=*u&bwy - epubs.surrey.ac.ukepubs.surrey.ac.uk/844568/1/10149051.pdf · and diaphragm walls step by...
Transcript of =*u&bwy - epubs.surrey.ac.ukepubs.surrey.ac.uk/844568/1/10149051.pdf · and diaphragm walls step by...
=*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