1 tayangan

Diunggah oleh DanishPawaskar

SAMPLE

- Pile Foundation Design - A Student Guide
- RRM Gravity Dam
- Composite Behavior of Geosynthetic Reinforced Soil Mass FHWA-HRT-10-077
- Pile Design Calculation
- Lecture20 Drilled Shafts
- Nonlinear Finite Element Analysis of Four-Pile Caps Supporting Columns Subjected to Generic Loading 04 July.pdf
- 91143345 Pile Cap Design
- Repair and Strengthening of Bridge Substructures
- Foundations Theory & Design
- Chapter2-Lateral Earth Pressure RE Wall
- How to Do Pile Driving With Percussion Method
- TC207
- Pile Driving Equipment
- Drilling Rigs
- Introduction to Geo.tech. Engineering-Sivakugan
- Pile Footings for Flats for Golden Green at Asr
- 826
- Pile Capacity
- The Optimal Shapes of Piles in Integral Abutment Bridge
- 1 viewNitPdf_1716598

Anda di halaman 1dari 88

COHESION-LESS SOIL

by

G. S. KAME

(Roll No. 08404803)

Prof. D. M. DEWAIKAR

and

Prof. DEEPANKAR CHOUDHURY

INDIAN INSTITUTE OF TECHNOLOGY BOMBAY

POWAI, MUMBAI -400076, INDIA.

January 2011

APPROVAL SHEET

Indian Institute of Technology Bombay

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.

Supervisor

Co-supervisor

ii

CONTENTS

No.

APPROVAL SHEET ii

LIST OF FIGURES vi-vii

LIST OF TABLES viii

NOMENCLATURE ix

ABSTRACT x

1.1 General 1

1.2 Problem Statement 2

1.3 Objective 2

1.4 Scope of the study 2

1.5 Significance of the study 2

1.6 Scheme of proposed work 3

2.1 General 4

2.2 Vertical plate anchor 4

2.2.1 Experimental Investigations 5

2.2.2 Theoretical Investigations 7

2.3 Active and passive thrust on a vertical retaining wall 12

3.1 General 14

3.2 Derivation of Ktters (1903) equation 14

Chapter 4 ACTIVE THRUST ON A VERTICAL RETAINING WALL WITH 22-36

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

HORIZONTAL COHESION-LESS BACKFILL

5.1 Introduction 37

5.2 Proposed method 37

5.2.1 Geometry of Failure Surface 41

5.2.2 Pole above wall top 41

5.2.3 Pole below wall top 41

5.2.4 Computation of soil reaction on the failure surface 42

5.2.5 Computation of soil reaction on plane failure surface AB 43

5.2.6 Computation of vertical and horizontal components of 43

reaction on curved failure surface EA

5.2.7 Components of resultant soil reaction on the failure surface 45

5.3 Magnitude of passive thrust 47

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

COHESION-LESS SOIL

6.1 Introduction 56

6.2 Proposed method 56

6.2.1 Geometry of the problem 57

6.2.2 Ktters (1903) equation 58

6.2.3 Magnitude of passive thrust 62

6.2.4 Trial and error procedure 63

6.3 Analysis of Anchor Plate 64

6.3.1 Vertical equilibrium 64

6.3.2 Moment equilibrium 65

6.3.3 Horizontal equilibrium 65

6.4 Discussion 65

Chapter 7 CONCLUSIONS AND SCOPE OF THE FUTURE WORK 68-69

7.1 General 68

7.2 Scope of the future work 69

REFERECES 70-76

LIST OF PUBLICATIONS 77

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

To study the failure mechanism around the vertical plate anchor in sand

To estimate the pullout capacity of vertical plate anchor embedded in sand

To study the pullout behavior of anchors embedded in sand

To define the influence of foundation geometry on the pullout behavior of vertical plate

anchors

To compare pullout capacity or breakout factors derived from the proposed methods with

the available existing theoretical methods vis--vis experimental results

Development of analytical methods for evaluation of pullout capacity/breakout factors

Application of Ktters (1903) equation to evaluate soil reaction on the failure surface

Evaluation of pullout capacity/breakout factors based on the trial failure surface

Analytical method is proposed for better understanding of pullout behavior of vertical plate

anchors in cohesion-less soil. A unique failure surface with reactive pressure distribution in

active and passive case is obtained in the analysis. The analysis takes into account the influence

of soil and anchor parameters. Ktters (1903) equation is employed to evaluate resultant soil

reaction to make the analysis statically determinate

The proposed work is divided into seven chapters as given below.

Chapter 1 gives introduction of vertical plate anchors. problem statement, scope and significance

of the work are briefly stated.

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.

Research into the behavior of soil anchors can take one of two forms, namely experimental or

numerical/theoretical based studies. The brief summary of existing research herein is separated

based on this distinction. Report is limited solely to the topic of vertical plate anchors in

cohesion-less soil or sand.

Majority of past research has been experimental based and as a result, current design practices

are largely based on empiricism. In contrast, very few thorough numerical analyses have been

performed to determine the ultimate pullout loads of anchors. As the area of application of

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

The above analysis is based on the method of stress characteristics and represents a more refined

analytical and numerical attempt to predict the ultimate capacity of the vertical plate anchors but

it ignores the active earth pressure distribution behind the anchor plate and the kinematic

behaviour of the material.

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.

and theoretical methods proposed by Ovesen and Stromann (1972), Neely & Stuart, (1973) was

carried out by Dickin (1983). Significant disparity was observed in the results because they were

based on two- dimensional analysis and their application to single anchors required a suitable

shape factor. Dickin and Leung (1985) observed that, effect of anchor shape on dimensionless

coefficients was due to side shear resistance. They observed failure planes radiating outward

involving a soil mass wider than a single anchor itself in the failed body. A dimensionless shape

factor, Sf to account for the influence of anchor geometry on the ultimate resistance was

introduced by them.

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%.

Naser (2006) analyzed pullout capacity of an anchor block using limit equilibrium approach

(Fig. 2.5). The ultimate pullout capacity of block anchor (Tu) was obtained from the equilibrium

of forces acting on the block by summing them along the horizontal direction and multiplying

the lateral earth pressures (passive and active) by the 3-D corrections factor M, to yield the

following equation.

Tu = M (Pph Pah ) + Ft + Fs + Fb (2.9)

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.

The analysis of bearing capacity of strip footings and thrust exerted by backfill on retaining wall

fall in the category of plain strain problems. Therefore, derivation of the required expression

starts with the basic equilibrium equations for plain strain condition. For the analysis, a soil

element, ABC is considered and the co-ordinate system is shown in Fig. 3.2. Considering no

body forces acting X direction, the equilibrium equations can be written as,

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,

major and minor principal stresses, 1 and 3 in the following form,

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 )

( 1 + 3 )

=m (3.5a)

2

( 1 3 )

And, = sin m (3.5b)

2

Substituting Eq. 3.5 into Eq. 3.3, following expressions for x , y and xy are obtained.

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

s1 s 2

(3.11a)

2 m sin sin cos( ) =0

s1 s 2

And

s1 s 2

(3.11b)

2 m sin cos + sin( ) + cos = 0

s1 s 2

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,

Therefore the normal stress, s can be written in terms of 1 and 3 as,

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

H = height of the retaining wall

h = height of point of application of active thrust from wall base

= 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 = 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 radius r1

From Fig. 4.3, which shows free body diagram of failure wedge EACD the following

information is generated.

RH, RV = horizontal and vertical components of soil reaction acting on the curved

part of the failure surface

X

Pole F

O

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

Geometry of failure surface

24

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, EACD

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.

OD OE DE H

= = = (4.1)

sin v sin m sin m

sin 135 +

2

sin v

OD = H

sin m

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

r = r0 e tan

Making necessary substitutions in Eq. 4.9 the following expression is obtained.

dp 45 t tan

2 p tan = r0 sec sin te 2

dt

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,

.r0 sec e

1 + 9 tan 2

p= (4.14)

3 tan sin L + cos L

.r0 sec e tan 2 + .e tan 2

{sin AD}

1 + 9 tan 2

L

4.2.4 Components of resultant soil reaction on the failure surface

The resultant soil reaction, R (Fig. 4.3) on the failure surface is given as,

R = p.ds (4.15)

28

m

RV = p cos(

0

L ).ds (4.16)

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 ]

.r0 sec 2 sec 2 [tan cos L + sin L ]

RV2 = (4.17c)

(

)

1 + 9 tan cos e

2

L

tan m

[tan cos( L m ) + sin ( L m )]

sec 2 [tan cos L + sin L ]

And

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)

.r02

RH 2 =

( )

4 1 + 9 tan 2 [tan sin 2 L + cos 2 L ] (4.18c)

tan m

[ tan sin ( L m ) + cos( L m )]

RH 3 =

(

1 + 9 tan 2

)

[ tan sin L + cos L ] (4.18d)

RH 4 =

(

1 + 9 tan 2 ) [ tan sin L + cos L ] (4.18e)

And

m

RH 5 =

sec [ tan sin L + cos L ] (4.18f)

2

In Fig. 4.3 the active Rankine thrust H1 acts at a distance .AC from point, C. Static equilibrium

3

of wedge EACD is considered.

Vertical force equilibrium condition gives

Pav = Pa sin = WACD + WADE Rv (4.19)

From which, Pa is obtained as,

WACD + WADE Rv

Pa = (4.20)

sin

Horizontal force equilibrium condition gives

PaH = Pa cos = RH + H 1 (4.21)

From which, Pa is obtained as,

30

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,

X0 = (4.24)

(

3 1 + 9 tan 2 ) ( )

e 2 tan m 1

Y0 = (4.25)

(

3 1 + 9 tan 2 ) ( )

e 2 tan m 1

Where, r1 is radius of arc of log spiral at the base of retaining wall, i.e. at = m

Axes, X and Y are another set of coordinate axes. Axis, X passes through the pole of log spiral

and is horizontal. Axis, Y is perpendicular to X axis and passes through the pole, O. With

reference to these axes, the coordinates, of centroid of log spiral are given 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).

friction, (degrees) 0 5 10 15 20 25 30

20 0.333 0.334 0.333 0.330 0.324 - -

25 0.333 0.334 0.334 0.332 0.328 0.323 -

30 0.333 0.334 0.334 0.333 0.331 0.327 0.322

35 0.333 0.334 0.335 0.334 0.332 0.330 0.326

40 0.333 0.334 0.335 0.334 0.333 0.332 0.329

45 0.333 0.334 0.335 0.335 0.334 0.333 0.331

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

Soil, Wall, Coulomb Caquot and Kerisel Soubra Lancellotta Proposed

(degrees) (degrees) (1776) (1948) (2002) (2002) Method

20 0.490 0.490 0.490 0.490 0.490

25 0.405 0.410 0.406 0.406 0.405

30 0 0.333 0.330 0.333 0.333 0.333

35 0.271 0.270 0.271 0.271 0.271

40 0.217 0.220 0.217 0.217 0.217

20 0.458 0.460 0.459 0.457 0.456

25 0.377 0.380 0.378 0.375 0.373

30 1/3 0.308 0.305 0.309 0.305 0.305

35 0.251 0.248 0.251 0.247 0.246

40 0.202 0.200 0.202 0.197 0.197

20 0.447 0.450 0.449 0.445 0.442

25 0.367 0.370 0.369 0.363 0.361

30 1/2 0.301 0.300 0.303 0.294 0.293

35 0.246 0.240 0.247 0.237 0.236

40 0.199 0.190 0.200 0.189 0.188

20 0.438 0.450 0.442 0.434 0.430

25 0.361 0.360 0.364 0.353 0.349

30 2/3 0.297 0.290 0.300 0.285 0.282

35 0.244 0.230 0.247 0.229 0.227

40 0.200 0.180 0.202 0.182 0.180

20 0.427 0.440 0.436 0.422 0.410

25 0.355 0.350 0.363 0.340 0.329

30 0.297 0.280 0.304 0.273 0.263

35 0.250 0.220 0.256 0.218 0.209

40 0.210 0.170 0.215 0.172 0.165

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

E G Log spiral failure surface

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

C G of area ADE

+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

K r0

+Y0 X

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

OD OE DE H

= = = (5.1)

sin v sin m sin m

sin135 -

2

In which, angles, m and v are as shown in the same figure.

From the above expression,

H sin v

OD =

sin m

The initial radius, OE = r0 of the log spiral is given as,

H sin 135 -

2

OE = r0 =

sin m

Also, from the equation of the log spiral,

OA = r0 .e m tan

And

AD = OA OD

Referring to Fig. 5.2(b) and considering triangle ODE,

OD OE DE H

= = = (5.2)

sin v sin A sin S sin S

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

OA = r0 .e m tan

And

AD =OA + OD

5.2.4 Computation of soil reaction on the failure surface

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.

surface EA

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,

Where,

e(3 tan )( L + ){3 tan sin ( L + ) cos( L + )}

p1 = (5.19)

( )

1 + 9 tan 2

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 resultant soil reaction, R on the failure surface is given as,

R = p.ds (5.22)

The vertical component, RV (Fig. 5.3) of resultant soil reaction is obtained as,

m

RV = p cos(

0

L + )ds (5.23)

m

RV = pr e 0

tan

cos( L + )secd (5.24)

0

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)

Rv WACD WADE

Pp = (5.29)

sin

Horizontal force equilibrium condition gives

PpH = Pp cos = RH + H1 (5.30)

RH + H1

Pp = (5.31)

cos

It may be noted that, both Eqs. 5.29 and 5.31 give the magnitude of unknown thrust, Pp. These

two equations will yield the same and unique value of Pp 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. 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,

X0 = (5.33)

3 (1 + 9 tan 2 ) (e 2 tanm 1)

Y0 = (5.34)

3 (1 + 9 tan 2 ) (e 2 tanm 1)

Where, r0 is radius of arc of log spiral at the base of retaining wall, i.e. at = 0.

Axes, X and Y are another set of coordinate axes. Axis, X passes through the pole of log spiral

and is horizontal. Axis, Y is perpendicular to X axis and passes through the pole, O. With

reference to these axes, the coordinates, of centroid of log spiral are given as,

X = Y0 sin + X0 cos (5.35)

Where, is the angle made by the axis, X0 with horizontal.

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.

2

WACD (OF + 3 DC) + WADE X

PpH (Ypp + FD ) = (5.37)

2

+ H 1 AC + FD + Pp V 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, PaH about

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

friction, 5 10 15 20 25 30 35 40

v -3.880 -9.264 -14.320 -19.065

25 Kp 3.404 3.705 4.001 4.287 4.560

v 6.568 1.254 -3.800 -8.622 -13.243

30 Kp 4.196 4.655 5.126 5.606 6.090 6.572

v 15.405 10.167 5.132 0.276 -4.427 -9.000

35 Kp 5.231 5.921 6.658 7.439 8.264 9.126 10.018

v 23.174 18.001 12.990 8.117 3.358 -1.303 -5.888

40 Kp 6.624 7.668 8.823 10.098 11.499 13.030 14.689 16.464

v 30.199 25.08 20.089 15.204 10.406 5.676 0.999 -3.639

-ve sign of angle v refers topole below the wall top

friction, ( degree) 5 10 15 20 25 30 35 40

20 0.225 0.226 0.229 0.234

25 0.235 0.235 0.237 0.241 0.248

30 .241 0.240 0.241 0.244 0.250 0.259

35 0.246 0.243 0.243 0.245 0.249 0.257 0.268

40 0.249 0.245 0.243 0.244 0.247 0.253 0.262 0.275

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).

O C B

D

0.00

Pp 45- /2 45- /2

m

H

v

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%.

Angle of soil friction, ( degree) 20 30 40

Angle of wall friction, ( degree) 1/2 1/2 1/2

Proposed Method 2.97 3.29 5.13 6.57 10.098 16.46

Coulomb (1776) 2.89 3.53 4.98 10.1 11.77 92.57

Caquot and Kerisel (1948) 2.60 3.01 4.50 6.42 9.00 17.5

Janbu (1957) 2.60 3.00 4.50 6.00 9.00 14.0

Sokolovski (1965) 2.55 3.04 4.62 6.55 9.69 18.2

Shield & Tolunay (1974) 2.43 2.70 4.13 5.02 7.86 11.00

Chen (1975) 2.58 3.14 4.71 7.11 10.07 20.90

Basudhar & Madhav (1980) 2.56 3.12 4.64 6.93 9.56 19.35

Kumar and Subba Rao (1997) 2.5 3.07 4.6 6.68 9.8 18.86

Soubra and Macuh (2002) 2.57 3.13 4.65 6.93 9.81 20.1

Lancellotta (2002) 2.48 2.70 4.29 5.03 8.38 11.03

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

In Fig. 6.1(a), the failure mechanism adopted in the analysis is shown. There are passive and

active states of equilibrium in which the failure surface consists of a log spiral followed by its

tangent that meets the ground surface. The anchor is flushing with the ground surface (basic

case, H/h = 1.0).

Oa

Op

D C B

L

45+ /2

45- /2 45- /2

Pa

H=h K Tu

J h/2

ha hp

A

Pp

E 90

dp G

dp

Normal to the tangent

Tangent

Normal

Fig. 6.1(a) Failure mechanism at ultimate load for continuous (strip) vertical plate anchor

in cohesion-less soil

t

Wp

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.

Pa = resultant active thrust

= angle of friction between soil and the plate anchor

= angle of soil friction

hp = distance of point of application of passive thrust, Pp from the anchor base

ha = distance of point of application of active thrust , Pa from the anchor base

Wp = weight of the anchor plate per meter

t = thickness of the anchor plate

N = upward soil reaction

The parameters Pp, Pa, hp and ha are computed using Ktters (1903) equation.

6.2.2 Ktters (1903) equation

O

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

In a cohesion-less soil medium with passive and active states of equilibrium under plane strain

condition, Ktters (1903) equation is given as,

dp d

+ 2 ptan = sin ( + ) for the passive states (Fig. 6.2(a)) (6.1a)

ds ds

And

dp d

2 ptan = sin ( ) for the active states (Fig. 6.2(b)) (6.1b)

ds ds

In the above equations,

dp = differential reactive pressure on the failure surface

ds = differential length of arc of failure surface

= angle of soil internal friction

d = differential angle and,

= inclination of the tangent at the point of interest with the horizontal

Kame, Dewaikar and Choudhury (2010 a & b) have reported a method based on the application

of Ktters (1903) equation for the estimation of active and passive thrusts on a vertical wall

retaining horizontal cohesion-less backfill. The unique failure surface consisting of a log spiral

and its tangent is identified on the basis of force equilibrium conditions and the point of

application of active/passive thrust is computed using moment equilibrium. In the proposed

analysis, this procedure is adopted to compute the values of Pp, Pa, hp and ha. The final

expression for the reactive passive pressure distribution at any point on the curved failure surface

using Ktters (1903) equation is obtained with the following expression.

tan (3 m 2 )

r0 Ksin + e

4 2

r sec e tan 3 tan sin ( + )

p = + 0

L

(6.2)

( )

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 starting radius of log spiral r0 as measured from point D (Fig. 6.3).

= spiral angle measured from the starting radius

59

m = maximum spiral angle

V = angle between vertical face of the wall and the starting radius r0 and

L = (90-V)

Op C B

D

0.00

Pp 45- /2 45- /2

m

H=h

v

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.

.r0 sece

1 + 9 tan 2

p= (6.3)

3 tan sin L + cos L

.r0 sece tan 2 + .e tan 2

{sin AD}

L

1 + 9 tan 2

Where, L = (45-/2) and AD = length of line AD (Fig. 6.4)

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

In Fig. 6.5, which shows free body diagram of failure wedge EACD the passive Rankine thrust

2

H1 acts at a distance .AC from point, C. Static equilibrium of wedge, EACD is then

3

considered.

Vertical force equilibrium condition gives

PpV = Pp sin = Rv WACD WADE (6.7)

Rv WACD WADE

Pp = (6.8)

sin

Horizontal force equilibrium condition gives

PpH = Pp cos = RH + H1 (6.9)

RH + H1

Pp = (6.10)

cos

X

Op F

Pole

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

hp = distance of point of application of passive thrust, Pp from the anchor base

= varying angle of inclination of reactive pressure with vertical

It may be noted that, both Eqs. 6.8 and 6.10 give the magnitude of unknown thrust, Pp. These

two equations will yield the same and unique value of Pp 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.

6.2.4 Trial and error procedure

In this procedure, first a trial value of V is assumed and corresponding weight of trial failure

wedge, EACD (Fig. 6.5) is computed. Using Eqs. 6.5 and 6.6, magnitudes of vertical and

horizontal components of soil reaction (RV and RH) are computed and from Eqs. 6.8 and 6.10,

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. Thus, in this method of

analysis, the unique failure surface (Fig. 6.6) 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.

Pole of log spiral

Op

O2

O1

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.

(Kame, Dewaikar and Choudhury, 2010 a & b)

Angle of friction Earth pressure coefficients

Soil, (degrees) Wall, (degrees) Ka (Active state) Kp (Passive state)

20 0.430 3.086

25 0.349 4.097

30 2/3 0.282 5.606

35 0.227 7.983

40 0.180 11.995

20 0.410 3.290

25 0.329 4.560

30 0.263 6.572

35 0.209 10.018

40 0.165 16.464

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.

Referring to Fig. 6.1(b), all the three equilibrium conditions are examined.

6.3.1 Vertical equilibrium

The forces involved are Pp sin and N in the vertically upward direction and Pasin and Wp in

the vertically downward direction. Since Pp > Pa and weight, Wp of the plate is small enough,

there is no equilibrium of the forces in the vertical direction. The reaction, N is zero and the plate

accelerates in the vertically upward direction. This agrees well with the experimental observation

(Naser, 2006).

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).

Tu = Pp cos Pa cos

(6.11)

As stated earlier, Pp and Pa are evaluated using Ktters (1903) equation and then Tu is

determined from the horizontal equilibrium condition.

The values of Tu are computed using the available experimental data and comparisons with the

available theoretical solutions are made.

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.

1. Ktters (1903) lends itself as a powerful tool in the limit equilibrium analysis of

vertical plate anchors.

2. Application of Ktters (1903) equation makes the analysis statically determinate and

facilitates the evaluation of horizontal and vertical soil reaction on the failure surface.

3. The assumption of a failure surface consisting of log spiral and its tangent yields

better predictions in comparison with available theoretical methods when compared

with experimental results.

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

Engrg. Div., ASCE, 104(2), 283286.

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.

lower bound break-out factors of strip anchors. Geotechnique, 44(2), 307-318.

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.

Gesellschaft, 12, 300.

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.

plates. J. Geotech. Eng. Div., ASCE, 101(7), 711715.

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.

rectangular and square anchors in cohesion-less soil. Int. J.of Geotech. Eng., 4, 79-87

Terzaghis mechanism. Int. J Geomech., ASCE 3(1),123128

wall-point of application. Soils and Foundations, 42(1), 915.

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.

Terzaghis Mechanism. International Journal of Geomechanics, 3(1), 123-128.

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.

Denmark.

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.

procedures of slices. Proc., 4th Int. Conf. on Soil Mechanics and Foundation Engineering.

London.

vertical retaining wall with cohesion-less backfill. Electronic journal of geotechnical

engineering (EJGE), U.S.A., 15(Q),1848-1863

vertical retaining wall with horizontal cohesion-less backfill. communicated to Soils and

Rocks, Int. J. of Geotech. and Geo-env. Eng., Brazil.

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.

617619

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.

bound solutions for plate anchors in sand,Geotechnique,43, 852-868

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.

cohesionless medium. Geotechnique, 39(3), 417-431.

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.

Makassar International Conference on Civil Engineering (MICCE2010), March 9-10, 2010,

Indonesia,

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.

Royal Society of London, 147

Geotechnique, London, England, 32(l), 9-23.

Geotechnique, London, England, 32(1), 25-41.

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.

Kame, G. S., Dewaikar, D. M. and Deepankar Choudhury. ( 2010-b). Passive thrust on a

vertical retaining wall with horizontal cohesion-less backfill. Soils and Rocks, Int. J. of

Geotech. and Geo-env. Eng., Brazil.

vertical plate anchors in cohesion-less soil. Int. J. of Geomechanics and Geoengineering,

Techno press, Korea.

77

- Pile Foundation Design - A Student GuideDiunggah olehMaurice Balkissoon
- RRM Gravity DamDiunggah olehLivian Teddy
- Composite Behavior of Geosynthetic Reinforced Soil Mass FHWA-HRT-10-077Diunggah olehyudhatebe
- Pile Design CalculationDiunggah olehhessian123
- Lecture20 Drilled ShaftsDiunggah olehAsia Ward
- Nonlinear Finite Element Analysis of Four-Pile Caps Supporting Columns Subjected to Generic Loading 04 July.pdfDiunggah olehPrantik Adhar Samanta
- 91143345 Pile Cap DesignDiunggah olehอภิรักษ์ มานะกิจศิริสุทธิ
- Repair and Strengthening of Bridge SubstructuresDiunggah olehLuis Alberto Riquelme Vidal
- Foundations Theory & DesignDiunggah olehbhavik91
- Chapter2-Lateral Earth Pressure RE WallDiunggah olehmoin
- How to Do Pile Driving With Percussion MethodDiunggah olehroypun
- TC207Diunggah olehgigarostom
- Pile Driving EquipmentDiunggah olehqwesrt
- Drilling RigsDiunggah olehMada Lina
- Introduction to Geo.tech. Engineering-SivakuganDiunggah olehRoshanRSV
- Pile Footings for Flats for Golden Green at AsrDiunggah olehraman_2107
- 826Diunggah olehhiyeon
- Pile CapacityDiunggah olehVasanthakumar
- The Optimal Shapes of Piles in Integral Abutment BridgeDiunggah olehninja
- 1 viewNitPdf_1716598Diunggah olehlaxmicc
- Lecture 22Diunggah olehSarah Cherry
- 20040504Diunggah olehRonald De Guzman
- 2010-0210Diunggah olehMelinda Gordon
- Three-Dimensional Dynamic Behaviour of Batter Pile GroupsDiunggah olehIJRASETPublications
- Geotech3 LS6 FoundationsDiunggah olehAisyah
- IJETR2286Diunggah olehanil kasot
- Murray_gDiunggah olehtrdnhiemg
- Initial load testDiunggah olehvivek
- Deep Comp ActionDiunggah olehFaiz Nasir
- Tensar_Basal_Reinforcement.pdfDiunggah olehFernanda Sequera

- BlackBook INDEXDiunggah olehDanishPawaskar
- Example 2 Parametric Study of Beams for Quantity of Steel Using Method of CurtailmentDiunggah olehDanishPawaskar
- Guidelines Thesis Dissertaion Report-Mumbai UniversityDiunggah olehbronin
- Srs Template 1Diunggah olehAnjali Singh
- Aviation Data Analysis ParametersDiunggah olehDanishPawaskar
- INTELLIGENT SYSEMS EXPERIMENT 2Diunggah olehDanishPawaskar

- Inwa.txtDiunggah olehdfgrtg454g
- Benson Autonomy How to Do It Here (1)Diunggah olehAlan
- The Local ChurchDiunggah olehMarko Petrović
- CRM Evaluation GuideDiunggah olehRamon Rincones
- Suicide PreventionDiunggah olehravigoa
- Statistical significanceDiunggah olehpinteaovidiu
- Ancient Stellar Races Involved With Humanity - Ashayana DeanneDiunggah olehRobson Honorato
- Scott Onstott - Taking Measure - Explorations in Number, Architecture and Consciousness.pdfDiunggah olehpep
- Trading Strategies in the Indian Stock Market 2461Diunggah olehSai Krishna
- Colloquium - Topological Insulators.pdfDiunggah olehDavid
- PSYLIFE_Chapter 1 Outline-1Diunggah olehJose Mitchell
- The Adventure of Sherlock Holmes by Arthur Conan DoyleDiunggah olehAnay J Panchal
- ALS Using the Lux Equation DN29A.pdfDiunggah olehJose Luis Alcazar Romero
- High Frequency TransformerDiunggah olehtosikur rahman
- Observation Notes FormDiunggah olehSonny Matias
- Complete each sentence with the correct form of the word given in brackets.docDiunggah olehCarla Maria Sotoc
- A Study Guide to Aristotle's PhysicsDiunggah olehcamestres baroco
- Distinction Between Organic Compounds Chemical TestDiunggah olehMission Security Services Pvt. Ltd
- Gérard Genette and the Pleasures of PoeticsDiunggah olehApple Xu
- Defendant’s Motion to Disqualify Judge Edward l. ScottDiunggah olehNeil Gillespie
- m.digestive SystemDiunggah olehgilissa
- SamplingDiunggah olehRezky Naufan Hendrawan
- Lutheranism and the Inerrancy of ScriptureDiunggah olehAdriano Da Silva Carvalho
- Husserl - Cartesian MeditationsDiunggah olehRoland Adorjani
- Sc Food GroupsDiunggah olehyrbag
- 020204 Paper ZeolitasDiunggah olehclaudia
- Elect Ivo RedesDiunggah olehKen Matsuda
- Suvarnaprashan Sanskar EnglishDiunggah olehLifecare Ayurveda Drnikul Patel
- Halo Al KanesDiunggah olehNirmal Sharma
- 15-06-10-SCDiunggah olehJam Zaldivar