stresses; displacement; density

1. INTRODUCTION

The generated soil vibrations may be very important, causing by the way great source of
annoyance for structures such as fatigue, and settlement phenomena; the interaction soil-
structure will be important phenomena to take into account while analyzing the real behavior
of a structure.
Many seismic table experimentations have been conducted on structures supported by
surface foundation [1-2]. These tests have shown that the foundation uplift play an insulating
role reducing the base forces.
Special tests much more appropriated are recommended for a consistent force case
(centrifuge test, cyclic triaxial test, etc.) [3-7]. However, these laboratory tests require the
confinement and the soil consolidation in order to recreate real site conditions. Within the
laboratory test category, the dynamic centrifuge tests have been currently used to study
complex boundary problems related to the soil structure interaction [8-9]. Also the effects of
foundation mass, foundation size, foundation shape, foundation embedment depth, and soil
density on load deformation have been explored by this technique. Experimental data collected
from these tests are used to compare the numerical predictions from different finite element
codes. The effect of the foundation mass, shape, soil density and the deformation load have
*
E-mail address of the corresponding author: bmslimane1@yahoo.fr (S. Belhadj-Mostefa)
576 S. Belhadj-Mostefa, A. Boumekik and F. Meribout

explicitly been shown in order to determine their impact on the dynamic response [10-13].
This article presents the experimental determination of the dynamic stresses distribution
generated at the soil foundation, measured during the test. The stresses are obtained directly
from the dynamic displacement collected trough a displacement captor, Figure 1. The
experimental approach used herein is similar to the simple case of foundation disposed on
homogenous elastic soil as it is specified by the European program framework (TRISEE 98)
[14] for dynamic cyclic tests performed on a superficial square foundation.

P(t)
1
2
3

6 7

Figure 1. Testing model for measurement of dynamic displacement

1: Repartition steel plate
2: Spring
3: Foundation prototype
4: Measuring rod with guiding system
5: Soil sample
6: Displacement transducer
7: Video recorder

The test will establish the comparison with cyclic tests conducted on dense (SSD) and
medium dense (SSMD) sands [15] under different vertical stresses.
The test series for the both sand types (index of density Id = 0.7 and Id = 0.9 ) have been
carried out in three points of the interface zone (Figure 2 ), (x=0 , y= 0) ; (x=3a/2 ,y=0 ) and
( x= −3a/2 , y= −3a/2 ) in order to characterize the soil dynamic response generated by the
foundation prototype under the cyclic loading for the following loading conditions:
amplitude of the excitation load P(t) = 1000 N, period = 0.6 s, contact pressure = 45 kPa.
emax  e
Id  (1)
emax  emin
 EXPERIMENTAL ANALYSIS OF THE DYNAMIC STRESS DISTRIBUTION... 577

e: index of voids in the natural state

emax : index of voids in the lightest state of the material
emin : index of voids of the material in its most compact state

3a
x

3a
Figure 2. Measurement points at the soil foundation interface

According to Brumund and Leonards [4] the allowable load value qu is taken within the
load interval (0.3, 0.5) qu determined from the engineering practice bases. Making sure that
the contact pressures remain elastic and proportional to the vertical displacements, the
ultimate vertical load capacity of the foundation may be estimated as follows:

w
qu  (2)
As
w: foundation weight
As: foundation area

2. EXPERIMENTAL MODEL

The harmonic load is generated directly from the vertical cyclic motion of the axles of
machine. The machine is able to simulate harmonic loading with different amplitudes and
various operating frequencies. The harmonic motion is transmitted to the soil sample
through the concrete foundation prototype, placed on the upper face of the soil sample. The
soil samples used in this study, are cubic shape with cross area As (0,6 x 0,6 m2) and a
height hs = 0,60 m , giving a soil volume Vs=0,205m3. The foundation prototype is square
shaped having an area 0.15x 0.15m2 and a thickness of 0,05m. Both foundation prototype
and soil sample are correctly centered with respect to the vertical axis, before testing.
The measurement collect is based on a simple experimental process (Figure1) that
consists of reading the resulting displacement established by the video recording; so the
linear relation between the forces and displacements is:

F=KU (3)
578 S. Belhadj-Mostefa, A. Boumekik and F. Meribout

K: soil stiffness.
U: displacement collected by the displacement captor.

The measurement of the dynamic displacement resulting from an excitation load F with a
corresponding operating frequency, recorded by video is used to establish the time related
variation of the dynamic stresses in the elastic field as follows:

KU
 (4)
As

σ : vertical stress under the foundation.

AS : foundation area:

The foundation is subdivided into nine sectors (Figure 3) according to the number of the
disposed displacement captors, every sector having an area 0,05x0,05 m2 .The measure of
the vertical stresses will be respectively the same at the corners , at the edges and at the
center .
3a

1.1 1.2 1.3

3a

3.3

Figure 3. Subdivision scheme of the foundation prototype

The conservation of the balance with the impact load induced by the machine is obtained
by a uniform distribution of vertical stresses under the prototype foundation; this state may
be expressed as:

F     .ds (5)

The impact load can be defined by a discrete extrapolation of the vertical stress at each
sector through the form as follows:
F   i si (6)

σi: stress within the sector i,

si: contact surface at the level of sector i
 EXPERIMENTAL ANALYSIS OF THE DYNAMIC STRESS DISTRIBUTION... 579

According to a uniform vertical stress distribution on every sector at each loading cycle
within the interface zone, that is to say that the distribution will be looked along the center or
along the edge as presented in (Figure 4), we notice that the value in the center is each time
superior.
a a a

Figure 4. Vertical stress repartition

3. CHARACTERISTICS OF THE SOIL SAMPLES

The soil samples come from north east of Algeria (Constantine) at quarry of Oued Zhor
for dense sand and quarry of Chaabet El Madbouh for medium sand, the essential physical
soil properties have been established at Fondasoil Laboratory (Soil Mechanic Laboratory at
Constantine) are presented in the Table 1.

Table 1: Geotechnical soil characteristics

Poisson’
Type Density Water content % Void index e
coefficient

Dense sand 2.09 0.33 19 0.51

Medium
1.99 0.37 25 0.67
dense sand

3. RESULTS

The Figure 5 shows the vertical displacement evolution for the two types of sand at the
foundation center, as a function of the number of cycles, under a contact pressure of 45 kPa.
The displacement captors disposed in the interface zone at the distance hs/10 from the
contact surface revealed that the displacement in the medium dense sand case is more
important than the displacement in the dense sand case, and the established curves show
approximately the same standard deviation as those elaborated by Bian and Nishimura [15]
for the same sand types and a loading value of 100 kPa. We also notice that the displacement
reduces with the increase of the density index (Id), Table 2. Our dense sand stands out
rapidly but with little displacement variations because of an important density.
580 S. Belhadj-Mostefa, A. Boumekik and F. Meribout

Figure 5. Vertical displacement at the foundation center versus the number of cycles (SSD:
dense dry sand; SSMD: medium dense dry sand)

Table 2: Curve Identification

Authors Series Density index Soil nature Loads

BIAN SSMD Id = 0.6 Medium dense sand 100 kPa

BIAN SSD Id = 0.9 Dense sand 100 kPa

Present study SSMD Id = 0.7 Medium dense sand 45 kPa

Present study SSD Id = 0.95 Dense sand 45 kPa

4. STRESS ANALYSIS

The Figures 6 and 7 show the vertical stress dimensionless variations under dynamic
loading in the interface zone. These stresses have been estimated from the experimental
measurement data in the interface zone for the previously specified points (Figure 2) for our
two types of sand.
 EXPERIMENTAL ANALYSIS OF THE DYNAMIC STRESS DISTRIBUTION... 581

Figure 6. Vertical stress variation versus the number of cycles for the medium dense sand.

Figure 7. Vertical stress variation versus the number of cycles for the dense sand

We notice from Figures 6 and 7 when the SSMD is considered that the centre stresses
increase considerably for the nine first cycles (500pa) and stabilize at the end (18th cycle) .
In the same manner for the edge stresses where the increase is very important for the three
first cycles (200Pa) reaching their stabilization rapidly. However, the corner stresses
diminish for the three first cycles before to reach their stabilization. It is also shown when
the SMD (dense sand) case is considered that the axial stress diminishes in a non negligible
manner until the 6th cycle (400Pa) before to stabilize. The corner stress exhibits the same
behavior with much more variation (200Pa) for the first six cycles before to stabilize. A
582 S. Belhadj-Mostefa, A. Boumekik and F. Meribout

stress enhancement is registered at the corner until the 6th cycle (200Pa) before the stability
is reached. The dense sand would have a tendency to transmit the stress excess towards the
foundation corner.

5. CONCLUSION

Laboratory tests presented within this paper are used to estimate soil dynamic stress
induced by a vibrating foundation prototype and thus for three specific points of the
foundation–soil interface zone. The testing prototype used gives satisfying results for
simulating a superficial foundation behavior under a cyclic and dynamic loading. The
influence of soil density has been studied and two types of sand have been considered (dense
and medium dense).The obtained results at different locations of the foundation prototype
show that an increase in the density for the medium dense sand due to the particle
retightening at the central zone level leading to an increase in its density. In the dense sand
case the axial overstresses would have the tendency to be transmitted to the corners because
of the initial confinement.
The dynamic settlement test gives the possibility to envisage specific tests in the design
process of industrial installations with reduced costs, enabling a better understanding of the
future soil behavior under vibrating machines. Besides the classic tests, the presented test
offer more facilities than in situ tests. Nevertheless, much more research work is needed
concerning the apparatus dimensions and its sophistication.

REFERENCES

1. Svinkin MR, Zhuchkova AY. Dynamic tests of foundations on type – I slump-prone

soils, Soil Mechanics and Foundation, Engineering, No. 1, 9(1972)33-6.
2. Combescure D, Chaudat Th. ICONS European program seismic tests on r/c walls with
uplift; CAMUS IV specimen – Individual Study, ICONS project Rapport,
SEMT/EMSI/RT/00-27/4, CEA, Direction des réacteurs nucleaires, Department of
Mechanic and Technology, Paris, France, 2000.
3. Lysme J, Richard FE. Dynamic response of footings to vertical loading, Journal of the
Soil Mechanics and Foundations Division ASCE, 92(1966)65-91
4. Brumund WF, Leonards GA. Experimental study of static and dynamic friction between
sand and typical construction materials, Journal of Testing and Evaluation, No. 2,
1(1973)162-5.
5. Tokimatsu K, Seed H.B. Evaluation of settlements in sands due to Earthquake shaking,
Journal of geotechnical Engineering Division ASCE, No. 8, 113(1987)861-78.
6. Adam D, Kopf Adam C. The dynamic load plate test with the light falling weight
device, Der Bauingenieur, 79(2004)1185-92.
7. Boumekik A, Hijeb M. The behavior of undrained saturated clay under cyclic loading,
Journal of Engineering and Applied Sciences, No. 4, 1(2006)375-82.
8. Elgamal AW, Dobry R. Dynamic and cyclic centrifuge model testing, Geotechnical
News, No. 3, 9(1991)31-3.
 EXPERIMENTAL ANALYSIS OF THE DYNAMIC STRESS DISTRIBUTION... 583

9. Gazetas G. Foundations Vibrations, Foundation Engineering Handbook, 2nd edition,

Vas Nostrand, Reinhold, Edited by Hsai-Yang, Chapter 15, 1991.
10. Boumekik A, Sahal N. Dynamic stress generated by vertical vibrations of a surface
footing on a soil layer. Proceedings of the 7th International Conference on Computing
in Civil and Building Engineering, (ICCC BE-VII), Soil Dynamics Vol. 3, 1997, pp.
1917-1924.
11. Wolf JP. Foundation Analysis Using Simple Physical Models, Prentice Hall, Englewood
Cliffs N.J. (1994).
12. Boumekik A, Messast S. Dynamic response of a pile foundation due to the transmission
of vertical vibrations from the adjacent soil, European Journal of Scientific Research,
No. 4, 14(2006)569-77.
13. Vrettos C. Vertical and rocking impedances for rigid rectangular foundation,
Earthquake Engineering and Structural Dynamics, 28(1999)1525-40.
14. TRISEE. 3D Site Effects and Soil-Foundation Interaction in Earthquake and Vibration
Risk Evaluation. Part 4 “Large-squale geotechnical experiments on soil-foundation
interaction “European Commission, Directorate General XII for Science, Res. and
Dev., 1998.
15. Bian H, Nishimura T. Influence de la saturation sur le comportement cyclique d’un
sable, Laboratoire de mécanique de Lille (UMR 8107), Ashikaga Institute of
Technology, Tochigi, Japan, 2007.