=*u&bwy

All rights reserved

INFORMATION TO ALL USERS The qua lity of this rep roduction is d e p e n d e n t upon the qua lity of the copy subm itted.

In the unlikely e ve n t that the au tho r did not send a c o m p le te m anuscrip t and there are missing pages, these will be no ted . Also, if m ateria l had to be rem oved,

a n o te will ind ica te the de le tion .

Published by ProQuest LLC (2017). C opyrigh t of the Dissertation is held by the Author.

All rights reserved.This work is protected aga inst unauthorized copying under Title 17, United States C ode

M icroform Edition © ProQuest LLC.

ProQuest LLC.789 East Eisenhower Parkway

P.O. Box 1346 Ann Arbor, Ml 481 06 - 1346

The Influence of the Construction Process

on Bored P i le s and Diaphragm Walls

A Numerical Study

by

Goksel Kutmen B.Sc.(ITU)

A thes is submitted to the Univers ity of Surrey

for the degree of Master of Philosophy in the

Department of C iv i l Engineering

November, 1986

To my Mother and Father

Summary

It is known that in s ta l la t io n of c a s t - in s i tu concrete structures

such as bored p i les and diaphragm walls may influence the properties

of the adjacent s o i l .

Insta l la t ion effec ts are important because they may change the

stresses applied to, and subsequent performance of cast- ins i tu bored

pi les and diaphragm walls.

These effects have been studied using the f i n i t e element method. In

part icu lar the generation of excess pore pressures during

construction and subsequent d iss ipat ion of these pore pressures have

been examined using coupled consolidation analyses.

A number of analyses featuring d i f fe ren t variat ions of the

construction process have been carried out. The resu lts show that

the in -s i tu stress regimes around bored p i les and diaphragm walls do

change due to in s ta l la t io n and that there is only part ia l recovery

with time.

Acknowledgements

The author is grateful to his supervisors

Dr C.R.I. Clayton and Mr M.J. Gunn for the ir

guidance, help and support throughout th is

study and would l ik e to acknowledge the

f inanc ia l support of the Turkish State

Railways (TCDD).

The author wishes to thank Miss M.C. Bryant

for her help, and for typing th is thes is .

The fol lowing abbreviations are used for the d i f ferent analyses

which were carried out in th is study:

D-NS/ND/12

D-BS/ND/12

P-NS/ND/12

P-NS/ND/0

P-NS/1D/12

P-NS/7D/12

P-BS/ND/12

P-BS/ND/0

P-BS/1D/12

Diaphragm wall , without any support, no delay in

concreting, 12 hours concrete setting time.

Diaphragm wall , with bentonite support, no delay in

concreting, 12 hours concrete setting time.

Cast- ins itu bored p i l e , without any support, no delay in

concreting, 12 hours concrete setting time.

Cast- ins itu bored p i l e , without any support, no delay in

concreting, immediate concrete set.

Cast- ins itu bored p i l e , without any support, 1 day delay

in concreting, 12 hours concrete setting time.

Cast- in s itu bored p i l e , without any support, 7 days

delay in concreting, 12 hours concrete setting time.

C ast- ins itu bored p i l e , with bentonite support, no delay

in concreting, 12 hours concrete setting time.

Cast- ins itu bored p i l e , with bentonite support, no delay

in concreting, immediate concrete set.

Cast- ins itu bored p i l e , with bentonite support, 1 day

delay in concreting, 12 hours concrete sett ing time.

P-BS/7D/12 Cast- ins itu bored p i l e , with bentonite support, 7 days

delay in concreting, 12 hours concrete setting time.

Contents

Summary ..................................................................................................................... i

Acknowledgements ............................................... i i

L is t of Abbreviations ..................................................................................... i i i

Contents .................................................................................................. iv

1 Introduction .................................................................................. .................... 1

1.1 Research Objective ............................................... ................................... 1

1.2 Thesis Organisation ................................................................ . .............. 1

2 Diaphragm Wall/Cast-insitu Bored P i le Insta l la t ion and i t s

Effects ................................................................................................................. 3

2.1 Construction Process .......... ............................................................. .. 3

2.1.1 Diaphragm Walls .............................................................. .......... 3

2.1.1.1 Excavation ............................... ................ .................. ... 3

2.1 .1.2 Concreting ..................................................................... 6

2.1.1 .3 Front Excavation .......................................................... 6

2.1.2 Cast:-insitu Bored P i les .............. 9

2.1.2.1 Excavation ...................................................................... 9

2 .1 .2.2 C o n c r e t i n g ............................................... 12

2.2 Changes in Ground Conditions due to Bored P i le and Diaphragm

Wall Insta l la t ion in Saturated Clays .......................... 14

2.2.1 Excavation Stage ....................................... . ........... 14

2.2.2 Concreting Stage ............ 15

2.2.3 Pore Pressure Equilibrium Stage ......................................... 16

3 F in i te Element Method in Geomechanics .............. 17

3.1 F in i te Element Formulation for Coupled Consolidation

Analysis ..................................................................................................... 17

3.2 CRISP Programs ......................................................................................... 25

3.2.1 CRISP Programs in General ...................................................... 25

3.2.2 CRISP Programs, Modif ications and Addit ional

Programs ......................................................................................... 26

4 Analysis 30

4.1 Material Properties .............................................................................. 31

4.2 Soil Model ................................................................................................. 32

4.3 Types of Analyses .................................................................................. 35

4.3.1 Plane Strain Analyses ................................................ 35

4.3.2 Axisymmetric Analyses ............................................................ 35

4.4 Geometry Data ................................................................................. 37

4.4.1 Size of the Problem ......................................... 37

4.4.2 F in i te Element mesh and Element Type ................................. 37

4.5 In-situ Stresses and Pore Pressures ............................................. 40

4.6 Simulation of the Construction Process ................................. 42

4.6.1 Excavation ..................................................................................... 42

4.6.2 Concreting .................. . 52

4.6.3 Consolidation ............................................................................ 55

4.7 Choice of Increment Numbers and Time Intervals ......................... 62

5 Results ............................................................................................................... 66

5.1 Diaphragm Walls ..................................................................... 66

5.1.1 Analysis of Diaphragm Wall Construction Without

Support (D-NS/ND/12) ................................................................ 66

5.1.2 Analysis of Diaphragm Wall Construction With

Bentonite Support (D-BS/ND/12) ........................................... 67

5.2 Cast- in s itu Bored Pi les ...................................................................... 79

5.2.1 The Analysis P-NS/ND/12 and the Analysis P-BS/ND/12 . 83

5.2.2 The Analysis P-NS/1D/12, the Analysis P-NS/7D/12, the

Analysis P-BS/1D/12, the Analysis P-BS/7D/12, the

Analysis P-NS/ND/0 and the Analysis P-BS/ND/0 ............ 107

6 Discussion ......................................... 112

6.1 Analysis .................................................................................... 112

6.1.1 Assumptions ....................................... 112

6.1.2 Numerical D i f f i c u l t i e s ..... ....................................................... 114

6.2 Results ..................................................................................................... 123

v

Bibliography ....................................................................................................... 139

Appendices ................................ 146

Appendix 1. The E f fect ive Stress Method for Estimating the Shaft

F r ic t io n of Pi les and 3 Values ......................................... 148

Appendix 2. Horizontal E f fect ive Stress Graphs ................................ 150

Appendix 3. Horizontal Total Stress Graphs .................................. 169

Appendix 4. Vert ica l E f fect ive Stress Graphs ..................................... 188

Appendix 5. Excess Pore Pressure Graphs ............................................... 207

7 Conclusions ........................... 136

1 Introduction

1.1 Research Objective

Cast- ins itu bored p i les and diaphragm walls have been widely used in

C i v i l Engineering pract ice because of their economical and practical

advantages.

The construction process of these structures is known to cause some

changes in the ground condit ions such as in the stress and pore

pressure regimes, moisture contents etc. The changes in the ground

condit ions are very important because they d i r e c t l y affect the

subsequent performance of c a s t - in s i tu bored p i les and diaphragm

walls. Although al l researchers agree that the ground condit ions do

change during the construction process, there is no agreement

whether the ground condit ions come back to their i n i t i a l state a

certain period after the completion of in s ta l la t io n .

This study endeavours to examine the changes in the stress and pore

pressure states due to the in s ta l la t io n of c a st - in s i tu bored pi les

and diaphragm walls step by step during the construction process and

diss ipat ion of excess pore pressures, using the f i n i t e element

method with coupled consol idation theory. :-

1.2 Thesis Organisation

The in s ta l la t io n process of ca s t - in s i tu bored p i le s and diaphragm

walls is described in Chapter 2. A short l i te ra tu re review of the

in s ta l la t io n ef fects on the surrounding soi l is also presented in

that chapter.

Chapter 3 consists of f i n i t e element formulations for the coupled

consolidation analysis and information about the computer programs

used in th is study.

Chapter 4 describes the f i n i t e element analyses which have been

carried out and the d i f fe rent boundary condit ions for the d i f fe rent

analyses. It also gives information about the material propert ies,

insitu stress and pore pressure regime, and soi l model considered in

the analyses.

Results, discussion and conclusions are presented in Chapter 5,

Chapter 6 and Chapter 7 respect ive ly .

In Appendix 1, the e f fec t ive stress method for estimating the shaft

f r i c t i o n of pi les and 3 values are presented.

Appendix 2, Appendix 3, Appendix 4 and Apendix 5 present the

horizontal e f fec t ive s tress , horizontal total s tress , vert ica l

ef fec t iv e stress and excess pore pressure graphs respect ive ly .

2 Diaphragm Wail/Cast- insi tu Bored P i le Insta l la t ion and i ts

Effects

In th is chapter a short l i te ra tu re review of the in s ta l la t io n

process of diaphragm walls, ca st- ins i tu bored p i les and its ef fects

on surrounding so i l is presented. The f i r s t part describes the

construction processes of both diaphragm walIs and cast- ins i tu bored

p i les . The second part endeavours to highlight the ef fects of the

in s ta l la t io n on surrounding soi l re ferr ing to the previous studies

done by other researchers.

2.1 Construction Process

2.1.1 Diaphragm Walls

A diaphragm wall can be described as an a r t i f i c i a l membrane of

f i n i t e thickness and depth constructed in the ground by means of a

process of trenching with the aid of a f lu id support. Because of

their practical and economical advantages, diaphragm walls have been

widely used .as retain in g structures, vert ica l load bearing elements

and impermeable membranes in recent years.

Construction is carr ied out .from the . ground level by means of

various a lternat ive kinds of mechanical devices which permit the

progressive excavation of a r e l a t i v e l y narrow trench in the ground.

The s ta b i l i z in g f lu i d is introduced simultaneously as the trenching

operation proceeds. When the excavation has been completed, the

trench is f i l l e d with concrete or some other kind of b ac k f i l l

material ,, thereby d isp lacing the trench-supporting f lu i d (usually

bentonite) from the bottom upwards.

2.1 .1 .1 Excavation

Whatever technique is employed, or whatever tools are used, the main

feature of the trench excavation is the usage of the supporting

f l u i d , normally bentonite, for various purposes. When bentonite is

used in the excavation, i t penetrates into the so i l at the sides of

3

the trench and has the a b i l i t y to form a membrane (termed a f i l t e r

cake) of low permeabil ity at the s o i l - l i q u i d interface (Fig 2 .1) .

The depth of penetration may vary depending on soi l type (Fig 2.2).

The f i l t e r cake allows development of the f u l l hydrostatic pressure

induced by the s e l f weight of the bentonite s lurry against the sides

of the trench. This combined with l imited penetration of bentonite

into the s o i l , is the main factor contributing to the s ta b i l i z a t io n

of the trench (Fuschsberger, 1974). The ground water table must be

more than 1.5m below the bentonite level in the trench in order to

give a net bentonite pressure head on the trench walls, and the so i l

layer must not be highly permeable (Fleming and S l iw insk i , 1977).

The methods of trench excavation used to make diaphragm walls can be

c la s s i f i e d into two groups according to the transportation of the

excavated earth to the surface (Hajnal et a l , 1984).

a) The Mechanical Transportation Method

In th is method, excavating tools such as the auger, bucket, shovel,

grab etc. cut the soi l in bulk and have to be brought up above

ground level to discharge the spoil . . The main task of the s lu rry in

th is case is the s ta b i l i z a t io n of the trench.

In. trench . excavation the grabbing tool is commonly operated on

either a rope or a ke l ly bar. Excavation is carried out by lowering

a grab f i t t e d with teeth and a cutting edge. The grab cuts into the

so il with i ts f u l l y opened jaws due to i ts pure s e l f weight. The

jaws are then closed either mechanically or h ydrau l ica l ly , enclosing

the soi l inside the grab. The soi l is brought to the surface by

l i f t i n g the grab to discharge i t (Fig 2.3). In soft cohesive soi l

the grab can be pushed into the ground by using a r ig id rod. This

may increase the effectiveness of grabbing.

b) The Mud Transportation Method

In the mud transportation method percussive, rotary tools loosen and

break the so i l into small pieces. The cuttings are mixed with the

bentonite suspension at the cutting face. The suspension bearing

the soi l cuttings is then transported to the ground le v e l . This is

4

Fig 2.

Fig 2.

1 F i l t e r cake formed by bentonite s lu rry (Fleming and Sl iw insk i , 1977)

• ‘‘ ‘Slurry soturoted :• . * zone;

°.?°.W poo * 9- °* Slurry % OpO* o .O’ *° oS. °*°n° 'O a*0 “ ..OAo«o‘ .O ..'•o V «9 O »° O O 0*° 0 °0.4.0+•o“o . ° L “„ " 0 *o o.O » «o .°Q°»Q O Q» ,*c' O " o V f o » o » V ® . <° o 'o °, 0,/0 * ° *°* •° O • “o * JO t ° 'o 4 '0 o°%'o*,°t? '/ • V 0 " %j * o .* » . * * 0 ° ° -• 2 ^ - o ' o ^ / . O " *

£a.‘‘: V S>and V

/lO-.V-Ky ' Gravel o*'°o*

o* °° *00;; o ‘.0";° ?p-o,

, . * * ° *o*U^Gravelly s i lt "0 !.♦ O o *0* 0 * 0

2 Depth of slurry penetration (Hajnal et al, 1984)

5

done by either d irect or reverse c i rcu la t io n of the bentonite

s lu r ry . These tools advance in s i tu and are not removed from the

cutting face unti l excavation is complete (Fuschsberger, 1974). 'The

function of the s lurry is to transport the sol id material to the

surface as well as s t a b i l i z a t io n of the trench (Fig 2.4).

2 .1 .1 .2 Concreting

The bentonite s tab i l ized trench is normally backf i l led with concrete

to construct a diaphragm wall for structural purposes. The concrete

used in th is process should have a uniform consistency and be of

high qual ity (see Federation of P i l in g S p e c ia l i s t s , 1973, for

d e t a i l s ) . The s ta b i l i z in g f l u i d has to be replaced e n t i re ly by the

concrete to avoid consequent unwanted effects on the diaphragm

w a l l ’s performance.

Concrete is placed from the bottom of the trench upwards as

concreting is done in the presence of bentonite s lu rry . A tremie

pipe is used to cast the concrete by gravity into the trench. It is

ra ised from the bottom of the trench in stages as the concrete level

r ises (Fuschsberger, 1974). The tremie is always kept immersed in

concrete to prevent the concrete mixing with bentonite s lu rry (Fig

2.5). Concrete pushes the s lu r ry upwards, scraping the f i l t e r - c a k e

o f f from the sides of the excavation (Fig 2.6). In order to do

t h i s , the concrete must be of a p l a s t i c consistency, behaving as a

heavy f lu i d since concrete which has a s ig n i f ica n t shear strength

may not exert s u f f i c ie n t pressure to displace the bentonite

suspension (Sliwinski and Fleming, 1974).

Reinforcement, i f necessary, is placed in the trench before

concreting.

2 .1 .1 .3 Front Excavation

When the in s t a l la t io n of the diaphragm wall into the ground is

completed, the front of i t is excavated to form the f in a l

structure . In th is stage ground anchors and/or props are in s ta l le d ,

i f required, as excavation proceeds.

6

Side-view

Fig 2.

Fig 2.

/■60rnp.9C|~|1.80|mp.SC|Plan

3 The mechanical transportation method (Hajnal et a l , 1984)

4 The mud transportation method (Hajnal et al, 1984)

7

vtCD "”** T3 u <i>oa •— c a E £

<uto q .

u n,'"0 c a> -O

E “ tn ± o a -c

xi {5 — to 2: o> r= o g0 ^ — 0

%X? D +s °-~>• <u C +J ;r.

ccc+-0.'

TO X.

>: a> <t% 0) O)

u

3 £ |to to — 3 2

=> h EI -O £ TOrtj b: X}V- O CL)Ol U W

E cucO C

T> 0).2 § Ep. S'S .2 p x> ^ q E^ TO TO

.2! to E cv- n TO £t/i TO

.a 0 0^0E xi .a ~ <7TO 4J TO. CO OIE -c t= 10H*

aE . -d

•— a> >. ci£ c § |c 'i y<S§.£K3

TJd) 1—OJ To»- o .to a.•2 o Se t) xi .2 «to E =TO TO ̂3TO ^ — .C O -EO +J C u .2 3

xi ETO OTO CL C

,2 i TO TO X ) TO C

§ -5 H o E al. c h- a “ 3l— o o a P «

8

Fig

2.5

Vari

ous

stag

es

of co

ncre

ting

by

trem

ie

.{Fl

emin

g an

d S

liw

insk

i,

1977

)

2.1.2 Cast-insitu Bored Piles

The need to transfer very large loads to the so i l from gigant ic

on-land or off-shore structures and the necessity of constructing

structures on weak foundations coupled with the s ig n i f i c a n t advances

in p i l in g technology have made p i l in g , espec ia l ly bored p i l i n g , very

popular during the past two decades.

Cast- ins i tu bored p i les are constructed by d r i l l i n g a hole in

the ground and f i l l i n g the hole with concrete. Reinforcement, i f

necessary, is placed into the hole just before the concreting.

2 .1 .2 .1 Excavation

Excavation can be done by a number of d i f ferent methods. Regardless

of the excavation method, the s t a b i l i t y of the hole has to be

maintained during the boring process. The walls of the hole may

cave in as the hole is advanced and create a larger hole than

intended in some types of ground or layer sequence i f no supporting

method is used. Sometimes the so i l caves in and f i l l s up the bottom

as fast as i t can be removed. In some cases, however, the soi l is

strong and stable enough to permit the excavation to remain stable

without any form of support. - -

Boring Methods

a) Percussion

In th is method, the hole is formed by dropping a digging tool (e.g.

a sludge pump) onto ground. The digging tool cuts into the s o i l ,

when l i f t e d , brings some spoil which is trapped in i t . The spoil is

discharged at the surface. The hole is dug in the ground by

constant repet i t ion of th is process.

For small diameter p i le s , a sludge pump in non-cohesive so i ls ,and a

similar tool without a f lap valve at the bottom of the tubular

section in cohesive s o i l s , is used. The tools used for large

diameter p i les are various forms of grab for sands, gravels and soft

9

clays, hammer grabs for harder c lays, soft rocks, and ch ise ls in

combination with grabs for the harder rocks.

A short length of temporary casing is always inserted at the top of

the hole. This prevents so i l and surface water f a l l i n g into the

hole. The casing may be extended as the hole deepens, i f necessary.

b) Rotary Boring

Boring is carried out by a rotary r i g . A short f l i g h t auger or a

d r i l l i n g bucket is used. Having been located over the p i le

posit io n , the d r i l l i n g tool is rotated and lowered into the ground.

When the f l ig h t s or bucket are f i l l e d with spoil the ke l ly and b i t

are l i f t e d . The spoil is discharged on the surface. The tool is

then reintroduced into the hole and boring is continued. This

process is repeated unti l the required depth has been reached.

Various techniques may be employed to support the sides of the

hole. Temporary casing and bentonite s lurry are common ones.

c) Flush D r i l l i n g

This method is l ik e that used in o i l well d r i l l i n g . The ke l ly and

d r i l l i n g tool are rotated and cut into the ground in order to form

the hole which is kept f u l l of water. The spoil is transported by

water which is cont inually pumped back out of the hole through an

opening in the d r i l l i n g to o l , up the ke l ly through the swivel and

suction pipe.

Bentonite s lu rry may be used instead of water. Temporary casing is

not necessary. The d r i l l i n g operation is a continuous process

without breaks for the removal of s p o i l .

Supporting Methods

Dif ferent supporting methods, which are used in order to overcome

the i n s t a b i l i t y of the hole can be divided into the fol lowing categories.

a) S lurry (Bentonite Suspension)

Bentonite s lurry is used in bored p i l in g to carry out a wide variety

of functions such as excavation support, excavation seal ing etc.

which were discussed in part 2 .1 .1 .1 . Bentonite makes possible

unhindered use of rotary boring r ig s with a high speed of

excavation, and the el imination of casing from the construction

procedure. Consequent economy in d r i l l i n g can be achieved.

If d r i l l i n g is carried out in the presence of bentonite, the sides

of the borehole are often irregular and the excavating tools may

cause overbreak in less compact strata . This may lead to an

excessive use of concrete, but i t has no effect on the p i le

performance (Fleming and S l iw in sk i , 1977).

Any unwanted effects of bentonite on the p i le behaviour can be

avoided by complying with the bentonite s lurry spec if icat ions which

are avai lable in the l i te r a t u r e (e.g. Sl iwinski and Fleming, 1974;

Hutchinson, et al 1974, Fleming and S l iw insk i , 1977).

b) Casing

The s t a b i l i t y of a p i le boring can be po s i t iv e ly achieved by using

casing. Depending on the boring method, casing is inserted either

while excavation progresses or afterwards. It can be temporary or

permanent, inserted to f u l l depth or p a r t i a l l y . Especia l ly in

unstable ground, casing is often driven to the f u l l depth of a p i le

before boring is carried out.

Casings are normally avai lable up to about 20m in length. Sectional

casing systems may be used for longer p i les in unstable ground but

excavation has to be carr ied out by a clamshell grab in th is case.

The boring process is therefore slower.

If the f u l l length of casing is inserted before excavation overbreak

is negl ig ib le and the sides of the borehole are even. I f , however,

boring is carried out below the bottom of casing in unstable s o i l s

large cav it ies may occur. This can cause problems at the concreting

stage (Fleming and S l iw in sk i , 1977).

11

In general, casing prevents the loosening of the surrounding soil

and helps to achieve a stra ight p i le with the required diameter.

c) Hollow-Stern Continuous F l ig h t Auger

A hollow-stern continuous f l i g h t auger is used in the boring process

and maintained in the borehole unti l concreting begins. Then it is

extracted from the hole gradually as the concrete is pumped down the

hollow stem of auger.

D i f f i c u l t i e s in f inding long augers l im it the usage of th is method

only to short p i le s .

d) Combined Methods

Under certain circumstances both casing and s lu rry may be. used as.

complementary techniques to s ta b i l i z e the hole.

2 .1 .2 .2 Concreting

The technique to be used to f i l l the p i le hole with concrete depends

on the support used. ■

If no support is used or no casing necessary, f i l l i n g the hole with

concrete is f a i r l y stra ight forward. Concrete is dropped by;:.gravity

(free f a l l method). High slump mixes are usually used so that the

concrete does not segregate.

When bentonite s lurry is present, the concreting operation is done

using a tremie pipe. The technique used to carry out concreting was

described in Section 2 .1 .2 . The concreting procedure is r e l a t i v e l y

simple in the absence of casing (Fig 2.5).

If temporary casing is used concreting is considered as a

complicated process. Concrete is either dropped by gravity or

placed using a tremie pipe. Water f i l l e d cav it ies outside the

casing may cause a reduction in a p i le diameter or a d iscont inu ity

in a p i le shaft when the casing is withdrawn. General ly, a vibrator

is employed to extract the casing so that any damage to the p i le

section through arching and loss of workabil ity can be avoided

(Fleming and S l iw insk i , 1977),

12

Fig 2.6 The removal of the f i l t e r cake by concrete

Lateral concrete pressure, kN/m’

Fig 2.7 Total stresses between fresh concrete and soi l (Clayton and Mi 1 i t i t s k y , 1983)

13

2.2 Changes in Ground Conditions due to Bored P i le and Diaphragm

Wall In s ta l la t io n in Saturated Clays

It is obvious that ground condit ions prior to the construction

process such as stress state , pore pressure state, moisture content

e t c . , change during the in s ta l la t io n due to stress r e l i e f ,

softening, swelling etc. But it is not easy to say whether these

ground condit ions wil l eventually come back to their i n i t i a l values

after the in s ta l la t io n or not. This is an important question yet to

be answered. Since the bearing capacity of bored p i les and the

loads applied to diaphragm walls depend on the so i l condit ions after

i n s t a l la t io n , they c e r ta in ly af fect the performance of these

structures. In the case of a diaphragm wall as a retaining

structure, wall movements during the front excavation are affected

by the stress state prior to the excavation.

Although there has been an increasing interest in the in s ta l la t io n

effects on bored p i le performance recent ly (Chadeisson, 1961; Endo,

1977; Zliwinski and Philpot, 1980; Bustamante and Gouvenot, 1979;

Curt is , 1980; Clayton and M i l . i t i tsky , 1983; Mil i t i t sky, 1983) there

has not been a large amount of research done on th is top ic . Recent

developments in ca lcu la t ing the wall movements using f i n i t e element

method (Potts and Burl and, 1983(a) and 1983(b); Potts and.Fourie,

1984; Hubbard et a l , 1984; Tedd et a l , 1984; Wood and Perrin, 1984)

have in i t ia ted a new interest in e f fects of in s ta l la t io n procedures

on diaphragm wall behaviour.

In th is section, in s ta l la t io n ef fects at the d i f fe rent stages of the

construction process wil l be discussed. Construction stages are

divided into three groups: 1. Excavation, 2. Concreting, 3. Pore

Pressure Equ i l ib ra t ion .

2.2.1 Excavation Stage

When a hole/trench is excavated, total horizontal stresses at the

s ides, and total vert ica l stresses at the base of the hole/trench

f a l l to lower values than the i n i t i a l ones, with values which depend

14

on the supporting methods used. This causes some changes in stress

and pore pressure states in the v i c i n i t y of the hole/trench.

Swelling and softening occur due to migration of water from the

surrounding so i l mass towards the hole/trench as a re su l t of the

depression of pore pressures. These effects are r e l a t i v e l y small in

magnitude i f any means of support is used during the excavaton.

When bentonite s lu rry is used during the excavation, the f i l t e r cake

which is formed at the sides of the hole/trench may present a

potential weakness in the so i l structure system because of its

r e l a t i v e l y low undrained shear strength and undesirable e f fec t iv e

strength parameters (Clayton and Mil i t i t s k y , 1983). Although there

is a p o s s i b i l i t y that the f i l t e r cake is removed by concrete tremied

into the base of the hole/trench, i t has been suggested that a very

thin bentonite layer wil l be l e f t in surface i r r e g u l a r i t i e s . In

c lays, however, only a few mil l imetres thick f i l t e r cake is found at

the sides of the hole/trench even before concreting (Fleming and

S l iw insk i , 1977) and i t is r e l a t i v e l y thin comparing with the

surface i r r e g u la r i t i e s . It has been concluded that the adverse

effects of bentonite s lurry on the load carrying capacity of bored

pi les and vert ica l load bearing diaphragm walls are in s ig n i f ica n t

(Chadeisson, 1961; Burland, 1963; Fernandez, 1965; Komornik and

Wiseman, 1967; Farmer et a l , 1970; O'Neil and Reese, 1972; Corbett

et a l , 1974; Fleming and S l iw in sk i , 1977; Fearenside and Cooke,

1978).

2 .2.2 Concreting Stage

After the completion of excavation, the hole/trench is. f i l l e d with

concrete. When the concrete is placed, the interact ion between clay

and concrete s ta rts . The e f fects of wet concrete on the surrounding

soi l have been investigated by a number of researchers (Meyerhof and

Murdock, 1953; Skempton, 1959; Mohan and Chandra, 1961; Burland,

1963; Tay lor , 1966; Chuang and Reese, 1969; Chandler, 1977;

Fearenside and Cooke, 1978; M i l i t i t s k y et a l , 1982; M i l i t i t s k y ,

1983; Clayton and M i l i t i t s k y , 1983).

15

Normally, high slump concrete with a high water/cement ra t io is

used. Before concrete sets, i t applies total stresses to the sides

of the hole/trench. Fig 2.7 which is taken from Clayton and

M i l i t i t s k y , 1983 demonstrates the measurements of pressures done by

DiBiagio and Roti , 1972 and Uriel and Otero, 1977. Lateral

pressures seem to be hydrostatic at shallow depth, but the increase

with depth is not hydrostatic at larger depths. The la tera l

pressures decrease with time as hydration and setting takes place.

Since the water pressure is higher in wet concrete than that in the

surrounding s o i l , water migration takes place from the concrete to

the surrounding s o i l . This causes an increase in the moisture

content of the adjacent s o i l , and consequently swelling occurs. An

increase up to 5 - 6% in moisture content between 2 - 7 cm from the

soi l/concrete interface i . e . swelling zone has been observed

(Meyerhof and Murdock, 1953; Clayton and M i l i t i t s k y , 1983).

2.2 .3 Peak Pressure Equ i l ib ra t ion Stage

During the excavation and. concreting stages the in -s i tu la tera l

e f fec t iv e stresses change because of stress r e l i e f and swell ing.

There have been two d i f fe rent opinions about whether the horizontal

ef fec t iv e stresses are re-established after in s ta l la t io n . Anderson

et a l , (1984) carried out some small scale laboratory p i le

in s ta l la t io n te s ts . They found as a resu l t of the tests that

i n i t i a l horizontal e f fec t iv e stresses are subsequently restored

within a r e l a t i v e l y short period. Accuracy of simulation of the

p i le in s ta l la t io n by small scale laboratory experiments is

questionable. Clayton and M i l i t i t s k y (1983) re-analysed some of the

resu lts presented by Whitaker and Cooke (1966), Taylor (1966) and

Combarieu (1975) and demonstrated that the shaft f r i c t i o n components

can increase over a period of many months. According to the re s u l t s

of the f u l l scale p i le test carried out by M i l i t i t s k y (1984) the

undisturbed horizontal e f fec t iv e stresses were not f u l l y recovered.

These in s ta l la t io n ef fects wil l influence the f in a l e f fect ive stress

on the shaft of a p i le and th is needs to be taken into account in

any e f fec t ive stress design method.

16

3 Finite Element Method in Geomechanics

Owing to its power and v e r s a t i l i t y , the f i n i t e element method has

become very popular in the last ten to f i f te en years for the

solution of various engineering problems. Like every powerful tool

i t has to be handled with care and resu lts have to be assessed using

engineering experience and judgement.

The f i n i t e element method has the potential for c lose ly predicting

real behaviour. This is the case when geometry and loading are well

defined and material properties are known accurately. Geotechnical

engineers, however, are not as fortunate as the others in being able

to obtain these v i ta l input data for the analysis. Uncertainties

lead them to make wide ranging idea l isat ions and assumptions.

Therefore the resu l ts should be regarded as real. "'-as the

idea l isat ions and assumptions which are used in the analysis.

Despite the d i f f i c u l t i e s in making appropriate ideal isat ions and

assumptions, the f i n i t e element method is s t i l l one of the advanced

tools avai lable for the geotechnical engineers. In part icular the

value of the analysis in ass is t ing the engineer to place bounds on

l i k e l y overal l behaviour and in assessing the influence of various

assumptions is not to be overlooked (Burland, 1978). It seems that

the f i n i t e element method w i l l continue being used and provide

geotechnical engineers with useful information about soi l behaviour

which is impossible to obtain otherwise at least unt i l better

analytical methods are developed.

3.1 F in i t e Element Formulation For Coupled Consolidation Analysis

In th is section, a coupled (Biot) consolidation formulation is

derived in two dimensions. What follows summarises a part of the

book ‘ C r i t i c a l State Soi l Mechanics via F in i te Elements' by B r i t to ,

and Gunn, 1986. Readers should re fer to th is book for a complete

treatment.

17

The basics of the f i n i t e element method and continuum mechanics are

assumed to be known by the reader. The notation used follows that

established by Zienkiewiez in his series of texts on the f i n i t e

element method (see for example Zienkiewiez, 1977).

A coupled consolidation formulation can be obtained from the two

main p r in c ip le s , equil ibrium and cont inuity , of continuum mechanics

and the e f fec t ive stress concept in so i l mechanics, using the

p r inc ip le of v ir tual work.

The equil ibrium pr inc ip le of continuum mechanics is expressed in two

dimensions by the fol lowing d i f f e r e n t ia l equations:

3ax + 9tyx =3x

3t

ay

xy + 3ay = u>3x sy

(3.1)

(3.2)

where ox , 0y are the normal stresses, xxy, xyx are the shear

stresses and o>x , wy are the body forces (Fig 3.1).

Equations (3.1) and (3.2) are mult ip l ied by arb it rary scalar

functions h and v, added together, and integrated over the area5of

the continuum.

(3.3)f hf3ox + 3 T l/yyx _ + V 9xxy + 90y - to dA = 0

L 3x ay J L 3X ay J

Terms h and v represent v i r tual horizontal and vert ica l displacement

funct ions. They are arb it rary functions of x and y. When h and v

are id ent i f ied as the v i r tua l horizontal displacement dx* and

v ir tual vert ica l displacement dy*, then Sih, _3v, _̂h + 8_v become

3x 3y 3y 3x

-ex*, -£y*s -Yxy* known as the components of the v ir tualstra in matrix in two dimensions.

If Equation (3.3) is integrated by Green's Theorem in two dimensions

and necessary subst itut ions as

18

X

o ;

/yx

<E

, Z , #

(E

“ a

O' ,

Fig 3.1 State of stresses in two dimensions

Fig 3.2 Posit ive a) normal, b) shear stra ins in two dimensions

19

d* =’ dx*

*ex a x

= - £ * = ey * > _a = ay

. df . y xy*_ Txy

' wx ' * nx a x

w = > I =

r «e ny 0y .

are made, the equation for displacements in a two dimensional

continuum media is obtained:

Je*T a dA = Jd*T T dS + Jd*T w dA

A S A . . . (3.4)

The Equation (3.4) can be expressed in an incremental form:

Je*T Aa dA = Jd*T AT dS + j d * T Aw dA

A S A . . . . (3.5)

A l l the equations so far are derived in terms of total stresses. To-

write the same equations in terms of e f fec t iv e stresses, the

following well known re la t ionsh ip between total stresses, e f fect ive

stresses and pore pressures is used. The equation is written in

incremental form:

or

Aa = Aa' + m Au

Aa = D'Ae + m Au

. . . . (3.6)

. . . . (3.7)

Aa = incremental total stress matrixAa' = incremental e f fec t iv e stress matrixTm - [1 1 0]

Au = incremental pore pressuresD' = D matrix which contains e f fect ive soi l properties

(E1, v 1)Ae = incremental stra in matrix

Au = incremental excess pore pressures (Au = Au)

20

The f i n i t e element d isc re t isa t io n of the problem is now introduced.

The displacements are assumed to vary over a f i n i t e element

according to:

d = N a . . . . (3.8)

and the excess pore pressures are assumed to vary over the same

element according to:

u = N b . . . . (3.9)

where a. are the displacements at the nodal points , b_ are the excess

pore pressures at the nodal points and N, N are the shape

functions.

As usual the stra ins are given by:

e = B a . . . . (3.10)

or in incremental form:

Ae = B A a . . . . (3.11)

Using Equation (3.7) and making the usual f i n i t e element

subst itut io ns,

_*T (BT D'B) dA.Aa + a*T (BTm N) dA.Ab

A

= a*T NT AT dS

Tis obtained. a* can be cancelled

equation is:

. . . . (3.12)

The f in a l form of the

K Aa + L Ab = NT AT dS

.... (3.13)

21

where

K = (B D'B) dA and L = (B m N) dA

The two dimensional d i f f e r e n t ia l equation of cont inuity is:

32u + ky 32u

Tu 9x Tw 3y_ d V

fit (3.14)

where u is excess pore pressure, kx and ky are permeabil ity of

the media in the x and y d i rect io n , is the unit weight of water

and V is the volumetric s t ra in .

To obtain the f i n i t e element matrix equations, another 'v i r tua l

p r i n c i p l e ' , which is similar to the v ir tual work principle. , is

applied to Equation (3.14). The f i r s t step is to multip ly the

continuity equation by an arb it rary scalar which can v„ary with x ;and

y. This scalar is id en t i f ied as an imaginary or v ir tual pore

pressure. Thus Equation (3.14) becomes:

jA

kx ^ + ky + dVYw 9x Yo) 3y eft

dA = 0(3.15)

Green's Theorem is now applied to Equation (3.15) and

-Jk x 3U7" 3u + k y 3u* 3u

Yw 9x Y ^^y 9y

f * d v{ dt

dA = 0

dA - u vn dS

. (3.16)

is obtained, where vn is the a r t i f i c i a l seepage ve loc i ty normal to

the boundary

w _ kx 3u n , kv 3u „-v n = nx + y . nyYo) 3x Yo) 3y

The f i n i t e element d is c r e t i s a t io n of the problem is then introduced

and necessary subst itut ions as

22

Iff9y

_3_U

8X

= E b # V = m e , k =kx 0

0 k,

are made.

*T N4 m B dA d(a)

dt

- b*T ET kE/Y© dA b

A

= b*T NT vn dS

. . . . (3.17)

The v ir tual pore pressure can be cancelled from th is equation and

the equation:

J d(a) - ® b. = [nT vn dSdt . . . . (3.18)

is obtained. Where

L = J b4 m N dA , $ =

A

ET kE/tu dA

Equation (3.18) is a f i r s t order d i f fe r e n t ia l equation which can be

integrated with respect to time., from time t to time t + At:

-t+At $ Jb dt

t

r t + i t LT d(a) dt -

dt

't+At I vn dS dt

t S . . . . (3.19)

In performing th is integration the fol lowing approximation can be

made:

j"t+At'b dt = {(1 - Q)bL + Offi?} At

23

where bj = b(t) and b2 = b(t + At)

A similar approximation is made for the integration of vn and the

equation becomes:

LT [a]t+At _ _ 0^bi + eb2}Att

mT {(1 - e)vni + 0vn2} At dS

. . . . (3.20)

Adopting a value of 0 = 1, making the subst itution in (3.20) and

defining Aa_ = j*(t + At) - _a(t), Alb = lb2 - bj., the f in a l form of the

cont inuity equation is then obtained:

LT Aa - $ At.Ab = $ A t .bx + r,T Vp2 dS. . . . (3.21)

F i n a l l y , Equations (3.13) and (3.21) can be used to establish a

solution at time t + At from the solution at time t . Thus the

solution can be moved a step forward in time from t » 0. The

Equations (3.13) and (3.21) together can be written:

. . . . (3.22)

K L Aa

I"""1"

i -

LT -Mt Ab_ 1k

l --1

e Ar i = nt A T dS

4—2 =i!JAr2 = J N vn2 dS + $ At.b^

S

The right-hand-side term Ar ̂ consists of the normal f i n i t e element

incremental load terms. The r ight-hand-side terms Ar_2 consist of a

load term corresponding to a prescribed seepage on the boundary and

an additional term ($ At.bty which is calculated as the solution

proceeds.

24

3.2 CRISP Programs

3.2.1 CRISP Programs in General

The computer programs known as CRISP were written and developed by

research workers in the Cambridge Univers ity Engineering Department

Soil Mechanics Group, start ing in 1975. The main authors of the

programs have been M. Zytynski (1975 - 1977), M.J. Gunn (since 1977)

and A. Br it to (since 1980). The programs were o r ig in a l l y cal led

'MZSOL' and in 1976 they were renamed 'CRISTINA'. The 1980 versions

of the programs were cal led 'CRISTINA 1980' and early in 1981 they

were renamed 'CRISP* (CRItical State Programs).

CRISP can carry out undrained, drained or coupled (Biot)

consolidation analysis of three dimensional or two dimensional plane

stra in or axisymmetric (with axisymmetric loading) so l id bodies.

The fol lowing soi l models are avai lable in CRISP:

Anisotropic e l a s t i c i t y , in-homogeneous e l a s t i c i t y (properties

varying with depth), c r i t i c a l state models (Cam Clay, Modified Cam

Clay) and e la s t i c - p e r f e c t ly p l a s t i c models (with Tresca, von Mises,

Mohr-Coulomb or Drucker Prager y ie ld / f a i lu r e surfaces).

The fol lowing element types are avai lable in CRISP:

Linear stra in t r ia n g le , cubic stra in t r ia n g le , l inear stra in

quadri lateral and the 20-noded brick element (with extra pore

pressure degrees of freedom for a consolidation analys is) .

CRISP uses an incremental (tangent s t i f fn ess) technique for

non-linear analysis. An option for updating nodal coordinates with

progress of analysis is ava i lab le .

CRISP has the fol lowing options for handling boundary condit ions:

Element sides can be given prescribed incremental values of

displacements or excess pore pressures. Loading may be applied as

nodal loads or pressure loading on element sides. Automatic

ca lculat ion of load simulating excavation or construction when

elements are removed or added.

25

3.2.2 CRISP Programs, Modifications and Additional Programs

In order to perform an analysis using the CRISP package, i t is

necessary to submit at least two computer jobs which run completely

separate computer programs: The 'Geometry Program' and the 'Main

Program'. The Geometry Program sorts out the geometry data, i . e .

the coordinates of nodal points associated with each f i n i t e element

and element co n nect iv i t ies , and produces a l ink f i l e which contains

the information about the geometry of the problem, a p lott ing data

f i l e which can be read by a separate plott ing program to plot the

f i n i t e element mesh and an output f i l e which is then sent to the

l ine printer to obtain a hard copy of the geometry data of the

problem. The Main Program reads the geometry data from the l ink

f i l e , the control data, material propert ies , insitu stresses, loads,

boundary condit ions etc. from the input f i l e and then carr ies out

the analysis and produces an output f i l e which contains the resu lts

and which is sent to the l ine printer to obtain a printout after the

completion of the analysis and plott ing data f i l e .

In th is analys is, version MP1 of the Main Program and version GP1 of

the Geometry Program have been used. Although these programs have

been run on the Cambridge Univers i ty IBM 3081 Computer successfu l ly ,

i t was necessary to make some modifications to run them on the

Univers i ty of Surrey PRIME Computer. F i r s t l y , a subroutine which

creates the necessary f i l e s for input, output, p lott ing etc. has

been added to the programs. ,

In versions MP1 and GP1, a l l the integer and real variables are

stored in a large one dimensional array. Since the Univers ity of

Salford Fortran compiler does not accept integer and real values in

the same array, there were two options to run the programs. , One was

to run the programs with the -NOCHECK option, the other was to

accommodate the integer and real values in d i f fe ren t arrays

throughout the programs. I f the f i r s t option was to be taken, when

the program crashed i t would be impossible to f ind out whether a

data error or an error in the programs caused the crash. After a

number of t r i a l runs, i t was decided to take the second option and

the necessary changes were made to overcome th is d i f f i c u l t y .

26

As explained in the next chapter, the f i n i t e elements are removed

from the hole/trench in an incremental manner to simulate the

excavation process. When bentonite s lu rry is used for supporting

the hole/trench during the excavation, the hydrostatic pressure of

the f lu id is applied to the base as well as the sides of the

hole/trench. In version MP1, when a pressure is applied to a side

of an element, i t stays there unti l the end of the analysis unless

the same pressure acting in the opposite di rect ion is applied to the

same side of the element. This operation, however, does not remove

the pressure load from the load array, i t remains there as a zero

pressure acting on the element side. The way that the program

removes the pressure loadings from the element sides causes a

program error when i t t r ie s to apply a zero pressure load on a side

of an element which has been removed from the mesh in previous

increment block. The necessary modif ications have been made to run

the program without a program error caused by the resu lt of the

above operation.

The main program has been modified to avoid unnecessary printed

output. With the modified version of the Main Program, i t is

possible to printout the resu lts only at the required increments.

Versions GP1 and MP1 do not have a pre-processor or a

post-processor. It was necessary to write two programs for th is

purpose. A small program has been written to create the f i n i t e

element mesh i . e . nodal coordinates, element connect iv i t ies . The

output from this program then has been added to the Geometry Program

input f i l e . The Main Program has been modified so that i t creates a

binary f i l e which contains the resu lts of the selected increments.

A plotting program has been written to plot the p las t ic zone, the

contours of the hor izonta l , vert ica l e f fec t iv e stress and excess

pore pressure f i e l d s around the hole/trench using the necessary

subroutines in the GINO package. The program also calculates the 3

values (see Appendix 1) at the contact area the pile/diaphragm wall

and s o i l . Since the horizontal e f fect iv e stresses are known at the

nine integration points in each element, an interpolat ion and then

an extrapolation process was needed to

27

obtain the horizontal e f fect ive stress values at the side of an

element. F i r s t , the values of the horizontal e f fec t iv e stress at

the four points in an element were obtained from the nine

integration points using Lagrangian Functions, then from these four

values, the horizontal e f fec t ive stresses at the side of the element

were extrapolated. The reason for using four points is that

stresses are more accurate at these points (Barlow, 1976,

Zienkiewicz, 1977).

The analysis has been carried out using the modifated versions of

MP1 and GP1 of Crisp Main Program, Crisp Geometry Program and

additional programs mentioned above. The in te r - re la t ionsh ip of the

programs is shown in Fig. 3.3.

28

INPUTGeometryData

INPUT Control Data

Material Prop.

Insitu S trs Loads

etc.

C R IS PGEO M ETRYPROGRAM

LinkFile

Printed Data

Out-put File

CR ISPMAINPROGRAM

CurrentResults

stop/start5* facility

PreviousResults

Printed

Out-put

INPUT PLOTTING PrintedControl Data PROGRAM Out-put

Contour GraphPlotting Plotting

Fig 3.3 System descript ion

29

4 Analysis

Although the influence of the construction process on the behaviour

of diaphragm walls and c a st - in s i tu bored pi les has been recognised

by researchers and pract is ing engineers, because of the cost of

computer time and, perhaps, the fa l l a c y of regarding these effects

in s ig n i f ic a n t , no d i rect attempt has been made to include the

construction process in the f i n i t e element analysis in any example

of its kind. Using reduced values of soi l parameters ( c 1, 0 ' , E'

etc .) seems to be the most straightforward way of modelling the

effects of the construction process in the analysis. Th is , however,

affects the behaviour of soi l not only near the wall/p i le but also

in the whole mesh.

In this study, the f i n i t e element method has been used to ca lcu la te

changes in hor izonta l , vert ica l e f fec t ive stresses, horizontal total

stresses and excess pore pressures during the construction process.

Some typical so i l properties of London Clay have been used in the

analys is, resu lts therefore are s t r i c t l y va l id for th is soi l type

only, but may be expected to be representative of s t i f f -

overconsolidated clays in general.

A number of analyses have been carried out to look at the ef fects of

d i f fe rent construction techniques and the delays of the construction

stages (Table 4 .1) .

Type Support Delay in Concreting

Setting Time of Concrete

Plane Strain

Axisymmetr ic

No Support

Bentonite Slurry

No Delay

1 Day

7 Days

Immediate

12 Hours

Table 4.1 Different combinations of the analyses

30

4.1 Material Properties

Since the aim of th is study is to examine the ef fects of the

construction process rather than the effects of the soi l parameters

on the behaviour of diaphragm walls and cast- ins i tu bored p i le s ,

only one set of soi l parameters, which is typica l for the London

Clay, has been used throughout the analyses.

The Main Program requires the fol lowing soi l parameters to carry out

the analysis:

1. Unit weight of so i l (If)

2. Ef fect ive e l a s t i c i t y modulus (E‘ )

3. E f fect ive Poisson's r a t io ( V )

4. E f fect ive angle of f r i c t i o n (0')

5. E f fect ive cohesion ( c ‘ )

6. Permeability c o e f f ic ie n ts (kx , ky)

7. Unit weight of water

The values of the soi l parameters used in the analyses have been

obtained mainly from Potts and Burland (1983,a). In th is report

they present the resu lts of the f i n i t e element analysis of the Bell

Common cut and cover tunnel diaphragm walls' front excavation. The

values of soi l parameters used were obtained from extensive s ite

invest igat ions, laboratory experiments, back analyses and past

exper ience.

Although it is common to use e l a s t i c i t y modulus varying with depth

for London Clay, in th is analysis an average value of the e l a s t i c i t y

modulus has been employed because the option of increasing modulus

with depth is not avai lable in the CRISP programs to carry out the

analyses using the E la s t i c (varying with depth) Perfect ly P last ic

so i l model. This approximation makes the soi l s t i f f e r near the

ground leve l , less s t i f f towards, the bottom of the mesh. Results

should be interpreted bearing th is fact in mind.

Densit ies of fresh concrete and bentonite s lurry which have been

used to calculate the excess pore pressure f i x i t i e s and applied

pressures to the boundaries have been obtained from the l i te ra tu re

(Hutchinson et a l , 1974).

31

The value of 2 adopted for the earth pressure c o e f f ic ie n t at rest

(Kq)> is in the range experimentally determined for the London

Clay (Skempton, 1961; Bishop et a l , 1965). K0 condit ions were

assumed to calculate the insitu horizontal e f fect ive stresses.

Material properties which have been d i re c t ly or in d i r e c t ly used in

the analyses are presented in table 4.2. Al l strength parameters

are in terms of e f fec t ive stress.

Material Properties

Value Unit

Water 10.0 kN/m3

Unit weight London Clay 20.0 kN/m3

(V) Bentonite 12.0 kN/m3

Concrete 23.0 kN/m3

E l a s t i c i t y Modulus (El ) 120000.0 kN/m2

Poisson1s Ratio (v‘ ) 0.2 1

Angle of F r ic t ion (0') 25.0 Degree

Cohesion (c l ) 0.0

CMz

Permeabi1i ty (kx , kj ) 1.0x10s m/sec

Table 4.2 Material Properties

4.2 Soi l Model

A number of const i tut ive laws which describe the re la t ions of

physical quantit ies i . e . stress , s tra in , time have been developed in

so i l mechanics. But none of them seem to represent the behaviour of

every kind of so i l under any loading and boundary condit ions. In a

part icu lar problem, some of them may give t o t a l l y nonsensical

r e s u l t s . Therefore, when an analysis of a so i l is to be carried out using either conventional or numerical methods, a model which can

represent the behaviour of the part icu lar so i l type under part icular

condit ions as c lose ly as possible, has to be chosen. This may mean

32

employing very complicated const i tut ive laws in some cases with a

consequent-rise in computer costs.

In general a l l models have either a theoretical (such as e l a s t i c i t y ,

p l a s t i c i t y ) or an experimental ( i . e . curve f i t t i n g ) basis . Linear

e l a s t i c , variable e l a s t i c , r i g i d - p e r f e c t l y p l a s t i c ,

e la s t i c - p e r f e c t ly p l a s t i c , e la s t i c - s t r a i n hardening/softening

p l a s t i c , c r i t i c a l state and e la s to - v isc o - p la s t ic are al l well known

models avai lable in so i l mechanics. Description and comparison of

these models can e a s i ly be found in the l i te ra tu re (e.g. Schofield

and Wroth, 1968; Davis, 1968; H i l l , 1971; Zienkiewicz, 1977;

Chr ist ian and Desai, 1977; Naylor, 1978; Owen and Hinton, 1980;

Naylor and Pande, 1981; Smith, 1982; Chen, 1984). The scope of th is

study excludes a comprehensive review of a l l the so i l models. Only

the model used in the analysis wil l be examined in d e t a i l . ‘

The selection of a so i l model’ fo r ; the analysis was l imited to the

options avai lable in the CRISP package. Since so i l does y ie ld ,

e la s t i c models were not considered. Moreover, e la s t i c models

generate unreal i s t i c a l l y very small pore pressure changes for the

p i le problem (Gunn,. 1984). C r i t i c a l State models were disregarded

because of the fact that they are not appropriate for

over-consolidated clays e .g . London Clay (with y ie ld ing occurring on

the dry s ide) . It was decided to employ a l inear e la s t i c -p e r f e c t ly

p la s t i c so i l model in the ana lysis. The Mohr-Coulomb fa i lu r e

cr i te r io n with an associated flow ru le has been adopted (Fig 4.1).

Despite having no theoret ical basis, experimental resu l ts strongly

support the Mohr-Coulomb f a i lu r e c r i t e r io n . It is the only simple

c r i te r io n of reasonable genera l i ty (Bishop, 1966). It can be argued

that associated flow rule i . e . a p las t ic potential para l le l to the

f a i l u r e surface, gives excessive and continuing volume change, thus

larger displacements then occur in r e a l i t y . A l te rn a t iv e ly , a

non-associated flow ru le which is considered as more capable of

predict ing the behaviour of so i l under a l l condit ions may be

employed (Zienkiewicz et a l , 1975a and 1975b). In the analysis ,

however, an associated flow ru le has been adopted because a solver

rout ine for asymmetric matrices was not avai lable in the package and

therefore an analysis could not have been carried out using the

non-associated flow ru le .

hydrostatic

axis

or

O'

Fig 4.1 Mohr-Coulomb failure criterion

34

The important feature of th is analysis is the generation of pore

pressures using the coupled consolidation (Biot) theory (see Section

3.1) . The pore pressure changes during the d i f fe rent stages of

bored p i le and diaphragm wall in s ta l la t io n can then be predicted.

4.3 Types of Analyses

Two d i f ferent types of analyses have been carried out to investigate

the effects of construction on the behaviour of diaphragm walls and

ca st - in s i tu bored p i les : 1. Plane s tra in , 2. Axisymmetric (with

axisymmetric loading).

4.3.1 Plane Stra in Analyses

Plane stra in analyses have been carried out to simulate diaphragm

wall construct ion. If one dimension of a so l id body normal to a

certain plane (e.g. xy) is large compared with the dimensions in

th is plane and the sol id body is subjected to loads in the same

plane only, then i t may be assumed that displacements in the

direct ion normal to that plane are neg l ig ib le and the in-plane

displacements are a function of the dimensions of that plane ( i .e .

ez , yXz and Yyz e9ual to zero), (Timoshenko and Goodier, 1951; Fig 4.2) .

Since diaphragm walls are constructed in panels, and not of an

in f i n i t e length and consequently arching may develop in the soi l

around the trench during the excavation, plane stra in analysis wil l

give larger deformations then could occur in r e a l i t y . If the

complexity of a three dimensional f i n i t e element analysis and

subsequent high computing costs are to be considered, plane stra in

analysis seems to be an a ttract ive way of tackl ing the problem.

4.3 .2 Axisymmetric Analyses

Axisymmetric analyses have been carried out to simulate the

construction of a bored p i l e . If a three dimensional sol id is

symmetrical about i t s centre l ine axis and is subjected to loads and

35

Fig 4.2 Plane stra in

Fig 4.3 Axisymmetry

36

boundary condit ions which are symmetrical about th is axis, then its

behaviour is independent of the circumferential coordinate 0, and

Tr8> Tz0 are eQual to zero (Timoshenko and Goodier, 1951; Fig

4 .3) . Therefore stresses and strains have the same value in every

plane in the radia l d i rect ion .

An axisymmetric analysis is appropriate for the analysis of a

c i rcu la r hole in the ground.

4.4 Geometry Data

4.4 .1 Size of the Problem

In the analysis, a uniform soi l deposit 40m in depth and 80m wide

(or in diameter in the case of the bored p i le construction) has been

considered (Figs 4.4 and 4 .5) . Owing to the symmetric nature of the

problem only ha lf of the so i l deposit mentioned above has been

modelled. The so i l mass has been restrained in the horizontal

direct ion by the sides and in the vert ica l di rect ion by the base

(Fig 4.6).

4 .4 .2 F in i te Element Mesh and Element Type

It is des irable to use as many elements as possible to obtain

accurate resu lts in a f i n i t e element analysis. On the other hand,

the use of too many elements should be avoided because of the

consequent escalation of the computer cost. Moreover, data

preparation becomes tedious when a large number of f i n i t e elements

are used.

Another way of improving the accuracy of resu lts is to use higher

order elements. It can be demonstrated that a dramatic improvement

in accuracy is obtained with the same degrees of freedom when high

order elements are used (Zienkiewicz, 1977). The use of high order

elements leads to p a r t i c u l a r l y e f f i c i e n t e la s to -p la s t ic solution

programs (Owen and Hinton, 1980). In these analyses a f i n i t e

element mesh which consists of 208 Linear Quadr i latera l Elements

37

Fig 4.4 Uniform soi l deposit considered in the diaphragm wall analysis

Fig 4.5 Uniform so i l deposit considered in the ca s t - in s i tu bored p i le analysis

38

Fig 4.6 F in i te element mesh used in the analyses

39

!

with l in e a r ly varying excess pore pressures, has been adopted (Fig

4.6). The mesh has been made f iner in regions where ra p id ly varying

stra ins and stresses and expected i . e . in the v i c i n i t y of the

hole/trench.

The element type used in the analysis has 8 nodes where

displacements are unknown. At 4 corner nodes pore pressures are

unknown as well (20 degrees-of-freedom). Every element has 9

integration points where stresses/strains are calculated (Figs 4.7

and 4.8) .

4.5 In-si tu Stresses and Pore Pressures

In an e la s to -p las t ic analysis the s t i f fn ess matrix of a f i n i t e

element wil l be dependent on the stress state within the element.

For th is reason, the program needs to know the ins itu stresses and

the pore pressures before the analysis s tarts .

In th is analysis, ins itu horizontal and vert ica l e f fec t iv e stresses

and pore pressures have been assumed to vary only with depth, not in

the horizontal d i rect ion.

In-si tu horizontal and ver t ica l e f fec t ive stresses and pore

pressures have been calculated using the fol lowing re la t ionsh ips :

u = Twh V s (Ys " Yw)h aft = KqV

Where u = pore pressure (kN/m2)

Yw = unit weight of water (10 kN/m3)

h = depth (m)

av ' = vert ica l e f fect iv e stress (kN/m2)

Ys = bulk unit weight of soi l (kN/m3)

aft = horizontal e f fec t iv e stress (kN/m2)

K0 = earth pressure c o e f f ic ie n t at re s t , taken as 2

40

4

O

( g -

7

o

d>

■ 02

Q displacements unknown

© pore pressures unknown

Fig 4.7 Linear stra in quadri latera l element with l in e a r ly varying excess pore pressures (8 nodes, 20 degrees-of-freedom)

Fig 4.8 Integration points in a l inear s tra in quadri la tera l element

41

It has been assumed-that the water table is at the ground level and

that the soi l is f u l l y saturated. A K0 value of 2 has been

adopted.

In-s i tu horizontal and ver t ica l stresses and pore pressures used in

the analysis are i l lu s t r a te d in Fig 4.9.

4.6 Simulation of the Construction Process

As was explained in Section 4.3, diaphragm wall and bored p i le

constructions have been simulated by carrying out plane stra in and

axisymmetric analyses resp ect ive ly . Having defined the analysis

type, a l l construction stages for both diaphragm wall and bored

p i le , have been assumed to be the same.

The f i n i t e element analysis has been divided into three stages to

simulate as c lose ly as possib le the construction process. The

construction stages are:

1. Excavation

2. Concreting

3. Consolidationr

4.6 .1 Excavation

In the analysis, excavation has been simulated by removing elements

from the mesh. While the excavation progresses, excess pore

pressures have been f ixed at the base and the sides of the

hole/trench, and the pressures have been applied there consistent

with the presence of bentonite (or no support). When a non l inear

or consolidation analysis is performed using CRISP i t is necessary

to divide either the loading or the time span of the analysis (or

both i f there is consolidat ion with non-linear material properties)

into a number of increments. Therefore the analyses have been

carried out in an incremental manner.

Fig 4.9 Insitu stresses and pore pressures

Excavation procedure consists of the fol lowing steps:

1. Excavation to a depth of 5m in 3 hours (removing 2 elements)

Fig 4.10a

2. Excavation from the depth of 5m to a depth of 10m in a further

3 hours (removing 2 elements) Fig 4.10b

3. Excavation from the depth of 10m to a depth of 15m in a further

3 hours (removing 2 elements) Fig 4.10c

4. Excavation from the depth of 15m to a depth of 20m in a further

3 hours (removing 2 elements) Fig 4.10d

5. Delay in concreting (no delay, 1 day, 7 days)

Because of the way the CRISP program works, every step has been

accommodated in an increment block which has 10 increments in th is

case. When performing an excavation analysis the program calculates

the implied loading due to the removal of elements. These implied

loadings wil l often be too large to be applied in a single increment

when the material behaviour is non l in ear . The use of an increment

block spreads these implied loads over several increments. The

s t i f fn ess of an element, however, is removed e n t i re ly in the f i r s t

increment of a block introducing an extra approximation to the

modelling of excavations.

Two types of excavation method have been considered in the

analyses. Excavations have been carr ied out either with no support

or with support from bentonite s lu r ry . Other means of support are

beyond the scope of th is study.

Excess Pore Pressure F i x i t i e s

Depending on the excavation method, the sides of the hole/trench

have been f ixed either as an impermeable boundary or as a boundary

with a known pressure head. Owing to the nature of the program, i f

the boundaries are impermeable there is no need to specify any pore

pressure boundary condit ions since a l l boundaries are automatical ly

assumed to be impermeable.

If bentonite s lu rry is not used during the excavation, then the

impermeable f i l t e r cake is not formed at the sides and base of the

44

Fig 4.

a

1

1

1

* ---------------------11

/ rrr

f

\ ./

/

z ' ............... u

INI l f '

Ijn! i| A - '■TIDr10-

ID-

ID-

'“v..•I'l I I

I,-,

f c -

lt>

10 Excavation stages

45

trench/hole, and these are assumed to be drained boundaries.

Therefore pore pressures have the value of zero there. Since the

program requires excess pore pressures as input, negative excess

pore, pressures have been applied in order to make the pore pressures

zero at the sides of the hole/trench. Excess pore pressures at the

base and the sides of the hole/trench at d i f fe rent stages of the

excavation process are i l lu s t r a t e d in Fig 4.11.

When an excavation is carried out using bentonite s lu r ry as support,

the sides of the hole/trench can be assumed to be impermeable

because of the bentonite s l u r r y ' s a b i l i t y of forming an impermeable

f i l t e r cake at the sides of the hole/trench.

In the analysis the sides of the elements at walls of the

hole/trench have been made impermeable during the excavation process

to simulate the e f fects of f i l t e r cake.

Pressures Applied to the Boundaries

In the case of excavation with support, i . e . bentonite s lurry is

present, the hydrostatic pressure of bentonite s lu r ry has been

applied to the sides of the hole/trench as excavation progresses.

Obviously there is no need to apply pressures when excavation is

carried out without any means of support. • .**•■

The hydrostatic pressure of the bentonite f lu i d has been applied to

the base as well as the sides of the hole/trench. In the program

when a pressure is applied to a side of an element, i t stays there

unt i l the end of the analysis unless i t is removed. Since the

excavation has been carr ied out in an incremental manner, the

pressures applied to the base of the progressing hole/trench had to

be removed before the elements at the bottom had been removed. This

d i f f i c u l t y has been overcome by having extra increment blocks just

after the incremental excavations. This operation gives an extra

approximation to the analysis (Fig 4.12).

Applied pressures at the base and the sides of the hole/trench in

d i f fe rent stages of the excavation process are i l lu s t ra te d in Fig

4.13.

46

31

33

15 32

11 31

13 30

12 29

11 28

10 27

9 2G

Q

NodeNumber

PorePressure kN/m2

Excess Pore Pressure kN/m2

34 0.0 0.0

33 0.0 -25.0

32 0.0 -50.0

15 0.0 -50.0

31

33

32

3)

13 30

12 29

11 28

10 27

9 28

NodeNumber

PorePressure kN/m2

Excess Pore Pressure kN/m2

34 0.0 0.0

33 0.0 -25.0

32 0.0 -50.0

31 0.0 -75.0

30 0.0 -100.0

13 0.0 -100.0

Fig 4.11 Excess pore pressure f i x i t i e s at d i f fe ren t stages of the excavation with no support

47

31

33

32

31

30*

23

28

10 27

26

NodeNumber

PorePressure kN/m2

Excess Pore Pressure kN/m2

34 0.0 0.0

33 0.0 -25.0

32 0.0 -50.0

31 0.0 -75.0

30 0.0 -100.0

29 0.0 -125.0

28 0.0 -150.0

11 0.0 -150.0 ,

31

33

32

31

33

29

28

27

9 26

d

Fig 4.11 Continued

NodeNumber

PorePressure kN/m2

Excess Pore Pressure kN/m2

34 • 0.0 0.0

33 0.0 -25.0

32 0.0 -50.0

31 0.0 -75.0

30 • 0.0 -100.0

29 0.0 -125.0

28 0.0 -150.0

• 27 0.0 -175.0

26 0.0 -200.0

9 0.0 -200.0

48

1st Inc. B lo ck

C. a t th e s ta r t o f 3rd Inc. B lo ck

Fig 4.12 Removal of the applied pressure from the base of the hole/ trench

49

NodeNumber

Incremental Pressure kN/m2

Cumulative Pressure kN/m2

34 0.0 0.0

33 30.0 30.0

32 60.0 60.0

15 60.0 60.0

b

NodeNumber

Incremental Pressure kN/m2

Cumulative Pressure kN/m2

34 ■ 0.0 0.0

33 0.0 30.0

32 0.0 60.0

31 30.0 90.0

30 60.0 120.0

13 60.0 120.0

Fig 4.13 Applied pressures at the different stages of excavationwith bentonite support

50

NodeNumber

Incremental Pressure kN/m2

Cumulative Pressure kN/m2

34 0.0 0.0

33 0.0 30.0

32 0.0 60.0

31 0.0 90.0

30 0.0 120.0

29 30.0 150.0

•28 60.0 180.0

11 60.0 180.0

Fig 4.13 Continued

.Node 1 Number

Incremental Pressure kN/m2

Cumulative Pressure kN/m2

34 0.0 0.0

33 0.0 30.0

32 0.0 60.0

31 0.0 90.0

30 0.0 120.0

29 0.0 150.0

28 0.0 180.0

27 30.0 210.0

26 60.0 240.0

9 60.0 240.0

51

4.6.2 Concreting

In the analysis the concreting process has been simulated by excess

pore pressure fixities and applying pressures at the base and the

sides of the hole/trench. The elements removed in the excavation

process could not be reintroduced to the mesh. This meant that the

stiffness of fresh concrete could not be modelled. Applying these

boundary conditions has been done in an incremental manner. The

concreting procedure has been accommodated in five incremental

blocks. Each incremental block consisted of 10 increments.

The concreting procedure consists of the following steps:

1. Concreting from the bottom to a depth of 15m in 1 hour

2. Concreting from the depth of 15m to a depth of 10m in 1 hour

3. Concreting from the depth of 10m to a depth of 5m in 1 hour

4. Concreting from the depth of 5m to ground level in 1 hour

5. Setting time (no time, 12 hours)

As will be discussed in Section 5.1, due to the extensive plastic

zone developed around the unsupported trench excavation, no further

attempts were made to carry out the analysis for the diaphragm wall

construction without any support used..

Excess Pore Pressure F ix itiesUnlike the excavation process, the base and the sides of the

hole/trench have been considered as drained boundaries with known

excess pore pressures regardless of whether concreting has been

carried out using bentonite slurry as support or not. The sides

which were at contact with bentonite slurry, however, have been kept

impermeable until bentonite slurry has been replaced by fresh

concrete. The filter cake has been assumed to be removed completely

by the fresh concrete during the concreting.

When bentonite slurry is not present, i.e. no support is used,

during concreting the difference between the hydrostatic pressures

of fresh concrete and the initial insitu pore pressures at each

nodal point level have been applied at that level as an excess pore

pressure fixity (Fig 4.14).

52

NodeNumber

InsituPorePressurekN/m2

TotalPorePressurekN/m2

ExcessPorePressurekN/m2

28 150.0 0.0 -150.0

27 175.0 57.5 -117.5

26 200.0 115.0 -85.0

9 200.0 115.0 -85.0

NodeNumber

InsituPorePressurekN/m2

TotalPorePressurekN/m2

ExcessPorePressurekN/m2

. 30 100.0 0.0 -100,0

29 125.0 57.5 -67.5

28 150.0 115.0 -35.0

27 175.0 172.5 -2.5

26 200.0 230.0 30.0

9 200.0 230.0 30.0

b

Fig 4.14 Excess pore pressure f ix i t ie s at the different stages ofconcreting (no support used)

53

NodeNumber

InsituPorePressurekN/m

TotalPorePressurekN/m2

ExcessPorePressurekN/m2

32 50.0 0.0 -50.0

31 75.0 57.5 -17.5

30 100.0 115.0 15.0

29 125.0 172.5 47.5

28 150.0 230.0 80.0

27 175.0 287.5 112.5

26 200.0 345.0 145.0

9 200.0 345.0 145.0

NodeNumber

InsituPorePressurekN/m

TotalPorePressurekN/m2

ExcessPorePressurekN/m2

| 34 0.0 0.0 0.0

33 25.0 57.5 32.5

32 50.0 115.0 65.0

31 75.0 172.5 97.5

30 100.0 230.0 130.0

29 125.0 287.5 162.5

28 150.0 345.0 195.0

27 175.0 402.5 227.5

26 200.0 460.0 260.0

9 200.0 460.0 260.0

Fig 4.14 Continued

54

In the case of concreting under bentonite slurry, the differences

between the hydrostatic pressures of fresh concrete plus the weight

of bentonite slurry on the fresh concrete and insitu pore pressures

have been applied as incremental excess pore pressure fixities (Fig

4.15).

When concreting has been finished, the base and the sides of the

hole/trench have been kept as drained boundaries with known excess

pore pressures for a while to simulate the setting process of

concrete. Although it is rather unrealistic, the concrete has been

assumed to set either immediately or 12 hours after concreting has

been completed. This is because it is impossible to change the

material properties of certain finite elements in the middle of the

analysis.

Pressures Applied to the BoundariesThe total hydrostatic pressure of the fresh concrete has been

applied to the base and the sides of the hole/trench if bentonite

slurry is not present. This has been done from the bottom to the

top in an incremental manner. Fig 4.16 shows these pressures at the

different steps of the concreting process.

When the concreting was carried out in the presence of the bentonite,

slurry the difference between the hydrostatic pressures of fresh

concrete and the hydrostatic pressures of the bentonite slurry were

applied to the base and the sides of the hole/trench as fresh

concrete displaced the bentonite slurry in an incremental manner.

Applied pressures at the different stages of the concreting process

are illustrated in Fig 4.17.

4 .6 .3 Consolidation

As was mentioned previously, concrete has been assumed to set

immediately after a certain time-lag when concreting has been

completed in the analysis. In reality, however, it is known that

concrete sets gradually. Its stiffness changes from a small value

to the value of, for example, 30GPa during the setting process.

55

NodeNumber

InsituPorePressurekN/m

TotalPorePressurekN/m

ExcessPorePressurekN/m

28 150.0 180.0 30.0

27 175.0 237.5 62.5

26 200.0 295.0 95.0

9 200.0 295.0 95.0

----- 31

— —

33

32

31

. A . - \

\ \ °< \ \ • &

\ q * \ v

30

29

28

27

26

NodeNumber

InsituPorePressurekN/m2

TotalPorePressurekN/m2

ExcessPorePressurekN/m2

30 100.0 120.0 20.0

29 125.0 177.5 52.5

28 150.0 235.0 85.0

27 175.0 292.5 117.5

26 200.0 350.0 .150.0

9 200.0 350.0 ! 150.0

Fig 4.15 Excess pore pressure fixities at different stages of concreting in the presence of bentonite slurry

56

NodeNumber

InsituPorePressurekN/m2

TotalPorePressurekN/m2

ExcessPorePressurekN/m2

32 50.0 60.0 10.0

31 75.0 117.5 42.5

30 100.0 175.0 75.0

29 125.0 232.0 107.5

28 150.0 290.0 140.0

27 175.0 347.5 172.5

26 200.0 405.0 205.0

9 200.0 405.0 205.0

NodeNumber

InsituPorePressurekN/m2

TotalPorePressure kN/m2 .

ExcessPor e ___ .PressurekN/m2

34 . 0.0 0.0 0..0

33 25.0 57.5 32.5

32 50.0 115.0 65.0

31 75.0 172.5 97.5

30 100.0 230.0 130.0

29 125.0 287.5 162.5

28 150.0 345.0 195.0

27 175.0 402.5 227.5

•26 200.0 • 460.0 260.0

9 200.0 460.0 260.0

d

Fig 4.15 Continued

31

33

32

31

30

29

28

27.... /9 26

Q

NodeNumber

Incremental Pressure kN/m2

Cumulative Pressure kN/m

28 0.0 0.0

27 57.5 57.5

26 115.0 115.0

9 115.0 115.0

NodeNumber

Incremental Pressure kN/m2

Cumulative Pressure kN/m2

30 0.0 0.0

29 57.5 57.5

28 115.0 115.0

27 115.0 172.5

26 115.0 230.0

25 115.0 230.0

b

Fig 4.16 Applied pressures at the different stages of concreting(no support used)

58

NodeNumber

Incremental Pressure kN/m2

Cumulative Pressure kN/m2

32 0.0 0.0

31 57.5 57.5

30 115.0 115.0

29 115.0 172.5

28 115.0 230.0

27 115.0 287.5

26 115.0 345.0

9 115.0 345.0

Node . Number

^Incremental Pressure kN/m2

Cumulative Pressure kN/m2

34 0.0 0.0

33 57.5 57.5

32 115.0 115.0

31 115.0 172.5

30 115.0 230.0

29 115.0 287.5

28 115.0 345.0

27 115.0 402.5

26 115.0 460.0

9 115.0 460.0

d

Fig 4.16 Continued

59

NodeNumber

Prior toConcreting,Pressuredue toBentonitekN/m

Pressure due to Concrete & Bentonite kN/m2

IncrementalPressurekN/m

28 180.0 180.0 0.0

27 210.0 237.5 27.5

26 240.0 295.0 55.0

9 240.0 295.0 55.0

b

NodeNumber

Prior toConcreting,Pressuredue toBentonitekN/m

Pressure due to Concrete & Bentonite kN/m2

IncrementalPressurekN/m

30 120.0 120.0 0.0

29 150.0 177.5 27.5

28 180.0 235.0 55.0

27 210.0 292.5 55.0

26 240.0 350.0 55.0

9 240.0 350.0 55;.0

Fig 4.17 Applied pressures at the different stages of concreting in the presence of bentonite slurry

60

NodeNumber

Prior to Concreti ng, Pressure due to Bentonite kN/m

Pressure due to Concrete & Bentonite kN/m2

IncrementalPressurekN/m2

32 60.0 60.0 0.0

31 90.0 117.5 27.5

30 120.0 175.0 55.0

29 150.0 232.5 55.0

28 180.0 290.0 55.0

27 210.0 347.5 55.0

26 240.0 405.0 55.0

9 240.0 405.0 . 55.0

NodeNumber

Prior toConcreting,Pressuredue toBentonitekN/m

Pressure due to Concrete & Bentonite kN/m

IncrementalPressurekN/m2

34 0.0 0.0 0.0

33 30.0 57.5 27.5

32 60.0 115.0 55.0

31 90.0 172.5 55.0

30 120.0 230.0 55.0

29 150.0 287.5 55.0

28 180.0 345.0 55.0

27 210.0 402.5 55.0

26 240.0 460.0 55.0

9 240.0 460.0 55.0d

Fig 4.17 Continued

61

Owing to the fact that it is not possible in the program to put the

elements which have previously been removed back into the mesh with

different properties, a different approach has been employed to

simulate the setting process.

According to this approach: first the base and the sides of the

hole/trench have been kept for a while as drained boundaries with

known excess pore pressures and with the hydrostatic pressure of

fresh concrete acting on them after concreting has been completed

(Fig 4.18a). Then suddenly the base and sides of the hole/trench

have been made impermeable and restrained in both the horizontal and

vertical directions (Fig 4.18b).

Although this approach may seem to be crude, it gives an insight to

the extreme cases when the duration of the first step is changed

from no time to 12 hours.

After the base and sides of the hole/trench have been made

impermeable and restrained in the horizontal and vertical

directions, it has been left for consolidation in order to observe

the change of stress regime as excess pore pressures dissipate.

4.7 Choice of Increment Numbers and Time Intervals

The program calculates the incremental displacements for each

increment using a tangent stiffness approach. It is necessary to

use as many increments as possible to obtain accurate results.

Moreover, when a too large increment has been applied close to

collapse the plastic multiplier, dx, may take negative values due to

numerical problems (theoretically, dx cannot take negative values).

Increasing the number of increments overcomes these difficulties,

but at the same time, however, it escalates the computer costs.

Therefore some compromise has to be made between accuracy and cost.

The increment numbers have been chosen bearing these facts in mind.

Every different stage of the construction process has been

accommodated in 10 increments (Table 4.3).

62

{'(/ p e r m e a b l e ,

'v. no re s tra in ts .

M W

ft%

!

a

Fig 4.18 Boundary conditions during the setting period of concrete

63

Inc.Number T ime Total T ime

10 3 hours 3 hours

Excavation20 3 it 6 ii

30 3 ii 9 ii

40 3 ii 12 n

50 1 hour 13 ii

60 1 u 14 ii

Concreting 70 1 n 15 ■<

80 1 ii 16 it

90 12 hours 28 it

100 5 days 5 days

110 : 10ii 15 ii

120 45 n 50 ii

Consolidation 130 100 n 150 it

140 150 ii 300 ii

150 300 H 600 ii

160 ■ 600 ii 1200 ii

Table 4.3 Duration of the construction stage

64

Although the time intervals for the analysis have been chosen

according to the common practice of diaphragm wall and cast-insitu

bored pile construction, the following factors have also been

considered as suggested in the CRISP users manual:

1. Excess pore pressures are assumed to vary linearly with time

during each increment.

2. In a non-linear analysis the increments of effective stresses

must not be too large.

3. It is a good idea to use the same number of time increments per

log cycle of time.

4. If a very small time increment is used near the start of the

analysis then the finite element equations will be ill

conditioned.

5. When a change in pore pressure boundary condition is applied

the associated time step should be large enough to allow the

effect of consolidation to be experienced by those nodes in the

mesh with excess pore pressure variables that are close to the

boundary.

65

5 Results

The results of the finite element analyses which have been carried

out in this study are presented in two main sections. The first

contains the results of diaphragm wall analyses and the second of

cast-insitu bored pile analyses. Horizontal and vertical effective

stresses, horizontal total stresses, excess pore pressures, plastic

zones and 3 values have been considered in each case. Because of

the difficulties in the interpretation of a large quantity of

printed output, all results are presented in the form of either

contoured effective stress fields, excess pore pressure fields or

effective stress, total stress and excess pore pressure graphs.

Although all possible care has been taken to produce precise

contoured effective stress and excess pore pressure fields, due to

the approximations which are involved in the contouring routine in

the GINO package, these contours may be smoother than they are

supposed to be.

5.1 Diaphragm Walls

Basically two different analyses have been carried out to

investigate the effects of the construction process of diaphragm

walls on., the surounding soil prior to the front excavation. Long­

term changes of the ground conditions have also been examined in the

case of load bearing diaphragm walls without any retaining function

i.e. no front excavation. Firstly a plane strain analysis was

carried out to simulate diaphragm wall construction without any

support, and secondly another plane strain analysis was carried out

to simulate a diaphragm wall construction with bentonite support.

5.1.1 Analysis of Diaphragm Wall Construction Without Support (D-NS/ND/12)

During the excavation stage of this analysis an extensive plastic

zone developed and large plastic deformations occurred. This was

not unexpected. These large deformations caused an error in

equilibrium up to 120 - 150% in both x and y directions in the

analysis. Therefore no further attempts were made to examine the

66

other stages of the construction, and the results of this analysis

have been disregarded. Fig 5.1 shows the extensive plastic zone

developed in the vicinity of the trench, excavation. In reality,

however, a smaller plastic zone may be expected because of the fact

that diaphragm walls are constructed in panels and are not of an

infinite length.

5.1.2 Analysis of Diaphragm Wall Construction With Bentonite Support (D-BS/ND/12)

In this analysis the following results of horizontal and vertical

effective stresses, horizontal total stresses, excess pore

pressures, plastic zones and 8 values were obtained.

Horizontal Effective StressesDuring the excavation horizontal effective stresses decreased by up

to 35% of their initial values around'-the trench. This was- due to

the stress relief at the sides of the trench. An increase in

horizontal effective stresses occurred at the bottom of the trench

(Fig 5.2, Fi,g.A2.2, Fig A2.3, Fig A2.4, Fig A2.5).

Horizontal effective stresses approached a value of zero near the

trench at, the concreting stage (F..ig_ 5.3,\ Fig A2.6, Fig A2.7,. Fig

A2.8, Fig A2.9). This was because the concrete was assumed to

behave like a heavy fluid and thus did not apply any horizontal

effective stresses to the sides and the base of the trench.

After the completion of concreting, the concrete was left as a heavy

fluid for 12 hours. This caused the depressed zone to extend (Fig

5.4(a), Fig A2.10). Compare Fig 5.3(d) and Fig 5.4(a), Fig A2.9 and

Fig A2.10. When the concrete set, horizontal effective stresses

recovered partially but even * 3.2 years after the setting of

concrete they did not come back to their initial values. Especially

towards the bottom of the trench the decrease in horizontal

effective stresses reached about 30% of their initial values. Fig

5.4(b), 5.4(c) and 5.4(d) show the horizontal effective stress

fields around the trench 5 days, 15 days and « 3.2 years after the

setting of the concrete respectively (see also Fig A2.ll, Fig A2.12,

Fig A2.13).

67

DIAPHRAGM

- PLASTIC Z

ONE

G>

(uj) iqdap

68

Fig

5.1

Plastic

zones

at the

excavation

stage

of analysis

D-NS

/ND/

12

DlAPHRAGfl

- EFF

HOR STR

S

69

Fig

5.2

Hori

zont

al

effective

stresses

at the

excavation

stage

of analysis

D-BS

/ND/

12

DIAPHRAGM

- EFF

HOR STR

S

71

Fig

5.4

Hori

zont

al

effective

stresses

a) 12

hours

after

conc

reti

ng,

b) 5

days

c) 15

days

d)

1200

days

(* 3

years)

after

concrete

set

in analysis

D-BS

/ND/

12

Horizontal Total StressesAlthough it is possible to calculate the horizontal total stress

changes during the construction process considering insitu pore

pressure,excess pore pressures and horizontal effective stresses

together, they are also presented for convenience. Graphs which

show horizontal total stress variations during the construction

processcan be found for the analyses D-BS/ND/12, P-NS/ND/12 and

P-BS/ND/12in Appendix 3.

Vertical Effective StressFig 5.5 shows the vertical effective stress isochrones during the

excavation. As the excavation progressed, the vertical effective

stresses around the trench rose by up to 50% of their initial values

(see also Fig A4.1, Fig A4.2, Fig A4.3, Fig A4.4, Fig A4.5).

The vertical effective stress fields during the concreting are

presented in Fig 5.6. The vertical effective stresses which had

risen at the excavation stage, partially dropped as the fresh

concrete was placed into the trench (see also Fig A4.6, Fig A4.7,

Fig A4.8, Fig A4.9).

The values of vertical effective stresses approached their initial

values 3.2 years after the concrete had set. (Fig 5.7, Fig A4.10,

Fig A4.ll, Fig A4.12, Fig A4.13).

Excess Pore PressuresNegative excess pore pressures were generated at the sides and

positive excess pore pressures at the bottom of the trench

respectively during the excavation (Fig 5.8, Fig A5.2, Fig A5.3, Fig

A5.4, Fig A5.5). Generation of negative excess pore pressures at

the sides of the trench was due to the stress relief caused by

excavation.

During the concreting, because of the fresh concrete positive excess

pore pressures were generated (Fig 5.9, Fig A5.6, Fig A5.7, Fig

A5.8, Fig A5.9).

72

DIAPHRAGM

- EFF

VER SIRS

(uj) iqdap

73

Fig

5.5

Vertical

effective

stresses

at the

excavation

stage

of analysis

D-BS

/ND/

12

DIAPHRAGM

- EPF

VER STR

S

74

Fig

5.6

Vertical

effective

stresses

at the

concreting

stage

of analysis

D-BS

/ND/

12

DIAPHRAGM

- EFF

VER STR

S

(ui) qqdap

75

Fig

5.7

Vertical

effective

stresses

a) 12

hours

after

conc

reti

ng,

b) 5

days

c) 15

days

d)

1200

days

(» 3

years)

after

concrete

set

in analysis

D-BS

/ND/

12

DIAPHRAGM

- EXS

POR PRE

S

76

Fig

5.8

Excess

pore

pressures

at the

excavation

stage

of analysis

D-BS

/ND/

12

DIAPHRAGH

- EXS

FOR PRE

S

(UI) I|4d9p

77

Fig

5.9

Excess

pore

pressures

at the

concreting

stage

of analysis

D-BS

/ND/

12

DIAPHRAGM

- EXS

POR PRE

S

78

Fig

5.10

Excess

pore

pressures

a) 12

hours

after

concreting

b)

5 days

c)

15 days

d)

1200

days

(~

3 years)

after

concrete

set

in analysis

D-BS

/ND/

12

After concreting had been completed and concrete had set, the excess

pore pressures dissipated over a period of 3.2 years (Fig 5.10, Fig

A5.10, Fig A5.ll, Fig A5.12, Fig A5.13).

Plastic ZoneA plastic zone developed near the trench as the excavation

progressed but it was not as extensive as the one in the case where

bentonite was not present. Fig 5.11 and Fig 5.12 show the plastic

zones which developed during excavation and concreting

respectively.

8 Values8 values were calculated after the excess pore pressures had

dissipated i.e. 3.2 years after the setting of the concrete (see

Appendix 1). Fig 5.13 shows the variation of 8 values with depth.

The average was found as 0.59 at the element centroids next to the

wall and the equation of a straight line which was fitted using the

Least Squares method was obtained as 8 - 0.751 - 0.0160H (where H is

the depth). 8 values at the nodes adjacent to the wall were also

calculated, but these values were disregarded for reasons explained

in section 5.12.

5.2 Cast-insitu Bored Piles .........

A number of analyses have been carried out to examine the effects of

the construction process as a whole and the delays in the particular

stages of construction on the surrounding soil and thus on the

performance of the cast-insitu bored piles.

The following analyses to simulate cast-insitu bored pile

construction have been carried out:

1 Without any support, no delay in concreting, 12 hours concrete

setting time (P-NS/ND/12)

2 With bentonite slurry support, no delay in concreting, 12 hours

concrete setting time (P-BS/ND/12)

3 Without any support, 1 day delay in concreting, 12 hours conrete

. setting time (P-NS/1D/12)

4 Without any support, 7 days delay in concreting, 12 hours

concrete setting time (P-NS/7D/12)

5 With bentonite support, 1 day delay in concreting, 12 hours.-.. .

concrete setting time (P-BS/1D/12)

79

DIAPHRAGM

- PLASTIC Z

ONE

(ui) tqdgp

80

Fig

5.11

Plastic

zones

developed

during

the

excavation

stage

of analysis

D-BS

/ND/

12

DIAPHRAGM

- PLASTIC Z

ONE

(UI) L|4d0p

81

Fig

5.12

Plastic

zones

developed

during

the

concreting

stage

of analysis

D-BS

/ND/

12

Fig 5.13 8 values varying with depth in analysis D-BS/ND/12

82

6 With bentonite support, 7 days delay in concreting, 12 hours

concrete setting time (P-BS/7D/12)

7 Without any support, no delay in concreting, immediate concrete

set (P-NS/ND/0)

8 With bentonite support, no delay in concreting, immediate

concrete set (P-BS/ND/0)

Only the results of the analysis P-NS/ND/12 and the analysis

P-BS/ND/12 will be presented in detail to avoid unnecessary

repetitions. The rest of the analyses will be considered in terms

of 3 values. Because of the similarities between the results of

P-NS/ND/12 and P-BS/ND/12, the presentation of the results of these

analyses will be in the same section.

5.2.1 The Analysis P-NS/ND/12 and the Analysis P-BS/ND/12

In these analyses the following results of horizontal and vertical

effective stresses, excess pore pressures, plastic zones and 3

values were obtained.

Horizontal Effective StressesIn both analyses, horizontal effective stresses decreased around the

hole during the excavation due to the stress relief. The decrease

in horizontal effective stresses in analysis P-NS/ND/12 was more

than in the analysis P-BS/ND/12 because of the fact that no support

was used in the latter analysis (Fig 5.14, Fig 5.15, Fig A2.14, Fig

A2.15, Fig A2.16, Fig A2.17, Fig A2.26, Fig A2.27, Fig A2.28, Fig

A2.29).

Fig 5.16, Fig 5.17, Fig A2.18, Fig A2.19, Fig A2.20, Fig A2.21, Fig

A2.30, Fig A2.31, Fig A2.32, Fig A2.33 show the changes of the

horizontal effective stress fields during concreting.

Approximately 3.2 years after the concrete had set, horizontal

effective stresses partially recovered in both cases. Horizontal

effective stresses in the analysis P-NS/ND/12 were lower than the

ones in the analysis P-BS/ND/12 (Fig 5.18, Fig A2.22, Fig A2.23, Fig

A2.24, Fig A2.25, Fig A2.34, Fig A2.35, Fig A2.36, Fig A2.37).

83

PILE

- EFP

HOR STR

S

84

Fig

5.14

Hori

zont

al

effective

stresses

at the

excavation

stage

of analysis

P-NS

/ND/

12

PILE

- EFF

HOR SIR

S

(uj) iqdap

85

’ Fig

5.15

Hori

zont

al

effective

stresses

at the

excavation

stage

of analysis

P-BS

/ND/

12

PILE

- EFF

HOR STR

S

C DO

OO

(uu) M+dBp

86

Fig

5.16

Hori

zont

al

effective

stresses

at the

concreting

stage

of analysis

P-NS

/ND/

12

PILE

- EFF

HOR STR

Si

(UJ) iqdap

87

Fig

5.17

Hori

zont

al

effective

stresses

at the

concreting

stage

of analysis

P-BS

/ND/

12

PILE

- EFF

HOR STR

S

fun uidao

88

Fig

5.18

Hori

zont

al

effective

stresses

a) 12

hours

after

concreting

b)

5 days

c) 15

days

d)

1200

days

(~ 3.2

years)

after

concrete

set

in analysis

P-NS

/ND/

12

PILE

- EFF

HOR STR

S

89

Fig

5.19

Hori

zont

al

effective

stresses

a) 12

hours

after

concreting

b) 5

days

c) 15

days

d)

1200

days

(~ 3.2

years)

after

concrete

set

in analysis

P^BS

/ND/

12

It should be noted that the zones which were affected by the

construction were smaller than the case of diaphragm wall

construction.

Vertical Effective StressesDuring excavation vertical effective stresses increased. In the

analysis P-NS/ND/12, at the bottom of the hole a high vertical

effective stress concentration zone occurred (Fig 5.20, Fig 5.21,

Fig A4.14, Fig A4.15, Fig A4.16, Fig A4.17, Fig A4.26, Fig A4.27,

Fig A4.28, Fig A4.29).

The vertical effective stresses obtained from the analysis

P-NS/ND/12 decreased whereas the vertical stresses as a result of

analysis P-BS/ND/12 did not change very much around the hole during

the concreting stage (Fig 5.22, Fig 5.23, Fig A4.18, Fig A4.19, Fig

A4.20, Fig A4.21, Fig A4.30, Fig A4.31, Fig A4.32, Fig A4.33).

When the excess pore pressures dissipated, the vertical effective

stresses partially recovered in analysis P-BS/ND/12. Those in the

analysis P^NS/ND/12, however, remained lower than their initial

values (Fig 5.24, Fig 5.25, Fig A4.22, Fig A4.23, Fig A4.24, Fig

A4.25, Fig A4.34, Fig A4.35, Fig A4.36, Fig A4.37).

Excess Pore PressuresThe excess pore pressures which were generated during the excavation

are presented in Fig 5.26, Fig 5.27, Fig A5.14, Fig A5.15, Fig

A5.16, Fig A5.17, Fig A5.26, Fig A5.27, Fig A5.28, Fig A5.29.

Fig 5.28, Fig 5.29, Fig A5.18, Fig A5.19, Fig A5.20, Fig A5.21, Fig

A5.30, Fig A5.31, Fig A5.32, Fig A5.33 show the changes of excess

pore pressures during the concreting.

Approximately 3.2 years after the setting of concrete all excess

pore pressures were dissipated (Fig 5.30, Fig 5.31, Fig A5.22, Fig

A5.23, Fig A5.24, Fig A5.25, Fig A5.34, Fig A5.35, Fig A5.36, Fig

A5.37).

Plastic ZonesDuring excavation a larger plastic zone developed in analysis

P-NS/ND/12 than in analysis P-BS/ND/12 (Fig 5.32, Fig 5.33). Fig

90

PILE

- EFF

VER STR

S

(uj) ipdap

91

Fig

5.20

Vertical

effective

stresses,

at the

excavation

stage

of analysis

P-NS

/ND/

12

PILE

- EFF

VER STR

S

(uj) U-ldap

92

Fig

5.21

Vertical

effective

stresses

at the

excavation

stage

of analysis

P-BS

/ND/

12

PILE

- EFF

VER SIR

S

(uj) M4dap

93

Fig

5.22

Vertical

effective

stresses-at

the

concreting

stage

of analysis

P-NS

/ND/

12

PILE

- EFF

VER STR

S

(\u\ uidaD

95

Fig

5.24

Vertical

effective

stress

es.a

) 12

hours

after

concreting

b)

5 days

c) 15

days

d) 1200

days

(» 3.2

years)

after

concrete

set

in analysis

P-NS

/ND/

12

PILE

- EFF

VER STR

S

96

Fig

5.25

Vertical

effective

stresses

a) 12

hours

after

concre

ting

b) 5

days

c) 15

days

d)

1200

days

(~ 3.2

years)

after

concrete

set

in analysis

P-BS

/ND/

12

PILE

- EXS

POR PRE

S

97

Fig

5.26

Excess

pore

pressures

at the

excavation

stage

of analysis

P-NS

/ND/

12

PILE

“EXS

POR PRE

S

98

Fig

5.27

Excess

pore

pressures

at the

excavation

stage

of analysis

P-BS

/ND/

12

PILE

- EXS

POR PRE

S

o><d>

(Ui) M4d s p

99

Fig

5.28

Excess

pore

pressures

at the

concreting

stage

of analysis

P-NS

/ND/

12

PILE

- EXS

POR PRE

S

( U i ) i q d s p

100

Fig

5.29

Excess

pore

pressures

at the

concreting

stage

of analysis

P-BS

/ND/

12

PILE

- EXS

POR PRE

S

101

Fig

5.30

Excess

pore

pressures

a) 12

hours

after

concreting

b) 5

days

c)

15 days

d)

1200

days

(«

3.2

years)

after

concrete

set

in analysis

P-NS

/ND/

12

PILE

- EXS

POR PRE

S

102

Fig

5.31

Excess

pore

pressures

a) 12

hours

after

concreting

b) 5

days

c) 15

days

d)

1200

days

(»

3.2

years)

after

concrete

set

in analysis

P-BS

/ND/

12

PILE

- PLASTIC Z

ONE

(rn) MTdap

103

Fig

5.32

Plastic

zones

developed

during

the

excavation

stage

of analysis

P-NS

/ND/

12

PILE

- PLASTIC Z

ONE

rt a§ § ac

oU.ci

_j j |----- n---o C3 Od 52 ocsi rn -j-o

(ui) q+dap

104

Fig

5.33

Plastic

zones

developed

during

the

excavation

stage

of analysis

P-BS

/ND/

12

PILE

- PLASTIC Z

ONE

105

Fig

5.34

Plastic

zones

developed

during

the

concreting

stage

of analysis

P-NS

/ND/

12

PILE

- PLASTIC Z

ONE

d

106

Fig

5.35

Plastic

zones

developed

during

the

concreting

stage

of analysis

P-BS

/ND/

12

5.34 and Fig 5.35 show the plastic zones during concreting. It

should be noted that the plastic zone was confined to a narrow band

around the hole in each case.

8 Values8 values which were calculated after the excess pore pressures

dissipated are presented in Fig 5.36 and Fig 5.37 from the analysis

P-NS/ND/12 and the analysis P-BS/ND/12 respectively. The average of

the 8 values at the element centroids next to the pile in analysis

P-NS/ND/12 (0.24) was found to be lower than the one in analysis

P-BS/ND/12 (0.40). 8 values at the nodes adjacent to the pile were

also calculated but they were disregarded for the reasons explained

in section 6.12.

5.2 .2 The Analysis P-NS/1D/12, the Analysis P-NS/7D/12, the Analysis P-BS/1D/12, the Analysis P-BS/7D/12, the Analysis P-NS/ND/0 and the Analysis P-BS/ND/0

These analyses were carried out to investigate effects of the delays

in concreting and the setting time of concrete on the horizontal

effective stresses adjacent to the pile after excess pore pressures

had dissipated. It was found that the delays in concreting caused a

. decrease in 8 values in the analyses where, bentonite slurry Used as

support, while 8 values in analyses where no support used were not

affected by the delay (Fig 5.38, Fig 5.39, Fig 5.40, Fig 5.41)’.

Relatively lower 8 values were obtained from analyses P-NS/ND/0 and

P-BS/ND/0 where concrete was assumed to set immediately (Fig 5.42,

Fig 5.43).

107

CO00CS5

t. rf Ql ft > <1/ < CO

<S5CM<s>G3<S5

<35 +■ lV CO' UO * ft

® CO <35 COCO CO

II • II *GO (35

<Uon n Si IIrf LU rf0. rf <L» ft > d>< oa

CS)CS)CM

CS)in

oCS)

UiQ>rs

cJ- f tdiCQ

-<3-<3-<q-_CL

_£i— <— —̂ ,3— <i— <3— d— <— *3— <r

-O-<S)L i n

CS)CM

i n CS) in

(Ui) qjdaQ

CS).CS)CS)

X

10CDS5cJ

©-fta»co

CM

(UI) i^daa

108

F1q

5.36

8 values

varying

with

depth

in analysis

Fl9

5.37

S values

varying

with

depth

in an

alys

is

P-NS/ND/12

P-BS

/ND/

12

cnCDcocaco

v + (ft • (N ca c-"

a ll t— i cao <D o' cn iiI— d Z t. d UJ (Li O > 0) O < oa

COincataca

£ Ico • co ca ca

cn" <9Si l|dU d<U> <u< CD

o

—'"Q~

C£>CM

IO <33 rto

CS3CM

(S3UO

CS3(S3

CS3L i n

<33<33

C5-Pcuco

(S3

(ui) q7daQ

X XCD cncn CDca CO<o ca<s ca

v + CO |(N CD

* caca in co <o

cna ii rt ll *

ca <oo (U aiD i cn l| cn i|h- d UJ dZ V, d CD C dUJ <U P o <U +->(_) > a> z > a»o < cn <3 < ca

<33CS3

CM

<SIX )

(S3<33

k1cu35CJ>

CUCQ

CS3.IX)

-O _CL

(S3 , (S3

(S3CM

IX ) (S3 i n (S3

(UJ) qqdea

109

Fig

5.38

8 values

varying

with

depth

in analysis

Fig

5.39

8 values

varying

with

depth

in an

alys

is

P-NS/1D/12

P-NS

/7D/

12

Z XCOv~ cn(S3 (S3(S3 (S3(53 (S3

If) + CO 1cn U3• 0 * CM

(S3 (S3 (S3 CMcn C-«a II > It •M (S3 (S3a a; OrO' cn i| cn i|tr d UJ dz t, d a t. dUJ di -H CD Qi *->u > di z > Qio < cd <J < CQ

CSDCSDCM

CSDLO

CDCSD

tnQJZ5

cJCU

OQ

co>>rdcrrd

c l<u*o

O -Q- ■Q—

-<5---<3---<3--- <3— <--=q-

-€+ ■cr

CSD■ to

C£>CSD

CS)CD

to CD If?

(ui) qideo

CS)

CDcr

’>»s_rd> CM coCU Q =3 1̂ .rd CO > CO

1on. o_

LOCD

x XCO CMCM CM(S3 O<53 (S3(S3 (S3*

03 + in 1cn CO• in

(S3 CO (S3 C-cn toa ii » IIt-H (S3 (S3CD QJ Qiac cn ii cn tiI— • d UJ dz t, d Q L. dU) di -H O Qi -r>(-3 > Qi Z > QiO < CQ <3 < CQ

_<3__ <1—cj--<3---<---<---<1---<1— «3--*3---<---<J---«cr

~o~

CSDCSD

CM

CSDin

CSDCSD

CSDLin

CSDCSD

CSDCM in CSD in CSD

tna»S3

cSVQiOQ

(u i) M^deQ

CO>3rdcrd

+->CLCU•a

+ j•I—3CDC

’>»

> CM rH

COcu oZ rHrd to> CQ

Ion cu o

CDU_

110

Xoo

Xco CS)

(U I) L|+daa

uiaio

c3-ftCU0Q

CO>)rOCfO

fta .a>X)-Cft

DOC>)s-fU>

oCOCU Q 3 ZT03 OO> ca

ica o_

co

LT3CD•r—Ll_

X Xin ftft G)S3 S3S3 S3

ft + ft 1CO* ▼— • ftS3 ft S3 ft

S3 COa II • II 'i— i S3 S3o CD 1 <Vos cn n cn i|I— Ci UJ rf

C rf a rfCD ai ft a QJ ftu > V z >o < ca < < CQ

CS)(S)C M

CS)LO

CS)CS)

CS)in

incursCS

>c$

f tQjCQ

(u i) qqdaQ

to

CCSc03

ftC lCU

X J

-Cft

CDc>»S-03>

OCOCU Q =5 ^

r— S , 03 OO > Z

Ica CL.

CM«rf*inCJ>

111

6 D i s c u s s i o n

The r e s u lt s of th is study, which are presented in Chapter 5, answer

some of the questions concerning the e ffe cts of the diaphragm wall

and c a s t - in s it u bored p ile in s t a lla t io n on the surrounding s o i l , and

subsequently on the performance of these s tru c tu re s. At the same

time, however, they r a is e fu rth er issues which cannot be answered or

even guessed properly without ad dition al experimental and numerical

research .

6 .1 A nalysis

6 .1 .1 Assumptions

In th is study, as in any other numerical study, some assumptions

have had to be made in order to sim ulate the in s t a lla t io n process.

The common problem in geotechnical engineering, u n ce rta in tie s in

obtaining s o il parameters, has a d ire c t e ffe c t on the r e s u lt s of

t h is study though a ll the necessary values of s o il p ro p erties were

obtained from the r e s u lt s of extensive f ie ld and laboratory t e s ts ,

and past-experience j- a va il able- in< -the - l it e r a t u r e .

As explained in Chapter 4, the s o il deposit was assumed to be

homogeneous " and is o t ro p ic /w ith a . constant - e la s t ic it y modulus,.:

P o isso n 's r a t io , earth pressure c o e ff ic ie n t at re s t (K0 ) , cohesion

and angle of f r ic t io n . A l in e a r ly el a s t ic -p e r f e c t ly p la s t ic s o il

model with a Mohr-Coulomb y ie ld c r it e r io n obeying the associated

flow ru le was employed in the analyses. I f London Clay had the

above p ro p ertie s and behaved e x a c tly according to the model

employed, l i f e would have been too easy for geotechnical engineers

and the r e s u lt s of t h is study could interpreted with confidence.

In r e a l it y , however, London Clay has d iffe re n t p ro p ertie s in

horizontal and v e r t ic a l d ire c t io n s and thus is not is o tro p ic and it s

e la s t ic it y modulus and K0 values do vary with depth. The e la s t ic it y

modulus has the value of zero at ground level and increases with

depth. Taking an average value for the e la s t ic it y modulus makes the

s o il too s t i f f near the su rfa ce . Since displacem ents are not the

main in te re st of t h is study, the e ffe c ts of an average e la s t ic it y

modulus may not be s ig n if ic a n t . The earth pressure c o e ffic ie n t at

re s t (K0) for London Clay v a rie s with depth and can have values of 3

near the surface decreasing to le s s than 1 at great depth.

Employing a value of 2 for Kq underestimates the horizontal

e ffe c t iv e stre sse s near the surface and therefore the horizontal

e ffe c t iv e stre ss leve l may be lower than can be expected in r e a l it y

in these re g io n s. Near surface strengths may be u n r e a lis t ic .

The Mohr-Coulomb y ie ld c r it e r io n with associated flow r u le , which

was employed in the analyses, p re d icts u n r e a lis t ic a l ly large

d ila ta n c y , though overconsolidated cla y s do d ila t e to a ce rta in

extent.

The analyses f a i l to sim ulate the gradual se ttin g of 'concrete

p ro p erly. I t is known that e la s t ic it y modulus of concrete changes

from a small value to the value of about 30 GPa. The concrete used

in bored p ilin g and diaphragm w a llin g has a high water-cement r a t io

and it s B value is very close to u n ity i f measured s h o rtly after

water is added to the mix. Therefore the strength of wet ..concrete

would be expected to be n e g lig ib le and wet concrete can be regarded

as a heavy f lu id .

The assumption that h y d ro static pressures are applied to the sid es

of the hole would not be u n r e a lis t ic at th is stage though at larger

depths pressures do not increase so fa s t due to the arching and most

l ik e ly s t if fe n in g of the deepest m aterial before the upper part of

concrete is placed (Fig 2 .7 , Clayton and M il it it s k y , 1983).

Concrete can be assumed to be r ig id r e la t iv e to the s t if fn e s s of the

surrounding s o il when i t se ts. In the analyses, the concrete was

modelled as a heavy f lu id fo r 12 hours and set immediately,

d isreg ard ing the gradual manner of s e ttin g . Two analyses where

concrete set immediately afte r completion of concreting were also

ca rrie d out.

Despite these short-com ings, t h is study is quite capable of g ivin g

an in sig h t into what goes on during the co n struction process of>

113

diaphragm w alls and c a s t - in s it u bored p ile s . E s p e c ia lly the use of

coupled co nsolid ation theory enables the e ffe c t iv e stre ss and excess

pore pressure changes to be examined throughout the construction

process. I f the d i f f i c u l t i e s , which are involved in an experimental

study to examine the same problem, are considered, the importance of

such a study can e a s ily be appreciated.

6 .1 .2 Numerical D if f ic u lt ie s

In t h is study, the values of e ff e c t iv e /t o t a l stre sse s and excess

pore pressures at one in te g ra tio n p o in t, which coincides with the

element ce n tro id , in every element were chosen to represent the

o v e ra ll behaviour of the s o il d ep o sit. A ll the r e s u lt s are

presented considering these valu e s.

Fig 6.2 Fig 6.3 and Fig 6.4 show the horizontal e ffe c t iv e stre sse s

at the in te g ratio n points at the depth of 11.25m (Fig 6.1) during

the in s t a lla t io n process in the a n a ly sis D-BS/ND/12. T h is was a

plane s tra in a n a ly s is . The values at the element cen troid s (t) and

at the other in te g ra tio n points (o) show a good agreement.

Co.ns id erring only, ceg.tr oi dal values seems, to be adequate to .re p re sen t

the behaviour of the s o il deposit at that depth. I f , however, the

horizontal e ffe c t iv e stre sse s at the in te g ra tio n points at the same

depth in the a n a ly sis P-NS/ND/1-2 are examined,- a flu c tu a tio n in the

values can be seen in the element adjacent to the p ile (Fig 6 .5 , Fig

6.6 and Fig 6 .7 ) . In a ll the axisymmetric analyses, such as

an a lysis P-NS/ND/12, the same kind of behaviour was observed.

T h is type of response is s im ila r to that reported by Naylor (1974)

who found s im ila r o s c il la t io n s using eight noded q u a d rila te ra ls in

the a n a ly sis of nearly incom pressible e la s t ic m a te ria l. Sloan and

Randolph (1982) found that the eight noded q u a d rila te ra l elements

can give poor estim ates of co lla p se loads in axisymmetric e la sto -

p la s t ic analyses. These observations can be explained by the

d if f ic u lt y these elements have in modelling no volume change

p re c is e ly . Various ways have been suggested of avoiding these

d i f f ic u lt i e s : reduced in te g ra tio n , high order tria n g u la r elements

114

11.2

5 m

0.75m 0 .5m i 0 .5m 0 .5m 0.5m 0 .5m 0.75m

element 28

w i M

0+0 ftp

4 f t ft4/

0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

Fig 6.1 Locations of the in te g ra tio n points considered in Fig 6 .2 , Fig 6 .3 , Fig 6.4 , Fig 6 .5 , Fig 6.6 and Fig 6.7

HOR

EFF

STRS

(k

Pa)

HOR

EFF

STRS

(k

Pa)

HOR

EFF

STRS

(k

Pa)

HOR

EFF

STRS

(k

Pa)

320.

210.

160.

1005.0m Excavat ion

0 • 00 • 00 » 00 » 00 * 0 0

.10 .80 1.20 1 .60 2-00 2.10 2?80 3?20 3?60 7.

100

320J

210.

160.

80J

10.0ni Excavation

Tl0 U2Q K60 2J00 2̂ 10 2.80 3.20 3.60 ?.

100.

320.

210.

160.

15.0m Excavation

0 ■ 0 0 • o0 * 0

0 • 0 0 • 0 o

.00 .10 .80 1 .20 1.60 2.00 2.10 2.80 3.20 3.60 1.

100

320.

210.

160,

80.

20.0m Excavation

• 0 • 0 0 • 0 0 • o 0 * 0

Tl0 J30 1L20 \ M 2 M 2~10 ±80 ±20 ±60 1.00

Distance From the wall axis <m)

Fig 6.2 H orizontal e ffe c t iv e stre sse s at the element centroids (•) and the other in te g ra tio n points (o) during the excavation process in a n a ly sis D-BS/ND/12 (see Fig 6.1 for the lo c a tio n s)

116

HOR

EFF

STRS

CK

Pa)

HOR

EFF

STRS

(k

Pa)

HOR

EFF

STRS

(k

Pa)

HOR

EFF

STRS

(k

Pa)

100

320.

210.

160.

5.0m Con cr et in g

o 0 • 00 • 00 a Oo

.00 .10 .8 1.20 1.80 2.00 2.10 2.80 3.20 3.80 1.

100

'320.

210.

160

80 J

0.

10.0m Concreting

0 00 0 .■ Oo • oo a o 0 a

.03

100

320.

210.

180.

.10 .80 1.20 1 .60 2.00 2.10 2.80 3.20 3.60 1.00

15.0m Concreting

o a 0 0o o « o o t o 0 • o

.10 .80 1.20 1 .80 2.00 2.10 2.80 3.20 3.80 1.

100_

320.

210.

180.

20.0m Concreting

0 o • oo • oo a o o

.00I r i r "1 1 1 1.10 .80 1.20 1.60 2.00 2.10 2.80 3.20 3.80 1.00

Distance From the wall axis <m)

Fig 6.3 H orizontal e ffe c t iv e s tre sse s at the element ce n tro id s (•) and the other in te g ra tio n points (o) during the concreting process in a n a ly sis D-BS/ND/12 (see Fig 6.1 for the lo c a tio n s)

117

HOR

EFF

STRS

(K

Pa)

HOR

EFF

STRS

(K

Pa)

HOR

EFF

STRS

(K

Pa)

HOR

EFF

STRS

(K

Pa)

100

320.

210.

160

80.

12 hours a f t e r c o n c r e t i n g

• o 0 • 0

o o

.00

100.

320.

210.

160.

©0 JB0 +20 +60 2 M 2.10 2.80 +20 +60 7

5 days a fte r concrete set

° 0 • 0 0 . • 0 0 • oo t 0 o o o

.10 .80 1 .20 1 .60 2.00 2.10 2.80 3.20 3.60 1,00

100

320.

210.

160.

80.

15 days a ft e r concrete set

0 # ° o t o 0 t 0 0 » 0 0 « 0 o

7l0 AM +20 +60 +00 +10 +80 +20 +60 7.

100.

320.

210.

160.

1200 days a fte r concrete set

0 0 0 I 00 • 00 0 00 • O o

.00 .10 .80 1.20 1.60 2.00 2.10 2.80 3.20 3.60 1.00

Distance From the wall axis Cm)

F ig 6.4 H orizontal e ffe c t iv e stre sse s at the element cen tro id s (•) and other in te g ra tio n points (o) a fte r the completion of concreting in a n a ly sis D-BS/ND/12 (see Fig 6.1 for the lo c a tio n s)

118

HOR

EFF

STRS

Ck

Pa)

HOR

EFF

STRS

(k

Pa)

HOR

EFF

STRS

(k

Pa)

HOR

EFF

STRS

(k

Pa) 320.

210.

160.

80

1005.0m Excavat ion

0 » 0 0 * 0 0 « 0 0 « 0 0 i 0 0 ■ 0

7l0 8̂0 ±20 ±60 ±08 ±10 ±80 ±20 ±60 1.00

100.,

320.

210

160J

80.

0.

10.0m Excavation

0 » 0 o » 0 o » 0 0 « 0 0 • 0

L10 ?80 ±20 ±60 ±00 ±10 ±80 ±20 ±60 1.

100

320.

210.

160.

0.

15.0m Excavation

0 n . o o • o 0 • 0

.00

100

320.

210

160.

80.

.10 .80 ±20 ±60 2.00 2.10 2.80 3.20 3.60 1.00

20.0m Excavation

0 t On ■ 00 * 0 o

.00 .10 .80 ±20 ±60 2.00 2.10 2.80 3.20 3.60 1.00

Distance From the pile axis

Fig 6.5 H orizontal e ffe c t iv e stre sse s at the element centroids (•) and other in te g ra tio n points (o) during the excavation process in a n a ly sis P-BS/ND/12 (see Fig 6.1 for the lo c a tio n s)

119

HOR

EFF

STRS

Ck

Pa)

HOR

EFF

STRS

Ck

Pa)

HOR

EFF

STRS

(k

Pa)

HOR

EFF

STRS

Ck

Pa) 320.

210.

1G0.,

100

0.

5.0m Concret ing

° » o 0 » o o » ° 0

.00 1̂0 -80 +20 + 60 2/00 2X10 + 88 +20 3'. 80 7.00

100

320.

210.

180.

80.

0.

10.0m Concreting

* ° O » 0 O * 0 0 *

.00 .10 .80 1.20 1 -80 2. 7.10 +80 +20 3.80 1.00

100„

320

210.

160.

15.0m Concreting

° o • 00 • 0 0 •

.00

100

320.

210.

180.

.10 .80 1.20 1.60 2.00 2.10 2.80 3.20 3.80 1.

0 • 0

20.0m Concreting

* 0 0 • 0 0 • 0

.00 [10 To +20 +60 +00 + 10 +80 +20 +60 7.00

Distance Pram the pile axis <m)

Fig 5.6 H orizontal e ffe c t iv e stre sse s at the element centro id s (•) and other in te g ra tio n points (o) during the concreting process in a n a ly s is P-BS/ND/12 (see Fig 6.1 for the lo c a tio n s)

120

HOR

EFF

STRS

(K

Pa)

HOR

EFF

STRS

(k

Pa)

HOR

EFF

STRS

(k

Pa)

HOR

EFF

STRS

(k

Pa) 320

210J

160.

80.

100_12 hours a f t e r c o n c r e t i n g

0 0 * oo . O o . o o #

0 0

r10 *80 ±20 ±60 ±00 ±10 ±80 ±20 ±60 1.00

100

320.

210.

160J

80.

0.

5 days a ft e r concrete set

0 0 • 0 0Q0 * 0 0 « 0 0 » 0

>0 J30 ±20 ±60 ±00 ±10 ±80 ±20 ±60 ±

100

320.

210J

160.

15 days a fte r concrete set

0 o * o o « o o . o o

80 .10 .80 ±20 ±60 2.00 2.10 2.80 3.20 3.60 1.

100

328.

210.

160

1200 days a fte r concrete set

O o , o 0 i 0 o » o o * 0

.00 .10 .80 ±20 ±60 ±00 ± 10 ±80 ±20 3'. 60 ±00

Distance From the pile axis Cm)

Fig 6.7 H orizontal e ffe c t iv e stre sse s at the element centroids (•) and other in te g ra tio n points (o) a fte r the completion of concreting in a n a ly s is P-BS/ND/12 (see Fig 6.1 for the lo catio n s)-

121

and the use of f in it e elements with 'mixed' v a ria b le s (displacem ents

and mean normal p re ssu re s). In fa ct the co n so lid atio n a n a ly sis

performed by the CRISP Program is s im ila r to the la s t option

described above and o s c illa t io n in stre sse s do not occur, except in

the element next to the p ile .

In order to c a lc u la te 8 va lu e s, however, i t is necessary to obtain

an estimate of the horizontal e ffe c t iv e stre sse s next to the p ile

i . e . on the side of the element which contains the stre ss

o s c il la t io n . I n i t i a l l y the fo llo w in g scheme was t r ie d , based on the

observation that accurate stre sse s are u su a lly obtained at the 2 x 2

in te g ratio n p o in ts, re g a rd le ss of the integ ratio n r u le used (N aylor;

1974; Barlow, 1976; Zienkiew iez, 19 77):

a) the stre sse s at 2 x 2 in te g ra tio n points were ca lcu la te d from

the values at the 3 x 3 in te g ra tio n points (which CRISP

c a lc u la te s ) using Lagrangian in te rp o la tio n polynomials

b) the stre sse s at the element sides were ca lcu lated by b ilin e a r

e xtrap o la tio n from the values at the 2 x 2 integation, points

(E s s e n t ia lly t h is is the method suggested by Cook (1981))

In p ra c tic e t h is method gave stre sse s which did not f i t in very w ell

with the o v e ra ll ra d ia l d is tr ib u t io n (the stre sse s seemed to .be-too

h ig h ). The values of stre sse s ca lcu lated at the in teg ratio n point

corresponding to the element centroids seemed to f i t in better with

the o v e ra ll stre ss d is tr ib u t io n and so these stre sse s were taken to

represent the stre sse s next to the p ile .

As a summary, the values at the element centroids have been assumed

to represent the behaviour of s o il during the in s t a lla t io n process.

A ll r e s u lt s are presented and interpreted in terms of these valu e s.

At present there is no experimental data or a lte rn a tiv e numerical

a n a ly sis r e s u lt s a v a ila b le to back t h is assumption. Therefore the

author is re lu c ta n t to make any comment on t h is without fu rth e r

experimental and numerical research which was beyond the scope of t h is study.

6 . 2 R e s u l t s

The changes in the ground conditions during the in s t a lla t io n

examined in th is study are very important because they have d ire ct

e ffe c ts on the shaft f r ic t io n of bored p ile s and load applied to

diaphragm w a lls . I f the e ffe c t iv e stre ss method is to be used for

the design of bored p ile s then in s t a lla t io n e ffe c ts should be f u l ly

understood.

The e ffe c t iv e stre ss method may seem simple and straig htforw ard (see

Appendix 1 ) , but estim ating the stre ss regime around the p ile after

in s t a lla t io n is not as simple and straightforw ard as the subsequent

c a lc u la tio n s in the method.

F ir s t of a ll i t ’ is necessary to estimate the value of K a fte r

in s t a lla t io n . O bviously in s it u horizontal e ffe c t iv e stre sse s are

changed due to the co nstruction process. I t is questionable whether

these horizontal e ffe c t iv e stre sse s come back to th e ir i n i t ia l

values i .e . whether K = K0 can be used to estim ate the shaft

f r ic t io n of p ile s . For driven p i le s , K = Kq gives a lower lim it for

shaft f r ic t io n because d riv in g a p ile into the ground causes an

increase in horizontal e ffe c t iv e stre sse s due to the compaction of

the ground. T h is , however, is not the case when a bored p ile is

in s t a lle d .

During the c a s t - in s it u bored p ile in s t a lla t io n , the process of

boring a shaft in c la y causes s tre s s r e l i e f and consequent y ie ld in g

around the hole. The undisturbed horizontal e ffe c t iv e stresse s

decrease as a r e s u lt of s tre s s r e l i e f and y ie ld in g . This can

c le a r ly be seen from the r e s u lt s of the analyses which were ca rried

out in th is study. E s p e c ia lly in the a n a ly sis where excavation was

ca rrie d out without any support, the decrease in horizontal

e ffe c t iv e stre sse s reached more than 50% of the i n i t i a l values (Fig

5 .14 , Fig A2.14, Fig A2.15, Fig A2.16 and Fig A 2 .1 7). The use of

'b ento nite as support had a p o s it iv e e ffe ct on reducing the stre ss

r e l i e f but did not prevent i t com pletely (Fig 5 .1 5 , Fig A2.26, Fig

A2.27, Fig A2.28 and Fig A 2.29). A ll analyses showed a zone of

p la s t ic y ie ld in g accompanying the stre ss r e l i e f (Fig 5 .32 , Fig

5 .3 3 ).

Another e ffe c t of s tre ss r e l i e f on the ground conditions is to

generate negative excess pore pressures around the hole causing the

groundwater to flow towards the depressed area. In the a n a lysis

P-NS/ND/12 where the hole was unsupported, excess pore pressures

reached a value as low as -190.0 kN/m2 at one point (Fig 5.26, Fig

A5.14, Fig A5.15 and Fig A 5.16 ). When bentonite s lu r r y was used as

support, because of it s a b i l it y to form an impermeable f i l t e r cake

on the sides of the hole, higher excess pore pressures were observed

than in a n a ly sis P-NS/ND/12 where no support was used. Although i t

was not p o ssib le to detect the changes in moisture content because

of the s o il model used, i t is obvious that an increase in moisture

content and softening w ill take place around the hole due to stre ss

r e l i e f coupled with groundwater flow towards that area.

The same e ffe c ts were observed in the diaphragm w all a n a ly sis

(D-BS/ND/12). The affected zone was larg er than that in the

c a s t - in s it u bored p ile analyses (P-NS/ND/12, P-BS/ND/12 etc. ) . The

e ffe c ts of lo ca l changes in boundary conditions did not remain

lo c a l. This was not unexpected because of the nature of the

geometry of the problem i . e . plane s t r a in .

I f there is a delay in co n cretin g , the e ffe c ts of excavation might

be expected to be more s ig n if ic a n t , due to the increased time

a v a ila b le for the s o il to s w e ll. Four analyses were ca rrie d out to

examine the e ffe c ts of the delay in concreting on the f in a l level of

horizontal e ffe c t iv e s tre s s (P-NS/1D/12, P-NS/7D/12, P-BS/1D/12,

P-BS/7D /12). According to the r e s u lt s of analyses P-BS/1D/12 and

P-BS/7D/12 where bentonite was used as support, the e ffe c ts of 1 day

delay in concreting were not very s ig n if ic a n t , w hile 7 days delay

caused a decrease in 8 values (Fig 5.40, Fig 5 .4 1 ). Delay in

concreting did not have any e ffe c ts on the 8 values obtained from

analyses P-NS/1D/12 and P-NS/7D/12 where no support was used (Fig

5 .38 , Fig 5 .3 9 ).

124

This unexpected behaviour can be explained by the stre ss states at

the end of the excavation stage and during the concreting process in

the analyses where the hole was excavated without any means of

support (P-NS/ND/12, P-NS/7D/12). I f the horizontal to ta l and

e ffe c t iv e stre sse s and pore pressures in an element next to the p ile

(e.g . element 28 at a depth of 11.25m) are examined c lo s e ly during

the excavation and concreting stages (Fig 6 .8 , Fig 6 .9 , Fig 6.10 and

Fig 6 .1 1 ), a drop in h o rizo n tal e ffe c t iv e stre sse s and a r is e in

pore pressures around the hole can be observed as a r e s u lt of the

delay in co n creting . When the wet concrete was placed in the hole,

i t did not apply any h orizontal e ffe c t iv e stre sse s to the sides of

the hole because i t was assumed to behave lik e a heavy f lu id .

Therefore the wet concrete caused the horizontal e ffe c t iv e stre sse s

around the hole to drop. In the case of the unsupported hole, with

no delay in co ncretin g, the placement of concrete led to a reduction

in horizontal e ffe c t iv e s tre s s e s . These stre sse s f e l l approxim ately

to zero before the concrete se t. In the case of the unsupported

hole with a delay in co n cretin g , however, the horizontal e ffe c t iv e

stre sse s had already fa lle n to about zero by the time the concrete

was placed and concreting then had l i t t l e e ffe c t . Therefore, at the

end of the concreting process the horizontal e ffe c t iv e stre sse s in

a n a lysis P-NS/7D/12 reached the same leve l as in a n a ly sis P-NS/ND/12

and more or le ss the same 3 values., were, obtained from these , two

analyses. Delay in concreting seems to have no e ffe c t on the f in a l

horizontal e ffe c t iv e stre ss state in the case of an unsupported

c a s t - in s it u bored p ile in s t a lla t io n for the p a rt ic u la r geometry,

s o il conditions and co nstructio n sequence considered in t h is study.

I t would be wrong to conclude that bentonite support had an adverse

e ffe c t on the recovery of the horizontal e ffe c t iv e stre sse s when

there was a delay in co n cretin g . In fa c t , the ra te s of the pore

pressure increase and the horizontal e ffe c t iv e stre ss decrease due

to the delay in concreting in a n a lysis P-BS/7D/12 where bentonite

s lu r r y was used as support are lower than in a n a ly sis P-NS/7D/12

where no support was used (Fig 6 .9 , Fig 6 .1 1 ) . T h is shows the

a b il it y of bentonite s lu r r y to reduce the e ffe c ts of s tre s s r e l i e f

by applying h yd ro static pressures and forming an impermeable f i l t e r

cake on the sides of the hole.

125

PORE

PRES

CkPa

) HOR

EFF

STR

S (k

Pa)

HOR

TOT

STRS

CkPa

)

oou_

“1 1 1 1 1 11 10 100 1000 10000 100000

TIME (Hours)Fig 6 . 8 H o r izo n ta l t o t a l s t r e s s e s , h o r i z o n t a l e f f e c t i v e s t r e s s e s

and pore p r e s s u r e s in e lemen t 28 during th e i n s t a l l a t i o np r o ce s s in a n a l y s i s P-NS/ND/12

126

PORE

PR

ES

(kPa

) HO

R EF

F ST

RS

CkPa

) HO

R TO

T ST

RS

(kP

a)

300_

210.

180_

120_

60_

0

-60

1

7 days delay

-0

0

0

□□

-+

O . Excavation Q. Concreting □ . C onsolidation

i — 10 100 1000

10 10 1000

10000 100000

10000 100000

TIME (H ours)

F ig 6 . 9 H o r izo n ta l t o t a l s t r e s s e s , h o r i z o n t a l e f f e c t i v e s t r e s s e s andpore p r e s s u r e s in e l em en t 28 during the i n s t a l l a t i o n p r o c e s sin a n a l y s i s P-NS/7D/12

127

PORE

PR

ES

(kP

a)

HOR

EFF

STRS

Ck

Pa)

HOR

TOT

STRS

C

kPa)

360„

300

210

180

120_

60-

0

-60

360.

300.

210.

180.

120.

60.

0.

-60.

1

360

300-

210

180_

120_

60_

0_

-60

1

— j—

10

“ i—

10\

100 1000

1 100 1000

10000 100000

TIflE (H ours)

10000 100000

TIME (H o u rs)

10000 100000

r . c 1n u . . , TIflE (H ours)ig 6.10 H orizontal to ta l s tre s s e s , h o rizo n tal e ffe c t iv e stre sse s

and pore pressures in element 28 during the in s t a lla t io n process in a n a ly sis P-BS/ND/12

128

PORE

PR

ES

(kPa

) HO

R EF

F ST

RS

(kP

a)

HOR

TOT

STRS

(k

Pa)

360

300_

210_

180

120

60J

-60.

360

300_

2T0_

18:

120_

60_

0.

-60.

~t— 10 100 1000

—i— 10 1000

10000 100000

TINE (H ours)

10000 100000

TINE (H ours)

Fig 6 .1 1 H o r iz o n ta l t o t a l s t r e s s e s , h o r i z o n t a l e f f e c t i v e s t r e s s e sand pore p r e s s u r e s in e l em ent 28 during th e i n s t a l l a t i o np r o c e s s in a n a l y s i s P-BS/7D/12

129

When concrete is placed into the hole water m igration takes place

from wet concrete towards surrounding s o il since the water pressure

is higher in concrete than that in the surrounding s o i l . H orizontal

e ffe c t iv e stre sse s drop to zero at the sides of the hole due to the

fa c t that concrete behaves lik e a heavy f lu id when it is wet. The

water which migrates from the concrete causes an increase in the

water content of the surrounding s o i l . According to the data

a v a ila b le in the lit e r a t u r e m oisture content increases vary from 1%

to 7 - 9% and a 2 - 8cm zone from the s o il concrete in te rfa ce is

affected (Mayerhof and Murdock, 1953; Rodin and Thomson, 1953;

Burl and, 1963; O'Neil and Reese, 1972; Chandler, 1977; Fearenside

and Cooke, 1978; M ilit it s k y , 1983). Young (1979) ca rrie d out small

sca le experiments on model bored p ile s (25, 28 and 50mm diameter)

in s ta lle d (using micro concrete) in a n is o tro p ic a lly consolidated

k a o lin . He found an increase in m oisture content around the p ile

during concreting. The higher the i n i t i a l moisture content, the

lower was the average increase in moisture content according to the

r e s u lt s of that study. He examined the v a ria tio n of the moisture

content at d iffe re n t p o sitio n s from the pi 1 e - s o il in te rfa ce during

the curing of concrete. A fter 60 days the moisture contents were

found s l ig h t ly below the o r ig in a l values though the small scale

experim ent's a b i l it y of sim ulating the long term e ffe cts

r e a l i s t i c a l l y may be questionable due to the sca le e f f e c t s . ................

M ilit it s k y (1983) examined the in te ra c tio n between fresh sand/cement

mortar and remoulded London Clay using a simple laboratory te st

arrangement. He looked into the e ffe c ts of water/cement r a t io of

mortar on the m oisture content changes in the s o i l . The moisture

content of the s o il increases as the water/cement r a t io of mortar

in cre ase s.

In the analyses, high excess pore pressures were generated around

the hole during the concreting process. T h is in d ica te s a water flow

from the concrete to the surrounding s o il and a consequent r is e in

the water content of s o i l . I f the water pressure head, perm eability

of s o il and curing time of concrete are taking into account, the

affected zone is expected to be 2 - 3cm th ick around the hole. The

130

s o il model used in t h is study is not capable of detecting any

moisture content changes, and e ffe c ts of the m oisture content

changes on the strength parameters of s o i l . Moreover, the f in it e

element mesh used in the analyses was not f in e enough around the

hole/trench for the clo se examination of water flow and it s e ffe c ts

on the stre ss and pore pressure regime. P o ssib ly , the C r it ic a l

State Model might be employed to examine these m oisture content

changes though it is not a s u ita b le model for overconsolidated

cl a y s.

Fig 6.12 shows the stre ss paths experienced by an element next to

the hole/tren ch (element 28 at the depth of 11.25m) throughout the

in s t a lla t io n process in analyses D-BS/ND/12, P-NS/ND/12 and

P-BS/ND/12. S tress paths l ik e the ones during the concreting

process are co n siste n t with some softening and sw e llin g .

A ll research ers seem to agree on the e ffe c ts of the in s t a lla t io n

process on the surrounding s o il so fa r described. There are,

however, d iffe re n t views about the re-establishm ent of horizontal

e ffe c t iv e stre sse s a ce rta in period a fte r concrete se t. Yong (1979)

and Anderson et al (1984) have concluded from the r e s u lt s of small

sca le lab oratory model p ile experiments that the ho rizo n tal

e ffe c t iv e stre sse s which are reduced due to boring are almost f u l ly

recovered with time. On the other hand, M ilit it s k y (1983) c a rrie d

out h a lf scale model p ile experiments in the London Clay and found

that the horizontal to ta l stre sse s around c a s t - in s it u bored p ile s

did not increase g re a tly above the le v e l produced by the h y d ro static

pressure of the concrete.

R esults of t h is study confirm Mil i t i t s k y 1s f in d in g s . In a l l

analyses which were c a rrie d out in th is study o nly p a rt ia l

re c o v e rie s in horizontal e ffe c t iv e stre sse s with time were observed

and horizontal to ta l stre sse s » 3.2 years after the concrete set

were found to be s l ig h t ly lower than the h y d ro static pressure of wet

concrete (Fig 6 .1 3 ) .

131

(0 + (p /2 (kPa)

Fig 6.12 Stress paths in element 28 during the in s t a lla t io n process in a n a ly sis a) D-BS/ND/12, b) P-NS/ND/12, c) P-BS/ND/12

132

cJCU

CQ CU I—1 CQ

I— ■ O

CUozsz

( W ) q ^ d a a

133

Fig

6.13

H

oriz

onta

l to

tal

stre

sses

at

the

elem

ent

cent

roid

s ne

xt

to

the

pile

/dia

phra

gm

wal

l *

3.2

year

s af

ter

the

conc

rete

se

t

The f in a l horizontal e ffe c t iv e stre ss regimes are presented in terms

of 8 values at the p i le - s o i l in te rfa ce for convenience. The values

at the p i le - s o il in te rfa c e , which were extrapolated from the

in te g ratio n points in each f in it e element adjacent to the p i le , were

disregarded for the reasons explained in Section 6 .1 .2 . The values

at these elements centroid s were considered instead.

8 values were ca lcu late d at d iffe re n t depths for each a n a ly s is .

Looking at the average values of 8 and th e ir v a ria tio n s with depth

the follow ing points can be concluded (Fig 5 .1 3 , Fig 5 .36 , Fig 5 .3 7 ,

Fig 5 .38 , Fig 5 .39 , Fig 5 .40, Fig 5 .4 1 , Fig 5.42, Fig 5.43 and Table

6 .1 ) :

In a ll cases apart from the diaphragm wall 8 values increase with

depth. The use of bentonite during the in s t a lla t io n has a p o s it iv e

e ffe c t on the recovery of horizontal e ffe c t iv e stre s s e s . Delay in

concreting causes some decrease in 8 values in the analyses where

bentonite s lu r r y used as support but i t has no e ffe c ts on 8 values

in the analyses where no support is used.

When it was assumed that the concrete set immediately a fte r the

completion of concreting., there was not enough time for high

p o s it iv e excess pore pressures to be generated. Since any change in

the horizontal e ffe c t iv e stre sse s can only be caused by the

d is s ip a t io n of excess pore pressures in the absence of external

load, a small recovery in horizontal e ffe c t iv e stre sse s was

observed.

I t should be noted that most of the p a rt ia l recovery of horizontal

e ffe c t iv e stre sse s took place w ithin 5 to 15 days after the concrete

had se t, though in p ra c tic e i t has been observed that increase in

shaft f r ic t io n goes on for up to 50 weeks a fte r in s t a lla t io n

(T a y lo r, 1966; Whitaker and Cooke, 1966; Combarieu, 1975).

According to the r e s u lt s of th is study for the c a s t - in s it u bored

p ile s 8 can be taken as 0.20 - 0.25 i f no support used or 0.35 -

0.40 i f bentonite s lu r r y is used as support during the

in s t a lla t io n . These values of 8 are lower than that obtained from

134

Analysi s Aver age 8

V a ria tio n of 8 with depth (H)

D-BS/ND/12 0.59 8 = 0.751 - 0.0160H

P-NS/ND/12 0.24 8 = 0.141 + 0.0160H

P-BS/ND/12 0.40 8 = 0.385 + 0.0169H

P-NS/1D/12 0.24 8 = 0.151 + 0.0090H

P-NS/7D/12 0.24 8 = 0.172 + 0.0069H

P-BS/1D/12 0.39 8 = 0.365 + 0.0028H

P-BS/7D/12 0.35 8 = 0.307 + 0.0046H

P-NS/ND/0 0.12 8 - -0.031 + 0.0155H

P-BS/ND/0 0.32; 8 = 0.242 + 0.0078H

Table 6.1 Average values of 8 in d iffe re n t analyses

the f ie ld observations made by various research ers (Skempton, 1969;

Whitaker and Cooke, 1966; Bur land et a l, 1966; Meyerhof, 1976) and

can be regarded as lower lim it s for the c a s t - in s it u bored p ile s

in s ta lle d in an overconsolidated c la y deposit under the co nd itions

considered in t h is study.

The r e s u lt s of the diaphragm w all a n a ly sis show that the i n it ia l

stre sse s around the diaphragm w all do change because of the

co nstruction process and they do not recover completely even after

excess pore pressures have f u l ly d is s ip a te d . Although t h is r e s u lt

is in contrast with the assumption, which has been made by various

research ers (Potts and Burl and, (1983a); Hubbard et a l, (1984);

Potts and F o u rie, (19 8 4 )), of taking in - s it u stre sse s as real

stre sse s acting on the diaphragm wall before the fro n t excavation,

f ie ld observations done by Tedd et al (1984) support th is r e s u lt .

Tedd et al (1984) observed a 20 - 30% decrease in in - s it u horizon tal

to ta l stre sse s near a secant p ile wall before the fro n t excavation.

The same reduction has been more or le ss found in th is study (Fig

A 3.12, Fig A 3.13 ). Therefore it is suggested that the f in a l stre ss

regime should be considered for estim ating wall movements and

bending moments, so that economy and sa fte y in design can be

achieved.

135

7 . Concl us ions

The in s t a lla t io n process of the c a s t - in s it u bored p ile s and

diaphragm w alls is known to a ffe ct the ground conditions such as

stre ss sta te , pore pressure sta te , moisture content etc. and s o il

p ro p erties in the v ic in it y .

In th is study, changes in stre ss and pore pressure regime during

construction i . e . short term e ffe c t s , and during the d is s ip a tio n of

excess pore pressures i . e . long term e ffe c ts were examined by

carryin g out a number of f in it e element analyses.

The main feature of the f in it e element a n a ly sis was the use of

coupled co n so lid atio n theory so that the generation of excess pore

pressures at any stage of the in s t a lla t io n process could be

observed.

A l in e a r - e la s t ic p e rfe c tly p la s t ic s o il model was employed. The

Mohr-Coulomb y ie ld c r it e r io n with an associated flow ru le was used.

A s o il deposit with p ro p ertie s s im ila r to London Clay was considered

and the r e s u lt s are, th e re fo re , s t r i c t l y v a lid for th is s o il type

o n ly, but may be expected to be re p re se n ta tiv e of s t i f f

overconsolidated cla ys in g e n e ra l...

Two d iffe re n t types of co nstructio n technique regarding the

supporting methods used were considered in the in s t a lla t io n of both

c a s t - in s it u bored p ile s and diaphragm w a lls : 1. No support used,

2. Bentonite s lu r r y used as support. The e ffe c ts of delay in

concreting and d iffe re n t se ttin g time of concrete were also examined

in the case of c a s t - in s it u bored p ile s .

The in s t a lla t io n process was simulated by removing elements and

applying appropriate boundary conditions to the f in i t e element

mesh.

The follow ing co nclusions can be drawn from th is purely numerical

study:

- During th e e x c a v a t i o n s t a g e h o r i z o n t a l t o t a l and e f f e c t i v e

136

stre sse s drop, and negative excess pore pressures generate

around the hole/tren ch due to stre ss r e l i e f and p la s t ic

y ie ld in g .

Water m igration from the surrounding s o il to the hole/trench may

be expected because of the pressure head d iffe re n ce between the

regions near the hole/trench and the le ss affected zones which

is produced as the excavation progresses.

The use of bentonite reduces the above mentioned e ffe c ts of

excavation but i t does not prevent them com pletely.

Excavating a hole/trench in the ground causes p la s t ic zones to

develop. Due to the extensive p la s t ic zones, an unsupported

trench excavation proves to be im possible under the ground

conditions described in Chapter 5.

Delay in concreting makes the excavation e ffe c ts more

s ig n if ic a n t .

During the concreting process pore pressures r is e in the regions

next to the h o le /tre n ch . Water m igration from the wet concrete

to the surrounding s o il takes p lace. Consequent softening and

sw elling in the s o il adjacent to the hole/trench should be

expected.

H orizontal e ffe c t iv e stre sse s around the hole/trench drop to

very low valu e s, near zero, before the concrete se ts.

A fter concrete se t, the ho rizo n tal e ffe c t iv e stre sse s p a r t ia lly

recover with time as the excess pore pressures d is s ip a te .

The values of h o rizo n tal e ffe c t iv e stre sse s around a c a s t - in s it u

bored p ile may be expected to be 30 - 50% of the in - s it u values

after excess pore pressures have d iss ip a te d .

The use of bentonite s lu r r y has a p o s itiv e e ffe c t on the

recovery of horizontal e ffe c t iv e s tre s s .

According to t h is a n a ly sis 8 = 0.40 and 8 = 0.25 can be regarded

as lower lim it s for the supported (with bentonite s lu r ry ) and

unsupported c a s t - in s it u bored p ile in s t a lla t io n re s p e c tiv e ly .

Most of the p a rt ia l recovery of horizontal e ffe c t iv e stresse s

takes place w ithin 5 to 15 days after the concrete has set.

The stre ss regime around a diaphragm wall before the fro n t

excavation is lower than the in s itu stre ss regime. Therefore

t h is stre ss regime should be considered for estim ating wall

movements and bending moments.

137

To the author's knowledge, th is is the f i r s t numerical study to

examine the in s t a lla t io n e ffe c ts of c a s t - in s it u bored p ile s and

diaphragm w alls using coupled co n so lid atio n theory. I t can be

regarded as a p re lim in a ry work in th is f ie ld .

This study endeavours to give a better in sig h t into what goes on in

the s o il around the c a s t - in s it u bored p ile s and diaphragm w alls

during the in s t a lla t io n process and after the d is s ip a t io n of excess

pore pressures.

The r e s u lt s prove that th is in s t a lla t io n causes a reduction in the

in s itu horizon tal e ffe c t iv e stre s se s. The magnitude of th is

reduction obtained from t h is study, however, may not be expected to

be re p re se n ta tive of the re a l s itu a tio n because of the assumptions

and s im p lif ic a t io n s involved in the analyses.

In order to obtain the p r a c t ic a lly a p p licab le red uction fa c to rs , i t

is necessary to c a rry out more elaborate numerical analyses. The

gradual settin g of concrete should be modelled as c lo s e ly as

p o ssib le since concreting has major e ffe c ts on the stre ss and pore

pressure regime around the h o le /tre n ch . The v a ria tio n s of Kq and

e la s t ic it y modulus with depth in London Clay ought to be taken into

account. A f in e r mesh may be used to detect the stre ss and pore

pressure changes, which are confined to a small area around the

hole, more a ccu ra te ly. Employing a s o il model which is capable of

sim ulating the re a l behaviour of the s o il can improve the

r e l i a b i l i t y of the r e s u lt s .

138

B i b l i o g r aphy

Anderson, W.F., Yong, K .Y ., Sulaiman, J.A . (1984); ‘ Laboratory

Testing of Bored and C a s t - in s itu Microconcrete P ile s in Clay to

Study Shaft Adhesion', Proc Jo in t I . Struct E ./B .R .E . Seminar on

Design of Concrete S tructu res - The Use of Model A n a ly sis , B uilding

Research Establishm ent, Garston, Watford.

Barlow, J. (1976); 'Optimal Stress Locations in F in it e Element

M odels', In t . J. Num. Meth. Engng, Vol 10, pp 243 - 252.

Bishop, A.W., Webb, D .L ., Lewin, P . I . (1965); 'Undisturbed Samples

of London Clay from the Ashford Common Shaft: S tre n g th -E ffe ctive

S tress R e la t io n s h ip ', Geotechnique, 15, No 1, pp 1 - 31.

Bishop, A.W. (1966); 'The Strength of S o ils as Engineering

M a te r ia ls ', S ixth Rankine Lecture, Geotechnique, 16, No 2,

pp 89 - 130.

B r it t o , A ., Gunn, M.J. (1986); 'C r it ic a l State S o il Mechanics v ia

F in ite Elem ents', (to be published by) E l l i s Horwood.

Burl and, J.B . (1963); D iscu ssio n , Proc Symposium on Grouts and

D r i l l in g Muds in Engineering P ra c t ic e , London, pp 223 - 225,

B utterw orths.

Burland, J .B . (19 73); 'Sh aft F r ic t io n of P ile s in Clay - a Simple

Fundamental Approach', Ground Engineering, Vol 6, No 3, pp 30 - 42.

Burland, J.B , (1978); 'A p p lic a tio n of the F in it e Element Method to

P re d ictio n of Ground Movements', Developments in S o il Mechanics - 1 ,

pp 69 - 101, Applied Science P u b lish e rs L td ., London.

Bustamante, M., Gouvenot, J.D . (1979); 'Facto rs Conditioning the

Actual Bearing Capacity of Bored P i l e s ', Proc 7th ICSMFE, Brighton,

Vol 2, pp 167 - 178.

139

Chadeisson, R. (1961); 'In flu e n c e of the Boring Methods on the

Behaviour of C a st-in -p la ce Bored P i l e s ' , Proc 5th ICSMFE, P a ris , Vol

2, pp 27 - 32.

Chandler, R .J. (1966); 'D is c u s s io n ', Large Bored P ile s , ICE, London,

pp95.

Chandler, R .J. (1968); 'The Shaft F r ic t io n of P ile s in Cohesive

S o ils in Terms of E ffe c tiv e S t r e s s ', C iv il Engineering and P ub lic

Works Review, Vol 63, pp 48 - 51.

Chandler, R .J . (1 9 77); D iscu ssio n , Abridged Proc of DOE and CIRIA

P ilin g Seminar, ICE, Nov 1976, CIRIA P u b lica tio n .

Chen, W.F. (1984); 'C o n s titu tiv e Modelling in S oil M echanics',

Mechanics of Engineering M a te ria ls , pp 91 - 120, John Wiley and

Sons.

C h ris t ia n , J .T . , D esai, C.S. (1 9 77); 'C o n s titu tiv e Laws for Geologic

M edia', Numerical Methods in Geotechnical Engineering, pp 65 - 115,

McGraw - H i l l , London.

Chuang, J.W ., Reese, L. C. (1969); 'Studies of Shearing Resistance

Between Cement Mortar and S o i l ’ , Research Report 89 - 3, Centre fo r

Highway Research, The U n iv e rs ity of Texas at A ustin.

Clayton, C .R . I . , M il it it s k y , J. (1983); ‘ In s t a lla t io n E ffe cts and

the Performance of Bored P ile s in S t i f f C la y ', Ground Engineering,

Vol 16, No 2, pp 17 - 22.

Combarieu, (1975); 'E s s a is de Chargement de Pieux Fores dans un

Limon A rg ile u x ', Bull L iaiso n des Ponts et Chausees, No 80, Nov -

Dec.

Cook, R.D. (1981); 'Concepts and A p p lica tio n s of F in ite Element

A n a ly s is ', 2nd Ed., W iley.

140

Corbett, B.O., D avies, R .V ., Langford, A.D. (1974); 'A Load Bearing

Diaphragm Wall at Kensington and Chelsea Town H a ll, London', Proc

Diaphragm Walls and Anchorages Conference, ICE, London, pp 57 - 62.

C u rt is , R .J. (1980); 'The In flu en ce of P ilin g Equipment and

In s t a lla t io n Methods in P ile Performance' Unpublished Report,

U n iv e rs ity of Surrey, C iv il Engineering Department.

D avis, E.H. (1968); 'Theories of P la s t ic it y and the F a ilu re of S o il

M asses', Soil Mechanics, pp 341 - 380, Ed. Lee, I . K . , Butterworth.

D iB iag io, E ., R o ti, J.A . (1972); 'Earth Pressure Measurements on a

Braced S lu rry Trench Wall in Soft C la y ', Proc 5th ECSMFE, Madrid,

Vol 1, pp 473 - 483.

Endo, M. (19 77); ‘ Relation Between Design and Construction in S o il

Engineering - Deep Foundations - Caisson and P ile Systems1, 'Proc 9th

ICSMFE, S p e c ia lity Session 3, Tokyo, pp 140 - 165.

Farmer, I.W ., Buckley, P .J .C ., S liw in s k i, Z. (1970); 'The E ffe ct of

Bentonite on the Skin F r ic t io n of C a st-in -p la c e P i l e s ' , Proc Conf

Behaviour of P ile s , ICE, London, pp 115 - 120.

Fearenside, 6 .R ., Cooke, R.W. (1978); 'The Skin F r ic t io n of Bored

P ile s Formed in Clay under B e n to n ite ', CIRIA Report No 77.

Federation of P ilin g S p e c ia lis ts (19 73); 'S p e c if ic a t io n for

C a st-in -p la ce Concrete Diaphragm W a llin g ', London.

Fernandez, R.L. (1965); W ritten C ontribution to Session 6, Div 4,

Deep Foundations, Proc 6th ICSMFE, Vol 3, pp 495 - 496.

Fleming, W.K., S liw in s k i, Z .J. (1 9 77); 'The Use and In flu en ce of

Bentonite in Bored P ile C o n stru c tio n ', CIRIA Report PG3.

Fuchsberger, M. (1974); 'Some P ra c tic a l Aspects of Diaphragm Wall

C o n stru c tio n ', Diaphragm W alls and Anchorages Conference, ICE,

London, pp 75 - 79.

141

Gunn, M.J. (1984); ‘The Deformation of Ground Near Tunnels and

T renches', Report, Cambridge U n iv e rs ity .

Gunn, M .J., B r it to , A. (1984); 'CRISP Users Manual', Cambridge

U n iv e rs ity .

H a jn a l, I . , Marton, J . , Regele, Z. (1984); 'C onstruction of

Diaphragm W a lls ', John Wiley and Sons.

H i l l , E.H. (19 71); 'The Mathematical Theory of P l a s t ic i t y ' , Oxford

U n iv e rs ity Press.

Hubbard, H.W., Po tts, D.M., M ille r , D., Burland, J.B . (1984);

'Design of the Retaining Walls for the M25 Cut and Cover Tunnel at

B e ll Common', Geotechnique, 34, No 4, pp 495 - 512.

Hutchinson, M.T., Dow, G .P ., Shotton, P .G ., Jack, B .J. (1974); 'The

P rop erties of Bentonite S lu r r ie s Used in Diaphragm W alling and T heir

C o n tro l', Diaphragm W alls and Anchorages Conference, ICE, London,

pp 33 - 39.

Komornik, A., Wiseman, G. (1967); 'Experience with Large Diameter

C a s t- in s itu P i l i n g ', 3rd Asian Regional Conf. SMFE, Vol 1, pp-200 -

204.

Meyerhof, G.G., Murdock, L .J . (1953); 'An In v e stig a tio n of the

Bearing Capacity of Some Bored and Driven P ile s in London C la y ',

Geotechnique, 3, No 7 , pp 267 - 282.

Meyerhof, G.G. (1976); 'Bearing Capacity and Settlement of P ile

Found ations', ASCE Jo ur. Geot. Eng. D iv ., 102, pp 197 - 228.

142

M ilit it s k y , J . , Jones, J .R ., Clayton, C .R .I. (1982); lA

Radiochemical Method of Studying the Moisture Movement Between Fresh

Concrete and C la y ', Geotechnique, 32, No 3, pp 271 - 275.

M ilit it s k y , J. (1983); ' In s t a lla t io n of Bored P ile s in S t i f f Clay:

an Experimantal Study of Local Changes in S oil C o n d itio n s ', PhD

T h e sis , U n iv e rsity of Surrey.

Mohan, D., Chandra, S. (1951); 'F r ic t io n a l R estistance of Bored

P ile s in Expansive C la y s ', Geotechnique, 11, No 4, pp 294 - 301.

Naylor, D .J. (1974); 'S tre sse s in Nearly Incom pressible M aterials by

F in ite Elements with A p p lica tio n to the C a lcu la tio n of Excess Pore

P re ssu re s', In t . Jour, for Num. Meth. in Engng, Vol 8, pp 443 - 460.

Naylor, D .J. (1978); 'S t r e s s -s t r a in Laws for S o i l ' , Developments in

S o il Mechanics - 1, pp 39 - 68, Applied Science P u b lish e rs L td .,

London.

Naylor, D .J ., Pande, G.N. (1981); 'F in it e Elements in Geotechnical

E n g in e e rin g ', Pineridge P ress, Swansea.

Owen, D ..R.J., Hinton, E. (1980); 'F in it e Elements in P la s t ic i t y ' ,

Pineridge Press, Swansea.

O 'N e il, M.W., Reese, L.C. (1972); 'Behaviour of Bored P ile s in

Beaumont C la y ', JSMFD, ASCE, Vol 98, SM2, pp 195 - 213.

P arry, R.H .G ., Swain, C.W. (1 9 77); 'A Study of Skin F r ic t io n on

P ile s in S t if f C la y ', Ground Engineering, Vol 5, No 5, pp 33 - 3 7.

Potts, D.M., Burland, J .B . (1983a); 'A Numerical In v e stig a tio n of

the Retaining Walls of the B ell Common T u n n e l', TRRL Supplementary

Report 783.

Potts, D.M., Burland, J.B . (1983b); 'A parametric Study of the

S t a b il it y of Embedded Earth Retaining S tru c tu re s ', TRRL

Supplementary Report 813.

143

Po tts, D.M., Fo u rie, A.B. (1984); 'The Behaviour of a Propped

Retaining W all: Results of a Numerical- Experim ent1, - Geotechnique,

34, No 3, pp 384 - 404.

S c h o fie ld , A.N., Wroth, C.P. (1968); 'C r it ic a l State S o il

M echanics', McGraw-Hill.

Skempton, A.W. (1959); 'C a s t - in s it u Bored P ile s in London C la y ',

Geotechnique, 19, No 4, pp 153 - 173.

Skempton, A.W. (1961); 'H orizontal Stresses in an O ver-consolidated

Eocene C la y ', Proc 5th ICSMFE, Vol 1, pp 351 - 357.

S liw in s k i, Z ., Fleming, W.G.K. (1974); 'P ra c t ic a l C onsiderations

A ffe ctin g the Construction of Diaphragm W a lls ', Diaphragm W alls and

Anchorages Conference, ICE, London, pp 1 - 10.

Sloan, S.W., Randolph, M.F. (1982); 'Numerical P re d ictio n of

C ollapse Loads Using F in ite Element Methods', In t. J. Num. Anal.

Meth. in Geomechanics.

Smith, I.M . (1982); 'Programming the F in ite Element method with

A p plicatio n to Geomechanics', John Wiley and Sons.

T a y lo r, P.T. (1966); 'Age E ffe cts on Shaft Resistance and E ffe ct of

Loading Rate on Load D is tr ib u tio n of Bored P i l e s ' , PhD T h e sis ,

U n iv e rs ity of S h e ffie ld .

Tedd, P ., Chard, B.M., C harles, J .A ., Symons, I . F . (1984);

'Behaviour of a Propped Embedded Retaining Wall in S t i f f Clay at

B e ll Common T u n n e l', Geotechnique, 34, No 4, pp 513 - 532.

Timeshenko, S ., Goodier, J.H . (1951); 'Theory of E l a s t i c i t y ' , 2nd

Ed., McGr aw H i l l .

U r ie l, S ., Otero, C.S. (1 9 77); 'S tre ss and S tra in Beside a C irc u la r

Trench W a ll', Proc 9th ICSMFE, Tokyo, Vol 1, pp 781 - 788.

144

Whitaker, T ., Cooke, R.W. (1966); 'An In v e stig a tio n of the Shaft and

Base Resistance of Large Bored P ile s in London C la y ', Proc Symp on

Large Bored P ile s , ICE, pp 7 - 49, London.

Wood, L .A ., P e rrin , A .J. (1984); 'O bservations of a Strutted

Diaphragm Wall in London C lay: a P re lim in ary Assessm ent',

Geotechnique, 34, No 4, pp 563 - 579.

Yong, K.Y. (1979); 'A Laboratory Study of the Shaft Resistance of

Bored P i l e s ', PhD T h e sis , U n iv e rs ity of S h e ffie ld .

Z ienkiew icz, O.C. (19 77); 'The F in it e Element Method in Engineering

S c ie n c e ', 3rd Ed, McGraw-Hill, London.

Z ienkiew icz, O .C., Humperson, C ., Lewis, R.W. (1975a); 'A U nified

Approach to S o il Mechanics In clu d in g P la s t ic it y and

V is c o - p la s t ic it y ', F in it e Element in Geomechanics, 1977,

pp 151 - 177, John Wiley and Sons.

Z ienkiew icz, O .C., Humperson,. C . , Lewis, R.W. (1975b); 'A ssociated

and Non-associated V is c o - p la s t ic it y and P la s t ic it y in S o il

M echanics', Geotechnique, 25, No 4, pp 671 - 689.

Z liw in s k i, Z .J . , P h ilp o t, T.A. (1980); ‘ Conditions for E ffe c tiv e End

Bearing of Bored, C a s t- in s itu P i l e s ' , Proc Conf Recent Developments

in the Design and Construction of P ile s , ICE, London, pp 57 - 64.

145

Appendices

A short review of the e ffe c t iv e stre s s method for estim ating the

sh aft f r ic t io n of p ile s is presented in Appendix 1.

Appendix 2, Appendix 3, Appendix 4 and Appendix 5 co n sist of

horizontal e ffe c t iv e s tr e s s , h o rizo n tal to ta l s tre s s , v e r t ic a l

e ffe c t iv e stre ss and excess pore pressure graphs re s p e c tiv e ly . The

graphs show the values of these v a ria b le s at the fo rty -e ig h t

d iffe re n t element cen troid s near the pile/diaphragm w a ll. The same

symbols represent the values at the same depth. Locations of

element centroids considered in the graphs are shown in Fig A .I.

The captions of the graphs in Appendix 2, Appendix 3, Appendix 4 and

Appendix 5 are abbreviated to avoid unnecessary r e p e t it io n s . The

fo llo w in g format is used throughout:

V a ria b le , A n a ly sis , Construction Stage

For example:

'Hor E ff S trs , P-BS/ND/12, 10.0m E xcavatio n1

means that the graph shows the ho rizo n tal e ffe c t iv e stre sse s at the

end of 10.0m excavation in c a s t - in s it u bored p ile co nstruction with

bentonite s lu r r y used as support (no delay in co ncreting , 12 hour

concrete se ttin g tim e).

146

2.5m

i

2.5

m t

2.5

m i

2.5m

,

2.5m

i

2.5m

,

2.5

m ,1

.25m

0 . 7 5 m 0 . 5 m 0 . 5 m 0 . 5 m 0 . 5 m 0 . 5 m 0 . 7 5 m

I

A A A A A A A

V V V V V 7 V

+ + + + + + +

X X X X X X X

a 0 D D 0 D □

0 0 0 0 0 0 0 1

0 0 0 0 0 0 0

$ % X $ & * &

Fig A .l Locations of the element cen troid s considered in Appendix 2, Appendix 3, Appendix 4 and Appendix 5

147

Appendix 1

The E ffe c tiv e S tress Method fo r Estim ating the Shaft F r ic t io n of

P ile s and 8 Values

The e ffe c t iv e stre ss method for estim ating the shaft f r ic t io n of

p ile s in c la y was proposed by Chandler (1966, 1968). Although th is

method has not been used in p ra c tic e , there has been an in creasing

in te re st in it ia t e d by Burland (19 77), Parry and Swain (1976, 1977a,

1977b) re c e n tly .

According to the e ffe c t iv e stre ss approach the shaft f r ic t io n is

given by:

t s = c ‘ + cty'tanS . . . . ( A l. l )

where !c ‘ is the e ffe c t iv e cohesion of c la y , is the horizontal

e ffe c t iv e stre ss acting on the p i le , 6 is the e ffe c t iv e angle of

f r ic t io n between c la y and the p ile sh a ft.

As a r e s u lt of remoulding during in s t a lla t io n the s o il has no

e ffe c t iv e cohesion. Therefore the equation A l . l becomes:

t s = 0 h 'ta n 6 . . . . (A l.2)

cr^1 can be expressed by

oh* = K.P (A l.3)

where K is the earth pressure c o e ff ic ie n t , p is the v e r t ic a l

e ffe c t iv e overburden p re ssure . Assuming proportional to the

v e r t ic a l e ffe c t iv e overburden pressure is questionable and re q u ire s

clo se examination and refinem ent (Burland, 19 77).

S u b stitu tin g equation A1.3 into equation A1.2, the f in a l form of

the expression for the shaft f r ic t io n is obtained:

ts = p .KtanS . . . . (A l . 4 )

148

I f the quantity Ktan6 is denoted by 8, then the equation A1.4

becomes:

xs = 8p ----- (A l.5)

8 looks s im ila r to the em pirical fa cto r a, but since 8 is re la te d to

the e ffe c t iv e stre s s parameters K and 5, i t is a fundamental

fa c to r.

In t h is study, 8 values were obtained from the follo w in g

re la t io n s h ip between horizon tal e ffe c t iv e stre sse s (crft), v e rt ic a l

e ffe c t iv e overburden pressures (p) e ffe c t iv e angle of f r ic t io n

between the c la y and the p ile shaft (5) with respect to 8 values:

<V8 = tan 6

p . . . . (A l.6)

6 was taken as 21.5° which is remoulded drained angle of f r ic t io n

for London Clay because it is common to assume that f a ilu r e takes

place in the remoulded s o il close to the shaft su rfa ce .

149

Hor

izon

tal

EPFe

ctiv

e St

ress

es

(kP

a)Appendix 2

H o r izo n ta l E f f e c t i v e S t r e s s Graphs

X10140_

35 J

30.

25.

20.

15..

10..

5..

'*1------ i--1----- 1------1----- 1---- 1------1----- f-----1-----1-----1-----!-----1 I IG 8 10 12 M 1G 18 20 22 21 2G 28 30 32 31 3S

X10-1

D istance From the p ile /w a ll a x is (m)

Fig A2.1 In - s it u horizontal e ffe c t iv e stre sse s

(r»dvj) sassajqs ^Aipajjg jioquozijoh151

Fig

A2.2

Ho

r Ef

f S

trs,

D

-BS/

ND

/12,

5.0

m

exca

vati

on

Fig

A2.3

Ho

r Ef

f S

trs,

D

-BS/

ND

/12,

10

.Om

exca

vati

on

( ridM) sa ssa x js 0An D0'± B 'iB ju o z ijoh152

Fig

A2.4

Ho

r Ef

f S

trs,

D

-BS

/ND

/12,

15.Om

ex

cava

tion

Fi

g A2

.5

Hor

Eff

Str

s,

D-B

S/N

D/1

2,

20.0

m

exca

vati

on

( 10+ 1) s a s s o j q s 9 A n : ) a dcl 3 t y q u o z u i o n

Fig

A2.6

Ho

r Ef

f S

trs,

D

-BS/

ND

/12,

5.0

m co

ncre

ting

Fi

g A2

.7 Ho

r Ef

f S

trs,

D

-BS/

ND

/12,

10

.0m

co

ncr

etin

g

( t ) ^ ) sa ssa jq s ^ A iq a jjd pquoziJOH154

Fig

A2.8

Ho

r Ef

f S

trs,

D

-BS/

ND

/12,

15

.0m

co

ncre

ting

Fi

g A2

.9

Hor

Eff

Str

s,

D-B

S/N

D/1

2,

20.0

m

con

cret

ing

(DdM) s a s s e j js aA !* P add3 ItijuoziJOH155

Fig

A2.1

0 Ho

r Ef

f S

trs,

D

-BS/

ND

/12,

12

hour

s af

ter

F19

A2-

n Ho

r Ef

f S

trs,

D-B

S/N

D/1

2,

5 da

ys

afte

r co

ncre

ting

co

ncre

te

set

(10+) sassejqs aAipajJ3 ppoziJoq156

^ O O CD

O <> n

h c:

o <>

Vi CU <i>

S3 CSD LO (S3 . LO CSDX M“ CO CD CM CM

( D J ? ) ) s a s s a j q s ^1300333 I b q u o z i j o j - |

COr-OO

ICSD'—■ £X v_x

LAXCSOi

CLQj

JC-+>

OL.o_a>uccS-OLA

« rH □

(odd) sossojqs aAipejJd pquoziJOH157

Fig

A2.1

4 Ho

r Ef

f S

trs,

P-

NS/

ND

/12,

5.0

m ex

cava

tion

F1

g AZ

.15

Hor

Eff

Str

s,

P-N

S/N

D/1

2,10

.0m

ex

cava

tio

n

1 10+ 1) s a s s a j q s a A i p a dd3 p u i o z p o H158

Fig

A2.1

6 Ho

r Ef

f S

trs,

P-

NS/

ND

/12,

15

.Om

exca

vati

on

Fig

A2.1

7 Ho

r Ef

f S

trs,

P-

NS/

ND

/12,

20

.0m

- ex

cava

tion

( D + D s a s s a j q s eAIP a dd3 l* nu021J0H159

Fig

A2.1

8 Ho

r Ef

f S

trs,

P-

NS/

ND

/12,

5.0

m co

ncre

ting

Fi

g A2

.19

Hor

Eff

Str

s,

P-N

S/N

D/1

2,

10.0

m

con

cret

ing

(tJcfti) sa ssa jq s ^ A ip a j j g pquoziJO H

i<s>><

( t )d> i ) 50SS0JJS a A ip 0 J d :3 P U IO Z IJ O H

160

Fig

A2.2

0 Ho

r Ef

f S

trs,

P

-NS

/ND

/12,

15.Om

co

ncre

ting

F

ig

A2.2

1 Ho

r Ef

f S

trs,

P-

NS/

ND

/12,

20

.0m

co

ncr

etin

g

(tu>j) 50sse*ns 0A i p a j j q ' 'iB'iuozuoH ' 161

Fiq

A2.2

2 Ho

r Ef

f S

trs,

P-

NS/

ND

/12,

12

hour

s af

ter

F’9

A2.2

3 Ho

r Ef

f S

trs,

P-

NS/

ND

/12,

5

days

af

ter

9 co

ncre

ting

co

ncre

te

set

</> +Js_ a)+-> t/ito cuif- +J4 - cu uj s_

ouoXLOCMCM<

(Dd>|) s a s s a jp i jg p q u o ziJo n

CO

xC5CU

Q.Qj

£0 S_Cj_01 Ucc5•+J-LAa

S-cu+j+-(t)to>>03■a

t o

CDz:COIQ_

tn +j5- a)+-> (Ato

cu +-> <+- cu UJ L_ o U c o o X u*3-CMCM<CCD

(Dd>l) sassajqs ^1330333 pquoziJQH

162

(D+j) sassejqs aArp0jjg pquozijon163

Fig

A2.2

6 Ho

r Ef

f S

trs,

P-

BS/N

D/1

2,

5.0m

ex

cava

tion

p

ig A

2.27

Ho

r Ef

f S

trs,

P-

BS/

ND

/12,

10.0

m

exca

vati

on

(Djvj) sessajqs 3ATP add3 pquozi joh164

Fig

A2.2

8 Ho

r Ef

f S

trs,

P-

BS/

ND

/12,

15

.Om

exca

vati

on

Fig

A2.2

9 Ho

r Ef

f S

trs,

P-

BS/

ND

/12,

20

.0m

ex

cava

tio

n

( T ) + ) sas sa jq s aA ! P add3 p q u o z i jo q165

Fig

A2.3

0 Ho

r Ef

f S

trs,

P-

BS/

ND

/12,

5.0

m co

ncre

ting

F

l9 A2

.31

Hor

Eff

Str

s,

P-BS

/ND

/12,

10

.0m

co

ncr

etin

g

(10+1) sa sso jq s 5ATP add3 p iu o zT jo |-|166

Fig

A2.3

2 Ho

r Ef

f S

trs,

P-

BS/

ND

/12,

15

.Om

conc

reti

ng

Fig

A2.3

3 Ho

r Ef

f S

trs,

P-

BS/N

D/1

2,

20.0m

co

ncr

etin

g

C DC O

M ~- C O 1

P MCSD

£- C O X

( S ) tA- c o . —>

C DXc s

- C M

C D

CU

- C M C L

M“ CU- CM j C

CM■+J

- C M £0CSD

uC l .

- C M

C OQ jU

- • . — C

C D

c t

LA

M -

• <—1Q

CM

CS)

CO

- C D

S- 4-) (Ot o

<1)4 - +J 4 - ai lu s_

o t_ c o o X uL OC O

CM<=cC O'f­lu

(t>dvj) sa ssa j^ s 0A i p 0d33 P 3 UOZIJOH

CD

LA

CU

C L

CU-C

£Oc_

C l .

a>ucc s

-*->IA

Q

t -cu4-3%toS-=3OXCM

aCOco

IX

toi-

4-> C O t o C

4- +J 4— a) tu t-

u s_ c o o

x u

C O

CM<

(iDd>i) sessaps 0A1300333 P3U0Z]JO[j167

cs>><

CS)f t i nco CS)cn inCM C S)CM

C QO')

h o TCM ® ■oo X

CS)• C O

C OCMCD•CMM~CMC MCMCS)CMC O

CQ

ft

CM

CS)

CO

r CQin cs>

inxCSQ j

ClQjJZft

oCi_

QjUCaf tin

□

in <s>

(D + ) sessajqs aA ip e j j 3 p p o z i j o n

C D

CU

CUJZ

£Oi-Cl_CUucC5-ftido

( d+ [ ) s a s s a j p 9 A i p a j j 3 p p o z i j o n

168

Fig

A2.3

6 Ho

r Ef

f S

trs,

P-

BS/

ND

/12,

15

days

af

ter

Fig

A2.3

7 Ho

r Ef

f S

trs,

P-

BS/

ND

/12,

12

00

days

af

ter

conc

rete

se

t -.

. •

conc

rete

se

t

Appendix 3

H o r iz o n t a l T o ta l S t r e s s Graphs

C!Q_

inminincu•ftCO

c5-fto

C JftcoNL.O

rr.

X10160.

55.

50.

15 j

10.

35.

30.

25.

20.

15

10

5 j

0.

-A

6 8 10 12 11 ie is 20 22 21 26 28 30 32 21 36X10-1

D istance From the p ile /w a ll a x is (m)

Fig A3.1 In - s it u horizontal to ta l stre sse s

169

<d+1) sassajqs PT°1 P juozijoh170

Fig

A3.2

Ho

r To

t S

trs,

D

-BS/

ND

/12,

5.0

m ex

cava

tion

Fi

g A3

.3

Hor

Tot

Str

s,

D-B

S/N

D/1

2,

10.0

m

exca

vati

on

( io+ l) s a s s a p s p q o j p q u o z i jo n

171

Fig

A3.4

Ho

r To

t S

trs,

D

-BS/

ND

/12,

15.0

m

exca

vati

on

Fig

A3.5

Ho

r To

t S

trs,

D

-BS/

ND

/12,

20

.0m

ex

cava

tio

n

Petti) sassejp V°T°1 VD1U02TJ0H172

Fig

A3.6

Ho

r To

t St

rs,

D-BS

/ND/

12,

5.0m

conc

reti

ng

Fig

A3.7

Ho

r To

t St

rs,

D-BS

/ND/

12,

10.0

m co

ncre

ting

(TD+i) sassajqs pquozi joh

173

Fig

A3.8

Ho

r To

t St

rs,

D-BS

/ND/

12,

15.0

m co

ncre

ting

Fi

g A3

.9

Hor

Tot

Strs

, D-

BS/N

D/12

, 20

.0m

conc

reti

ng

POD

(D p i) sessajqs p q o j p p o z u o n

( Ddvi) s a s s e j js W o j pTUozijo|( 175

Fig

A3.1

2 Ho

r To

t St

rs,

D-BS

/ND/

12,

15 da

ys

afte

r Fi

g A3

.13

Hor

Tot

Strs

, D-

BS/N

D/12

, 12

00

days

af

ter

conc

rete

se

t co

ncre

te

set

CD

UT • < V C5CU

C L

Cb

Oc_Cl.CUuc:o-ft UT

■. —( CQ

cro-ft03>03OX(V

Eo

CDzCDIO-UT$-

-ftoooi-o3C

cn<

(tjdM) sassaj^S W ° i p n u o z u o H176

(t>d>j) sessajqs p p i pquoziJOH

177

Fig

A3.1

6 Ho

r To

t St

rs,

P-N

S/N

D/1

2,15

.Om

exca

vati

on

• Fi

g A3

.17

Hor

Tot

Strs

, P-

NS/N

D/12

, 20

.0m

exca

vati

on

( ) Sassers TD3°J. pguozijoj-j178

Fig

A3.1

8 Ho

r To

t St

rs,

P-NS

/ND/

12,

5.0m

conc

reti

ng

. Fi

g A3

.19

Hor

Tot

Strs

, P-

NS/N

D/12

, 10

.0m

conc

reti

ng

CD

CS)

CS

Qj.— *• i— IC L

QjJ S

£O4_

Cj_.

QjUCcS-ftUIcn

(D+j) sa ssa jq s P P i p p o z i j o n

CD r CO

-CO 1CM CS

£-CO Xcs U!

-c oX

oo-CM

cs

CUCD *

-CM CLM- CU

- CM -C-ft

CM _-CM tr.

OCS

-CM cTQj

CO U_,— ftci

CD • ft

M“

UTcn

Xinin

CS)in

CM

cs>

.CDinM~ CS)M" inco (3)co in

CMCS)CM

in cs in cs in!

( D + ) sassajqs PPJ. pquozi joh179

Fig

A3.2

0 Ho

r To

t St

rs,

P-NS

/ND/

12,

15.0

m co

ncre

ting

fi

g A3

.21

Hor

Tot

Strs

, P-

NS/N

D/12

, 20

.0m

conc

reti

ng

(lo+i) s e s s a j p P P i piuoziJO|-j

CD

(a pi) sassajp p p j. pquoztuoH180

( Ddvi) se ssa jg s V0T°1 P iuoztjoh

CD-o>M-C)CM■OoCDO'DOD04CDCM

M~C M

CMC M

CSDCM

CO

CD

M-

04

CS

^00

CD

CS) £mxC5Cj

—>

CLCL»

JZ-i->£Oi-Cjl.CL)Uccr>-t-4LA

• r— (cn

(t-dvj) sassajig pioj puiozijoj-j

181

Fig

A3.2

4 Ho

r To

t St

rs,

P-NS

/ND/

12,

15 da

ys

afte

r '

Fig

A3.2

5 Ho

r To

t St

rs,

P-NS

/ND/

12,

1200

da

ys

afte

r co

ncre

te

set

conc

rete

se

t

(tj+D sassa 'jqs p i o | pquoziJO|-j

Fig

A3.2

6 Ho

r To

t St

rs,

P-BS

/ND/

12,

5.0m

exca

vati

on

Fig

A3.2

7 Ho

r To

t St

rs,

P-BS

/ND/

12,

10.Om

ex

cava

tion

<r>d>D s a s s e jp P P 1 piuozTJO jj

COr co

ho-)

<shcoool-Cs!

Ofthoft

( D + ) sessejqs p i o q p q u o z i joj-j

CDCD

M“n TCM ® C O •><

<33CD

COCM

CDC M

•ft■CM

CM-CM .

<33•CM

CO

C D

M~

CM

<33

- CO

;-CD

£UI

•r— IXo

(©+) sassjis P T °1 idtuozijoh185

Fig

A3.3

2 Ho

r To

t St

rs,

P-BS

/ND/

12,

15.Om

co

ncre

ting

Fi

g A3

.33

Hor

Tot

Strs

, P-

BS/N

D/12

, 20

.0m

conc

reti

ng

C D■ O')

M“•COCM•COCSDCDCOCM

CD-CMM-CM

CMCMCSDCM

CD

CD

CM

CSD

CSDX

CU

• —HCl

Qj-C

o

Cl.OjUCC5 -+-> Ul * ,—( a

CSDx

CSDCD LTDin CSDID LOM~ CSD LOCO

CSDCD

LOCM

CSDCM

LTD CD L O CS)t - cd

(iDcj>j) s a s s 0J 3s p g o j P 3UOZIJOH

Fig

A3.3

4 Ho

r To

t St

rs,

P-BS

/ND/

12,

12 ho

urs

afte

r A3

.35

Hor

Tot

Strs

, P-

BS/N

D/12

, 5

days

af

ter

conc

reti

ng

conc

rete

se

t

COr- CO

<53><CS3CO

LOLO (33IO inft- <33 in03 C3303 inCMi rC53 inCM —

ft-•03 £CM ® XC5303COCMCO•CMV1 OslCMCM(33CM

CO

xj—

CM

CO

-CO

(3 in rco

cuCL

Cjjz

ocuucC3

-ftin• «— IQ

O)ft<ftrcfinnsT3OOCM

COCOIC l .

oo ft S_ CU

f t in CO 03f t f t O CU

I— S-u s_ c o o nr uxCO

(S303<

(tidq) se ssa jq s P P 1 p iu o z ijo u

Appendix 4

Vertical Effective Stress Graphs

CL.

inCUIQinQj1--ftCOOj>

Cl .

c5U-ft

Qj

XI0120.

18_

16.

11.

12.

10..

3.

6-

4.

2

0. i r T 1-----1---- 1---- 1G 8 10 12 H 1G 18 20 22 21 26 28 30 32 31 3G

X I 0 '1

Distance Prom the pi le/wall axis (m)

Fig A4.1 In-situ vert ica l e f fec t iv e stresses

188

COr-OD

(Dd>|) S0SS0J3S 3AI330JJ3 P D T 3J0A

co

CDX

cujC

£odCi_

CUUCci-OLOdi

CO4->rd>n3OXCDEO

LO

CM

COCQIO(/)5-

+->CO

4 -4 —LUS~O)

CM

<+<c

(Dd>|) S0SS0J3S GAT3D0dJ3189

(iD+j) sassajp a A i p a+3 p a i p a ^

: 190 ,

Fig

A4.4

Ve

r Ef

f St

rs,

D-BS

/ND/

12,

15.0

m ex

cava

tion

Fi

g A4

.5

Ver

Eff

Strs

, D-

BS/N

D/12

, 20

.0m

exca

vati

on

(U+I) sassajqs aA! P aJJ3 p o t v e A

, 191

Fig

A4.6

Ve

r Ef

f St

rs,

D-BS

/ND/

12,

5.0m

conc

reti

ng

Fig

A4.7

Ve

r Ef

f St

rs,

D-BS

/ND/

12,

10.0

m co

ncre

ting

C D- C O

— CO t

CM- C O X

I0'

CSD- C O

C D “ CM

C D - CM

CMhoiCSD

- CM

CM

CSD

- CO

(DdVj) 50S50J3S S A n ^ J jg 1D3I3J0A192

Fig

A4.8

Ve

r Ef

f St

rs,

D-BS

/ND/

12,

15.0

m co

ncre

ting

Fi

g A4

.9

Ver

Eff

Strs

, D-

BS/N

D/12

, 20

.0m

conc

reti

ng

CD

CSX £■w

in♦Xcs

CUJZ

os..o_cuucd■ftm

o

(t>+) sassajp aA! P aJJ3 jDDtpa/y193

Fig

A4.1

0 Ve

r Ef

f St

rs,

D-BS

/ND/

12,

12 ho

urs

afte

r Fi

g A4

.ll

Ve

r^ff

^tr

s,

D-BS

/ND/

12,

5 da

ys

afte

r co

ncre

ting

(t)+D sassajjs aAipajjg p o T p e A194

ig A4

.12

Ver

Eff

Strs

, D-

BS/N

D/12

, 15

days

af

ter

Fig

A4>1

3 Ve

r Ef

f st

rS}

D-BS

/ND/

12,

1200

da

ys

afte

r co

ncre

te

set

conc

rete

se

t

(3X

CD

(3 CD +

CU

CLCU

JZ

OCl!CUuccsfttn

£3

CMCM

<3 CO CM

CM <3 CO CD <3

(fl+ l) s a s s jq s SA Tpajjg p :> ip aA

f

C3

X

C3X

in

cu• r— tCLCj

JZ

O(11cuucCJftcn

- <— iO

<3CM

CO CO M" CM (3 CO CD CM C3

(iD+i) sessejp aAtpajjg

195

Fig

A4.1

4 Ve

r Ef

f St

rs,

P-NS

/ND/

12,

5.0m

exca

vati

on

' Fi

g A4

.15

Ver

Eff

Strs

, P-

NS/N

D/12

, 10

.0m

exca

vati

on

( D+j) sassa jig ^at300333 p on ja / \196

Fig

A4.1

6 Ve

r Ef

f St

rs,

P-NS

/ND/

12,

15.0

m ex

cava

tion

Fi

g A4

.17

Ver

Eff

Strs

, P-

NS/N

D/12

, 20

.0m

exca

vati

on

( n+'l) s e s s a ^ s ^AiiD0dJ3 pD iiJaA

CD-CO

- C O

CSD I" COCO-CMCDCM

M-CMCMCM

CSDCM

M~-°°T

CM ® -CO ><

Fig

A4.1

8 Ve

r Ef

f St

rs,

P-NS

/ND/

12,

5.0m

conc

reti

ng

Fig

A4.1

9 Ve

r Ef

f St

rs,

P-NS

/ND/

12,

10.0

m co

ncre

ting

< 0 + ) sessejqs aA n aajj3

198

Fig

A4.2

0 Ve

r Ef

f St

rs,

P-NS

/ND/

12,

15.0

m co

ncre

ting

Fi

g A4

.21

Ver

Eff

Strs

, P-

NS/N

D/12

, 20

.0m

conc

reti

ng

to +->d O) +J oo toQJ

4 - +->QJ U d O d C

OJ o >■ o

( T0d>i) s a s s a j is 3 A ip a jj3 l ^ n j a A

( odtt) s e s s a j T S 0 A t p a j j 3 p m p a / \

199

CD-CDft

- CD

CM-CD

CSCO

(D+D s a s s a j p aA !P add3 P a H j a A

CD

in

Cj♦ rHCLQJ

J C

£O1_Cl_CL)Ucc5-ftft■»— iCD

( D + ) sassajqs aA i p add3 P a ! P aA200

Fiq

M

24 Ve

r Ef

f St

rs,

P-NS

/ND/

12,

15 da

ys

afte

r F

i9 A4

.25

Ver

Eff

Strs

, P-

NS/N

D/12

, 12

00

days

af

ter

conc

rete

se

t co

ncre

te

set

(0+ 1) sassajqs aAipejjg p:>!PaA201

Fig

A4.2

6 Ve

r Ef

f St

rs,

P-BS

/ND/

12-,

5.0m

exca

vati

on

Fig

A4,2

7 Ve

r Ef

f St

rs,

P-BS

/ND/

12,

10.0

m ex

cava

tion

(DcM) 50S50J3S aAi3Dajj3 p D ip aA

(Dpi) S a s s e r s e A tp a j ja l ^ H ^ A202

Fig

A4.2

8 Ve

r Ef

f St

rs,

P-BS

/ND/

12,

15.0

m ex

cava

tion

fig

A4

.29

Ver

Eff

Strs

, P-

BS/N

D/12

, 20

.0m

exca

vati

on

<33

CDOOft-03 £

Oft 2^ x<3303

COOftCDOftft"

■Oft

Oft■Oft<33■Cft

Oft

<33

CLCU J3 • ftaocuuccsftinCD

, , i i ......i------1----- 1— —i---- 1----- 1-----1-----1----- r..... r00 CO ft- Oft <53 CO CD ft- Oft <33 CO CD ft" Oft (S3

X Cft Oft Oft Oft Oft .—■ ■>— .— .— .—

( ) s e s s a j p eA JP add3 P P P ^ A

Fig

A4.3

0 Ve

r Ef

f St

rs,

P-BS

/ND/

12,

5.0m

conc

reti

ng

Fig

A4.3

1 Ve

r Ef

f St

rs,

P-BS

/ND/

12,

10.0

m co

ncre

ting

Fig

A4.3

2 Ve

r Ef

f St

rs,

P-BS

/ND/

12,

15.0

m co

ncre

ting

Fi

g A4

.33

Ver

Eff

Strs

, P-

BS/N

D/12

, 20

.0m

conc

reti

ng

( D + ) sassajp -̂ 1300333 p : > n jaA205

Fig

A4.3

4 Ve

r Ef

f St

rs,

P-BS

/ND/

12,

12 ho

urs

afte

r Fi

g *4

.35

Ver

Eff

Strs

, P-

BS/N

D/12

, 5

days

af

ter

conc

reti

ng

conc

rete

se

t

(!Dd>|) S0SS0J3S poipa/\206

Fig

A4.3

6 Ve

r Ef

f St

rs,

P-BS

/ND/

12,

15 da

ys

afte

r Fi

g A4

.37

Ver

Eff

Strs

,'

P-BS

/ND/

12,

1200

da

ys

afte

r co

ncre

te

set

conc

rete

se

t

*

Appendix 5

Excess Pore Pressure Graphs

ciCL.

UTOiC.3UTUlCUL.Q_Qj<O

CL.

XI01 20.

18.

.18.

14.

12.

10

—h

G ~8 10 12 11 1G 18 20 22 24 2G 28 30 32 31 3GXI 9_1

Distance From the pi le/wali axis (m)

Fig A5.1 In-situ pore pressures

207

(lOdH) sajnssejd a jo j ssaaxg

Fig

A5.2

Ex

s Po

re

Pres

', D-

BS/N

D/12

, 5.0

m ex

cava

tion

A5

.3

Exs

Pore

Pr

es,

D-BS

/ND/

12,

10.0

m ex

cava

tion

(10+) s a jn s s a j j a jo d ssaoxg

CD

csX

ft

cuJZ

£OS.

Cl .

CUUcdft

* r HCQ

ft

d5:CUsz-t—^eos_.

Cl.CUucdft

• i*HCQ

cnu_

(13+) sajnssajj ajoj ssaoxg

209

A5.4

Ex

s Po

re

Pres

, D-

BS/N

D/12

, 15

.0m

exca

vati

on

Fig

A5.5

Ex

s Po

re

Pres

, D-

BS/N

D/12

, 20

.0m

exca

vati

on

CSD

Xin

CL)JS£Oi_

Cj_CLiUCd-1-JIA

O

(DdH) sejnssa^d a j o d s s a a x g

210

Fig

A5.6

Ex

s Po

re

Pres

, D-

BS/N

D/12

, 5.0

m co

ncre

ting

Fi

g A5

.7

Exs

Pore

Pr

es,

D-BS

/ND/

12,

10.0

m co

ncre

ting

(Ddd) sajnssejd ajod ssaaxg

211

Fig

A5.8

Ex

s Po

re

Pres

, D-

BS/N

D/12

, 15

.0m

conc

reti

ng

Fig

A5.9

Ex

s Po

re

Pres

, D-

BS/N

D/12

, 20

.0m

conc

reti

ng

C N C S CNI ftI

CO • CO"ft"-coC N•cocs•coCO

■CN

CO■CN

ftC N

CNCN

C SCN

CO

CO

ft

C N

C S

OO

C S

X

COI COI CS CN Nt- CO CO CSCN

C N M— C D C N C N CN I I I

r co

in

CUjc

£oCl .

QjUCdftft* r— |cm

( io + ) se jn ssa j j a jo j ssaax^

- CO

CD■CO

ftCO

cs■coooCN

CD■CN

CSCN

-OO

( 10+ ) sa jnssa+ auoj ssaaxg212

Fig

A5.1

0 Ex

s Po

re

Pres

, D-

BS/N

D/12

, 12

hour

s af

ter

Fig

A5.l

l Ex

s Po

re

Pres

, D-

BS/N

D/12

, 5

days

af

ter

conc

reti

ng

conc

rete

se

t

to

oX

cs3:

Qj

SOS ,ftQjUCcs-t-Jtna

S-QJftft<tfin

<L5noOoCM

Oz:ooc aiQinO) ft L- <L> o_ in<u <d s- ft o aj ft t- u in c : x o

u j o

c o

LO<CD•n*

LU

(n+i) sa jnsse j j ajo,-) ssaaxg

( 'ocj>j) se jnssaJ+ aoioj ■ ssesxg

CD L O C D L O1 CD

Ii r

in cd— CMI IL OCMI

CSD00 inooi

(Dd>i) sajnssajj ajod sseaxg

214

CD-co

-oo 1CM-co

CSDX

CD-CDCD-CMCD-CMM- - CM04-04CD-CMCD- H—CD- r-M"- '—CM“CSD— V-

-CD

-CD•CDI

LA

O)■ r— LC L

CUJZ

aiucCDLA■ rHa

Fig

A5.1

4 Ex

s Po

re

Pres

, P-

NS/N

D/12

, 5.0

m ex

cava

tion

Fi

g A

5.i5

Exs

Pore

Pr

es,

P-NS

/ND/

12,

10.0

m ex

cava

tion

(D+l) s e jn s s a j j e jo j ssaoxg 215

Fig

A5.1

6 Ex

s Po

re

Pres

, P-

NS/N

D/12

, 15

.0m

exca

vati

on

Fig

A5.1

7 Ex

s Po

re

Pres

, P-

NS/N

D/12

, 20

.0m

exca

vati

on

(D+) sajnssaJd a jo j ssaaxg216

Fig

A5.1

8 Ex

s Po

re

Pres

, P-

NS/N

D/12

, 5.0

m co

ncre

ting

Fi

g A5

.19

Exs

Pore

Pr

es,

P-NS

/ND/

12,

10.0

m co

ncre

ting

(td+ ) sajnssajd ajoj ssaaxg

217

Fig

A5.2

0 Ex

s Po

re

Pres

, P-

NS/N

D/12

, 15

.0m

conc

reti

ng

Fig

A5.2

1 Ex

s Po

re

Pres

, P-

NS/N

D/12

, 20

.0m

conc

reti

ng

X10"

(Dd>i) sajnssajd ejod ssaaxg218

Fig

A5.2

2 Ex

s Po

re

Pres

, P-

NS/N

D/12

, 12

hour

s af

ter

Fig

A5.2

3 Ex

s Po

re

Pres

, P-

NS/N

D/12

, 5

days

af

ter

conc

reti

ng

conc

rete

se

t

f-CDCOI

ICEOX

( 10+ ) sa jnssa j^ a jo^ sseaxg

( D + ) sejnssajj ajoj ssaax^219

Fig

*5.2

4 Ex

s Po

re

Pres

, P-

NS/N

D/12

, 15

days

af

ter

Fig

A5.2

5 Ex

s Po

re

Pres

, P-

NS/N

D/12

, 12

00

days

af

ter

conc

rete

co

ncre

te

set

(D+l) sajnssajj ajoj ssaaxg

220

Fig

A5.2

6 Ex

s Po

re

Pres

, P-

BS/N

D/12

, 5.0

m ex

cava

tion

Fi

g A5

.27

Exs

pore

Pr

es,

P-BS

/ND/

12,

10.0

m ex

cava

tion

CD

CUCLCUJSfa£Od

Li-CUUCcsUIO

(iD̂ vi) saunssejj a jo j ssaax^

(Djvi) sa jn sse jj a jo j ssaax^221

Fig

A5.2

8 Ex

s Po

re

Pres

, P-

BS/N

D/12

, 15

.0m

exca

vati

on

pl9

A5.2

9 Ex

s Po

re

Pres

, P-

BS/N

D/12

, 20

.0m

exca

vati

on

(D+f) sajnssajd ajod ssaoxg222

Fig

A5.3

0 Ex

s Po

re

Pres

, P-

BS/N

D/12

, 5.0

m co

ncre

ting

Fi

g A5

.31

Exs

Pore

Pr

es,

P-BS

/ND/

12,

10.0

m co

ncre

ting

Fig

A5.3

2 Ex

s Po

re

Pres

, P-

BS/N

D/12

, 15

.0m

conc

reti

ng

Fig

A5.3

3 Ex

s Po

re

Pres

, P-

BS/N

D/12

, 20

.0m

conc

reti

ng

(Dd>|) sajnssajj ajoj ssaax^

224

Fig

A5.3

4 Ex

s Po

re

Pres

, P-

BS/N

D/12

, 12

hour

s af

ter

Fig

A5.3

5 Ex

s Po

re

Pres

, P-

BS/N

D/12

, 5

days

af

ter

conc

reti

ng

conc

rete

se

t

CO

CSD><

ID

CLQjJZ

Oi-o_cuuea

-ftin□

>>a)"OooCM

CM

COcoiC l

IDCU f tS— cu

CU CU L- ftO

C L

LU <J

XOOLO<C n

Ll

(0+1) se jnssa jg ajod ssaaxg