Professional Documents
Culture Documents
COHESION-LESS SOIL
by
G. S. KAME
(Roll No. 08404803)
The Second stage annual progress report for Ph.D. program entitled analysis of vertical plate
anchors in cohesion-less soil submitted by Mr. G. S. Kame (Roll No. 08404803) can be
accepted for evaluation.
ii
CONTENTS
iii
HORIZONTAL COHESION-LESS BACKFILL
4.1 Introduction 22
4.2 Proposed method 22
4.2.1 Computation of soil reaction on the failure surface 25
4.2.2 Computation of soil reaction on plane failure surface AB 26
4.2.3 Computation of vertical and horizontal components of 27
reaction on curved failure surface EA
4.2.4 Components of resultant soil reaction on the failure surface 28
4.2.5 Magnitude of active thrust 30
4.3 Trial and error procedure 31
4.4 Centroid of log spiral 32
4.5 Point of application of active thrust 33
4.6 Discussion 33
4.6.1 Point of Application of Active Thrust 33
4.6.2 Distribution of reactive pressure over failure surface 34
4.6.3 Comparison with other solutions 35
iv
5.4 Trial and error procedure 47
5.5 Centroid of log spiral 48
5.6 Point of application of passive thrust 49
5.6.1 Pole above wall top 49
5.6.2 Pole below wall top 50
5.7 Discussion 50
5.7.1 Point of application of passive thrust 51
5.7.2 Distribution of reactive pressure over failure surface 52
5.7.3 Comparison with other solutions 53
v
LIST OF FIGURES
Figure Caption Page No.
No.
2.1 Basic case - failure surface of a vertical anchor plate in cohesion-less soil 5
(Ovesen and Stromann 1972)
2.2 Surcharge method (Neely et al., 1973) 8
2.3 Equivalent free surface analysis (Neely et al., 1973) 9
2.4 Failure modes and zones of plastic yielding for rough vertical anchors in 10
cohesion-less soils (=20), Merifield et al. (2006)
2.5 Free body diagram of block anchor (Naser 2006) 11
3.1 Ktters (1903) equation for a curved failure surface 15
3.2 Element considered in the derivation of expressions for stress at a point 15
3.3 Orientation of stresses 1 and 3 with respect to x andy 17
3.4 Vectorial resolution of elemental lengths dx and dy in the direction of 18
slip lines
3.5 Stresses acting on an infinitesimal length ds along the slip line 20
3.6 Free body diagram of the failure surface 21
4.1 Retaining wall with a horizontal cohesion-less backfill - failure 23
mechanism
4.2 Failure surface adopted in the proposed analysis 23
4.3 Free body diagram of wedge EACD 24
Geometry of failure surface
4.4 Reactive pressure distribution on the failure surface for active case 26
4.5 Trial procedure for locating pole of the log spiral 31
4.6 Reactive pressure distribution on the failure surface 34
5.1 Retaining wall with a horizontal cohesion-less backfill - failure 37
mechanism
2.2a Failure surface adopted in the proposed analysis with pole located above 38
the wall top
vi
5.2b Failure surface adopted in the proposed analysis with pole located below 39
the wall top
5.3a Free body diagram of failure wedge EACD, with pole above the wall top 39
5.3b Free body diagram of failure wedge EACD, with pole below the wall top 40
5.4 Reactive pressure distribution on the failure surface for passive case 42
6.1a Failure mechanism at ultimate load for continuous (strip) vertical plate 57
anchor in cohesion-less soil
6.1b Free body diagram of the anchor plate 57
6.2a Reactive pressure distribution on the failure surface for passive case 58
(Kame, Dewaikar and Choudhury, 2010-b)
6.2b Reactive pressure distribution on the failure surface for active case 58
(Kame, Dewaikar and Choudhury, 2010-a)
6.3 Reactive pressure distribution on the failure surface for the passive case 60
using Ktters (1903) equation (Kame, Dewaikar and Choudhury,
2010-b)
6.4 Reactive pressure distribution on the failure surface for the active case 60
using Ktters (1903) equation (Kame, Dewaikar and Choudhury,
2010-a)
6.5 Free body diagram of failure wedge EACD (passive state) 62
6.6 Trial procedure for locating pole of the log spiral 63
vii
LIST OF TABLES
Table Caption Page
No. No.
4.1 Variation of Hr with and 34
5.1 Passive earth pressure coefficient and location of pole of the log spiral 51
5.2 Variation of Hr with and 51
5.3 Comparison of Kp values 53
5.4 Comparison of Kp values 55
6.1 Earth pressure coefficients using proposed method (Kame, Dewaikar and 64
Choudhury, 2010 a & b)
6.2 Anchor & Soil Parameters 66
6.3 Comparison of pullout capacity experimental results and semi-empirical 67
methods for vertical (strip) plate anchors
6.4 Cumulative frequency distribution of errors for vertical plate anchors 67
viii
NOMENCLATURE
inclination of the tangent at the point of interest on log spiral with the horizontal
unit weight of soil
p unit weight of plate material
B anchor length
p reactive pressure on the failure surface
RH, RV components of resultant soil reaction acting on the curved part of the failure
surface
Tu ultimate pullout capacity of vertical plate (strip) anchor per unit width
r0 starting radius of the log spiral at the wall base
r radius of the log spiral at any general point
r1 maximum radius of the log spiral
spiral angle measured from the starting radius
m maximum spiral angle
v angle between vertical face of wall and the starting radius r0
ix
ABSTRACT
In the present investigations, methods are proposed for the evaluation of active and passive
thrusts on a vertical retaining wall with horizontal cohesion-less backfill and ultimate pullout
capacity of a shallow laid vertical plate anchors in cohesion-less soil.
In case of a vertical wall retaining horizontal cohesion-less backfill, a method based on the
application of Ktters equation is proposed for the complete analysis of active and passive earth
pressures. The unique failure surface consisting of log spiral and its tangent is identified on the
basis of force equilibrium condition. One distinguishing feature of the proposed method is its
ability to compute the point of application of active and passive thrusts using moment
equilibrium. Another distinguishing feature is the prediction of distribution of reactive pressure
along the failure surface. The results show a close agreement with some of the available
solutions.
For vertical plate strip anchors in cohesion-less soil, the ultimate pullout capacity is analyzed
with the consideration of active and passive states of equilibrium in the soil. Ktters equation is
used to compute the active and passive thrusts (along with their points of application) which are
subsequently used in the analysis in which, all the equation of equilibrium are properly
interpreted. A comparison of the results with the experimental results vis--vis available
theoretical/empirical solutions shows that, the proposed analysis provides a better estimate of the
pullout capacity.
Key words: Ktters equation, active earth pressure coefficient, passive earth pressure
coefficient, log spiral, point of application, horizontal cohesion-less backfill, vertical anchor
plates, pullout capacity
x
NOTES
Chapter 1
Introduction
1.1 General
One of the earliest examples of soil anchor is in supporting transmission towers. This is the only
application which gave inspiration for a lot of initial research into anchor behavior.
Initially these towers were supported by large dead weight concrete blocks where the required
uplift capacity was achieved solely due to the self weight of concrete. This simple design came
at considerable cost and as a result, research was undertaken in order to find a more economical
design solution.
Anchors are designed primarily to transmit outwardly directed loads imposed on the foundations
of structures. Until the middle of nineteenth century, anchors were primarily used for stabilizing
fairly light weight structures. With the design and construction of large suspension bridges, very
large loads were transmitted to the bridge foundation. In order to support these loads permanent
anchoring systems in rock medium were gradually developed and constructed.
Nowadays, earth anchors of various types are used for structures like transmission towers,
bulkheads, tall chimneys, jetty structures and underground tanks to transmit pullout/uplift forces.
In deep waters, anchors are required to hold offshore platforms in position by transmitting the
pullout/uplift forces to the ocean bed. The mooring systems for floating bodies, submerged
platforms and oil or gas pipelines laid on the seabed require anchors of different types.
Soil or ground anchors are a lightweight foundation system designed and constructed specifically
to resist any pullout force or overturning moment placed on a structure. In general, soil anchors
are classified as direct embedded anchors or plate anchors, helical anchors, grouted anchors and
pile anchors. They can be square, circular or rectangular in shape and are commonly used as
foundation systems for structures requiring pullout/uplift resistance.
Plate anchors are generally constructed from steel or concrete and may be circular, square or
rectangular in shape. They may be placed horizontally, vertically, or at an inclined position
depending on the load orientation or type of structure requiring. They are also classified as
shallow or deep anchors depending on the mechanism of failure in the soil mass.
Foundations such as plate or pile anchors are often used to support a variety of land structures
such as guyed lattice towers, transmission towers, tension cables for suspension bridges, tent-
type roofs and marine structures such as floating platforms and tension leg platforms. These
structures are often subjected to wind loading which causes pullout forces much greater than the
weight of the structure itself. In addition to wind, marine structures are also subjected to wave
forces. Thus, in several pullout cases plate anchors and/or pile anchors subjected to vertical,
inclined or horizontal pullout loads are required.
1.3 Objective
Chapter 2 presents a critical review of the available literature on theoretical and experimental
aspects of pullout capacity of vertical plate anchors and methods for estimation of active and
passive earth pressures.
In Chapter 4, method for estimation of active thrust on a vertical retaining wall with horizontal
cohesion-less backfill is presented.
Chapter 5 describes method for estimation of passive thrust on a vertical retaining wall with
horizontal cohesion-less backfill.
Chapter 6 presents a method for the estimation of pullout capacity of vertical plate strip anchors
in cohesion-less soil.
In Chapter 7, conclusions drawn from the present investigations and scope of the future work are
presented.
Chapter 2
Review of Literature
2.1 General
Structures like transmission towers, retaining walls, wing walls, abutments, sheet piles, chimneys
etc. are subjected to large lateral loads due to wind and/or lateral earth pressure and associated
moments induce tensional forces in the foundation. Structures like transmission towers are
constructed in large numbers and therefore, the investigation of pullout/uplift resistance of
foundation becomes necessary in order to achieve a rational and economical design. Soil anchors
are used in different shapes such as circular, square or rectangular as foundation systems for
structures requiring uplift resistance. Depending upon the depth of embedment in soil, plate
anchors are classified as shallow or deep anchors. In case of shallow anchors, failure surface
reaches the ground surface; whereas in case of deep anchors, due to limiting settlement
consideration, failure surface does not reach the ground surface.
The driving force behind the research on behavior of pullout/uplift anchors is due to extensive
application of soil anchors in supporting transmission towers (Balla 1961). Initially these towers
were supported by large mass-concrete blocks where the pullout capacity was supplied entirely
by self-weight of concrete. This simple design incurred considerable which initiated further
research.
Here a study of vertical anchor plate is presented, which derives its load bearing capacity from
the active and passive resistance of cohesion-less soil.
anchors expanded to include the support of more elaborate and large structures more research is
being carried out using experimental as well as analytical methods to provide an economical and
competitive alternative to these mass foundations. A lot of work is available on horizontal
anchors but vertical anchors subjected to horizontal or inclined pull have received very little
attention apart from the general earth pressure theory proposed by Hansen (1953) and important
work described by Das et al. (1975), Rowe (1982), and Dickin (1983)
2.2.1 Experimental Investigations
Buchholz (1930) observed that, critical embedment ratio was the ratio beyond which, the failure
surface did not reach the soil surface. He defined this ratio as H/h = 7. Hueckel (1957) has
reported that, ultimate resistance of inclined plates with the pull in horizontal direction was
smaller than that of the vertical one, regardless of their angle of inclination.
45 + /2 45 + /2 45 /2 45 /2
Pa
Tu(B) H=h
PaH PpH
Pp
Active Rankine zone
Logarithmic spiral
Logarithmic spiral
Fig. 2.1 Basic case - failure surface of a vertical anchor plate in cohesion-less soil
(Ovesen and Stromann 1972)
Ovesen and Stromann (1964 and 1972) used the failure mechanism proposed by Hansen (1953)
to estimate the earth pressures for the case of a continuous shallow plate anchor flushing with the
cohesion-less ground surface, termed as the basic case (H/h=1.0). The failure mechanism
consists of Rankine (1857) and logarithmic spiral zones (Terzaghi, 1943) as shown in Fig. 2.1.
Based on the above failure mechanism and laboratory model tests, the ultimate pullout capacity,
per unit width of a strip anchor in cohesion-less soil was estimated as,
Tu = R0vTu(B) (2.1),
Where Tu(B) is the ultimate holding capacity as estimated by the following expression from
Fig. 2.1 with horizontal force equilibrium.
Tu(B) = PpH PaH (2.2)
In the above expression, PpH and PaH are the horizontal components of the passive and active
thrusts, which can be estimated using the earth pressure coefficients reported by Caquot and
Kerisel (1948). The parameter, R0v in Eq. 2.1 is given as (Dickin and Leung 1985),
C0v + 1
R0 v = (2.3)
C0 v + H / h
Where C0v = 19 for dense sand and 14 for loose sands.
Neely et al. (1973) performed laboratory tests on anchor plates in dry sand and ultimate
resistances of these plates were examined using both limit analysis and the method of stress
characteristics. Results of tests on rigid anchor plates in terms of Mq, a dimensionless force
coefficient, were expressed as M q = Tu / Bh 2 . Mq varied strongly with geometry and for this, a
dimensionless parameter known as shape factor (Sf) was introduced that depended on B/h and
H/h ratios.
Das and Seeley (1975) conducted several laboratory model tests to determine the ultimate
pullout resistance of shallow vertical anchors and suggested a simple semi-empirical relation for
the pullout resistance in a non-dimensional form as the ratio of Tu /Bh2 for square and
rectangular anchors. Ultimate pull out capacity for a single anchor was expressed by the
following relation.
( )
Tu = 4.59 10 5 S 3.22 (H / h ) Bh 2
n
(2.4)
Where S is the shape factor which is a function of H/h and is angle of soil friction in degrees.
The value of n varies linearly from 1.8, for B/h = 1 to about 1.68 for B/h = 5.
The capacity of deeper vertical anchors in medium dense sand was investigated by Akinmusuru
(1978) for square, circular and rectangular anchors. On the basis of experimental findings, the
variation of Tu/h3, a non-dimensional anchor load at ultimate failure with , a non-dimensional
embedment coefficient ( = H/h) was presented in the form of a chart, where Tu is ultimate
pullout load for an anchor length of 10h. Akinmusuru (1978) clearly defined the critical
embedment depth as the one corresponding to = 6.5.
Dickin and Leung (1983,1985) conducted both centrifuge and conventional chamber tests and
reported very thorough investigations on the behavior of vertical square and rectangular anchors
in dense sand. The variations of breakout factor Nq, and the force coefficient Mq with
embedment ratio were separately reported in the form of a chart with Nq = Tu/BhH and Mq =
Tu/Bh2.The results obtained by them suggested potentially serious over predictions of pull-out
resistance and underestimations of the failure displacements. Such errors arose due to the
characteristic stress- dependent behavior of dense soils.
Hoshiya & Mandal (1984) investigated the capacity of square and rectangular anchors in loose
sand. The size of box used for testing was very small, which was only 300mm wide and 400mm
long. This was likely to introduce edge effects into the results. They concluded that, anchor
break-out factor increased with depth up to a certain embedment ratio before reaching a constant
value thereafter.
Naser (2006) carried out theoretical as well as experimental studies on the ultimate pullout
capacity of a block anchor of concrete embedded in sand and observed that, anchor thickness
contributed to the pullout capacity through base friction forces. This contribution was not
significant as compared to the passive resistance. Uplifting and tilting of the block was also
observed.
2.2.2 Theoretical Investigations
Terzaghi (1943) evaluated the resistance of vertical strip anchor plates assuming Rankine (1857)
states of passive and active pressures. This approach was adopted in the British civil engineering
code of practice. The net resistance of a vertical anchor, Tu is given as (Pp- Pa), where Pp and Pa
are the passive and active thrusts (kN/m) acting on the anchor plate.
Teng (1962) estimated holding capacity of a vertical (strip) plate anchor embedded in granular
soil at relatively shallow depth ( h/H 1/3 to 1/2), based on Rankines (1857) theory of lateral
earth pressures. He obtained the expression for ultimate holding capacity as Tu = Pp Pa , where
Pp and Pa are the passive and active pressure thrust (kN/m) acting on anchor plate. For anchors
with limited width B, the frictional resistance developed along the vertical faces of the failure
surface was also taken into account and the expression for ultimate holding capacity (Tu) was
reported as,
1
[ ]
Tu = TuB + K 0 K p + H K a H 3 tan
3
(2.5)
Where, K0 is the coefficient of earth pressure at rest, Ka is the Rankines active earth pressure
coefficient and Kp is the Rankines passive earth pressure coefficient.
In case of shallow strip anchors, Meyerhof (1968, 1973) used the passive and active coefficients
of earth pressure proposed by Caqout and Kerisel (1948) and Sokolovskii (1965) and proposed
the following simple relationship for ultimate holding capacity per unit width of a continuous
(strip) vertical plate anchor.
Tu = 1 / 2H 2 K b (2.6)
Where, Kb is the pullout coefficient that can be obtained from a graph using soil friction angle.
Neely et al. (1973) determined the theoretical resistance of continuous (strip) vertical anchor
plates in cohesion-less soils by two methods. In the first method, failure surface was assumed to
be consisting of a logarithmic spirals and its tangent inclined at (45-/2) to the horizontal as
shown in Fig. 2.2. Soil above top of the anchor was considered to act as a simple surcharge, q
{(H-h)} and therefore, the method was termed as surcharge method.
O q = (H h)
H (45-/2) (45-/2) D
h Straight line
C
Logarithmic spiral
A
Fig. 2.2 Surcharge method (Neely et al., 1973)
Shearing resistance of the soil above the anchor top is ignored when H/h is small; therefore the
method was subsequently modified by considering shear strength above the top of anchor plate
when H/h is considerable and was defined as the Equivalent Free Surface method. The assumed
failure surface in soil (Fig. 2.3) is an arc of logarithmic spiral with pole at top of the wall. OB is a
straight line which is an equivalent free surface. The shearing resistance of upper layers of soil
was included in the calculation by making use of equivalent free surface concept proposed by
Meyerhof (1951) in connection with bearing capacity of shallow foundations. The normal and
shear stresses along OB (p0 and s0, respectively) were calculated using Rankine (1857) active
stresses on the vertical surface, OA above the top of the anchor plate as shown in Fig. 2.3.
A B
p0 =
s0= s0= p0 tan m
O C
H
90-
h
Logarithmic spiral
Rowe and Davis (1982a, 1982b) reported a two-dimensional finite element analysis
incorporating an elasto-plastic soil model. For a continuous vertical plate anchor assumed to be
thin and perfectly rigid, the resistance is as given by the following expression.
M q = F R RR RK (2.7)
Where, F is the capacity factor of a smooth anchor resting on soil which deforms plastically at a
constant volume ( = 0), with coefficient of earth pressure at rest, K0 = 1 and R, RR and RK are
correction factors for the effects of sand dilatancy, anchor plate roughness and initial stress state
respectively. The theoretical data was presented in the form of design charts.
Finite element method is also used by various researchers such as Vemeer and Sutjiadi (1985),
Tagaya et al. (1983, 1988), Dickin and King (1997) and Sakai and Tanaka (1998). Unfortunately,
only limited results are available from these studies. Tagaya et al. (1983, 1988) reported two-
dimensional plane strain and axi-symmetric finite element analyses using the constitutive law of
Lade and Duncan (1975).
Upper and lower bound limit analyses are also reported by Murray and Geddes (1987, 1989),
Basudhar and Singh (1994) and Smith (1998) to estimate the capacity of vertical strip anchor
plates. Basudhar and Singh (1994) obtained estimates with a generalized lower bound procedure
based on finite element method and non-linear programming similar to that of Sloan (1988). The
solutions proposed by Murray and Geddes (1987, 1989) are based on kinematically admissible
failure mechanisms (upper bound).
Kumar and Rao (2004) extended the concept of the equivalent free surface to determine the
seismic horizontal pullout capacity of shallow vertical strip plate anchors buried in sand using
the method of stress characteristics and the results were expressed in the form of non-
dimensional charts.
Merifield et al. (2006) presented the results of a rigorous numerical work (finite elements
coupled with upper and lower bound limit analyses) to estimate the ultimate pullout capacity Tu,
for vertical anchor plate in the cohesion-less material. For comparison purposes, numerical and
theoretical results of the break-out factor were presented in the form analogous to Terzaghis
(1943) equation of the bearing capacity of shallow foundations.
Tu = HN (2.8)
Where, N is the anchor break-out factor that can be obtained from a graph using soil friction
angle.
Fig. 2.4 Failure modes and zones of plastic yielding for rough vertical anchors in
cohesion-less soils (=20), Merifield et al. (2006)
10
The failure mode (Fig. 2.4) reported by Merifield et al. (2006) for vertical anchors indicates that
the soil retained behind the anchor can significantly affect the estimated capacity of shallow
anchors. This is particularly the case for loose sands, where the development of a significant
active zone behind the anchor is observed. Changing the interface roughness from perfectly
rough to perfectly smooth can lead to a reduction in the anchor capacity by as much as 65%.
Where, Ft, Fb and Fs are the effective friction forces at the top, bottom and at two side of the
block, N is the normal force, Pph is the effective horizontal passive thrust and Pah is the effective
horizontal active thrust. For Coulomb(1776) and log spiral theories, Fb = 0 (as N=0). Pullout
capacity of block anchor with Rankines theory (1857), corrected for the 3-D effect with the
contribution of friction, showed a close agreement with experimental results.
Goel et al. (2006) estimated the breakout resistance of inclined anchors in sand using limit
equilibrium approach for different soil friction angles with varying relative depth ratio and
anchor inclination. They observed that, the resistance increased continuously with the inclination
11
of the anchor with respect to vertical. Hanna et al. (2007) has also reported analytical studies
using the limit equilibrium technique to predict the pullout capacity of shallow helical and plate
anchors in sand based on the failure mechanism observed in laboratory testing. They proposed an
empirical expression to determine the critical depth of anchors, which separated the shallow from
the deep.
Kumar and Kouzer (2008) used upper bound limit analysis and finite elements for the estimation
of vertical uplift capacity of horizontal anchors. The collapse load was expressed in terms of
non-dimensional uplift factor which was found to be increasing continuously with increase in
both embedment ratio and friction angle of sand.
In majority of the earlier studies, a failure mechanism based on the experimental observations
was assumed and the pullout capacity was then determined by considering equilibrium of the soil
mass retained by the anchor plate.
In case of earth retaining structures such as sheet piles, retaining walls, wing walls, abutments
and bulkheads, which are very common in engineering practise. While retaining earth, these
structures including vertical plate anchors are subjected to lateral earth pressures. Vertical plate
anchors mainly derive the pullout capacity due to active and passive earth pressure thrusts
therefore literature review on methods for determination of active and passive earth pressure
coefficient is also included here.
Coulomb (1776) and Rankine (1857) proposed methods for the estimation of active and passive
earth pressure on retaining walls based on the assumption of a plane failure surface. Caquot and
Kerisel (1948) and Kerisel & Absi (1990) proposed a log spiral mechanism and presented their
results in the form of charts. Lancellotta (2002) provided an analytical solution for the active
earth pressure coefficients, based on the lower bound theorem of plasticity. Soubra and Macuh
(2002) used an approach based on rotational log-spiral failure mechanism with the upper-bound
theorem of limit analysis for the analysis of active earth pressure.
Terzaghi (1943) suggested a failure mechanism, in which the failure surface consisted of a log
spiral originating from the wall base, followed by a tangent, meeting the ground surface at an
12
angle corresponding to Rankine's active and passive state. Several other research workers have
adopted this failure mechanism.
Janbu (1957), Rahardjo and Fredlund (1983), Chen and Li (1998) used the method of slices for
computing active pressure coefficients in respect of a cohesion-less soil by considering soil mass
in the state of limit equilibrium. Ching et al. (1994) used discrete element analysis for active and
passive pressure distribution on the retaining wall.
Sheilds and Tolunay (1973), Basudhar and Madhav (1980), and Kumar and Rao (1997) used
method of slices for computing passive pressure coefficients in respect of a cohesion-less soil by
considering soil mass in a state of limit equilibrium.
Morgenstern and Eisenstein (1970) compared the values of passive earth pressure coefficient Kp
calculated with the theories proposed by Caquot and Kerisel (1948), Hansen (1953), Janbu
(1957) and Sokolovski (1965).They concluded that with the assumption of a plane failure
surface, Coulombs (1776) theory overestimated the passive resistance.
Soubra and Macuh (2002) used an approach based on rotational log-spiral failure mechanism
with the upper-bound theorem of limit analysis for the analysis of passive earth pressures.
From the review of literature, it is observed that, Ktter (1903) was first to derive the slip line
equations for the case of plane deformations in the granular soil medium. He proposed a
differential equation for evaluating the distribution of soil reactive pressure on a failure surface,
based on Coulombs (1776) law of failure and stress equilibrium equations.
Using these equations Balla (1961), Saeedy (1987) and Ghali and Hanna (1994) determined the
distribution of shear stress along the failure surface to obtain the uplift capacity of horizontal
plate anchors. Ktters (1903) equation which is valid for plane strain condition was successfully
used for the analysis of a retaining wall by Dewaikar and Halkude (2002) and for the evaluation
of bearing capacity factor, N by Dewaikar and Mohapatro (2003).
13
Chapter 3
Derivation of Ktters Equation
3.1 General
Ktters (1903) equation gives solution for determining the distribution of the soil reaction
pressure, p exerted by cohesion-less soil media along the arc of the failure surface (Fig. 3.1) and
for the passive state of equilibrium, it is as given,
dp d
+ 2 p tan = sin ( + )s (3.1)
ds ds
Where,
dp = differential reaction pressure on the failure surface
ds = differential length of failure surf ace
= angle of soil internal friction
d = differential angle
= angle of failure plane formed by inclination of tangent at the point of interest with the
horizontal.
x
+ xy =0 (3.2a)
Y
y xy
+ + = 0 (3.2b)
14
ds
Slip
90 Curved
failure surface
dp
Tangent Normal
Y y
A xy
B
x
/2 + ds1
1
+
C
2 2 ds2
Fig. 3.2 Element considered in the derivation of expressions for stress at a point
15
Where, x and y are the normal (co-ordinate) stresses, xy is the shear stress and is the soil
unit weight. Referring to Fig. 3.3, the orientation of major principal stress 1 and minor principal
stress, 3 with respect to x and y is obtained. The direction of x makes an angle,
1 1
x = ( 1 + 3 ) + ( 1 3 ) sin( 2 ) (3.3a)
2 2
1 1
y = ( 1 + 3 ) ( 1 3 ) sin( 2 ) (3.3b)
2 2
1
xy = ( 1 3 ) cos( 2 ) (3.3c)
2
In which, is the angle made by slip line, ds1 , with vertical as measured in clockwise direction.
Since the soil in the failure zone is in he passive state of equilibrium, the relationship between
1 and 3 is as given below.
1 + sin
1 = 3 (3.4)
1 sin
1 3
Letting, m = =
(1 + sin ) (1 sin )
Substitution of Eq. 3.6 into Eq. 3.2, leads to the following set of equations.
16
/4 ( /2)
x
1
xy
xy y
/4 ( /2)
m
+ sin sin( 2 ) m + cos( 2 ) m
(3.7a)
+ 2 m sin cos( 2 ) sin( 2 ) =0
And
m
sin sin( 2 ) m cos( 2 ) m
(3.7b)
2 m sin cos( 2 ) + sin( 2 ) + = 0
The infinitesimal element of the slip line, or ds1 or ds2 makes an angle, / 4 / 2, with the
direction of major principal stress, 1, as shown in Fig. 3.1. In order to obtain stresses acting on
the slip line, it is necessary to change the co-ordinate system using the following relations.
ds1 ds 2
= + (3.8a)
ds1 d ds 2 d
And
ds1 ds2
= + (3.8b)
ds1 d ds2 d
17
The vectorial resolution of elemental lengths, dx and dy in the direction of slip lines is shown in
Fig. 3.4. Referring to this figure, the following relationships are obtained.
dx ds1 ds2
= =
sin( )
sin sin +
2 2
Therefore,
ds1 sin( ) ds 2 cos
= , and = (3.9a)
dx cos dx cos
dy ds2 ds1
= =
sin
sin sin ( )
2 2
dx
()
(90+)
-ds1 ds1
ds 2
dy
-ds2
(a) (b)
Fig. 3.4 Vectorial resolution of elemental lengths dx and dy in the direction of slip lines
Therefore,
ds1 cos( ) ds sin
= and 2 = (3.9b)
dy cos dy cos
Substitution of Eq. 3.9 into Eq. 3.8, leads to the following expressions.
1
= sin( ) + cos (3.10a)
x cos s1 s2
18
1
And = cos( ) sin (3.10b)
y cos s1 s2
And
Solving Eq. 3.11a and 3.11b simultaneously, following equations are obtained.
m
cos 2 m sin = cos( ) (3.12a)
s1 s1
m
And cos + 2 m sin = sin (3.12b)
s 2 s 2
The sliding surface makes an angle,( / 4 / 2 )with the direction of major principal stress, 1
as shown in Fig. 3.2 and thus, the normal stress, s acting on sliding surface makes an angle,
1 1
s = ( 1 + 3 ) + ( 1 3 ) cos 2 + (3.13)
2 2 4 2
19
p
s
3
(/4+/2)
1
Fig. 3.5 Stresses acting on infinite length ds along slip line
With the substitution of Eq. 3.5, the above equation is transformed to,
s = m + m sin cos( / 2 + )
or s = m cos 2 (3.14)
The resultant stress, p, which is acting on the sliding surface, is located at an angle, to the
normal stress, s (Fig. 3.4). Hence, s can be expressed as,
s = p cos
Substituting the above relation in Eq. 3.14, the following expression for p is obtained.
p = m cos (3.15)
Kinematically admissible sliding surface is one that makes an angle, / 4 / 2 with the major
principal stress, 1 , as measured in clockwise direction from 1 . Therefore, with the substitution
of Eq. 3.15, Eq. 3.12b is transformed to the following form.
p
+ 2 p tan = sin (3.16)
s s
In general, there are two slip lines through every point in the medium. These slip lines intersect
mutually at an acute of (90 ).
Referring to Fig. 3.6, following relationship for the curved failure surface is obtained.
(90 o a) = (90 o + ) (3.17)
Therefore,
= (a + ) and = a (3.18)
20
Substituting the values of and in Eq. 3.16, the following equation is obtained.
dp da
+ 2 p tan = sin( a + ) (3.19)
ds ds
The above equation is identical to Ktters (1903) equation. Matsuo (1967) used the same
equation for the analysis of uplift resistance of footing.
Y
B X
W
slip line
(90-)
curved failure surface H H
(90+)
This equation can be used to evaluate the resultant shearing resistance acting on the curved
sliding surface for plane strain conditions in passive state of equilibrium for cohesion-less soils.
However, this equation is valid also for cohesive soils (Jacky 1996).
The applicability of Ktters (1903) equation to the analysis of limit equilibrium problems is
demonstrated with respect to following cases.
Case I: retaining wall with plane failure surface (Coulombs, 1776 mechanism) by Dewaikar and
Halkude (2002a)
Case II: open cuts by Dewaikar and Halkude (2002b)
Case III: bearing capacity factors, Nq and Nc by Dewaikar and Mohapatra (2003)
In the next chapter, active thrust on a vertical retaining wall with horizontal cohesion-less
backfill is discussed.
21
Chapter 4
Active thrust on a vertical retaining wall with horizontal
cohesion-less backfill
4.1 Introduction
Earth retaining structures such as sheet piles, retaining walls, wing walls, abutments and
bulkheads are very common in engineering practices. While retaining earth, these structures are
subjected to lateral earth pressures.
From the review of literature, it is observed that, Ktters (1903) equation has been employed
(Balla, A., 1961 and Matsuo, M., 1967) to evaluate soil shearing resistance on a curved failure
surface. Dewaikar and Mohapatro (2003) used Ktters (1903) equation for computation of
bearing capacity factor, N for shallow foundations.
In the proposed investigations, a method is developed using Ktters (1903) equation for the
computation of active thrust and its point of application for a vertical wall, retaining horizontal
cohesion-less backfill, using mechanism suggested by Terzaghi (1943). The distribution of soil
reaction on the failure surface is also evaluated.
Ktters (1903) equation is used to evaluate vertical and horizontal components of the soil
reaction on the failure surface. This equation is valid for a plane strain condition.
In Fig. 4.1, is shown a vertical retaining wall DE, with a horizontal cohesion-less backfill. The
failure surface consists of log spiral EA, that originates from the wall base, with tangent, AB
meeting the ground surface at an angle, 45+ /2, where, is the angle of soil friction. At A, there
is a conjugate failure plane AD, passing through the wall top. Thus, as seen from the figure,
ABD is a Rankine zone and pole of the log spiral lies on the line AD or its extension and this is
also shown in Fig. 4.2.
22
O
m
ro
d
D C B
45+/2 45+/2
Straight line
r failure surface
r1
H
v
A
Tangent to curve
J
Log spiral
G
E
Fig. 4.1 Retaining wall with a horizontal cohesion-less backfill - failure mechanism
O F
+Yo -X
X0
m B
D C
45+/2 45+/2
Straight line
r Ypa failure surface
v
H C G of area AOE
A
Pa
h
Log spiral
+Y +Xo
23
2 DC
m
3
D C
W ACD
2 AC
3
Ypa
P aV W ADE
H1
P aH
A
m
h
RH
E RV
R
24
In Fig. 4.3, line AC represents the Rankine wall and force, H1 as described above, is the force
exerted on this wall by the backfill it retains. With this consideration and also considering that
pole of the log spiral lies above the top of wall, on line AD, the dispositions of various forces are
shown in Fig. 4.3.
And
H sin 135 +
2
OE = r1 =
sin m
The initial radius, OA = r0 of the log spiral is given as
OE
r0 = m tan (4.2)
e
Similarly,
AD = r0 OD
4.2.1 Computation of soil reaction on the failure surface
Ktters (1903) equation basically refers to the distribution of reactive pressure on the failure
surface, in a cohesion-less soil medium and for the active state of equilibrium (Fig. 4.4) is as
given below.
25
dp d
2 p tan = sin( ) (4.3)
ds ds
In which,
dp = differential reactive pressure on the failure surface
ds = differential length of arc of failure surface
= angle of soil internal friction
d = differential angle
= unit weight of soil and,
= inclination of the tangent at the point of interest with the horizontal
O
Slip
Curved failure surface
d
Tangent
ds
dp
Normal
Fig. 4.4 Reactive pressure distribution on the failure surface for active case
The failure surface as shown in Fig 4.1 has two parts; EA, which is curved and AB, which is a
straight line. Ktters (1903) equation is used to obtain the distribution of reactive pressure on
both these parts.
For a plane failure surface, d/ds=0, and Eq. 4.3 takes the following form.
dp
= sin( ) (4.4)
ds
Integration of the above equation gives,
p = sin ( )s + C 1 (4.5)
Eq. 4.5, gives distribution of reaction on the plane failure surface, AB. The distance, s is
measured from point B (Fig. 4.1). The integration constant, C1 is evaluated from the boundary
26
condition that pressure, p is zero at point B, which corresponds to s = 0. With this condition, C1
is zero and Eq. 4.5 becomes,
p = sin( )s (4.6)
In the above equation, = 45 + /2 and with this substitution one obtains,
p = sin 45 s (4.7)
2
At point A (Fig. 4.1), p is given as,
p = sin ( )AB (4.8)
4.2.3 Computation of vertical and horizontal components of reaction on curved failure
surface EA
In Eq. 4.3, let ( ) = t
Therefore d = dt
Then Eq. 4.3 is written as,
dp d
2 p tan = sin t
ds ds
Multiplying throughout by ds/d and rearranging, the above equation becomes,
dp ds
2 p tan = sin t (4.9)
d d
From Fig. 4.1, the angle, is evaluated in terms of log spiral angle as given below.
= 45 +
2
Therefore, d = d
For a log spiral,
ds = rd sec
ds
= rdsec and = r sec
d
Now, = 45 + = + t
2
From which,
= 45 t
2
27
Or
dp
2 p tan = r0 sec sin te( L t ) tan (4.10)
dt
Where,
L = 45
2
The solution of above differential equation is obtained as,
p = .r0 sec e (3 L 2 ) tan [ p1 + C 2 ] (4.11)
Where,
3 tan sin ( L ) cos( L )
p1 = e 3( L ) tan (4.12)
(1 + 9 tan 2 )
C2 is the constant of integration and it is obtained from the boundary condition that at point, A
(Fig. 4.1) reaction is as calculated from Eq. 4.8. Then C2 is evaluated as,
3 tan sin L cos L sin L
C2 = e 3 L tan (4.13)
1 + 9 tan 2 r0 sec
With the above value of C2, pressure distribution on curved surface is obtained as,
The resultant soil reaction, R (Fig. 4.3) on the failure surface is given as,
R = p.ds (4.15)
28
Or
m
RV = pr e
0
tan
sec cos( L ).d (4.17)
0
After substituting the value of p from Eq. 4.14 and integrating, RV is obtained in the following
form.
RV = RV1 + RV2 + RV3 (4.17a)
In which, the values of parameters RV1, RV2 and RV3 are as per the expressions given below.
3 tan e
2 tan m
[tan sin 2( L m ) + cos 2( L m )]
sec 2 [tan sin 2 + cos 2 ]
L L
.r0 sec 1
2
{ }
2
RV1 = + e 2 tan m 1
(
4 1 + 9 tan tan
2
)
(4.17b)
2 tan m tan cos 2( L m )
1 e sin 2( )
+
sec
2 L m
[tan cos 2 L sin 2 L ]
RV 3 =
.r0 sin L AD e
tan m
[tan cos( L m ) + sin ( L m )] (4.17d)
sec [tan cos L + sin L ]
Similarly, the horizontal component, RH (Fig. 4.3) of soil reaction is given as,
m
RH = pr e
0
tan
sec sin ( L ).d (4.18)
0
After substituting the value of p from Eq. 4.14 and integrating, RH is obtained as,
29
The values of parameters RH1, RH2, RH3, RH4 and RH5 are as given below.
sec 2 e 2 tan m 1 ( )
3 .r 2
RH1 =
(
0
4 1 + 9 tan 2 tan
e
)
2 tan m
[tan cos 2 ( L m ) sin 2 ( L m )]
[tan cos 2 L sin 2 L ]
(4.18b)
RH 5 =
sec [ tan sin L + cos L ] (4.18f)
RH + H1
Pa = (4.22)
cos
It may be noted that, both Eqs. 4.20 and 4.22 give the magnitude of unknown thrust, Pa. These
two equations will yield the same and unique value of Pa only when the equilibrium conditions
correspond to those at failure, which are uniquely defined by a characteristic value of V and this
value can be determined by trial and error procedure.
In this procedure, first a trial value of V is assumed and corresponding weight of trial failure
wedge, EACD (Fig. 4.3) is computed. Using Eqs. 4.17 and 4.18, magnitudes of vertical and
horizontal components of resultant soil reaction, RV and RH are computed and from Eqs. 4.20 and
4.22, values of Pa are determined. If the trial value of V is equal to its characteristic value
corresponding to the failure condition, the two computed values of Pa will be the same and
otherwise, they will be different.
O2
Pole of log spiral
O1
O
m
D C B
45+ /2 45+ /2
Final failure surface
Trial 1
Trial 2
H
Straight line
A
Pa v
Log spiral
Fig. 4.5 Trial procedure for locating pole of the log spiral
31
For various trial values of V , computations are carried out till the convergence is reached to a
specified (third) decimal accuracy.
Thus, in this method of analysis, the unique failure surface (Fig. 4.5) is identified by locating the
pole of log spiral in such a manner that, force equilibrium condition of failure wedge, EACD is
satisfied. This approach is different from other analyses in which, Pa is obtained from the
consideration of its maximum value.
2Pa
Ka = (4.23)
H 2
Values of active earth pressure coefficient, Ka are computed for different values of soil friction
angle, and angle of wall friction, .
These calculations are performed with reference to Fig. 4.2. Axis, X0 is taken along the line that
joins pole, O of log spiral to the wall base. Axis, Y0 is perpendicular to the axis, X0 and passes
through pole of the log spiral. With respect to these axes, coordinates of the centroid of area
inscribed in log spiral are calculated as,
32
Moment equilibrium condition is now used to compute the point of application of active thrust
by considering moments of forces and reactions about the pole of log spiral.
2
WACD (OF + 3 DC) + WADE X
PaH (Ypa + FD ) = (4.28)
2
+ H 1 AC + FD PaV OF
3
In which, the terms on the right hand side of the above expression represent moment of the force
H1, moment of weight of soil in the failure wedge, EACD and moment due vertical component
of active thrust, PaV about the pole, O. The terms on the left hand side of the above expression
represent moment due horizontal component active thrust, PaH about the pole, O. From the above
equation, YPa (which is the distance of point of application of Pa from wall top), is obtained as
2
WACD (OF + 3 DC) + WADE X
1
Ypa = (4.29)
PaH 2
+ H 1 AC + FD PaV OF-PaH FD
3
4.6 Discussion
The basic purpose of this analysis was to compute active pressure coefficient, Ka, location of
point of application of active thrust and study their variation with respect to the parameters
involved in the analysis. It was found convenient to express the height, h of point of application
of active thrust from the wall base in terms of its ratio with respect to height, H of the retaining
wall, in a non-dimensional form (Hr = h/H). Values of Hr and active pressure coefficient, Ka are
computed for various combinations of soil friction angle, and angle of wall friction, .
One distinguishing feature of the proposed method is its ability to compute the point of
application of active thrust using moment equilibrium. This has not been possible with other
33
existing methods. In Table 4.1, computed values of Hr are shown. They vary over a very narrow
range, from 0.333 (for = 20 and = 0) to 0.331 (for = 40 and = 30).
45+ /2 0.0
45+ /2
= 2
= 1kN/m3
= 45
= 30
Pressure distribution
along straight line
H failure surface
Pa 0.317 kN/m2
Pressure distribution
along curved failure
surface (arc of
a log spiral)
0.454 kN/m2
Another distinguishing feature of the proposed analysis is its capability to predict the distribution
of reactive pressure on the failure surface using Ktters (1903) equation. This is shown in
Fig. 4.6 for = 45 and = 30. The pressure distribution varies linearly over the straight part of
the failure surface followed by curvilinear variation over the log spiral part with a maximum
ordinate at the wall base.
34
35
In Tables 4.2, a comparison of Ka values as obtained by the proposed method with other solutions
is shown. For = 0 there is a perfect matching with Coulombs solution. With increasing and
, Coulombs values become higher than the proposed values and the difference increases with
increasing for a given . However, for a practical situation ( 40 and / between1/3 to 2/3)
the maximum difference is 11.11%.
There is a very close agreement between the Ka values of the proposed method and those
reported by Caquot and Kerisel (1948). The maximum difference is of the order of 4%.
The values reported by Lancellotta (2002) are also in a very close agreement with the proposed
solution with a maximum difference of only 1.1%.
The values of Ka reported by Soubra and Macuh (2002) are higher than the proposed values; the
difference increasing with higher and values, with a maximum of 12.2%.
It may be noted that, for the failure mechanism consisting of log spiral and its tangent, which is
adopted in the proposed analysis, the Ka values are unique; since they are evaluated from the
identification of a unique failure surface that satisfies force equilibrium condition.
The proposed method also enables the computation of point of application of active thrust using
moment equilibrium and reactive pressure distribution on the failure surface.
In the next chapter, passive thrust on a vertical retaining wall with horizontal cohesion-less
backfill is discussed.
36
Chapter 5
Passive thrust on a vertical retaining wall with horizontal
cohesion-less backfill
5.1. Introduction
Earth retaining structures such as sheet piles, retaining walls, wing walls, abutments and
bulkheads are very common in engineering practices. While retaining earth, these structures are
subjected to lateral earth pressures.
From the review of literature, it is observed that, Ktters (1903) equation has been employed
(Balla, A., 1961 and Matsuo, M., 1967) to evaluate soil shearing resistance on a curved failure
surface. Dewaikar and Mohapatro (2003) used Ktters (1903) equation for computation of
bearing capacity factor, N for shallow foundations.
In the proposed investigations, a method is developed using Ktters (1903) equation for the
computation of passive thrust and its point of application for a vertical wall, retaining horizontal
cohesion-less backfill, using the failure mechanism suggested by Terzaghi (1943). The distribution of
soil reaction on the failure surface is also evaluated.
5.2 Proposed method
r K r0
O
d r1
m D C B
45- /2 45- /2
H
v Straight line
failure surface
r0 A
Fig. 5.1 Retaining wall with a horizontal cohesion-less backfill - failure mechanism
37
Fig. 5.1 shows a vertical retaining wall DE, with a horizontal cohesion-less backfill. The failure
surface consists of log spiral EA, that originates from wall base, with tangent, AB meeting the
ground surface at an angle, (45- /2), where, is the angle of soil internal friction. At A, there
is a conjugate failure plane AD, passing through the wall top. Thus, as seen from the figure,
ABD is a passive Rankine zone and pole of the log spiral lies on the line AD or its extension and
this is also shown in Figs. 5.2(a) and 5.2(b).
From Fig. 5.1, the following information is generated.
H = height of the retaining wall
= inclination of the tangent to the log spiral at point G with the horizontal
= spiral angle measured from the starting radius
r0 = starting radius of the log spiral at the wall base (at = 0)
r = radius of log spiral at point G corresponding to the spiral angle
m = maximum spiral angle
r1 = radius of the maximum spiral angle at = m
v = angle between vertical face of the wall and the starting radius r0
X r1 K r0
Pole of log spiral O F
-X
m X0
D C B
+Y0
Ypp Pp 45- /2 45- /2
Straight line
r0 H v failure surface
h A
+Y +X 0
Fig. 5.2 (a) Failure surface adopted in the proposed analysis with
pole located above the wall top
38
Figs. 5.2(a) and 5.2(b) show the location of pole of the log spiral when it is located above and
below the wall top respectively.
From Fig. 5.2(b), the following additional information is generated.
A = angle between vertical face of the wall and line OD when pole is located below
the wall top
S = angle between the radius r0 and line OD when pole is located below the wall top
D C B
A O 45- /2 r1 45- /2
Ypp Pp -X
F s
H m
v
h
r0 X0
A
Straight line failure surface
E
C G of area Log spiral failure surface
AOE
+X 0 +Y
Fig. 5.2 (b) Failure surface adopted in the proposed analysis with
pole located below the wall top
X
F
Pole O
2
m DC
D 3 C
WACD
PpV 2
Ypp AC
WADE 3
H H1
PpH
v
h 90 A
RH
E
R
RV
Fig. 5.3 (a) Free body diagram of failure wedge EACD, with pole above the wall top
39
X
2
DC
D 3 C
PpV WACD
Pole
F s O 45- /2
A
Ypp 2 AC
H WODE 3
PpH m
WOEA H1
v
h
90 A
RH
E
RV
R
Fig. 5.3 (b) Free body diagram of failure wedge EACD, with pole below the wall top
From Figs. 5.3(a) and 5.3(b), which show free body diagrams of failure wedge EABCD, the
following information is generated.
PpH, PpV = horizontal and vertical components of resultant passive thrust, Pp
RH, RV = horizontal and vertical components of resultant soil reaction acting on the
curved part of the failure surface
H1 = active thrust exerted by the backfill on the Rankine wall AC
WACD = weight of soil in the failure wedge, forming a part of the Rankine zone
WADE = weight of soil in the zone, EAD of the failure wedge, EABCD
In Fig 5.3(a), line AC represents the Rankine wall and force, H1 as described above, is the force
exerted on this wall by the backfill it retains. With this consideration and also considering that
pole of the log spiral lies above the wall top on line AD, the dispositions of various forces are
shown in the same figure.
In Fig. 5.3(b), which refers to location of pole, O below the wall top, in addition to forces
mentioned earlier, forces, WODE and WOEA together represent the weight of portion EAD of the
failure wedge, EACD, as shown in the same figure.
40
And
AD = OA OD
From which, the initial radius, OE = r0, of the log spiral is given as,
H sin A
OE = r0 =
sin S
And
Hsin v
OD =
sin S
41
ds
Slip
90 Curved
failure surface
dp
Tangent Normal
Fig. 5.4 Reactive pressure distribution on the failure surface for passive case
Ktters (1903) equation basically refers to the distribution of reactive pressure on the failure
surface, in a cohesion-less soil medium and for the passive state of equilibrium (Fig. 5.4), it is as
given below.
dp d
+ 2 ptan = sin ( + ) (5.3)
ds ds
In which,
dp = differential reactive pressure on the failure surface
ds = differential length of arc of failure surface
= angle of soil internal friction
d = differential angle
= unit weight of soil and,
= inclination of the tangent at the point of interest with the horizontal
42
The failure surface as shown in Fig. 5.1 has two parts; EA, which is curved and AB, which is a
straight line. Ktters (1903) equation is used to obtain the distribution of reactive pressure on
both these parts.
For a plane failure surface, d/ds = 0 and Eq. 5.3 takes the following form.
dp
= sin ( + ) (5.4)
ds
Integration of the above equation gives,
p = sin( + )s + C1 (5.5)
Eq. 5.5 gives distribution of reaction on the plane failure surface, AB. The distance, s is
measured from point B (Fig. 5.1). The integration constant, C1 is evaluated from the boundary
condition that, pressure, p is zero at point B, which corresponds to s = 0. With this condition, C1
is zero and Eq. 5.5 becomes,
p = sin( + )s (5.6)
In the above equation, = 45 - /2 and with this substitution one obtains,
p = sin 45 + s (5.7)
2
At point A (Fig. 5.1), p is given as,
p = sin( + )AB (5.8)
The distance, AB depends upon the location of pole of log spiral, i.e., whether it lies below or
above the wall top.
Multiplying Eq. 5.3 throughout by ds/d and rearranging, the following equation is obtained.
dp ds
+ 2 p tan = sin t (5.9)
d d
In which, t = ( + ) , with d = dt (5.10)
From the geometry of log spiral,
43
ds
= r sec (5.11)
d
From Fig. 5.1, the angle, is evaluated in terms of log spiral angle, as given below.
= (90 V )
With (90 V ) = L , is written as,
= L and d = d (5.12)
= t + L (5.13)
After making necessary substitutions in Eq. 5.9 the following equation is obtained.
dp ds
+ 2 ptan = sint (5.14)
dt d
Using Eq. 5.11, the above equation is written as,
dp
+ 2 ptan = sintrsec (5.15)
dt
With r = r0e tan the above equation is transformed to,
dp
+ 2 ptan = sin tr0 e tan sec (5.16)
dt
Substitution of the value of from Eq. 5.13, in Eq. 5.16 gives the following equation.
dp
+ 2 p tan = sec r0 e (t + L ) tan sin t (5.17)
dt
The solution of above differential equation is obtained as,
C2 is the constant of integration and it is obtained from the boundary condition that, at
Point A (Fig. 5.1) with = m , reaction is as calculated from Eq. 5.8.
44
K cos sin 4 + 2 1 + 9 tan
(3 tan )( m L + )
2
( )
e
C2 = (5.20)
( 2
)
1 + 9 tan 3 tan sin ( m L + )
cos( m L + )
With the above value of C2, pressure distribution on the curved surface is given as,
tan (3 m 2 )
r0 Ksin + e
4 2
r sec e tan 3 tan sin ( + )
p = + 0
L
(5.21)
( )
1 + 9 tan 2 cos( L + )
r0 sec e
tan (3 m 2 )
3 tan sin ( m L + )
(
1 + 9 tan 2 )
cos( m L + )
Where, K is the parameter indicating location of the pole of the log spiral along line AO in terms
of radius r0 measured from point D (Fig. 5.1).
The expression for K is given as,
OD / r
K = 1 m tan 0 , for pole above wall top (Fig. 5.2 (a))
e
And
OD / r
K = 1 + m tan 0 , for pole below wall top (Fig. 5.2 (b))
e
The vertical component, RV (Fig. 5.3) of resultant soil reaction is obtained as,
m
RV = p cos(
0
L + )ds (5.23)
45
After substituting the value of p from Eq. 5.21, RV is obtained in the following form after
carrying out integrations.
RV = RV1 + RV2 + RV3 (5.25a)
Where,
RV1 = r02 K e 3tanm sin + e -tanm sin sin m (5.25b)
4 2 4 2 4 2
3 tan 2tanm
e [ sin 2 + sin 2( )
]
r02 sec 2 sec
m
(5.25c)
RV 2 =
(
4 1 + 9 tan 2 )
1
[
e 2m tan 1 +
1
] [ ( )]
e 2m tan cos 2 cos 2 2m
tan sec
And
- m tan
3tan m e sin sin m
r sece
2
4 2 4 2
RV 3 = 0
(5.25d)
(
1 + 9 tan 2 )
3 tan sin + cos +
4 2 4 2
Similarly, the horizontal component, RH (Fig. 5.3) of soil reaction is given as,
m
RH = pr e
0
tan
sin ( L + )sec d (5.26)
0
After substituting the value of p from Eq. 5.21, RH is obtained in the following form after
carrying out integrations.
RH = RH1+ RH2+ RH3 (5.27a)
Where,
RH1 = r02 K e 3tanm sin + cos m e - tanm cos (5.27b)
4 2 4 2 4 2
46
1 e 2 m tan cos(2 )
3 tan [
e 2 m tan 1 ] 1
r0 sec tan sec cos(2 2 m )
2 2
RH2 = (5.27c)
(
4 1 + 9 tan
2
)
sec
{[
1 e 2tanm sin 2 + [sin (2 m ] 2 )] }
And
3 tan sin + cos +
r02 sec e 3tanm 4 2 4 2
RH3 = (5.27d)
(1 + 9 tan 2 )
cos m e -tanm cos
4 2 4 2
5.3 Magnitude of passive thrust
2
In Fig. 5.3, the active Rankine thrust H1 acts at a distance .AC from point, C. Static
3
equilibrium of wedge, EACD is then considered.
Vertical force equilibrium condition gives
PpV = Pp sin = Rv WACD WADE (5.28)
In this procedure, first a trial value of V is assumed and corresponding weight of trial failure
wedge, EACD (Fig. 5.3) is computed. Using Eqs. 5.25 and 5.27, magnitudes of vertical and
47
horizontal components of soil reaction (RV and RH) are computed and from Eqs. 5.29 and 5.31,
values of Pp are determined. If the trial value of V is equal to its characteristic value
corresponding to the failure condition, the two computed values of Pp will be the same;
otherwise, they will be different.
For various trial values of V, computations are carried out till the convergence is reached to a
specified (third) decimal accuracy.
Pole of log spiral
O
O2
O1
D C B1 B2 B
Pp m
45- /2 45- /2 Trial 1
Trial 2
Trial 3
H
v Straight line
A Final failure surface
Log spiral
E
Fig. 5.5 Trial procedure for locating pole of the log spiral
Thus, in this method of analysis, the unique failure surface (Fig. 5.5) is identified by locating the
pole of log spiral in such a manner that, force equilibrium condition of failure wedge, EACD is
satisfied. This approach is different from other analyses in which, Pp is obtained from the
consideration of its minimum value.
The passive earth pressure coefficient, Kp is expressed as,
2Pp
Kp = (5.32)
H2
Values of passive earth pressure coefficient, Kp are obtained for different values of angles of soil
internal friction, and wall friction, .
These calculation are performed with reference to Fig. 5.2(a) (for pole of the log spiral above the
wall top) and Fig. 5.2(b) (for pole of the log spiral below the wall top) respectively. Axis, X0 is
48
taken along the line that joins pole, O of the log spiral to the wall base. Axis, Y0 is perpendicular
to the axis, X0 and passes through pole of the log spiral. With respect to these axes, coordinates
of the centroid of area inscribed in the log spiral are given as,
Moment equilibrium condition is now used to compute the point of application of passive thrust
by considering moments of forces and reactions about the pole of the log spiral.
49
the pole, O. From the above equation, Ypp (which is the distance of point of application of Pp
from the wall top), is obtained as,
2
WACD (OF + 3 DC) + WADE X
1
Ypp = (5.38)
PpH 2
+ H1 AC + FD + Pp V OF - PpH FD
3
5.6.2 Pole below wall top
Referring to Fig. 5.3(b), by taking moments of forces and reactions about the pole, O the following
equation is obtained.
2 2
WACD ( 3 DC OF) WODE 3 OF + WOEA X
PpH (YPP DF) = (5.39)
2
+ H1 AC DF PpV OF
3
In which, the terms on the right hand side of the above expression represent moment of weight of
soil in the failure wedge, EACD, moment of the force H1 and moment due to vertical component
of the resultant passive thrust, PpV about the pole, O. The term on the left hand side of the above
expression is the moment due to horizontal component of the resultant passive thrust, PpH about
the pole, O. From Eq. 5.39, YPP (which is the distance of point of application of PP from the wall
top), is obtained as,
2 2
WACD ( 3 DC OF) WODE 3 OF + WOEA X
1
YPP = (5.40)
PpH 2
+ H 1 AC DF PPV OF + PpH DF
3
The height, h of the passive thrust, Pp from the wall base is obtained as,
h = H Ypp (5.41)
5.7 Discussion
The basic purpose of this analysis was to compute passive pressure coefficient, Kp, location of
point of application of passive thrust and study their variation with respect to the parameters
involved in the analysis. It was found convenient to express the height, h of point of application
50
of passive thrust from the wall base in terms of its ratio with respect to height, H of the retaining
wall, in a non-dimensional form (Hr = h/H).
In Table 5.1, values of passive earth pressure coefficient, Kp along with angle v (angle defining
position of the pole of the log spiral on line AD) are shown for various combinations of soil
friction angle, and angle of wall friction, . For = 20, pole of the log spiral is located below
the wall top for all the values of . For = 25 and it goes below the wall top for higher value of
wall friction angle .
Table 5.1 Passive earth pressure coefficient and location of pole of the log spiral
51
One distinguishing feature of the proposed method is its ability to compute the point of
application of passive thrust using moment equilibrium. This has not been possible with other
existing methods. In Table 5.2, computed values of Hr are shown. They vary over a very narrow
range, from 0.225 (for = 20 and = 5) to 0.275 (for = 40 and = 40).
A
= 40
H = S2 m E 90
G
= 1kN/m3 3.07 kN/m
= 40
= 30
m = 59.33
v = 5.68
10.58 kN/m Normal = 40
Tangent
0 5 10 15 20 25 30 35 40 45 50 55 59.32
( degree)
Ordinate
10.58 9.16 7.96 6.93 6.07 5.35 4.75 4.26 3.87 3.56 3.33 3.16 3.07
(kN/m2)
Another distinguishing feature of the proposed analysis is its ability to predict the distribution of
reactive pressure on the failure surface using Ktters (1903) equation. This is shown in Fig. 5.6
for = 40 and = 30. The pressure distribution varies linearly over the straight part of the
failure surface followed by curvilinear variation over the log spiral part with a maximum
ordinate at the wall base.
52
In Table 5.3, computed values of KP for =20, 30 and 40 and =/2 and are compared
with other available solutions.
The values computed by Coulombs theory (1776) up to = 30 and =/2 are lower than the
proposed values in the range 2.69 to 2.92%, and up to = 30 and =, they are higher than the
proposed values in the range 7.29 to 53.73%. For = 40 and = 40, they tend to be very high
with no possible comparison.
The values reported by Chen (1975) are based on limit analysis. Up to = 40 and =20, they
are lower than the proposed values in the range 0 to 13.13%, and for = 40 and = 40, they
are higher than the proposed values by 26.97%.
Comparison with the values, which are based on rotational log spiral failure mechanism with the
upper-bound theorem of limit analysis reported by Soubra and Macuh (2002) shows that, these
values are lower than the proposed values in the range 2.85 to 13.5% and higher in the range
5.48 to 22.11%.
53
The values reported by Caquot and Kerisel (1948) are based on limit equilibrium of a log spiral
mechanism. Up to = 40 and =20, they are lower than the proposed values in the range 2.28
to 12.45%, and for = 40 and = 40, they are higher than the proposed values by 6.32%.
The values reported by Kumar and Subba Rao (1997) are based on method of slices. Up to =
40 and =20, they are lower than the proposed values in the range 2.95 to 15.83%, and for =
40 and = 40, they are higher than the proposed values by 14.58%.
The values reported by Sokolovski (1965) are based on the method of characteristics. Up to =
40 and =20, they are lower than the proposed values in the range 0.3 to 14.1% and for =
40 and = 40, they are higher than the proposed values 10.57%.
Comparison with the values, which are based on limit equilibrium analysis and reported by
Basudhar and Madhav (1980) shows that, these values are lower than the proposed values in the
range 5.3 to 13.8% and higher in the range 5.4 to 17.5%.
With the analytical solution based on the lower bound theorem of plasticity, the KP values as
reported by Lancellotta (2002) are lower than the proposed values in the range 16.37 to 32.98%.
Similarly, the KP values as reported by Janbu (1957) which are based on limit equilibrium
analysis are lower than the proposed values in the range 8.7 to 14.95%.
The values of KP, as reported by Shields and Tolunay (1973) are also based on limit equilibrium
analysis. These values are lower than the proposed values in the range, 17.93 to 33 %.
The above comparison shows that proposed values are fairly close to some of the available
solutions, except those of Shields and Tolunay (1973) and Lancellotta (2002).
In Table 5.4, more data giving Kp values computed by Coulombs theory (1776), Caquot and
Kerisel (1948), Kumar and Subba Rao (1997), Soubra and Macuh (2002), Lancellotta (2002) and
by the proposed method is reported. It may be noted that, for the failure mechanism consisting of
log spiral and its tangent, which is adopted in the proposed analysis, the Kp values are unique;
since they are evaluated from the identification of a unique failure surface that satisfies force
equilibrium condition. The proposed method also enables the computation of point of application
of passive thrust using moment equilibrium and reactive pressure distribution on the failure
surface.
54
Angle of
friction Passive earth pressure coefficient Kp
( degree)
Caquot Kumar Soubra
Soil Wall Coulomb and and and Lancellotta Proposed
(1776) Kerisel Subba Rao Macuh (2002) Method
(1948) (1997) (2002)
20 2.040 2.04 2.04 2.40 2.040 2.58
25 2.464 2.46 2.46 2.46 2.464 3.100
30 0 3.000 3.03 3.00 3.00 3.000 3.700
35 3.690 3.69 3.69 3.69 3.690 4.600
40 4.599 4.59 4.60 4.60 4.599 5.700
20 2.414 2.35 2.38 2.39 2.349 2.844
25 3.124 3.03 3.06 3.07 2.993 3.605
30 1/3 4.143 4.00 4.02 4.03 3.886 4.655
35 5.680 5.28 5.42 5.44 5.168 6.162
40 8.147 7.25 7.58 7.62 7.087 8.425
20 2.635 2.60 2.50 2.57 2.477 2.967
25 3.552 3.46 3.40 3.41 3.222 3.854
30 1/2 4.977 4.78 4.60 4.65 4.288 5.126
35 7.357 6.88 6.60 6.59 5.879 7.043
40 11.771 10.38 9.80 9.81 8.378 10.098
20 2.888 2.65 2.73 2.75 2.582 3.086
25 4.079 3.56 3.72 3.76 3.413 4.097
30 2/3 6.105 5.00 5.26 5.34 4.633 5.606
35 9.962 7.10 7.78 7.95 6.510 7.983
40 18.717 10.72 12.24 12.6 9.573 11.995
20 3.525 3.01 3.07 3.13 2.699 3.290
25 5.599 4.29 4.42 4.54 3.627 4.560
30 10.095 6.42 6.68 6.93 5.026 6.572
35 22.971 10.20 10.76 11.30 7.250 10.018
40 92.586 17.50 18.86 20.10 11.026 16.464
55
Chapter 6
Pullout capacity of vertical plate anchors in cohesion-less soil
6.1 Introduction
Generally, earth anchors are used to transmit tensile forces from a structure to the soil. Their
pullout capacity is obtained through the shear strength and dead weight of the surrounding soil.
Plate anchors may be made of a steel plate and precast or cast in situ concrete slab. These
anchors can be installed by excavating the ground to the required depth followed by back filling
and compacting with a good quality soil.
The analysis of ultimate pullout capacity of vertical anchor plates is quite similar to that for
shallow and deep horizontal anchors. In case of shallow anchors, the embedment ratio is such
that, failure surface reaches the ground surface at limit equilibrium; whereas in case of deep
anchors, the embedment ratio is such that, failure surface does not reach the ground surface at
limit equilibrium (Das, 1990).
The proposed analysis is confined to shallow laid anchors in cohesion-less soil.
In the present analysis, for a vertical strip plate anchor (basic case, H/h = 1.0) a total of seven
experimental studies namely, Ovesen and Stromann (1964 and 1972), Neely et al. (1973), Das
and Seeley (1975), Akinmusuru (1978), Dickin and Leung (1983 and 1985), Hoshiya & Mandal
(1984) and Murray and Geddes (1987) are referred for comparison with the proposed solution.
6.2 Proposed method
In the proposed analysis of the estimation of pullout capacity of a strip anchor in cohesion-less
soil, all the three equation of equilibrium are utilized to obtain the required solution. Both
passive and active states of equilibrium on the two sides of anchors are considered in the
analysis. The active/passive thrusts along with their points of application are evaluated using
Ktters (1903) equation. This equation has been used by other researchers such as Dewaikar and
Mohapatro (2003) for computation of bearing capacity factor, N, Deshmukh et al. (2010) for the
estimation of breakout capacity of horizontal rectangular/square anchors in cohesion-less soils,
Rangari et al. (2010) for the computation of seismic vertical uplift capacity factor fd for
horizontal strip anchors and Kame, Dewaikar and Choudhury (2010 a) for the estimation of
active thrust on a vertical retaining wall with horizontal cohesion-less backfill.
56
Pa
H =h A Tu
ha h/2
hp
Pp
N tan
57
In Fig. 6.1(b), free body diagram of the strip anchor is shown from which, the following
information is generated.
d
ds
Slip 90
Curved failure surface
dp
Tangent Normal
Fig. 6.2(a) Reactive pressure distribution on the failure surface for passive case
(Kame, Dewaikar and Choudhury, 2010-b)
O
Slip
Curved failure surface
d
Tangent
ds
Normal dp
Fig. 6.2(b) Reactive pressure distribution on the failure surface for active case
58
Where, K is the parameter indicating location of the pole of the log spiral along line AO in terms
of starting radius of log spiral r0 as measured from point D (Fig. 6.3).
= spiral angle measured from the starting radius
59
A
= 40
H = 2 m E 90
= 1kN/m3 G 3.07 kN/m
= 40
= 30
m = 59.33
v = 5.68
10.58 kN/m
Normal = 40
Tangent
Fig. 6.3 Reactive pressure distribution on the failure surface for the passive case using
Ktters (1903) equation (Kame, Dewaikar and Choudhury, 2010-b)
Pole of log spiral
Oa
v
m
r0 d
B D
0.0 45+ /2
45+ /2
= 2
= 1kN/m
3
= 45
A = 30
Pressure distribution along
straight line failure surface H=h
0.317 kN/m
Pa
Pressure distribution along curved E
failure surface (arc of a log spiral)
0.454 kN/m
Fig. 6.4 Reactive pressure distribution on the failure surface for the active case using
Ktters (1903) equation (Kame, Dewaikar and Choudhury, 2010-a)
60
Similarly, the final expression for reactive active pressure distribution at any point on the curved
failure surface based on the application of Ktters (1903) equation is given by the following
expression.
The distribution of reactive pressures on the failure surface in both cases using Ktters (1903)
equation are as shown in Figs. 6.3 and 6.4. The magnitudes of passive and active thrusts on the
vertical plate and their points of application are thus obtained using Ktters (1903) equation.
The resultant soil reaction, R (Fig. 6.5) on the failure surface is obtained as,
R = p.ds (6.4)
The vertical and horizontal components, RV and RH of resultant reaction are obtained as,
m
RV = p cos.ds
0
(6.5)
m
RH = p sin.ds
0
(6.6)
Where, ds is the length of failure surface and is the varying angle of inclination of reactive
pressure with vertical (Figs. 6.2 a & b). The detailed calculations for estimation of active thrust
are reported in the paper Active thrust on a vertical retaining wall with cohesion-less backfill
published in the Electronic Journal of Geotechnical Engineering, (EJGE), U.S.A. (Vol. 15(Q),
Page 1848-1863 , Kame, Dewaikar and Choudhury, 2010-a) and the detailed calculations for
estimation of the passive thrust are reported in the paper Passive thrust on a vertical retaining
wall with horizontal cohesion-less backfill communicated to Soils and Rocks, An International
Journal of Geotechnical and Geo-environmental Engineering, Brazil. (Kame, Dewaikar and
Choudhury, 2010-b)
61
m 2DC/3
D C
WACD
PpV
Y pp 2AC/3
WADE
H=h H1
PpH
v
hp 90 A
RH
E G
R
RV
Fig. 6.5 Free body diagram of failure wedge EACD (passive state)
Where,
PpH, PpV = horizontal and vertical components of resultant passive thrust, Pp
RH, RV = horizontal and vertical components of resultant soil reaction acting on the
curved part of the failure surface
H1 = passive thrust exerted by the backfill on the Rankine wall AC
WACD = weight of soil in the failure wedge, forming a part of the Rankine zone
62
WADE = weight of soil in the zone, EAD of the failure wedge, EABCD
Ypp = the distance of point of application of Pp from the wall top
D C B1 B2 B
Pp m
45- /2 45- /2 Trial 1
Trial 2
Trial 3
H=h
v Straight line
A Final failure surface
Log spiral
E
Fig. 6.6 Trial procedure for locating pole of the log spiral
63
Values of the passive thrust are obtained for different values of angles of soil friction, and wall
friction, . In Table 6.1 some of the computed values of passive earth pressure co-efficient, Kp
( 2 Pp / H 2 ) are reported.
Similar procedure (Kame, Dewaikar and Choudhury (2010 a) is adopted for determination of
active thrust for different values of angles of soil friction, and wall friction, and in Table 6.1
some of the computed values of active earth pressure co-efficient, Ka ( 2Pa / H 2 ) are reported.
64
The forces N and Ntan now can be considered to be zero and the moment equilibrium is
considered about the point, A as shown in Fig. 6.1(b). The only forces that contribute to the
moment equilibrium are Pa and Pp and since Pp > Pa, clearly moment equilibrium is also not
satisfied. The plate rotates at limit equilibrium and this agrees with the experimental observation
(Naser, 2006).
6.4 Discussion
In Table 6.2, the data of tests conducted by various researchers on the vertical plate anchors in
cohesion-less soil for the basic case (H/h=1.0) is reported.
In Table 6.3, the experimental values of Tu (kN/m) are reported in column 2 of the table. In the
same table, the values of Tu as computed using various theoretical/empirical solutions are
reported along with the results obtained with the proposed method.
From Table 6.3, it is seen that, the proposed method provides a better estimate of the pullout
capacity as compared to the other methods. For example, when compared to experimental value
reported by Neely (1973), the proposed method gives an error of +8.9% while the errors in
respect to the methods proposed by Ovesen (1973), Das (1975), Akinmusuru (1978), Dickin &
Leung (1983,1985), Hoshiya & Mandal (1984) and Murry & Geddes (1989) are -44.3%, 16.6%,
-22.8%, 11.2%, -34.8% and 36.5% respectively. This observation is further substantiated by the
data generated in Table 6.4, which gives cumulative frequency distribution of errors. While
generating this data only absolute value of the error is considered.
65
The proposed method gives absolute error in the range, 0 to 25% in 4 out of 7 cases and in
remaining cases the range is 25% to 100%.
Table 6.2 Anchor & Soil Parameters
t H B p Material of
Author (degrees) (degrees) (m) (kN/m3) (m) (m) (kN/m3) the plate
Ovesen and Stromann
42.0 38.7 ? 16.770 0.250 1.000 ? Plexiglas
(1972)
Neely
38.5 21.0 0.006 15.900 0.051 0.255 77.0 Steel
(1973)
Das
34.0 34.0 0.003 15.920 0.038 0.191 28.0 Aluminum
(1975)
Akinmusuru
35.0 29.0 0.003 15.550 0.038 0.380 77.0 Steel
(1978)
Dickin & Leung
41.0 29.0 0.003 16.000 0.050 0.250 28.0 Aluminum
(1983, 1985)
Hoshiya & Mandal
29.5 29.5 0.005 14.120 0.025 0.152 28.0 Aluminum
(1984)
Murry & Geddes
43.6 10.6 0.006 16.500 0.051 0.508 77.0 Steel
(1989)
In 1 out of 6 cases Ovesens (1972) method gives absolute errors in the range, 1% to 25% and in
remaining 5 cases the errors are as high as 25% to 100%.
Neelys (1973) method gives absolute errors in the range 1% to 25% in three out of 6 cases and
in remaining 3 cases the range is 25% to 100%.
The methods proposed by Das (1975), Akinmusuru (1978) and Dickin & Leung (1983.1985)
give absolute errors in the range, 1% to 25% in 3 out of 6 cases and in remaining cases, the errors
range is 25% to 100%.
Similarly, the methods proposed by Hoshiya & Mandal (1984) and Murry & Geddes (1989) are
giving absolute errors in the range, 1% to 25% in 2 cases and in remaining cases, the range is
25% to 100%.
The capability of the proposed method is related to the passive and active earth pressure
coefficient values which are unique for the failure mechanism consisting of log spiral and its
tangent.
66
Table 6.3 Comparison of pullout capacity experimental results and semi-empirical methods for vertical (strip) plate anchors
Experi Ovesen and
Proposed Neely Das Akinmusuru Dickin & Leung Hoshiya & Mandal Murry & Geddes
- Stromann
method (1973) (1975) (1978) (1983, 1985) (1984) (1989)
mental (1972)
Author
Tu Tu % Tu % Tu % Tu % Tu % Tu % Tu % Tu %
(N/m) (N/m) Error (N/m) Error (N/m) Error (N/m) Error (N/m) Error (N/m) Error (N/m) Error (N/m) Error
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
Ovesen (1972) 2545 4377 72.0 2545 0 3925 54.2 4577 79.8 2915 14.5 4200 65 2559 0.5 5156 102
Neely (1973) 41.7 45.4 8.9 22.3 -44.3 41.7 0.0 48.6 16.6 30.9 -22.8 44.6 11.2 27.2 -34.8 54.7 36.5
Das (1975) 20.3 16.3 -19.4 7.7 -63.2 17.4 -14.2 20.3 0.0 12.9 -38.2 18.6 -11 11.3 -44.1 22.8 9.3
Akinmusuru (1978) 25.6 32.6 27.3 15.8 -38.4 34.5 34.6 40.2 57.0 25.6 0 36.9 44.1 22.5 -12.2 45.3 76.9
Dickin &Leung (1983, 1985) 42.0 60.3 43.6 23.8 -43.4 39.3 -6.5 45.8 9.0 29.2 -30.6 42 0 25.6 -39.1 51.6 22.8
Hoshiya & Mandal (1984) 4.0 3.7 -9.2 3.8 -10.9 6.2 53.4 7.2 78.8 4.6 8.3 6.6 56 4.0 0.0 8.1 91.6
Murry & Geddes (1989) 108.2 99.7 -7.8 62.5 -42.3 82.3 -23.9 96.0 -11.2 61.2 -43.5 88.1 -18.6 53.7 -50.4 108.2 0
Table 6.4 Cumulative frequency distribution of errors for vertical plate anchors
Proposed Ovesen and Hoshiya &
Neely Das Akinmusuru Murry & Geddes
method Stromann Dickin & Leung Mandal
(1973) (1975) (1978) (1989)
(1972) (1983, 1985) (1984)
Absolute
Cum. Cum. Cum. Cum. Cum. Cum. Cum. Cum.
% error
Frequ- Frequ- Frequ- Frequ- Frequ- Frequ- Frequ- Frequ- Frequ- Frequ- Frequ- Frequ- Frequ- Frequ- Frequ- Frequ-
ency ency ency ency ency ency ency ency ency ency ency ency ency ency ency ency
0-10 3 3 1 1 2 2 2 2 2 2 1 1 2 2 2 2
10-20 1 4 1 2 1 3 2 4 1 3 3 4 1 3 0 2
20-30 1 5 0 2 1 4 0 4 2 5 0 4 0 3 1 3
30-40 0 5 1 3 1 5 0 4 1 6 0 4 2 5 1 4
40-50 1 6 3 6 0 5 0 4 1 7 1 5 1 6 0 4
>50 1 7 1 7 2 7 3 7 0 7 2 7 1 7 3 7
67
Chapter 7
Conclusions and Scope of the future work
7.1 General
The investigations that are reported in this report are confined to the pullout capacity of
vertical plate strip anchors using limit equilibrium approach. The proposed analysis is
essentially based on the use of Ktters (1903) equation, by virtue of which, no simplifying
assumptions are required to make the analysis statically determinate.
The applicability of Ktters (1903) equation to the analysis of limit equilibrium problem is
successfully demonstrated for vertical plate strip anchor embedded in cohesion-less soil. The
proposed analysis is confined to shallow laid anchors.
The main contributions that emerge from these investigations are as follows:
1. A protocol is developed for the evaluation of pullout capacity of vertical plate strip
anchors in cohesion-less soil with the interpretation of all the equilibrium equations.
2. Ktters (1903) equation is effectively utilized to compute the resultant soil reaction.
3. A method based on the application of Ktters (1903) equation is proposed for the
complete analysis of active and passive earth pressure on a vertical wall retaining
horizontal cohesion-less backfill.
4. Ktters (1903) equation lends itself as a powerful tool in the analysis to enable the
computation of point of application of active and passive thrusts and reactive pressure
distribution on the failure surface for a vertical wall retaining horizontal cohesion-less
backfill.
68
4. In case of active thrust on vertical wall retaining horizontal cohesion-less backfill
there is a very close agreement between the Ka values of the proposed method and
those reported by Caquot and Kerisel (1948). The maximum difference is of the order
of 4%.
5. In case of passive thrust on vertical wall retaining horizontal cohesion-less backfill the
values reported by Caquot and Kerisel (1948) are based on limit equilibrium of a log
spiral mechanism. Up to = 40 and =20, they are lower than the proposed values
in the range 2.28 to 12.45%, and for = 40 and = 40, they are higher than the
proposed values by 6.32%.
6. For the failure mechanism consisting of log spiral and its tangent, which is adopted in
the proposed analysis, the Ka and Kp values are unique; since they are evaluated from
the identification of a unique failure surface that satisfies force equilibrium condition.
7. In case of vertical plate strip anchor when compared to experimental value reported
by Neely (1973), the proposed method gives an error of +8.9% while the errors in
respect to the methods proposed by Ovesen (1973), Das (1975), Akinmusuru (1978),
Dickin & Leung (1983,1985), Hoshiya & Mandal (1984) and Murry & Geddes
(1989) are -44.3%, 16.6%, -22.8%, 11.2%, -34.8% and 36.5% respectively.
8. The proposed method gives absolute error in the range, 0 to 25% in 4 out of 7 cases
and in remaining cases the range is 25% to 100%.
1. For the failure mechanism consisting of log spiral and its tangent, computations of
active and passive thrusts on a vertical wall retaining horizontal cohesion-less backfill
with uniform surcharge using Ktters (1903) equation.
2. Computation of ultimate pullout capacity for shallow laid vertical plate strip anchors
in cohesion-less soils using Ktters (1903) equation.
69
References
Balla, A. (1961). The resistance to breaking out of mushroom foundations for Pylons. Proc
of the 5th Int. Conf. on Soil Mechanics and Foundation Engineering, Paris, France, 1, 569
576.
Basudhar, P. K., and Madhav, M. R. (1980). Simplified Passive Earth Pressure Analysis,
Journal of the Geotechnical Engineering Division, ASCE, Vol. 106, No. GT4, pp. 470-474.
Biarez, I., Boucraut, L.M. and Negre, R. (1965). Limiting Equilibrium of Vertical Barriers
Subjected to Translation and Rotation Forces. Proceedings of the 6th International
Conference on Soil Mechanics and Foundation Engineering, Montreal, Canada, Vol. II, 368-
372.
Caquot, A. I., and Kerisel, J. (1948). Tables for the calculation of passive pressure, active
pressure, and bearing capacity of foundations. Libraire du Bureau des Longitudes, de
Lecole Polytechnique, Paris Gauthier- villars, Imprimeur-Editeur, 120.
Chen, and Li. (1998). Evaluation of active earth pressure by the generalized method of
slices. Can. Geotech. J., 35, 591599
Ching, et al. (1994). Discrete element analysis for active and passive pressure distribution
on retaining wall. Computers and Geotechnics., 16, 291-310.
Coulomb, C. A. (1776). Essais sur une application des regles des maximis et minimis a
quelques problems de statique relatits a l'architecture. Mem. Acad. Roy. Pres. Divers, Sav.,
Paris 5, 7.
70
Das, B. M. (1975). Pullout resistance of vertical anchors. J. Geotech. Engrg. Div., ASCE,
101(1), 8791.
Das, B. M., Seeley, G. R. and Das, S. C. (1977). Ultimate Resistance of Deep Vertical
Anchors in Sand. Soils and Foundations, Jap. Eng. Soc., June, 52-56.
Dewaikar, D. M., and Halkude, S. A. (2002b). Active thrust on bracing system of open cuts
in cohesionless soil- point of application. Indian Geotechnical Journal., 32, 407-420.
Dewaikar, D.M. and Halkude, S.A. (2002). Influence of wall friction and horizontal seismic
acceleration on seismic earth pressure behind retaining wall. Numerical methods in
Geotechnical Engineering, Mestat (ed.), Presses de IENPC/LCPC, Paris, 1041-1046.
Dickin, E.A. and King, GJW. (1997). Numerical modelling of the loaddisplacement
behaviour of anchor walls. Comp Struct., 63,849858.
Dickin, E.A., Leung, C.F., (1983). Centrifugal model tests on vertical anchor plates.
Journal of Geotechnical Engineering, ASCE, 109 (12), 15031525.
Dickin, E.A., Leung, C.F., (1985). Evaluation of design methods for vertical anchor plates.
Journal of Geotechnical Engineering, ASCE, 111 (4), 500520.
71
Duncan, J. M. and Mokwa, R. L. (2001). Passive earth pressures: theories and tests. ASCE
J Geotech Geoenviron Eng., 127(3) 248-257.
Goel, S., Shalini, and Patra, N. R. (2006). Break out resistance of inclined anchors in sand.
Geotech. and Geol. Eng., 24, 15111525
Hanna, A., Tahar, A. and Mohab, S. ( 2007). Pullout resistance of single vertical shallow
helical and plate anchors in sand. Geotech. and Geol. Eng., 25(5), 559-573
Hanna, T. H., (1980). Design and Construction of Ground Anchors, Construction Industry
Research and Information Association (London) Report No. 65, 2nd ed., Oct.
Hanna, T. H., and Carr, R. W., (1971). The Loading Behavior of Plate Anchors in Normally
and Overconsolidated Sands, Proceedings, 4th European Conference on Soil Mechanics and
Foundation Engineering, Budapest, Hungary, 589-600.
Hueckel, S., (1957). Model Tests on Anchoring Capacity of Vertical and Inclined Plates,
Proc., of 4th International Conference on Soil Mechanics and Foundation Engineering,
London, U.K., 2, 203-206.
Jacky, J., (1936). Stability of earth slopes. Proc., of 1st International Conference on Soil
Mechanics and Foundation Engineering, Harvard, 2, 200-207.
72
Krisel, J., Absi, E. (1990). Tables de pousse et de bute des terres, 3rd ed., Presses de lE
cole Nationale des Ponts et Chausses, Paris, 1990.
Ktter, F. (1903). Die Bestimmung des Drucks an gekrmmten Gleitflchen, eine Aufgabe
aus der Lehre vom Erddruck Sitzungsberichte der Akademie der Wissenschaften. Berlin,
229233.
Kumar, J. and Kouzer, K. M. (2008). Vertical uplift capacity of horizontal anchors using
upper bound limit analysis and finite elements. Can. Geotech. J., ASCE, 45, 698-704.
Kumar, J. and Rao, M. (2004). Seismic horizontal pullout capacity of shallow vertical
anchors in sand. Geotech. and Geological Eng., 22, 331349,
Kumar, J., and Subba Rao, K.S. (1997). Passive pressure coefficients, critical failure
surface and its kinematic admissibility. Geotechnique, 47, 185192.
Kumar, J., Kanakapura, S., and Rao, S., (1997). Effect of cohesion and surcharge on pullout
capacity of vertical anchors, IGJ, 27(2),182-192.
Lade, P. V. and Duncan, J. M., (1975). Elasto-plastic stress-strain theory for cohesionless
soil. J. Geotech. Eng. Div. ASCE 101, 1037-1060.
Matsuo, M. (1968). Study on the uplift resistance of footings (II). Soils & Found., 8(1),
18-48.
Matsuo, M.(1967). Study on the uplift resistance of footing (I). Soils and Foundations,
7(4), 1-37.
Merifield, R.S., and Sloan, S.W. (2006a). The ultimate pullout capacity of anchors in
frictional soils. Canadian Geotechnical Journal, 43(8), 852868.
Meyerhof, G.G. (1951). The ultimate bearing capacity of foundations. Geotechnique, 2(4),
301332.
73
Meyerhof, G. G., (1973). Uplift Resistance of Inclined Anchors and Piles. Proc. of the 8th
International Conference on Soil Mechanics and Foundation Engineering, Moscow, USSR,
No. 2(1),167-172.
Meyerhof, G. G., and Adams, J. I. (1968). The ultimate uplift capacity of foundations.
Canadian Geotechnical Journal, 5(4), 225-244.
Morgenstern, N. R. and Eisenstein, Z., (1970). Methods of estimating lateral loads and
deformations. Proc., ASCE Speciality Conf. on Lateral Stresses in the Ground and Design
of Earth-Retaining Structures, Ithaca, N.Y., 51102.
Murry, E. J., and Geddes, J. D. (1987). Uplift of anchor plates in sand. Journal of
Geotechnical Engineering, ASCE, 113(3), 202-215.
Naser, Al-Shayea. (2006). Pullout capacity of block anchor in unsaturated sand. Proc. of
the Fourth International Conference on Unsaturated Soils., ASCE,403-414
Neely, W. J., Stuart, J. G. and Graham, J. (1973). Failure loads of vertical anchor plates in
sand. J. Soil Mech. and Found. Div., ASCE, 99(9), 669685.
Niroumand, H. and Kassim, Kh.A. (2010). Analytical and numerical studies of vertical
anchor plates in cohesion-less soils. Electronic journal of geotechnical engineering (EJGE),
15(L), 1140-1150
Ovesen, N. K. and Stromann, H. ( 1972). Design Method for Vertical Anchor Slabs in
Sand. Proceedings of Specialty Conf. on Performance of Earth and Earth-Supported
Structures, 1-2, 1418-1500.
Rahardjo, H., and Fredlund, D.G. (1984). General limit equilibrium method for lateral earth
force. Canadian Geotechnical Journal, 21: 166175.
74
Rangari, S., Choudhury, D. and Dewaikar, D. M. (2010). Computation of seismic vertical
uplift capacity factor fd for horizontal strip anchors using Ktters equation. Proceeding of
Int. conf. on disaster prevention technology and management, Chongqing, China, 3(4), 610,
October.
Saeedy, H. S. (1987). Stability of circular vertical earth anchors. Can. Geotech. J., 24(3),
452-456.
Sakai, T., and Tanaka, T. (1998). Scale effect of a shallow circular anchor in dense sand.
Soils and Found., Japan, 38(2):93-99.
Sakai, T., Erizal, and Tanka, T. (2007). Experimental and numerical study of uplift behavior
of shallow circular anchor in two layered sand. Journal of Geotechnical and
Geoenvironmental Engineering, ASCE, 133(4), 469-477.
Shields, D. H., and Tolunay, A. Z., (1973). Passive Pressure Coefficients by Method of
Slices. Journal of the Soil Mechanics and Foundations Division, ASCE, 99, 1043-1053.
Sloan, S. W. (1988). Lower bound limit analysis using finite elements and linear
programming. Int. J. for Numerical and Analytical Methods in Geomechanics, 12, 61-67
Soubra, A.H. and Macuh, B. (2002). Active and passive earth pressure coefficients by a
kinematical approach. Proc. of the Institution of Civil Engineers Geotechnical Engineering.
155, April 2002, Issue 2, 119131
Tagaya, K., Scott, R.F., and Aboshi, H. (1988). Pullout resistance of buried anchor in sand.
Soils and Foundations, 28(3), 114130.
75
Tagaya, K., Tanaka, A. and Aboshi, H. (1983). Application of finite element method to
pullout resistance of buried anchor. Soils and Found., 23(3), 91-104.
Teng, W. C. (1962). Foundation design, Prentice-Hall, Engeiwood Cliffs, New Jersey, USA.
Terzaghi, K. (1941). General wedge theory of earth pressure. ASCE Trans., 68- 80.
Terzaghi, K. and Peck, R. B. (1967), Soil mechanics in engineering practice, John Wiley and
Sons, Inc., New York, N.Y.
Vermeer, P.A., and Sutjiadi, W. (1985). The uplift resistance of shallow embedded
anchors. Pro. 12th Intl. Conf. on Soil Mech. Found. Engg. San Francisco, California,
3,1635-1638.
76
List of the Publications
From the research work carried out and presented in this report for the current Ph.D. program,
the following technical papers have been published or accepted/communicated for
publication.
Paper published in referred journal
Kame, G. S., Dewaikar, D. M. and Deepankar Choudhury. ( 2010-a). Active thrust on a
vertical retaining wall with cohesion-less backfill. Electronic journal of geotechnical
engineering (EJGE), U.S.A.,Vol. 15(Q),1848-1863.
77