See

discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/225336514

2D Numerical Simulations of Soil Nail Walls

Article in Geotechnical and Geological Engineering July 2010

DOI: 10.1007/s10706-009-9292-x

CITATIONS READS

20 1,441

2 authors:

Vikas Pratap Singh G. L. Sivakumar Babu

Indian Institute of Science Indian Institute of Science
12 PUBLICATIONS 48 CITATIONS 228 PUBLICATIONS 1,354 CITATIONS

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

EFFECT OF COIR FIBERS ON ENGINEERING PROPERTIES OF SOIL IN THE CONTEXT OF SUSTAINABLE

DEVELOPMENT View project

All content following this page was uploaded by G. L. Sivakumar Babu on 03 March 2014.

The user has requested enhancement of the downloaded file.

Geotech Geol Eng (2010) 28:299309
DOI 10.1007/s10706-009-9292-x

ORIGINAL PAPER

2D Numerical Simulations of Soil Nail Walls

Vikas Pratap Singh G. L. Sivakumar Babu

Received: 11 November 2008 / Accepted: 10 December 2009 / Published online: 25 December 2009
 Springer Science+Business Media B.V. 2009

Abstract In practice, numerical simulations of soil 1 Introduction

nail walls are often carried out to assess the
performance and stability. In the present study, The performance of soil nail walls is significantly
implications of the use of advanced soil models, affected due to the complex mutual interaction
such as hardening soil model and hardening soil with between its main componentsthe native soil, the
small-strain stiffness model to simulate the behavior reinforcement (nails) and the facing. Additionally,
of in situ soil on the overall response of simulated soil various other factors such as the construction
nail wall have been studied, and compared with sequence, the installation method of nails, the
respect to the analysis using conventional and most connection between the nails and the facing, are also
prevalently used Mohr-Coulomb soil model. Further, likely to influence the behaviour of the soil nail
influence of the consideration of bending stiffness of walls. Conventional soil nailing design procedures
soil nails on the simulation results has been exam- (e.g. FHWA 2003) based on limit equilibrium
ined. Results of the simulations indicated that the use methods fail to address such issues. Consequently,
of advanced models is desirable for cases of soil nail in practice, to study the complex soil-structure
walls constructed in soft soils and when lateral wall interaction and to assess the performance of soil nail
displacements are critical to the adjoining structures. walls, often numerical simulations are performed
Incorporation of bending stiffness of nails is found using rigorous computational codes based on numer-
important from the consideration of facing failure ical techniques such as finite element method (e.g.
modes of soil nail walls. Smith and Su 1997; Zhang et al. 1999; Fan and Luo
2008), discrete element method (e.g. Kim et al. 1997)
Keywords Soil nailing  Finite element  and finite difference method (e.g. Sivakumar Babu
Numerical simulation  Material models  et al. 2002).
Bending stiffness It is well established that the accuracy of numer-
ical simulations depends significantly on the consti-
tutive soil model used (e.g. Brinkgreve et al. 2006)
and the selection of the appropriate corresponding
V. P. Singh  G. L. Sivakumar Babu (&)
model parameters (e.g. Calvello and Finno 2004).
Department of Civil Engineering, Indian Institute
of Science, Bangalore, Karnataka 560 012, India Mohr-Coulomb model, being the first order soil
e-mail: [email protected]; [email protected] model, is most commonly used for numerical simu-
V. P. Singh lations in the soil nailing applications (e.g. Kim et al.
e-mail: [email protected]; [email protected] 1997; Smith and Su 1997; Zhang et al. 1999;

123
300 Geotech Geol Eng (2010) 28:299309

Sivakumar Babu et al. 2002; Fan and Luo 2008). Few according the allowable stress design procedure given
researchers have used advanced soil models such as in FHWA (2003). Table 1 summarises the geometric
Duncan-Chang hyperbolic model (e.g. Briaud and configuration and other design details of the soil nail
Lim 1997), SP modela modified Mohr-Coulomb wall. PLAXIS (2006) is used to carry out the finite
model incorporating strain softening behavior (Cheuk element based simulations of the soil nail wall
et al. 2005), Drucker-Prager yield criterion (Ng and considering it as a plane strain problem and account-
Lee 2002) and Hardening soil model (Liew and Khoo ing for the long term behaviour using drained
2006) for simulating soil behavior in the numerical conditions. Numerical simulations of the soil nail
modeling of soil nail walls. Brinkgreve et al. (2006) wall are performed considering MC-model, HS-
reported that the high stiffness of soils at very small- model and HSsmall soil models and observations
strains (\10-5) may play an important role in the are made regarding global stability, displacements of
applications such as embankments, foundations and the excavation base, lateral deformations and axial
excavations with engineering strain levels ([10-3; forces in the nails after each construction stage.
defined as range of shear strains that can be measured Given below is the brief description about the various
in conventional laboratory tests such as triaxial and model parameters required in MC-model, HS-model
oedometer tests without special instrumentation). and HSsmall soil models used to simulate soil nail
Benz (2007) developed HSsmall model accounting wall. Primary objective of the study being to bring
for the small strain stiffness of the soil. From the out the implications of the use of different soil
observations reported in the literature, it is evident models, typical values of the various soil model
that the use of advanced soil models provides more parameters for the study are adopted from Brinkgreve
realistic response of the simulated structures, but at et al. (2006) and are summarised in Table 2.
the same time, may require greater capacity of the
computational machine, may increase computational 2.1 Mohr-Coulomb Model (MC-model)
time drastically, may need sophisticated geotechnical
investigation and demand for more experienced and MC-model is an elastic perfectly plastic model,
judicious selection of the model parameters. Thus, it which combines Hookes law and the Coulombs
becomes desirable to assess the implications of the failure criterion. It is a first order model for soils
use of advanced soil models over conventional Mohr- which requires the five basic input parameters namely
Coulomb model for any particular application. In the Youngs modulus E and Poissons m for soil elasticity,
present study, two advanced soil models namely,
HS-model (Hardening soil model) and HSsmall Table 1 Soil nail wall geometry and other parameters
model (Hardening soil with small strain stiffness
Parameter Value
model) are benchmarked with respect to the most
prevalently used MC-model (Mohr-Coulomb model) Vertical height of the wall H (m) 10.0
for the numerical simulations of the soil nail walls. Face batter a (deg) 0.0
In addition to the above, the present study brings Backslope angle b (deg) 0.0
out the implications of the consideration of bending Nailing type Grouted
stiffness of nails and finite element mesh density on Grouted nails and facing
the simulation results of the soil nail walls. Following Material model Elastic
section discusses the methodology and the various Yield strength of reinforcement fy (MPa) 415.0
material parameters adopted for the study. Elasticity modulus of reinforcement En (GPa) 200.0
Elasticity modulus of grout (concrete) Eg (GPa) 22.0
Diameter of reinforcement d (mm) 20.0
2 Methodology and Material Parameters Drill hole diameter DDH (mm) 100.0
Length of nail L (m) 7.0
For the purpose of illustration and better understand-
Inclination wrt horizontal i (deg) 15.0
ing, a typical 10 m high soil nail wall with vertical
Spacing Sh 9 Sv (m 9 m) 1.0 9 1.0
face and horizontal backfill is considered for the
Facing thickness t (mm) 200.0
study. Design of the soil nail wall is carried out

123
Geotech Geol Eng (2010) 28:299309 301

Table 2 Soil model parameters (Bringreve et al. 2006)

Parameter MC HS HSsmall

Cohesion c (kN/m2) 10.0 10.0 10.0

Friction angle u (deg) 27.5 27.5 27.5
Dilatancy angle w (deg) 0.0 0.0 0.0
Unit weight c (kN/m3) 19.0 19.0 19.0
Modulus of elasticity of soil E (kN/m2) 30,000
ref
Secant stiffness in standard drained triaxial test E50 (kN/m2) 20,000 20,000
ref
Tangent stiffness for primary oedometer loading Eoed (kN/m2) 20,000 20,000
ref
Unloading/reloading stiffness Eur (kN/m2) 60,000 60,000
Reference shear modulus Gref
0
2
(kN/m ) 75,000
Reference stress for stiffness pref (kN/m2) 100.0 100.0 100.0
Shear strain at which Gsecant = 0.7 G0, c0:7 0.0001
Poissons ratio m 0.3 0.2 0.2
Power for stress level dependency of stiffness m 0.5 0.5
For HS and HSsmall models m = mur (unloadingreloading)

soil friction angle u and soil cohesion c for soil 2.3 Hardening-Soil with Small Strain Stiffness
plasticity, and the dilatancy angle w. Since the pre- Model (HSsmall model)
failure stiffness behavior is assumed to be linear
elastic, the model has a limitation in terms of The HSsmall model (Benz 2007) is a modification of
prediction of the deformation behavior before failure the Hardening Soil model that accounts for the
(Callisto et al. 1999). increased stiffness of soils at small strains. At low
strain levels most soils exhibit a higher stiffness than
2.2 Hardening-Soil Model (HS-model) at engineering strain levels; this stiffness varies non-
linearly with strain. This behaviour is described in the
Schanz et al. (1999) provides information about the HSsmall model using two additional strain-history
formulation, various model parameters and verifica- material parameters: (a) the initial or very small-
tion of HS-model. It is an advanced soil model strain shear modulus G0 and (b) the shear strain level
capable of simulating both soft and stiff soils. Similar c0.7 at which the secant shear modulus G is reduced to
to MC-model, failure in HS-model is defined by 70% of G0.
means of Mohr-Coulomb failure criterion. Unlike the
MC-model, the HS-model accounts for the increase
in stiffness with pressure. The model uses a power 3 Finite Element Simulations
law formulation for stress-dependent stiffness similar
to the one used in the Duncan-Chang hyperbolic As mentioned earlier, soil nail wall is modeled as a
model (Duncan and Chang 1970). The HS-model plane strain problem and long term behavior is
requires 11 input parameters, i.e. three reference simulated using drained analysis conditions. 15-no-
stiffness parameters; the triaxial loading stiffness E50, ded triangular elements are used for generating finite
the triaxial unloading stiffness Eur and the oedometer element mesh of appropriate density. Coarse mesh
loading stiffness Eoed. A power m parameter for the density is adopted globally, which is refined to fine
stress-dependent stiffness formulation, a Poissons density in the vicinity of the soil nail wall (Fig. 1).
ratio for loading and unloading mur, the Mohr- Mesh boundaries are placed far enough so as to
Coulomb strength parameters u and c, the dilatancy minimise the influence of mesh boundaries on the
angle w, a K0-value, and a parameter called the results of the numerical simulation (Briaud and Lim
failure ratio Rf which determines the strain level at 1997). Figure 1 shows the simulated soil nail wall
failure and pref a reference stress for stiffnesses. with dimensions and various parameters including in

123
302 Geotech Geol Eng (2010) 28:299309

nail; A 0:25pD2DH is the total cross-sectional area

of grouted soil nail; Ag A  An is the cross-
sectional area of grout cover; An 0:25pd2 is the
cross-sectional area of reinforcement bar and DDH is
the diameter of drill hole. If, Sh is horizontal spacing
of soil nails, knowing the equivalent modulus of
elasticity Eeq (Eq. 1) for the grouted soil nail, the
axial and bending stiffnesses can be determined using
Eqs. 2 and 3, respectively.
 
Eeq pD2DH
Axial stiffness EA kN=m 2
Sh 4
Fig. 1 Numerically simulated 10 m high soil nail wall
 
  Eeq pD4DH
Bending stiffness EI kNm2 =m 3
situ soil properties, mesh boundaries and fixity Sh 64
conditions.
3.2 General Procedure for Numerical Simulation
3.1 Equivalent Nail Parameters
Given below are the general steps followed in the
numerical simulation of the soil nail wall.
Most commonly, in the plane strain simulation of the
soil nail walls; rectangular shaped structural elements In the input program, geometry of the soil nail
are used to simulate soil nails. Depending upon wall (including nails and facing elements layout),
whether in the numerical simulations the bending boundary conditions, construction stages, material
stiffness of soil nails is to be considered or not, the properties are defined. This is followed by the
most important input material parameters for the generation of finite element mesh with desired
structural elements simulating soil nails are the axial mesh density and initial gravity stresses using
stiffness EA and/or the flexural rigidity (bending K0-procedure (i.e. at-rest condition).
stiffness) EI. In practice, plate structural elements or Accomplishment of physical modeling in the
geogrid structural elements (which only require the input program is followed with the calculation
axial stiffness EA as input) are being used for soil nail program. Staged construction option is used to
wall simulations using Plaxis. Both the plate and simulate cconstruction of the soil nail wall in five
geogrid structural elements are rectangular in shape stages indicated as E1, E2,,E5 in Fig. 1. In each
with width equal to 1 m in out-of-plane direction. construction stage, excavation depth of 2 m in
Since, the soil nails are circular in cross-section and situ soil is simulated by deactivating the corre-
placed at designed horizontal spacing, it is necessary sponding soil cluster and the installation of nail
to determine equivalent axial and bending stiffnesses and facing is performed by activating the corre-
for the correct simulation of circular soil nails as sponding structural element. Calculation type was
rectangular plate structural elements. set to the plastic analysis.
For the grouted nails, equivalent modulus of After each excavation stage, factor of safety for
elasticity Eeq shall be determined accounting for the global stability is determined using Phi/c reduc-
contribution of elastic stiffnesses of both grout cover tion option as the calculation type. In the Phi/c
as well as reinforcement bar. From the fundamentals reduction or strength reduction technique (Matsui
of strength of materials, Eeq can be determines as and San 1992) critical failure mechanism is
    identified automatically, which is normally
An Ag
Eeq En Eg 1 assumed in the conventional analysis. Dawson
A A
et al. (1999) showed that the factors of safety
where: Eg is the modulus of elasticity of grout calculated from this approach are very close to the
material; En is the modulus of elasticity of nail; Eeq is values obtained from conventional methods in
the equivalent modulus of elasticity of grouted soil geotechnical analysis. They also indicated that

123
Geotech Geol Eng (2010) 28:299309 303

this method is more general and flexible and it is negligible influence on the evaluation of global
more advantageous particularly when the failure stability of the soil nail wall using numerical
mechanism is complex as in the case of soil nail simulations.
walls.
4.2 Influence of Material Models on Base Heave
of Excavation
4 Soil Models and Response of Soil Nail Wall
FHWA (2003) identifies basal heave or bearing
For the soil nail wall defined by parameters given in capacity failure as one of the external failure mode
Tables 1 and 2, and using the methodology presented for soil nail walls. This failure mode may be of
earlier; various important aspects of the soil nail concern when a soil nail wall is excavated in soft
walls are studied. Observations are made from the soils. Because the wall facing does not extend below
numerical simulations of the soil nail wall using the bottom of the excavation, the unbalanced load due
different material models namely MC-model, to the excavation may cause the bottom the excava-
HS-model and HSsmall model, and are discussed in tion to heave and stimulate a bearing capacity failure
the following sub-sections. It is worth mentioning of the foundation. Figure 3 shows upward heave or
that for the same finite element mesh density and the displacement of the excavated soil in front of the
geometric parameters, the overall calculation time for soil nail wall face with construction stages for the soil
the numerical simulation of the soil nail wall nail wall simulated using three different material
increased to about 45 times by the use of HS-model models. For each construction stage, base heave
and to about 910 times by the use of HSsmall model shown in Fig. 3 is the maximum value of the vertical
in comparison to the MC-model. upward displacement of excavation base BC (see
Fig. 1). From Fig. 3, it is evident that MC-model
4.1 Influence of Material Models on Global over-estimates the base heave of the excavation face
Stability to almost twice as that predicted by HS-model or
HSsmall model. This observation regarding over-
Figure 2 shows the trend in the variation of the factor prediction of base heave is in good agreement with
of safety for global safety with construction stages of literature (e.g. Callisto et al. 1999; Brinkgreve et al.
the soil nail wall simulated using three different 2006) and may be attributed to the consideration of
material models. From Fig. 2, it is evident that all the linear elastic pre-failure soil behavior assumed in
three material models capture similar response. In MC-model formulation. Thus, whenever, soil nailing
other words, use of advanced soil models has walls in soft soil conditions are numerically

Fig. 2 Global factor safety of soil nail wall with construction

stage Fig. 3 Excavation base heave with construction stage

123
304 Geotech Geol Eng (2010) 28:299309

stages, cumulative plastic strains in the soil nail

system increase and thereby reduce the stiffness of
the retained soil mass significantly, and (b) assump-
tion of the linear elastic pre-failure behaviour of the
soil in MC-model.

4.4 Influence of Material Models on Axial Force

Development in Soil Nails

Figures 5 and 6 show the maximum axial tensile

forces developed in the soil nails with construction
stages of the soil nail wall and with the depth of
embedment of the soil nails, respectively. The soil
nail force in Fig. 5 is the maximum axial force
Fig. 4 Lateral displacement of soil nail wall with construction
developed among the nails installed up to the
stage
construction stage in consideration, whereas, the soil
nail force in Fig. 6 is the maximum axial force
simulated, it is advisable to use advanced models developed in a given nail along its length. From
such as HS-model or HSsmall model in-lieu of MC- Fig. 5, it is evident that all the three soil models used
model. The above aspect may also be useful from the in the study predicts almost similar response of the
consideration of stability of soil nail walls during maximum axial force developed in soil nails with
construction stages. HSsmall model predicts excava- construction stages. However, it may be noted that
tion heave even lesser than HS-model (though not the MC-model underestimates the axial force devel-
very significantly) attributing to the role of increased oped in the soil nails. Similar observations can be
stiffness of soils at very small strains (Brinkgreve made from Fig. 6 regarding the development of axial
et al. 2006; Benz 2007). force in the soil nails embedded at different depths.
The dashed line in Fig. 6 shows the theoretical value
4.3 Influence of Material Models on Lateral of the maximum axial force Tmax in a nail embedded
Displacement of Soil Nail Wall at depth z below ground surface obtained using
expression: Tmax = KaczShSv, where Ka is the active
Figure 4 shows the maximum lateral displacement of earth pressure coefficient (FHWA 2003). This expression
the soil nail wall with construction stages for the soil
nail wall simulated using three different material
models. For each construction stage, lateral displace-
ment shown in Fig. 4 is the maximum value of the
horizontal displacement (which usually occurs at top)
of vertical wall face AB (see Fig. 1). It may be
observed from the Fig. 4 that up to 60% of the
construction stage, lateral displacements of soil nail
wall predicted by MC-model are more than (though
not very significantly) those predicted by HS-model
or HSsmall model. However, beyond 60% construc-
tion stage, maximum lateral displacements of the soil
nail wall predicted by MC-model are significantly
less (almost 50% less for the fully constructed soil
nail wall i.e. 100% construction stage) than those
predicted by HS-model or HSsmall model. These
observations may be attributed to the following two Fig. 5 Development of maximum axial force in nails with
reasons: (a) that with the increasing construction construction stage

123
Geotech Geol Eng (2010) 28:299309 305

from the extensive experiments (such as national

research project Clouterre) and other works related to
soil nailed retaining structures in France over
10 years, stated that, at failure bending and shear
resistances of grouted nails are mobilised, however,
the influence of bending stiffness and shear on the
global safety factor is small (less than 15%). Jewell
and Pedley (1992) concluded that the effects of
bending and shear resistances can be ignored in the
design and analysis of soil nailing with marginal
conservatism. In practice, ignoring the effects of
shear and bending resistances of soil nails, soil
nailing analysis and design has been radically sim-
Fig. 6 Variation of maximum axial force in nails with plified and this approach is commonly accepted (e.g.
embedment depth (100% CS) FHWA 2003).
In practice, bending stiffness of the soil nails is
is based on the concept that a soil nail embedded at often considered in the numerical simulations of the
depth z below ground surface is subjected to a soil nail walls (e.g. Fan and Luo 2008). On the other
maximum lateral earth pressure Kacz acting on a nail hand, few researchers prefer to ignore the contribu-
influence area defined by the product of the vertical tion of bending and shear capacity of soil nails in the
and horizontal nail spacings (i.e. nail influence analysis (e.g. Liew and Khoo 2006). In the light of
area = ShSv). Since, the theoretical evaluation of fact that consideration of bending and shear resis-
the axial force in nails embedded at different depths tance of soil nails is conservatively ignored in the
do not account for the load re-distribution due to the conventional analysis and design of soil nail walls, it
construction stages, the theoretically expected axial is found desirable to study the implications of the
forces are found to be considerably more than the consideration of bending stiffness of soil nails on the
computed axial forces (see Fig. 6). In addition to the analysis of the soil nail walls. As mentioned previ-
development of axial force, observations were made ously, plate structural element in Plaxis can be used
regarding the development of maximum bending to perform analysis of soil nail walls considering
moments and the maximum shear forces in nails and bending stiffness of soil nails as they require both
it is found that use of different soil models does not axial stiffness EA and bending stiffness EI as the
have any considerable influence. Thus, in general, main material parameters. On the other hand, geogrid
MC-model can be considered capable of capturing structural elements can be used to model soil nails
the response of stress elements in soil nails reason- with considering bending stiffness of soil nails as
ably accurately. they require only axial stiffness EA as the main input
parameter.
In order to study the implications of the consid-
5 Role of Bending Stiffness of Soil Nails eration of bending stiffness of soil nails in analysis of
soil nail walls based on numerical simulations, two
Incorporation of bending and shear resistances of series of simulations of the soil nail wall of 10 m
nails in the analysis and design of soil nail walls had height are performed, one with the use of plate
been a much debatable issue reported in the literature. structural elements to simulate soil nails and the other
For example, Juran et al. (1990) reported that inclined with the use of geogrid structural elements to
nails (10150) would tend to undergo a local rotation simulate soil nails. Various material properties and
to approach the horizontal direction of maximum soil other parameters indicated in Tables 1 and 2 are used
extension, and therefore, the effect of bending and soil behaviour is simulated using MC-model. At
stiffness has significant effect on the development each construction stage of the wall, observations are
of nail forces. Schlosser (1991), based on his made with regard to the global factors of safety,
multicriteria theory in soil nailing and observations maximum lateral (horizontal) displacement of walls,

123
306 Geotech Geol Eng (2010) 28:299309

maximum axial tensile developed and development

of bending moment and shear force in nails (for plate
elements only). Response corresponding to the con-
sideration of bending stiffness is indicated by Plate
and ignoring bending stiffness is indicated by Geo-
grid in the plots of the observations form numerical
simulations.
Figure 7 shows the trend of variation of global
factor of safety of the soil nail wall with construction
stage. It is evident from Fig. 7, that for the fully
constructed soil nail wall (i.e. 100% construction),
consideration of bending stiffness of nails in the
Fig. 8 Trend of lateral displacements of soil nail wall with
analysis has negligible influence on the global construction stage
stability. However, it is worth noting from the
Fig. 7 that consideration of bending stiffness of nails
in the analysis has significant influence during the
construction stage. Geogrid structural elements
resulted in significantly less factors of safety for
global stability in comparison to the plate structural
elements. Alternatively, it can be interpreted that
bending stiffness plays important role in the stability
of soil nail walls during the construction stage.
Figure 8 shows the trend of maximum lateral
displacements of the soil nail wall with construction
stage. It is evident from Fig. 8 that the almost similar
displacement response is captured by both geogrid
elements and plate elements. This implies that the
Fig. 9 Development of maximum axial force with construc-
consideration of bending stiffness has negligible tion stage
influence on deformation analysis of the analysis of
soil nail walls.
Figures 9 and 10 show the development of max-
imum axial force in nails with construction stage. average the maximum axial force developed in nails
From Figs. 9 and 10, it can be observed that on an simulated using geogrid elements is found to be 15%
more in comparison to that developed in nails using
plate elements. In other words, lesser axial developed
in nails simulated using plate elements may be
credited to the contribution of bending stiffness of the
nails. This observation is in good agreement with
the literature (Schlosser 1991). Figure 11 shows the
variation of axial force along the nail length for nails
embedded at different levels. Very close resemblance
among the axial forces variation along nail length is
evident from Fig. 11 for the nails simulated using
geogrid and plate elements.
Development of maximum bending moment and
maximum shear force in nails with construction
stages are shown in Fig. 12. It is evident from Fig. 12
Fig. 7 Trend of global factor of safety of soil nail wall with that bending and shear capacities of soil nail start
construction stage mobilising with increasing construction stages.

123
Geotech Geol Eng (2010) 28:299309 307

Fig. 10 Maximum axial force developed in nails with

embedment depth (100% CS)
Fig. 13 Variation of shear force and bending moment along
nail length (alternate nails shown)

that bending moments and shear forces are concen-

trated near the face of the wall. This provides an
insight into the facing failure modes of the soil nail
walls. Generally, soil nail are rigidly connected with
the facing and in such cases it is desirable to evaluate
the facing design. Improper design may lead to the
bending and/or shear failures of soil nails at or near
the facing. Ignoring bending stiffness of soil nails
may overlook this practical aspect of the soil nail wall
analysis completely.
Thus, from the above observations it is apparent
Fig. 11 Variation of axial force along nail length (alternate that the consideration of bending stiffness of soil nails
nails shown) provides better insight into the analysis of soil nail
walls using finite element simulations. Since, finite
elements simulations are often performed to investi-
gate the cause of failure or to assess the performance
of soil nail walls, it is desirable that bending stiffness
shall be considered in the simulations, despite the fact
that it is conservatively ignored in the conventional
design procedures.

6 Influence of Mesh Density on Numerical

Simulations

Another important aspect of the numerical simulation

Fig. 12 Development of maximum bending moment and of any structure is the density of finite element mesh
maximum shear force in nails with construction stage adopted. To study the influence of mesh density on
the analysis of the soil nail wall, numerical simula-
Figure 13 shows the variation of bending moments tions are performed adopting mesh densities ranging
and shear forces along the nail length for nails from very coarse to very fine, and the results
embedded at different levels. It is interesting to note corresponding to the analysis are presented in

123
308 Geotech Geol Eng (2010) 28:299309

Table 3 Influence of mesh density on finite element simulation

Mesh density Elements per Global factor Max. hzl. displacement Calculation time normalised
unit volume of safety FS (mm) wrt medium mesh density

Very coarse 0.39 1.610 20.93 0.46

Coarse 0.60 1.598 22.31 0.61
Medium 0.98 1.592 22.86 1.00
Fine 2.08 1.553 24.79 2.24
Very fine 4.14 1.521 28.35 6.18
FS values correspond to the fully constructed wall. If FS is to be determined after each construction stage, calculation time may vary
even more drastically

Table 3. It is to be noted that this analysis is of advanced soil models is found to be essential when
performed using MC-model. From Table 3, it can very small lateral displacement of the soil nail wall is
be observed that global factor of safety varies likely to influence the stability and deformation
significantly from 1.61 for very coarse mesh to 1.52 requirements. Also, when soil nail walls are founded
for very fine mesh. Also, maximum lateral displace- in soft soils, contrary to MC-model, use of advanced
ments varied from 20.93 mm for very coarse mesh to soil models provides more appropriate estimate of the
28.35 mm for very fine mesh. Similar trends are basal heave of the excavation. Considering the
observed for the stress parameters in nails such as importance of the selection of suitable model param-
development of axial force, bending moment and eters and drastic increase in the calculation time in
shear force. Though, denser mesh may result in more case of advanced soil models, for general practical
accurate analysis, it is important to note that applications where lateral displacements are not
increasing the mesh density results in drastic increase critical, use of MC-model is found reasonable for
in the overall calculation time (see Table 3). Thus, the numerical simulations based analysis of soil nail
appropriate mesh density shall be used depending walls.
upon the degree of accuracy required and the capacity Since, often computational tools are employed to
of the computing machine. In general, coarse mesh assess the stability and performance of the soil nail
density globally and fine mesh density in the vicinity walls, results of the present study indicate that
of the soil nail wall may be used. consideration of the bending stiffness of soil nails in
the numerical simulations of soil nail walls may
provide better insight into the facing failure modes
7 Concluding Remarks and is found significant during the construction stage.
Further, influence of mesh density on the numerical
This study provides a comparison of the use of analysis of soil nail walls is studied and use of
advanced soil models such as HS-model and HSsmall appropriate finite element mesh density is highlighted.
model in lieu of the most prevalently used MC-model
in the numerical simulations of soil nail walls. Acknowledgments The work presented in this study is a part
of the research project Guidelines for Soil Nailing Technique in
Implications of the consideration of bending stiffness Highway Engineering (R-86) financed by the Ministry of
of soil nails in the analysis using computational tools Shipping, Road Transport and Highways, Government of India,
have also been brought out. Results of the finite New Delhi. The authors express thanks to the Ministry for
element based plane strain simulations of a typical funding and providing necessary support for the project.
soil nail wall indicate that the advanced soil models
considered in the present study have nominal influ-
References
ence on the overall soil nail wall stability analysis and
the development of stress elements such as axial Benz T (2007) Small-strain stiffness of soils and its numerical
forces, bending moments and shear forces in the soil consequences. PhD Thesis, Institut fur Geotechnik, Uni-
nails. However, in agreement with the literature, use versitat Stuttgart, Germany

123
Geotech Geol Eng (2010) 28:299309 309

Briaud JL, Lim Y (1997) Soil nailed wall under piled bridge Kim JS, Kim JY, Lee SR (1997) Analysis of soil nailed earth
abutment: simulation and guidelines. J Geotech Geoen- slope by discrete element method. Comput Geotech
viron Eng 123(11):10431050 20(1):114
Brinkgreve RBJ, Bakker KJ, Bonnier PG (2006) The relevance Liew SS, Khoo CM (2006) Soil nail stabilisation for a 14.5 m
of small-strain soil stiffness in numerical simulation of Deep excavation at uncontrolled fill ground. In: Pro-
excavation and tunneling projects. In: Proceedings of 6th ceedings of 10th International Conference on Piling and
European Conference in Geotechnical Engineering, Graz, Deep Foundations, Amsterdam, The Netherlands
Austria, pp 133139 Matsui T, San KC (1992) Finite element slope stability analysis
Callisto L, Amorosi A, Rampello S (1999) The influence of by shear strength reduction technique. Soils Found
pre-failure soil modelling on the behaviour of open 32(1):5970
excavations. In: Proceedings of 12th European Confer- Ng CWW, Lee GTK (2002) A three-dimensional parametric
ence on Soil Mechanics and Geotechnical Engineering, study of the use of soil nails for stabilising tunnel faces.
Amsterdam, Netherlands, pp 8996 Comput Geotech 29(8):673697
Calvello M, Finno RJ (2004) Selecting parameters to optimize PLAXIS (2006) Plaxis user manual (version 8.6). Delft Uni-
in model calibration by inverse analysis. Comput Geotech versity of Technology & Plaxis BV, The Netherlands
31(5):411425 Schanz T, Vermeer PA, Bonnier PG (1999) The hardening soil
Cheuk CY, Ng CWW, Sun HW (2005) Numerical experiments modelformulation and verification. In: Proceedings of
of soil nails in loose fill slopes subjected to rainfall Plaxis Symposium on Beyond 2000 in Computational
infiltration effects. Comput Geotech 32(4):290303 Geotechnics, Amsterdam, Rotterdam, Balkema, pp 281
Dawson EM, Roth WHA, Drescher A (1999) Slope stability 296
analysis by strength reduction. Geotechnique 49(6):835 Schlosser F (1991) Discussionthe multicriteria theory in soil
840 nailing. Ground Eng, November, pp 3033
Duncan JM, Chang CY (1970) Non-linear analysis of stress and Sivakumar Babu GL, Murthy BRS, Srinivas A (2002) Analysis
strain in soil. J Soil Mech Found Eng 96(SM5):16291653 of construction factors influencing the behavior of soil
Fan CC, Luo JH (2008) Numerical study on the optimum nailed earth retaining walls. Ground Improv 6(3):137143
layout of soil nailed slopes. Comput Geotech 35(4):585 Smith IM, Su N (1997) Three dimensional finite element
599 analysis of nailed soil wall curved in plan. Int J Numeric
FHWA (2003) Geotechnical engineering circular No 7soil Analytic Meth Geomech 21(9):583597
nail walls. Report FHWA0-IF-03017. US Department of Zhang M, Song E, Chen Z (1999) Ground movement analysis
Transportation, Federal Highway Administration, Wash- of soil nailing construction by three-dimensional (3-D)
ington DC finite element modeling (FEM). Comput Geotech 25(4):
Jewell RA, Pedley MJ (1992) Analysis for soil reinforcement 191204
with bending stiffness. J Geotech Eng 118(10):15051528
Juran I, Baudrand G, Farrag K, Elias V (1990) Kinematical
limit analysis for design of soil-nailed structures. J Geo-
tech Eng 116(1):5472

123

View publication stats