Sunteți pe pagina 1din 8

Available online at www.sciencedirect.

com

ScienceDirect
Procedia Engineering 172 (2017) 218 225

Modern Building Materials, Structures and Techniques, MBMST 2016

Experimental and Numerical Analysis of Direct Shear Test


Neringa Dirgliena*, arnas Skuodisa, Andrius Griguseviiusb
a
Department of Geotechnical Engineering,
b
Department of Structural Mechanic, Vilnius Gediminas Technical University, Saultekio al. 11, Vilnius LT-10223, Lithuania

Abstract

An experimental and numerical investigations of the direct shear test have been performed under constant vertical stress (q =
const) and constant sample volume (h = const). During the determination of soil shear strength in a laboratory by different test
methods soil is loaded in a different way. This fact has an influence on stress-strain distribution in the sample. The finite-element
method analysis also shows that during direct shear tests distribution of stress and strain in the sample is non-uniform. If we
know a real distribution of stress and strain in the sample, it is possible to determine the soil shear strength and deformation
parameters in a more precise way or to rate the influence of different factors on soil properties.
2017 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
2016 The Authors. Published by Elsevier Ltd.
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
Peer-review under responsibility of the organizing committee of MBMST 2016.
Peer-review under responsibility of the organizing committee of MBMST 2016
Keywords: direct shear test; finite-element method, constant vertical stress; soil constant volume; stress-strain distribution.

1. Introduction

Shear strength is the principal engineering property of soil, which controls the stability of a soil mass under
structural loads. Accurate determination of the soil shear strength parameters (angle of internal friction and
cohesion) is a major interest in the design of different geotechnical structures. These parameters can be determined
either in the laboratory or in the site. The triaxial compression and direct shear tests are the most common tests for
determining the angle of internal friction and cohesion values in the laboratory. Special care must be taken to
establish loading condition actually existing in the ground and to duplicate this condition in the laboratory.

* Corresponding author. Tel.: +370-5-2745220; fax: +370-5-270 0112.


E-mail address: Neringa.Dirgeliene@vgtu.lt

1877-7058 2017 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
Peer-review under responsibility of the organizing committee of MBMST 2016
doi:10.1016/j.proeng.2017.02.052
21 Neringa
Neringa
Dirglien
Dirglien
et al.et/ Procedia
al. / Procedia
Engineering
Engineering
172 (2017)
172 (2017)
218 218
225 225 21
9 9

Boundary conditions are not distinct when analysing soil samples by direct shear test [1]. Not all vertical stress
apllied on top of shear box specimen is transmitted on soil. It is not obvious the regularity of change of normal
stresses on a shear plane [26]. If a direct shear test is analysed, the stress distribution depends on the following:
manner of vertical load transmission, position of the mobile part of shear ring, horizontal displacement of the mobile
part of the ring. Some factors are not evaluated during the results interpretation, for example, friction between soil
and device metal parts [3,710].
The numerical methods enable the determination of material parameters that would have been difficult to
measure in the experimental study [4,5,11].
The aim of the current investigation is to analyse stress-strain distribution in the sample during direct shear
testing depending on the manner of the vertical load transmission. The experimental direct shear test on sand have
been performed under constant vertical stress (q = const) and constant sample volume (h = const). Experiments
were simulated performing the numerical analysis by finite element method program COSMOS/M.

2. Laboratory experiments

The air-dry sand of the Baltic Sea coastal area was used to perform the experiments for determining the shear
strength parameters. The average density of particles (s) value of marine sands is 2.66 Mg/m and ranges from 2.65
to 2.67 Mg/m [9].
The vertical stress magnitudes of 100, 200, 300 kPa have been applied. The soil is sheared under the constant
horizontal displacement velocity of 0,5 mm/min until the horizontal deformation reaches the limit of 9 mm. The
shear test is performed under two different cases (methods), namely: when constant vertical stress (q = const.) is
applied; when constant sample volume (h = const.) is applied.
The shear tests have been performed for loose sand (density = 1,491 g/cm3). The peak soil shearing strength has
been determined according to the maximum ratio of tangential and normal stresses, id est. according / = max [9].

3. Analysis of obtained results

The characteristic investigated sand shear graphs for loose soil (Figs. 12) have been processed for different
testing methods, id est. for q = const and h = const [9].
In the case of initially loose sand when q = const there is no significant particle interlocking to be overcome and
the tangential stress increases gradually to an ultimate value without a prior peak, accompanied by a decrease in
volume (Fig. 1 a). In the case h = const shear stress increases and decreases immediately and after remains almost
steady (Fig. 1 b).
a b

Fig. 1. Tangential stress at the applied normal stress when: a) q = const; b) h = const.

From Fig. 2 a one can find that the applied vertical stress of magnitudes of 100, 200, 300 kPa remained constant
during all the test time. Aiming to keep the constant sample volume at the beginning of testing, the vertical loading
primarily has been reduced whereas tangential stress has been increased. Later during shearing the vertical stress
was decreased aiming to constrain the sample contraction in a vertical direction. Simultaneously tangential stress
was decreased (Fig. 2 b).
a b

Fig. 2. Stress paths under constant vertical stress condition: a) q = const b) h = const.

4. Numerical analysis of soil sample during direct shear test

A numerical modelling was carried out by finite element method program COSMOS/M. Linear analysis was
assumed to simulate the elastic behaviour of sand. The accepted parameters of sample: elasticity modulus 20 MPa,
Poissons ratio 0,3, mass density 1,491 Mg/m3. The sample, 7,14 cm in diameter and 3,39 cm in height, was
discretised into tetrahedral elements. Two principal diagrams were designed: 1) in the first case vertical pressure 400
kPa (as in experiment q=const) acts on the top of the model (the displacement of the sample top surface nodes are
free in vertical direction), the nodes of upper lateral walls of sample are free in vertical direction, the nodes of lower
lateral wall of sample are free in vertical and horizontal directions and plane of bottom are free in horizontal
direction; 2) in the second case displacement uniformly acts vertically uy=const (2 mm) on the top surface of the
sample (the displacement of the sample top surface nodes are free in vertical direction), the nodes of upper lateral
wall are free in vertical and lower lateral walls of sample are free in vertical and horizontal directions, the nodes of
bottom plane are free in horizontal direction.
The results of numerical modeling are shown in Figs. 314. Application of vertical load (Fig. 3 a) shows, that
vertical stress distribution on the sample top is uniform when q = const (Fig. 4 a), and in the shearing plane
obtained smaller vertical stress (Fig. 5 a). The shear plane is loaded by ~95 % of the total vertical load. In the second
case when uy=const at the beginning of the test vertical stress is not uniform on the sample top (Figs. 3 b, 4 b). In the
sample sides the vertical stress is higher ~9 % than in the sample center on the soil top, the vertical stress at the
shear plane is distributed non-uniform similarly in case q = const (Fig. 5 b). The vertical displacements when acts
only vertical stress are showed in Figs. 6 a, b. The vertical displacements distribution on the sample top (Fig. 7 a, b)
and at the rings parts interface (Fig. 8 a, b) are almost smooth in both loading case. In the case when constant
vertical loading (q = const) and horizontal displacement (5 mm) are applied, obtained vertical stress on the sample
top is higher about 8 % in the left side than in the right side of the soil sample (Figs. 9 a, 10 a). Analyzing uy=const
case we see contrary results, the vertical stress on the sample top is higher about 54 % in the right side than in the
left side of the soil sample (Figs. 9 b, 10 b). When q=const at the rings parts interface we can see the soil in the
right side of the sample is lifted, and in the contrary side the soil moves down (Fig. 11 a). In case uy = const stress
distribute almost uniform (Fig. 11 b). Vertical displacements on sample top when vertical load and horizontal
displacement are applied in case q = const has small difference (Figs. 12 a, 13 a), in case uy = const are uniform
(Figs. 12 b, 13 b). The vertical displacements distributes almost equally on the sample top in both cases. In both
loading cases the vertical displacements on the interface of rings parts are similar (Figs. 14 a, 14 b).
a b

Fig. 3. Vertical stress when acts only vertical load: a) q = const; b) uy=const.

b
a

Fig. 4. Vertical stress distribution on the sample top when acts only vertical stress: a) q = const, b) uy=const.

a b

b)

Fig. 5. Vertical stress distribution on the interface of sample parts when acts only vertical load: a) q = const, b) u y=const.
a b

Fig. 6. Vertical displacements when acts only vertical stress: a) q = const, b) uy=const.

a b

Fig. 7. Vertical displacements distribution on the sample top when acts only vertical stress: a) q = const, b)

uy=const. a b

Fig. 8. Vertical displacements distribution on the interface between ring parts when acts only vertical load: a) q = const, b) u y=const.
a b

Fig. 9. Vertical stress when vertical stress and horizontal displacement are applied: a) q = const, b)

u y=const. a b

Fig. 10. Vertical stress on the sample top when vertical stress and horizontal displacement are applied: a) q = const, b) u y=

const. a b

Fig. 11. Vertical stress on the interface between rings parts when vertical stress and horizontal displacement are applied:
a) q = const, b) uy=const.
a b

Fig. 12. Vertical displacements when vertical stress and horizontal displacement are applied: a) q = const, b)

uy=const. a b

Fig. 13. Vertical displacements on sample top when vertical stress and horizontal displacement are applied: a) q = const, b)

u y=const. a b

Fig. 14. Vertical displacements on the interface of sample parts when vertical stress and horizontal displacement are applied:
a) q = const, b) uy=const.
5. Conclusions

x During the direct shear tests the soil is loaded in a different way, it has an influence on stress-strain
distribution in the sample.
x Not all applied vertical stress on the top of sample is transmitted to the shear plane.
x In the case when constant vertical loading (q = const) and horizontal displacement (5 mm) are applied
obtained vertical stress on the sample top is higher about 8 % in the left side than in the right side of the soil
sample.
x Analyzing uy=const case we see contrary results, the vertical stress on the sample top is higher about 54 %
n the right side than in the left side of the soil sample.

Acknowledgement

An equipment and infrastructure of Civil Engineering Scientific Research Center of Vilnius Gediminas Technical
University was employed for investigations.

References

[1] S. Heng, H. Ohta, T. Pipatpongsa, M. Takemoto, S. Yokota, Constant-volume direct box-shear test on clay-seam materials, in: Proc of the
17th Southeast Asian Geotechnical Conference, May 1013, Taipei, Taiwan, 2010, pp. 8387.
[2] C. Thornton, L. Zhang, Numerical Simulations of the Direct Shear Test, Chemical Engineering &Technology 26(2) (2003) 153
156. [3] V. Kostkanova, I. Herle, Measurement of Wall Friction in Direct Shear Tests, Acta Geotechnica (2012) 1-10.
[4] R. Ziaie Moayed, S. Tamassoki, E. Izadi, Numerical Modeling of Direct Shear tests on Sandy Clay, International Journal of Civil,
Environmental, Construction and Architectural Engineering 6(1) (2012) 2731.
[5] D.H. Kang, J. Choo, T. S. Yun, Evolution of pore characteristics in the 3D numerical direct shear test, Computers and Geotechnics 49 (2013)
5361.
[6] A. Amiejus, N. Dirglien, A. Norkus, . Skuodis, Comparison of sandy soil shear strength parameters obtained by various construction
direct shear apparatuses, Archives of Civil and Mechanical Engineering 14(2) (2014) 327334.
[7] C. OSullivan, L. Cui, J.D. Bray, Three-Dimensional Discrete Element Simulations of Direct Shear Tests. In: 2 nd International PFC
Symposium, 2004 10 282004 10 29, Kyoto, Japan.
[8] S.H. Liu, D.A. Sun, H. Matsuoka, On the interface friction in direct shear test, Computers and Geotechnics 32(5) (2005) 317325.
[9] . Skuodis, A. Norkus, N. Dirglien, A. Kavrus, Sand shearing peculiarities using direct shear device, Procedia Engineering 11th
international conference on modern building materials, structures and techniques, May 16-17, 2013, Vilnius, Lithuania. Amsterdam: Elsevier
Science Ltd, 2013. ISSN 1877-7058. 57, pp. 10521059.
[10] K. Thermann, C. Gau, J. Tiedemann, Shear strength parameters from direct shear tests influencing factors and their significance, in:
International Association For Engineering Geology and the Environment: Engineering Geology for Tomorrow's Cities, 10th,
2006, Nottingham: The Geological Society of London.
[11] J.M. Royo, S. Melentijevic, Comparison of laboratory direct shear test results with the numerical analysis, Numerical Methods in
Geotechnical Engineering Hicks, Brinkgreve&Rohe (Eds), London: CRC Press, 2014, pp. 199204.

S-ar putea să vă placă și