-
EXPERIlMENTAL lNVESTXGATION OF TAPERED PILES
Jin Qi Wei
Faculty of Engineering Science Department of Civil
Engineering
Submitted in partial ful fillment of the requirement for the
degree of
Master of Engineering Science
Faculty of Graduate Studies The University of Western
Ontario
London, Ontario August, 1998
O Jin Qi Wei 1998
-
National Library Bbliithèque nationale du Canada
Acquisitions and Acquisitions et Bibliographie Services senrices
bibliographiques 395 Weüiington Street 395. rue W d l i OüawaON K 1
A W OttawaûN K1AON4 canada Canada
The author has granted a non- L'auteur a accordé une licence non
exclusive licence allowing the exclusive permettant à la National
Lïbcary of Canada to Bibliothèque nationale du Canada de reproduce,
loan, disîniiute or sel reproduire, prêter, distri'buer ou copies
of this thesis in microform, vendre des copies de cette thèse sous
paper or electronic formats. la forme de microfiche/fïlm, de
reproduction sur papier ou sur format électronique.
The author retains ownership of the L'auteur conserve la
propriété du copyright in this thesis. Neither the droit d'auteur
qui protège cette thèse. thesis nor substantial extracts fiom it Ni
la thèse ni des extraits substantiels may be pruited or othenirise
de celle-ci ne doivent être imprimés reproduced without the
author's ou autrement reproduits sans son permission.
autorisation.
-
ABSTRACT
Tapered piles, which have greater top cross-sections than bottom
ones, have not
ofien been considered as a design option because of the lack of
laiowledge about their
behaviour. In this study, the characteristics of tapered piles
performance were established
from experimental investigation. A relatively large laboxatory
facility for testing model
instrumented steel piles was developed. In this facility, the
sample soi1 was contained in
a steel chamber and pressurized to mode1 the laterai stress dong
the different "segments"
of the pile. The effects of the pile taper on its response to
axial compressive, tensile and
cyclic, and lateral loads were examined. The results of the
study confirnied their
efiiciency over piles of uniform section with the same materid
input in al1 loading modes
considered in this study. It was concluded that tapered piles
represent a more equitable
distribution of pile material in several respects. A procedure
was developed to calculate
the shaft resistance of tapered and straight-sided wall piles
based on the experimental
observations. The shafi resistance for straight-sided wall piles
estabiished 60rn the
experimental results compared well with the theoretical
prediaions using the standard
design procedure, hence connmiing the validity of the
experimental resdts.
KEYWORDS: Tapered piles, Experimental model testing, Axial
response, Uplift
loading, Lateral response, Cyclic loadhg, Load transfer, Modulus
of subgrade reaction.
-
I would like to express my sincere gratitude and appreciation to
my Supervisor,
Dr. M. H. El Naggar, for his guidance, encouragement and support
during the course of
study to this thesis.
Thanks are due to Mr. Wilbert Logan for his help in the setup of
the data
acquisition system for this research and to Mr. Gary Lusk for
his assistance as well. Ms.
Trudy Laidlaw designed the soi1 chamber, her help is greatly
appreciated.
Sincere thanks are extended to the facdty and staff of the
Department of Civil
Engineering and my fellow graduate students for their assistance
and companionship.
1 am most gratefùl to my husband and my children for their love,
understanding
and patience throughout the work.
This thesis is dedicated to my parents, Wang Lianying and Wei
Guangcai.
-
TABLE OF CONTENTS
Page
CERTIFICATE OF EXAMZNATION
ABSTRACT
ACKNO WLEDGMENTS
TABLE OF CONTENTS
LIST OF TABLES
LIST OF FIGURES
C W T E R I INTRODUCTION
1 . 1 OVERALL REVIEW
1.2 OBJECTIVES AND SCOPE
2. 1 PILE LOAD CAPACITY 2. 1. 1 Axial Bearing Capacity
2. 1 . 1 . 1 Static analysis method 2. 1. 1.2 Pile load testing
method 2. 1 . 1.3 Dynamic analysis method
2.1.2 Upiift Load Capacity 2. 1. 3 Lateral Load Capacity 2. 1.4
Effect of Cyclic Loadiog
2 . 2 RESPONSE ANALYSIS METHODS FOR SINGLE PILES 2.2. 1 Elastic
Analysis Method 2. 2. 2 Subgrade Reaction Method 2. 2. 3 Finite
Element Analysis Method
2 .3 RELEVANT STUDES ON TAPERED PILES 2.3. 1 Field Testing
Resuits 2 .3 .2 Laboratory Testing Observations 2. 3. 3 Theoretical
Analysis Results
-
Page
2. 4 MOTNATIONS
C . T E R 3 TESTING FACILITES AND PREPARATION
INTRODUCTION
TESTING FACILITES 3.2. 1 Testing Piles 3 .2 .2 Strain Gauge
Installation 3 .2 . 3 Soil Sample 3 . 2 . 4 Soil Chamber (VLPSC ) 3
. 2 . 5 Soil Pressure Transducers 3 .2 .6 ûther Test Equipment
TESTING PREPARATION
EFFECT OF PILE INSTALLATION METHOD
CHAPTER 4 AXIAL COMPRESSIVE RESPONSE OF TAPERED PILES
INTRODUCTION
TESTING PROCEDURE
TESTING RESULTS 4 .3 . 1 Load-Displacement and Bearing
Capacity
4 . 3 . 1. 1 First group of tests 4 .3 . 1.2 Second group of
tests
4 . 3 . 2 Load Distribution 4.3.3 Unit Load Transfer 4. 3 . 4
Pile Tip Resistance
DISCUSSION: ARCHING PHENOMENON
SUMMARY
CHAPTER 5 UPLIFT BEHAVIOUR OF TAPERED PILES
5 . 1 INTRODUCTION
5 .2 TESTING PROCEDURE
-
Page
5 . 3 TESTING RESULTS 5.3. 1 Uplift Load - Displacement S. 3.2
Ultimate Uplifi Load
5.3.2. 1 hosesandstatus 5 .3 .2 .2 Relatively medium dense sand
status 5.3.2.3 EEect of placement method
5 . 3 . 3 Pile Head Stiffiiess 5. 3.4 Load Distribution 5 . 3 .
5 Shafi Friction 5.3.6 Downward and Uplifi Shaft Friction
5- 4 DISCUSSION: RESIDUAL STRESSES
C W T E R 6 RESPONSE OF TAPERED PILES S W C T E D TO LATERAL
LOAD
6.1 INTRODUCTION
6 . 2 TESTING PROCEDURE
6 . 3 TESTING RESULTS 6.3 . 1 Load - Deflection 6 . 3 . 2
Ultimate Lateral Load 6 . 3 . 3 Bending Moment Distribution 6 . 3 .
4 Maximum Bending Moment 6. 3. 5 Soi1 Resistance 6. 3 .6 Pile
Deflection 6.3.7 py Curve
6 . 4 PREDICTED ULTIMATE LATERAL LOAD
6 .5 PREDICTED py CURVES 6.5. 1 Analfical Background 6 .5 .2
Observations
vii
-
Page CHAPIER 7 CYCLIC RESPONSE OF AXIALLY LOADED
TAPERED PILES
7. 1 INTRODUCTION
7.2 TESTING PROCEDLJFE
7 . 3 TESTING RESULTS 7.3. 1 Cyclic Load-Displacement
7.3. 1. 1 Zero confinuig pressure 7.3. 1.2 20 kPa connning
pressure 7. 3. 1.3 40 kPaconfiningpressure 7 .3 . 1.4 60 kPa
confining pressure
7 . 3 . 2 Pile Head Stiffhess 7. 3. 3 Effect of Cyclic Load
Amplitude 7 . 3 . 4 A c c d a t e d Pile Head Movement 7 . 3 . 5
Rate of Pile Movernent
CItQPTER 8 VALIDITY AND APPLICATION OF THE STUDY
8 . 1 INTERPRETATION OF TESTING RESULTS
8 . 2 VALDITY OF MODEL TESTING RESULTS
8 . 3 SHAFT RESISTANCE OF TAPERED PILES 8.3. 1 Relatively Medium
Dense Sand Case 8 . 3 . 2 Loose Sand Case
8 . 4 DISCUSSION 8.4. 1 The State of Stress Inside of the Soi1
Charnber 8 . 4 . 2 Boundary Effect
C W T E R 9 CONCLUSIONS AND RECOMMENDATIONS
viii
-
Table
LIST OF TABLES
Description
Geometries of three piles
Sand properties
Test arrangement
Axial compression resdts (first group of tests)
Axial compression resdts (second group of tests)
Pile tip resistance at Qu (second group of tests)
Uplifi results (fbt group of tests)
Uplifi resuits (second group of tests)
Lateral loading resdts
Variation of K with deflection and relative density
(Geosofi)
Amplitude of cyclic load applied at pile head
Pile head stifkess
Page
-
Figure
LIST OF FIGURES
Description
Geometries of three piles
Strain gauge circuits a: Axial loading test b: Lateral loading
test
Strain gauge installation
Grain size distribution
Variable lateral pressure sand column (VLPSC)
Current - pressure relationships for soi1 transducers Applied
confining pressure vs meanwd stress around pile S
Axial loading equipment
Uplift loading equipment
Laterd loading equipment
Oblique view of the testing facility
Load-settlement curves of pile Tl with different confîning
pressures ( k t group of tests)
Load-setdement curves of pile T ï with different confining
pressures (fust group of tests)
Construction of the offset limit load (after Canadian Foundation
Engineering Manual, 1992 )
Load-settlement curves of pile T 1 with different confuiing
pressures (second group of tests)
Load-settlement curves of pile S with different confining
pressures (second group of tests)
Load distribution dong the pile with different values of load
applied at pile heads of Tl and T2 (first group of tests)
Load distribution d m g the pile with different values of load
applied at pile heads of Tl and S (second group of tests)
Page
-
Figure Description
Unit load transfer to the soil when ultimate load was reached at
piles Tl and T2 (fint group of tests)
Unit load t d e r to the soil when uitirnate load was reached at
piles Tl and S (second group of tests)
Variation of unit load transfer to the soi1 for piles Tl and T2
at different confining pressures applied (fim group of tests)
Variation of unit load transfer to the soil for piles Tl and S
at different confinhg pressures applied (second group of tests)
The distribution of the ultimate load between the pile point and
the pile shaft for various applied confining pressures
Load- upward movement curves of piles at different confinùig
pressure values in fîrst group of tests a: Tl b: T2
Load- upward movement curves of piles at different connnllig
pressure values in second group of tests a: Tl b: S
The effect of confining pressure on the uplift pile head
stiflhess a: First group of tests b: Second group of tests
5-4 (a) Load distribution along the pile at different load
increments applied at pile head of T 1 in first group of tests
5-4 (b) Load distribution along the pile at different load
increments applied at pile head of T2 in first group of tests
5-5 (a) Load distribution dong the pile at different load
increments applied at pile head of Tl in second group of tests
5-5 (b) Load distribution along the pile at different load
increments applied at pile head of S in second group of tests
5-6 Shaft friction for piles Tl and T2 at dBerent connning
pressure values in first group of tests
5-7 Shaft friction for piles Tl and S at different confining
pressure values in second group of tests
Page
45
48
50
-
Figure Description
5-8 Variation of shaft fiction at different confining pressure
values in f is t group of tests a: Tl b: T2
5-9 Variation of shafi fiction at Merent confining pressure
values in second group of tests a: Tl b: S
5- 10 (a) The comparison of shaft fiction at dtimate uplift and
compressive capacity in first group of tests, pile Tl
5- 10 (b) The cornparison of shaft fiction at uîtimate uplift
and compressive capacity in first group of tests, pile T2
5-1 1 (a) The comparison of shafi fi-iction at dtimate uplift
and compressive capacity in second group of tests, pile Tl
5-1 1 (b) The comparison of shaft friction at ultimate uplift
and compressive capacity in second group of tests, pile S
Load-displacement curves at the loading point in the push
fonvard phase for different piles a: Pile S b: Pile M c: Pile
Tl
Lateral load capacity versus confining pressure for three
piles
Moment distribution dong pile S
Moment distribution dong pile T2
Moment distribution dong pile Tl
Normalized moment distribution dong pile S
Nomalized moment distribution dong pile T2
Normalized moment distribution almg pile Tl
Variation of maximum bending moment with applied load for three
piles
Page
-
Figure Description
6- 10 Bending moment dong the pile sh& for the three piles
subjected to the same load at different connning pressure
6-1 1 Soil resistance dong pile shaft for three piles under a
typical load
6-12 Soil resistance along the pile shaft for three piles under
ultimate load
6- 13 (a) Pile deflection along the pile shafi under a typical
load
6- 13 (b) Pile deflection dong the pile shaft under ultimate
load
6-1 4 The effect of pile taper angle on p y c w e s at 0.4 m
depth
6- 15 The effect of confining pressure on p-y curves for
different piles
6- 1 6 Degradation of modulus of horizontal subgrade reaction
with deflection
6- 17 Unrestrained laterdly-loaded pile (after Poulos and Davis,
1980)
6- 1 8 Predicted p-y curves for piles Tl and S in relative dense
sand (Geosoft)
7-1 (a) Characteristics of cyclic load at different values of
confining pressure ( T 1 , R a n d S )
7-1 (b) Characteristics of cyclic load applied to pile Tl at 20
kPa confiring pressure
7-2(a) Load-movement curves for piles S, Tl and T2 at zero
confining pressure
7-2(b) Load-movement c w e s for piles S, Tl and T2 at 20 kPa
confuiing pressure
7-2(c) Load-movement curves for piles S, T1 and R at 40 kPa
confining pressure
7-2(d) Load-movement curves for piles S, Tl and T2 at 60 kPa
confIIiù1g pressure
7-3 Pile head load-movement curves in the first and the tenth
cvcles
Page
1 O0
101
102
1 O3
1 O4
105
1 O6
107
108
108
119
119
120
121
122
123
124
-
Figure Description Page
7-4 Pile head stiffness at different confining pressure for
piles Tl and S 126
7-5 (a) Load-pile movement curve of Tl at Merent loading value
under 20 kPa confining pressure 127
7-5(b) Cornparison of pile Tl head Stifbess at 20 kPa confining
pressure subjected to dflerent load amplitude 127
Accumulation of displacement for piles Tl and S under different
confining pressures 128
Pile Tl head movement under cyclic load with different amplitude
129 a: Accumuiation b: Rate of displacement
Average unit load tramfer a: in compression b: in tension
138
(a) Unit tip resistance (b) Measured and prototype vertical
stress 138
Cornparison of pile shaft resistance established fiom experiment
and theory a: in compression b: in tension 139
Variation of Kt for pile T l with lateral pressure in relative
dense sand 139
The shaft resistance of prototype pile established from
experimental results with relative medium dense sand (a) in
compression (b) in tension
Variation of l& for pile Tl and T2 with lateral pressure in
loose sand 141
The compression shaft resistance of prototype piles in loose
sand established fiom experimental results
(a) in compression (b) in tension
-
1.1 OVERALL REVIEW
Pile foundations are used extensively to support both inland and
offshore
structures, including important structures such as nuclear power
plants and oil-drilling
pladorms. Piles are usually loaded axially in compression to
tramfer stnichual loads to
deeper competent soil layers. In some structures, like
transmission towers and jetty
structures, pile foundations resist uplifi loads. Piles are also
fiequentiy used to support
structures subjected to lateral forces and moments such as
offshore structures, harbour
structures high rise bbuildings and bridge abutments.
Piles are generally classified according to the pile material
(timber, steel or
concrete), the method of installation (driven, cast-in-place,
bored, etc.), or are categorized
in terms of the load transfer mechanism. (a) Friction piles: the
load capacity depends
mostly on the amount of fiictionai resistance developed at the
interface between pile and
soil. (b) End-bearing piles: the loading capacity relies
primarily on the concentrated soil
resiçtance at the pile tip for developing the resistance to
axial load.
D i f f m t iypes of piles with different shapes such as circle,
square or rectangle
cross sections are used in practice. Piles are mody used with
straight-sided wdls. Most
of the design procedures and guidelines have been developed for
straight-sided wall piles
with littie or no reference to tapered piles, aIthough tapered
piles have the potential for
substantial cost advantages over straight-sided wall piles
(Robinsky et al, 1964 and
Rybnikov 1990). Tapered piles are not widely considered as a
design option due to the
-
lack of knowledge about their static and dynamic behaviour and
the lack of appropriate
design tools similar to those available for straight-sided wall
piles.
1.2 OBJECTIVES AND SCOPE
The objectives of the shidy are to explore the static response
of tapered piles
subjected to axial, lateral and cyclic loads, and to provide a
procedure for the design of
tapered piles. A research program for studying pile performance
in cohesionless soils was
developed with emphasis on the pile shape effect on its capacity
and displacement. Both
tapered and straight-sided wdl piles were examined in order to
obtain a complete
comparative picture of pile actions.
1.3 ORGANIZATION OF THESIS
This thesis consists of nine chapters:
Chapter 1 presents a general introduction and the objectives of
this thesis;
Pile's ultùnate load behaviour, displacement analysis methods
and relevant studies of
tapered piles are reviewed in Chapter 2;
Chapter 3 contains a description of the experimental setup, test
piles, sand properties
and test preparations;
In Chapter 4, the experimentd procedure and results for the
axial compressive load
are presented three piles were tested with six different
confining pressures;
Chapter 5 presents the experimental work and results for three
piles subjected to axial
temile Ioad;
-
In Chapter 6, the experimental data and andysis for lateraily
loaded piles are
descri bed;
The response of tapered piles and straight-sided wall pile
subjected to cyclic load is
descnbed in Chapter 7;
Chapter 8 presents the validity and application of mode1 test
results; and
Chapter 9 gives the conclusions and recommendations.
-
This literature review covers some s~rdies on straight-sided
wall piles and tapered
piles. Pile load capacity, the analysis methods of single piles
and relevant research works
on tapered pile are reviewed.
2.1 PILE LOAD CAPACITY
2. 1. 1 Axial Bearing Capacity
The traditional study of single piles is directed towards the
static load canying
capacity, assinning that the displacement or deformation is
acceptable if an appropriate
safety factor is used in determining the allowable loads. There
are three methods to
estimate or check the ultimate load capacity of a single pile:
static analysis method, pile
loading test method and dynamic analysis method.
2. 1. 1. 1 Static anabsis method
The axial load canying capacity of single pile Qu is determined
in practice as the
sum of the shaft resistance, Qs, fiom the pile-soi1 interface
and the tip resistance, Qb, less
the weight of the pile, Wp, i. e. Qu is calculated as
follows:
Q ~ = Q ~ + ~ ~ - w ~ = J f ~ d z + f b ~ b - w ~ (2 - 1)
where f, = ultimate unit shaft fiction; C = pile perimeter; fb =
dtirnate unit base
resistance; Ab = area of pile base; Wp = pile weight. The h c t
i o n f, a n d 5 are caicdated
-
fiom ernpirical correlation with standard geotechnical soi1
properties, such as the
undrained shear strength for cohesive soils and the fiction
angle for cohensionless soils,
and the overburden pressure ( Meyerhof, 1 976).
For piles in sand or gravel, effective stress analysis of
ultimate load capacity is
appropnate. When the cohesive component of drained strength is
ignored, the ultimate
unit shaft fiictionf, and d h a t e unit base resistancefb can
be expressed as foilows:
I;= Ks &tan 6 ( 2 - 2 )
fb = Nq ~ b ' ( 2 - 3 )
where &= vertical effective stress adjacent to the pile and
a,b'= vertical effective stress
at the pile base. This approach will be m e r described in
Chapter 8.
2. 1. 1.2 PiIe load testing method
Pile load testing is usually camied out to quanti@ pile
load-settiement behaviour
and to detemine the uitimate bearing capacity as a check on the
value obtained fiom
theoretical calculations. A number of empirical d e s have been
used to serve as criteria
for detemiinhg ultimate load f?om the load test. Examples are
the Davission criterion, the
Brinch-Hansen criterion and the Chin criterion (Canadian
Foundation Engineering
Manual, 1992). The Davission cntenon defines the pile offset
limit load as the load that
produces a movement, rn, of the pile head equal to:
where: m = the movement of the pile head at the offset limit
load, mm, A = elastic
shortenhg of the pile, mm, b = diameter of the pile at the tip,
m. The other two criteria
were described in Canadian Foundation Engineering Manual (1
992).
-
Many research achievements have also been made towards the
understanding of
characteristics of compressive piles as repoxted by Nordlund
(1963), Coyle and Reese
(1 966), Bane j e e and Davies (1 978), and O'Neill et. al. (1
982).
2. 1. 1 .3 Dynamic analysis method
The capacity of a single pile can be estimated by means of
dynamic methods. The
objective with the dynamic methods of pile test is to relate the
dynamic pile behaviour
(acceleration or driving resistance) to the ultimate static pile
resistance. Pile axial capacity
can be obtauied based on dynamic monitoring, wave-equation
analysis or dynamic
formula. However, dynamic anaiysis rnethods are better used to
provide general guidance
due to its high dependence on competent person, local expenence
and relevant
assumptions. For more detailed information about the dynamic
analytic rnethods, see
Poulos and Davis (1980).
2. 1 .2 Uplift Load Capacity
There were considerable debates over the relative magnitude of
pile shaft capacity
in tensile (uplift) loading as compared with compressive loading
case. Generally, it is
assumed that the pile shaft capacity is identical under both
tensile and compressive
loading case. However, there is widespread experimental and
numerical evidence that in
sand, the straight-sided shaft capacity is significantly lower
for tensile loading than for
compressive loading (e.g. Chattopadhyay and Pise, 1986 , Nicola
and Randolph, 1993).
Pile uplift load testing is u s d y carried out to obtain pile
uplifi resistance. In
practice, the interpretation methods used for estimating
ultimate pullout load fiom pile
-
puilout load- pile upward movement ciwe are: (1) failure load
may be taken as the Ioad
value that produces a net upward pile top movement of 6.25 mm;
(2) upward failure load
occun at the intersection point of tangents on the load-movement
curve; or (3) upward
failure load is the value at which upward movement suddenly
increases.
2.1.3 Lateral Load Capacity
Broms (1964% b) provided solutions for the ultimate lateral
resistance of a pile
assuming a distribution of the lateral pile-soi1 pressure and
considering the statics of the
problem. He considered two modes of failure: yielding of the
soi1 dong the length of the
pile (short-pile failure) and yielding of the pile itself at the
point of maximum moment
(long-pile failure). Meyerhof (1995) accounted for the ef6ects
of load eccentricity and
inclination on the ultimate lateral capacity.
Pile lateral load testing is performed to assess the
load-deflection behaviour of a
pile. Methods for detemwiing failure load fiom lateral load
testing Vary depending on the
tolerable movement of the structure supported by the pile. The
general critena are: (1)
ultimate load can be taken al 6.25 mm Meral movement or
deflection; or (2) ultirnate
load can be considered at the point of intersection of tangents
on the load-movement
curve.
2.1.4 Effect of Cyclic Loading
The performance of the foundation piles under cyclic loading is
an important
factor in the design of piles for offshore structures,
transmission towen, and some ta11
buildings. Two major types of cyclic loading are considered in
the design of piles. The
-
first type is dynamic or random loading whereby the dynarnic
component is significant
compared to the rea of the forces, such as earthquake load. The
second type is non-
dynamic or approximately systematic loading whereby a steady but
slow variation of the
load is applied, as shown in wind and wave loads or tidai
effects on piles resisting uplifi
forces. The cyclic loading has three possible effects on pile
performance:
1. Accumulation of permanent displacement;
2. " Shake down" phenornenon reported by Poulos (1982), the pile
defiection stabilizes
and react elastically to any further load cycle;
3. A possible reduction or " degradation" of pile resistance,
especially shaft resistance;
It is acknowledged that the pile response to cyclic loaduig
depends on the
characteristics of the pile-soi1 system. Chan and Hama (1980)
investigated the effects of
the load amplitude, the type and the number of load cycles on a
pile's response and
concluded that the response of the pile to cyclic load is
cornplex. Podos (1 988,1989)
used soil degradation factors, which are defined as the ratios
of soil properties after cyclic
loading to properties for static loading, to calculate the
reduction of pile load capacity due
to cyclic axial loading, and showed their significance.
2.2 RESPONSE ANALYSIS METHODS FOR SINGLE PILES
Several approaches have been developed for the response analysis
of axidly and
laterally loaded piles. These approaches assume either the
theory of elasticity or the
theory of subgrade reaction. The former includes Poulos and
Davis (19681, Mattes and
Podos (1969), Poulos (1971), Randolph (1981), Pise (1984) and
Budhu and Davies
(1988), and the latter includes Coyle and Reese (1966), Kraft
Jr. et al (1 98 1) and O'Neil
-
et al (1982). The load-settlement or loaddeflection behaviour of
axially or laterally
loaded piles is highly nonlinear and hence requires a nonlinear
analysis. Poulos and Davis
(1980) and Budhu and Davies (1987) modified the elastic
solutions to account for
nonlinearity using yield factors. The modulus of subgrade
reaction approach was
extended to account for the soil noniinearity. This was done by
introducing p y curves
(Matlock, 1970; Reese et al., 1974; Reese and Welsh, 1975;
Prakash, et al 1996).
2.2.1 Elastic Analysis Method
The elastic d y s i s of pile load-displacement behaviour under
static axial or
lateral loading was based on the elastic theory for both the
pile and surrounding soil. The
soil is considered as an elastic continuum, the pile is assumed
to be a thin strip and
divided into nurnber of elements. Factors such as soi1 yield,
soil layer depth, soil
inhomogeneity, stifk bearing stratum, and enlarged pile base
were considered, the
dificulty of the application of the elastic method to practical
problems is detemiining the
appropriate soil modulus.
2.2.2 Subgrade Reaction Method
The subgrade reaction method or load-tramfer method,
correspondhg to lateral or
axial loaded pile, is based on the curves of soil resistance us.
pile displacement. These
curves are commonly texmed as t- z, q-z curves, denoting load
transfer vs setdement
curves dong the pile or at pile tip, respectively; and p-y
curves, denoting lateral resistance
vs deflection. An apparent advantage of this method is that it
can incorporate inelastic soil
behavior by using nonlinear curves while not complicating the
analysis. The weakness of
-
this model is that the continuous nature of the soil medium is
ignored and the pile
reaction at a point is simplly related to the deflection at that
point.
The corresponding experimental curves have been develo ped for
clay (Co y le and
Reese 1966, O'Neil et al 1982, Brown et al 1987) and sand (Brown
et ai 1988). In
addition , an attempt was made to derive theoretical t- z curves
(Kraft Jr. et ai 198 l), in
which the theoretical analysis provided a bais for t-z criterion
that couid be applicable to
a variety of pile and soil conditions. Rakash et al. (1996)
developed a method to predict
the load-deflection relationship @-y curve) for single piles
embedded in sand, cons ide~g
soil nonlinearity using subgrade reaction, based on the analysis
of 14 Ml-scale lateral pile
Ioad tests results.
2.2.3 Finite Element Analysis Method
The finite element method could be used for the response
analysis of single piles.
It gives a better insight in the pile foundation behaviour,
provided that adequate modeling
of the soil and the pile- soil interface takes place. The finite
element method was used by
Ellison et al (1 97 l), Desai (1 974), Kuhlemeyer (1 979) and
Randolph (1 98 1 ). Ellison et al
(1971) developed a general procedure to collocate the behavior
of elements with the
cornplex foundation-soi1 system., where the h i t e elemect
model was used to predict the
load capacity and load-deformation of a bored pile in stiff
clay. Desai (1974) applied the
finite element method to predict load-deformations and stress
distributions for
compressive steel pipe piles in sandy soils. Kuhlemeyer (1979)
analyzed laterally loaded
piles and highhghted the dynamics solution. Randolph (1 98 1)
examined pile static lateral
-
load behaviour and eeated the soil as an elastic continuum with
a linearly varying soil
modulus.
2.3 RELEVANT STUDIES ON TAPERED PILES
A number of studies have been directed toward the response of
individual
straight- sided wall piles with littie attention being paid to
tapered piles. The vast rnajority
of the research on tapered piles focuses on the load-carrying
capacity.
The previous studies of tapered piie subjected to axial
compressive load include
full-scale field testing, laboratory testing and analytical
procedures. Field testing results
are reported in Norlund (1 963), Appolonia and Haribar (1963)
and Rybnikov (1990).
Laboratory testing observations include those conducted by
Robinsky et al. (1964) and
Bakholdin (1971). Ladanyi and Guichaoua (1985) and Kodikara and
Moore (1993)
suggested analytical solutions for tapered pile response; Poulos
and Sim (1990)
conducted a theoretical anaiysis with five different pile types
to assess their cyclic load
capacity.
2.3. 1 Field Testing Results
Norlund (1963) described a pile test program and a method for
computing the
ultimate axial resistance of a pile in cohesioniess soils was
developed. The test data
demonstrated the signifiant effect of pile taper, the roughness
and the shape of the pile
sdace , and the volume of soil displaced by the piie on the pile
bearing capacity.
Rybnikov (1 990) shidied the bearing capacity of
bored-cast-in-place tapered piles through
a field experimental investigation. He suggested that the
tapered piles had a specific
-
bearing capacity that exceeded the specific bearing capacity of
straight-sided wall pile
having the same length by 20.30%.
2.3.2 Laboratory Testing Observations
Robinsky et al. (1964) investigated the effect of the shape and
volume of piles
installed in sand on their capacity. In this study, instnunented
model straight-sided wall
and tapered piles were driven into sand at different embedment
depth to diameter ratios.
These tests revealed that the intensity of unit load transfer
through the pile walls changed
continuously as the piles were advanced. Furthemore, tapered
piles were found to be
appreciably more efficient than straight-sided wail piles.
Robinsky and Momion (1964)
studied the effect of pile taper on the displacement and
compaction of cohesiodess soi1
adjacent to fiction tapered piles. It was found that in
relatively homogeneous
cohesionless soils, a tapered pile with most of the load being
carried by skin fiiction can
support considerably greater loads than a sûaight-sided wall
pile with a larger point.
2.3.3 Theoretical Analysis Results
Ladanyi and Guichaoua (1985) compared the response of tapered
piles, straight-
sided wall piles and comgated piles in permafkost soils. They
showed that tapered piles
were the safea because they display strain hardening
characteristics as opposed to the
brittle failure that occurred in other types of piles. They also
developed a model for the
analysis of tapered piles wherein the soil resistance was
modeled by two components; the
first component was the fiction and adhesion dong the shaA
(shearing resistance) and the
second component was due to the lateral soil reaction mobilized
by the hole expansion
-
resdting fiom the pile penetrating the ground. A simila. mode1
was presented by
Kodikara and Moore (1993) for the analysis of the axial response
of tapered piles, it
accounted for the nonlinearity conditions dong the pile-soi1
interface. Podos and S im
(1 990) conducted a theoreticai analysis with five different
pile types to assess their cyclic
Ioad capacity. They concluded that the pile taper could have a
favorable effect on its
cyclic performance as it reduced stress concentration.
2.4 MOTIVATIONS
In tapered piles, which have the great top cross section than
the bonom one, an
increase in the side resistance can be expected when there is a
slip of the pile relative to
the ground. This occurs because the ground adjacent to the pile
is then forced to expand
radially, so that additional lateral pressures are developed and
lead to an increase in the
shear stresses across the pile-soi1 interface. The flexural
effects of deflection and bending
moment of a pile subjected to a lateml load at the top are also
highest at the top and
decrease rapidly with depth. Hence, it is also expected that
tapered piles represent a more
efficient distribution of the pile matenal in this loading mode
as well.
In order to obtain a complete comparative picture of pile
action, it is essential to
midy both tapered and straight-sided wall pile. A program was
initiated to study fiiction
piles in cohensioniess soils, with particular emphasis on the
effect of shape on pile
capacity. Four phases are included in this program, the fia
phase examined the pushing
down bearing capacity of the tapered piles (Wei and EL Naggar,
1998); the second phase
dealt with the effect of the pile taper on tensile loading; the
thKd phase examined the
-
response of test piles to the lateral loading and the forth
phase explored pile behavior
under uniforni axial cyclic loading.
-
CHAPTER 3
TESTING FACILITIES AND PREPARATION
3.1 INTRODUCTION
The loading tests were performed on mode1 piles installed in dry
sand in a
laboratory setup. The sand was enclosed in a steel chamber that
ailowed the application of
variable confining pressure to the sand. The description of the
experimentd setup,
hcluding the test piles, instrumentation, soil sample, soil
chamber, loading equipment,
and testing preparation are given below.
3.2 TESTING FACILITIES
3.2.1 Testing Mes
Three inmumented structural steel piles of equal length and
average embedded
diarneter but different taper angles were used in this mdy. Two
piles were tapered wirh
different taper angles and the third was a straight-sided wall
pile. Tapered pile number 1 ,
Tl, had a taper angle = 0 . 9 5 ~ ~ while tapered pile number 2,
Tî, had a taper angle = 0.6'.
The piles were 1.52 m in length with diameters varying between
160 and 200 mm and a
wall thickness of approximately 6.4 mm. The length to diameter
ratio for these piles was
approximately 9, typical of rigid piles. The geometrical
properties of piles are given in
Table 3-1, and are shown schematically in Figure 3-1. The piles
were fitted with slip - on
flanges at their heads to facilitate loading.
-
3.2.2 Strain Gauge Installation
The piles were instrwnented with electrical resistance saain
gauges (CEA-06-
XOUW- 120). Full and half bridge electrical resistance strain
gauge circuits were used in
axial and lateral loading tests, respectively, as shown in
Figure 3-2.
Six pairs of sirah gauges were attached to the exterior walls of
the piles using M -
Bond 200 adhesive. They were distributed over the length of the
pile such that the first
strain gauge was 0.3m from the pile head (approximately at the
surface of the sand) and
the last main gauge was close to the pile tip , 0.05 m fiom the
pile tip. The rest of the
main gauges were spaced as shown in Figure 3-3.
3.2.3 Soi1 Sample
The soi1 w d in the tests consisted of coarse, anguiar particies
of air dned sand.
The p i n size distribution for the sand is shown in Figure
3-4.
A standard test of the sand showed it had a maximum unit weight
of 18.35 kN/m3
and minimum unit weight of 15.83 w / m 3 at a moisture content
of 0.25 per cent. Two sets
of tests were perfomed on piles in this study for axial static
load test. In the fint set, the
sand was loose with relative density, Dr = 18.4%. In the second
set, the sand was
compacted by applying a 100 kPa confining pressure and then
releasing the pressure to
zero before starting the testing procedure. The initial relative
density in this case was
calculated as Dr = 32.7%. Grain size analysis and other data
relating to the sand are given
in Table 3-2. The relationship between angle of fiiction and
relative density of sand are
adopted fiom Das (1995).
-
3.2.4 Soil Chamber (VLPSC)
To mode1 the effixt of varying confining pressure on the
response of piles
installed in cohesioniess soil a device termed the Variable
Laterai Pressure Sand Column
or VLPSC was used and is depicted in Figure 3-5. This device was
designed and used by
Moore et al (1995).
The sand was contained in a steel chamber, VLPSC, 1 Sm in
intenor diameter,
with 10 mm thick walls, and 1.445m in depth. The top and bottom
of the steel chamber
were covered by ngid steel plates. The top plate has a 397mm
access hole for the test pile
and svain gauges leads to the data acquisition system. The edge
of the access hole is
stiffened by a steel r i . fitted with an arrangement to
facilitate the attachent of the
loading b e . The loading fiame (reaction fiame) was made of
steel channel sections
with a fitting to fasten the hydraulic jack used for both axial
and laterai loading.
The steel charnber was Iined with an air bladder so that sand
inside the steel
chamber could be pressurized. The air bladder was used to vary
the confining pressure to
simulate the typical embedded depth of piles. A manifold was
applied to control the air
pressure through lines comected to the air bladder. The pressure
variation was fiom O to
100 kPa.
3.2.5 Soil Pressure Transducer
Three soi1 pressure transducers were used in this study. Typical
Current-Pressure
relationships of soil pressure tmnsducers are showed in Figure
3-6. The variation of the
stresses around the test piles were obtained before loading
tests. Two soil transducers, Pl
and P2, were placed 150 mm fiom the pile tip; one to measure the
vertical stress at the
-
pile tip elevation and the other to masure the lateml stress at
the level of 200 mm above
the pile tip. A third soi1 transducer, P3, was installed in the
sand 200 mm under the sand
surface, 150 mm fkom the pile to measure the lateral stress ~f
the sand. The relationship
between applied confining pressures and measured stresses a r o
d the tested piles are
illustrated in Figure 3-7.
3.2.6 Other Testing Equipment
Other testing equipment used in this study are listed as
follow:
Blackhawk Holoram Hydraulic Jack, 178 kN (20 TON) capacity, 50.8
mm (2 in.) stroke
Simplex R106 Hydraulic jack ,89 kN (10 TON ) capacity, 127 mm
(Sin.) stroke
Simplex Hydraulic Hand pump, RP6A, 0-68.95 MPa (10,000 Psi )
capacity
S train,ce~e
FL25U- 2SGKT 1 1 1.25 kN (25,0001b) loadcell
Donc Digital Readout Mode1 420
Data Acquisition System includes
Compter @oric 245A, &ta logger)
Strain gauge conditioner (UC- 19)
Beckmann "logger" software
-
3.3 TESTING PREPARATION
More than seventy testing configurations were considered on pile
testing to
establish the behaviour of tapered piles in cohesionless soil,
as indicated in Table 3-3.
Two sets of tests were performed in this study for investigating
the behaviour of tested
piles under axial loading. In the first set, the sand was w d in
a loose state, while in the
second set, the sand was relatively medium dense due to the
application of a 100 kPa
c o n f i g pressure before testing started. Lateral and axial
cyclic loading tests were
conducted with loose sand. The same preparation procedure
described below was
followed for al1 performed tests.
The sand was spread in patches in the lab and air dried, it was
placed into the
VLPSC to a depth of about 400 mm using a min technique. The pile
was then placed at
the center of the chamber guided by a m e to assure centric
vertical alignrnent. The pile
was slightly embedded in the sand such that the total embedrnent
depth of the pile would
be 1.2m afier filling the soil chamber, and the first strain
gauge was approximately at the
suface of the soil. As mentioned forgoing two pressure
transducers were placed 150 mm
fiom the pile tip. Afier securing the pile and pressure
transducers in place, more sand was
added around the pile until the chamber was filled to capacity.
A third pressure
transducer was installed in the sand 200 mm under the sand
surface, 150 mm fkom the
pile. The surface of the sand was leveled and the top cover
plate was placed and screwed
to the chamber.
The reaction frame was placed across the access hole at the top
plate and tightly
screwed to the rim of the access hole. A reference beam was then
attached to the edge of
the chamber. Dia1 gauges were installed on the reference beam
(separated fiom the
-
loading system) to meamre the pile head settlement or
deflections. A hydraulic jack and
a load ce11 were placed between the pile head and the reaction
fiame and adjusted to
ensure centric loading. A data acquisition systern was connected
to the sûain gauges and
the load ce11 to read and record the strains and load appiied
simultaneously durulg testing
every ten seconds. The loading equipment for axial downward,
axial upward and laterai
loading are shown in Figures 3-8 to 3- 10, respectively, the
axial downward and upward
loading equipment were used to conduct the axial cyclic loading
tests. A typical oblique
view of the testing facility is shown in Figure 3-1 1.
The testing procedure descnbed and the pile load testing were
repeated three
times for each testing configuration. The difference between the
pile ultimate load in the
three sets was Iess than 10%. The results reported herei.
represent one set of results that
was the closest to the average of the three sets.
3.4 EFFECT OF PILE INSTALLATION METHOD
The pile placement method may have important influence on pile
performance.
When a pile is driven into sana the mil is usuaily cornpacted by
displacement and
vibration. In loose sand, the load capacity of a pile is
increased as a result of the increase
in relative density caused by the driving. Detailed
investigations of extent of compaction
of sand and the increase in relative density around the pile
have been carried out by
Robinsky and Monison (1 964).
Aiizadeh and Davisson (1 970) described a pile load testing
program conducted to
determine the lateral Ioad-deflection behavior for individual
vertical and batter piles, the
-
flexurd behavior of piles and the effect of sand density on the
pile response. The results
showed the significant effect of the relative density of sand on
the pile's behaviour.
To ensure that the placement method has no influence on the
relative performance
of the piles, the same instailation procedure was w d for aii
piles. In this procedure, as
was described, the pile was placed in the center of soi1 chamber
and then the sand was
poured into i t Therefore, no densification was due to the
placement method because no
soi1 displacement occurred. This installation method is more
like a "bored pile" case and
hence, the results obtained do not represent the case of driven
piles. However, it is
expected that the effect of the taper will be more pronounced in
dnven piles. This is
because the wedging effect of tapered piles would result in more
densification of the sand
narounding the pile that in tum would lead to a better
performance under axial and
lateral loads as shown Iater.
-
Table 3-1 Geometries of Three Piles
Table 3-2 Sand Properties
'pi le Diameter (outside. ml IThickness(mrn) 'Length(rn) S~night
sided wall pile. S Tapemi pile. T2 Tapend pile, Tl
Ettécrive diameter
Utirl'om~r~ coetticicnt
ASThn* Sieve Desipotion No. IO No. 20 No. 40 No. 60 No. 100
Mas~murn densi' Initial relative densin
Percentage f i bv weight 84.17% 45.96% 13.53% 4.66% 1.85%
D,o=0.35mrn
C,=2.85 26"-30"
1.615kg/rn3 28"-35"
l.873k~@? 18.4% (first group of tests) & 32.7% (second group
of tests)
1.524 1.524 1.524
0.1683 0. & 968~Q0.165 1
_0.2032&0.1524
13 2"( tint poup or tests) &3S0(second p u p of tests) ASTM*
1993
7.1 12 6.35 6.35
Table 3-3 Test Arrangement
Apyilicd contininp pressiirr
Coinpression Lcwsr ssind Relritivrly rnaiium dense sanù
~rnsion Lwse sand Rrlritivelv medium dense sond
h e m 1 Lmse sand
C'!.clic Loc~sc smd T1.TZ.S TI.T2.S TI.T2,SI T1.T2.S I
O kPa
TI. ft
20kPn
T 1 . n
4OWa
T1.n
60kPa I 8 0 U a
TI.T2 S. Tl
T 1 . n S. TI
T 1 . n . S
100L;Pa
S. TI
TI,T2 S. TI
T1.R.S
S. Ti
1
S. Tl
T1.Tz.S
S. Tl
Tl. T2 S. Tl
TI.T2,S
S. Tl
T 1 . n S. TI
T l .
S. TI
S. Tl
T1.Tz.S
-
Dummy / \ Active / Signai 1
A c t C ive T
Act ive
Figure 3-1
T Act ive
Strain griuge circuits a: mial loading test b: Lateral loading
test
-
Latera
L
I iood-
Strain lndicator
F e 3 - Strain gauge installation
Grain size in miIlimeters
Grain size distribution
-
O -! 1
0 2 3 4 5 6 7 8 9 1 0 1 1 1 2 1 3 1 4 1 5 Pressure. psi
Figure 3-6 Typicai current-pressure relationship for soi1
transducer (1 psi = 6.895 kPa)
Applied conlining pressure (kPa)
Figure 3-7 Applied confining pressure vs. measured stress around
pile S
-
To hydraulic
Figure 3-8
Figure 3-9
I /Reaction frame J
Load cell
Hydroulic jack
Bearing plate
Axial loading equipment
Uplifl loading equipmtnt
-
I ' VLPSC . . I I 1 . . I I I I I I 1 I I I
holdeî
beam
Figure 3-10 Lateral loading equipment
-
Figure 3-1 1 Oblique view of the testing facility
-
CHAPTER 4
AXIAL COMPRESSIVE RESPONSE OF TAPEIRED PILES
4.1 INTRODUCTION
Piles have been used to bansfer structural loads to deeper
competent soil layers,
allowing construction in areas where the soil conditions near
the ground surface are
unfavorable.
To investigate the effect of the pile taper on the pile axial
response, two sets of
compressive loading tests, as defhed in Chapter 3, were
performed in this study. The test
procedure and observations are discussed in this chapter,
including failure load
detemination, load distribution dong the pile, unit load tramfer
and the distribution of
pile shafl and tip resiçtance, etc.
4.2 TESTING PROCEDURE
At the beginning of pile load testing, al1 the instruments were
reset to zero. The
fist axial loading was performed with zero confining presstire
fiom the air bladder. The
pile was loaded in 20 increments. each about 10% of the expected
pile capacity. Each
load increment was maintained for 2.5 minutes. The measurements
of load and strains
were recorded every 10 seconds through the data acquisition
system. The dial gauge
readings were taken at the middle and the end of time interval
for each loading increment.
After the axial loading (downward) was completed, the uplift
test was conducted.
Uplifi capacity was less than the compression bearing capacity
for the same pile under the
same confining pressure and was attained at a smaller
displacement. Hence, the pile had
-
to & pdled up to the initial embedded depth (1.2m) &er
the uplift test. This was done to
ensure that the embedment depth was the same for ail loading
tests. Also, any residual
stresses due to previous loading may dissipate due to that
process. The pressure was then
increased by an additional 20 kPa, and the process repeated
until the pile had been tested
at dl desired values of confining pressure.
in each group of tests, the confinùig pressure was varied fkom O
kPa to 100 kPa.
However, before ninning the second group of tests, a connning
pressure of 100 kPa was
applied to the sand for two hours and then released to O kPa.
The top cover plate was
then removed and the drop of the sand sudace was measured and
was found to be 30 mm.
More sand was added to f iU the chamber up again and the top
cover plate was placed
back on the steel chamber. Therefore, the sand in the second set
of tests could be
wnsidered medium dense. The same test procedure used in the
first set was followed in
the second set. The axial load tests in the fkst group of tests
were successfully completed
with 0-60 kPa confinhg pressure only. The testing at higher
confining pressure was
intemipted due to a problem with the air bladder.
4.3 TESTLNG RESULTS
4.3.1 Load-Displacement and Bearing Capacity
4.3.1.1 First group of tests
The load applied at the pile head and the displacement of the
pile head were
measured during the first set of tests and plotted in Figures
4-1 and 4-2. Figure 4-1 shows
the load-displacement curves at different values of confining
pressure for tapered pile,
Tl, while Figure 4-2 shows the load-displacement curves for
tapered pile, R. It can be
-
noted fiom Figure 4-1 and 4-2 th, as expeaed, the effect of the
confining pressure was
to increase the pile capacity for both piles. Also, it may be
noted that the axial stiffiess
(load divided by displacement) of pile Tl, with greater taper,
at any loading increment
was greater than the axial sfiffness of pile T2 at the same
level of loading for d l values of
confining pressure.
The ultimate load for each pile was detemiined fiom the load
displacement c w e s
using Davisson cntenon (Canadian Fomdation Engineering Manual,
1992), illustrated in
Figure 4-3. The results are compared in Table 4-1 based on the
pile bearing capacity ratio,
KQ, and the coefficient of effective utilization of the pile
material, KV. The ratio KQ is
defined as the ratio of the bearing capacity of the two
investigated piles, while the ratio
KV is defmed as the ratio of the specific bearing capacity (pile
capacity per unit volume)
of the two piles. It cm be observed fiom Table 4-1 that pile Tl
displayed higher axial
capacity for all values of confinhg pressure. This increase was
manifested in terms of KQ
and KV values higher than 1. These values implied that the axial
capacity of pile Tl is 17
to 27% higher than the axial capacity of pile T2 for the given
values of confining
pressure, with the highea increase at a confïning pressure equal
to 20 kPa. The same
trend was observed in the coefficient KV, which implied that the
taper of the pile
increased the efficiency of the utilization of the pile
material. These results are consistent
with the results obtained by Rybnikov (1990), who found that the
specific bearing
capacity of a tapered pile was higher by a factor of
approximately 1.3 compared with a
straight-sided wall pile.
-
4.3. 1.2 Second group of tests
Figures 4-4 and 4-5 present the load and displacement
measurements at the pile
heads at different values of confining pressure during the
second set of tests. Figure 4-4
presents the load-displacement curves for tapered pile, Tl,
while Figure 4-5 shows the
load-displacement curves for dght -s ided wall pile, S. It can
be noted fiom the figures
that the confining pressure was to increase the pile head
sti&ess, for both piles.
However, this effect was less significant in this set of tests
(dense sand) than it was in the
first set of tests (loose sand), especially at higher values of
confining pressure (80 and 100
kPa). This observation could be used as an argument to support
the arching effect for
piles installed in cohesionless soil. Further, It may be noticed
that it was even less
significant for the pile Tl, which draws part of its capacity
fiom the lateral resistance of
the soil. The comprison of the response of the tapered pile, Tl,
and the straight-sided
wall pile, S, shows that Tl displayed a stiffer response at dl
loading increments for al1
values of confinhg pressure.
In Table 4-2, the axial capacity of Tl and S were compared based
on KQ and KV.
It c m be seen from the cornparison that the axial capacity of
the tapered pile was higher
than the axial capacity of the straight-sided wall pile for al1
values of confining pressure.
The ratio KQ varied between 1.05 to 1.37, with the maximum value
occurring at a
confining pressure of 40 kPa. The ratio KV varied between 1 .O9
to 1.42, with the
maximum value occurring at a confining pressure of 40 kPa. Both
KQ and KV increased
as the confining pressure increased until a confining pressure
of 40 kPa was reached. For
higher values of confining pressure, both KQ and KV decreased as
the confining pressure
increased.
-
4 3 . 2 Load Distribution
The forces transmitted at different locations were caiculated
fiom strain gauge
readings, i.e.
4, =E, E A , (4- 1)
where q, is the pile axial load at the location of strain gauge
i, E is the strain measurement
of snain gauge i, E is the elastic modulus of the pile matenal
and A, is the pile cross-
sectional area at the location of sûah gauge i.
Figures 4-6 and 4-7 show the load distribution dong the piles
under various load
increments for the first and second sets of tests, respectively.
It may be observed fiom the
two figures that the general trend of the load distribution was
the same for al1 three piles
at al1 loading increments. It can also be noticed that most of
the load was transferred to
the soii through the pile shaft with a small contribution fiom
the pile tip. A closer look at
these figures showed that tapered pile Tl transferred more load
to the soi1 along its upper
portion than did both tapered pile T2 and straight-sided wall
pile S.
4.3.3 Unit Load Transfer
The unit load transfer of the pile shaft was calculated from the
main
measurements during the pile test. When the ultirnate load of
the pile was reached, the
readings of the strain gauges were recorded and used to
calculate the load distribution
along the pile. The difference between the force calculated fiom
any two sets of strain
gauges dong the pile wall represented the total load transferred
to the surrounding soi1
between the two points. Dividing this ciifference by the
correspondhg surface area, the
average of unit load tnuisfer was obtained
-
wheref;, is the average unit load transfer between stations i
and j and Sv is the surface area
of the pile between stations i and j. The unit load transfer c w
e was obtained fkom the
values of unit load transfer dong the pile.
Figure 4-8 shows the comparison of the unit load transfer for
piles T l and T2 in
the f is t set of tests, while Figure 4-9 shows the comparison
between Tl and S in the
second set of tests. These two figures iliustrate clearly that,
as expected, the unit load
tramfer increased as the confining pressure increased, for dl
piles. However. for higher
confinhg pressure (Le. greater than 60 kPa) the unit load
tramfer for tapered pile Tl
leveled off at a maximum of 40 kPa. These results were in good
agreement with the
results obtained by Robinsky e t al. (1964). It may be noted
fiom the wo figures that the
effect of the taper was to increase the unit load transfer
especially for the topmon part of
the pile and the lower confining pressure range. Furthemore,
comparing the unit Ioad
transfer for the tapered pile Tl in the two sets of tests
(Figures 4-8 and 4-9), it could be
observed that the initial sand density has a significant effect
on the unit load transfer in
the lower confining pressure. This effect was less significant,
however, in the higher
confiking pressure range.
Figure 4-1 0 shows die unit load transfer curves for
different values of confining pressure, derived fkom the
£ïrst
tapered piles T l and T2
set of tests, while Figure
11 shows the unit Ioad transfer for piles Tl and S, derived fiom
the second set of tests.
These two figures confirrn that the intensity of the load
transfer increased with an increase
in the confining pressure until a maximum value of unit load
transfer was reached. The
-
increase in the confining pressure beyond this point has no
effect on the load tramfer
through the pile shaft. This observation suggests there was a
limiting fiction value that
depends on the of shearing resistance of sand and the pile
Wction angle. Similar
conclusions were made by Yoshimi and Kishida (1 98 1) in their
experimental study on the
characteristics of the fiiction between sand and metal
surface.
4.3.4 Pile Tip Resistance
Table 4-3 compares the distribution of the ultimate load between
the pile point
and the pile shaft for various applied confining pressures
plotted in Figure 4-12. It can be
observed fiom the figure that the contribution of the pile tip
was higher for the maight-
sided wall pile than for the tapered pile. For zero confiring
pressure, the tapered pile T1
derived approximately 58% of its support from the pile tip while
the straight-sided wall
pile denved about 66%. As the confining pressure value
increased, the pile tip
contribution decreased and the shaft contribution increased.
This is in a good agreement
with Abendroth and Greimann (1 WO), where the end-bearing
capacity of the mode1
fiction piles was as large as 70% and 49% of the total vertical
capacity for the piles
ernbedded in loose and compacted sand, respectively.
4.4 DISCUSSION: ARCHING PHENOMENON
The effect of pile taper can be explained by the arching
phenomenon. The pile
compression test was accompanied by some sand loosening dong the
pile walls, which
was believed to cause a complex systern of arching in the soil s
w r o ~ d h g the piles
(Robinslq, 1964). The pile shape and the initial soil density
detemüned the systern of
-
arching and the efficiency and capacity of the piles. Tests with
the tapered pile permitted
the development of wide arches, thus transferring the pile load
to a greater volume of
sand than in the tests with the straight-sided wall pile. With
the kght -s ided wdl pile,
the load had to be carried by a smaller volume of sand. The
failure stresses in the sand
surrounding a &&t-sided wall pile were tt.s reached at a
lower total load than in the
case of the tapered pile.
At low confinhg pressure values, the axial capacity of both
piles increased
significantly with the increase of the appiied confining
pressure. At hi& confining
pressure values, this effect is much less significant. The
experimentd investigation by
Sirnonini (1996), on the pile behaviour offered a possible
explanation for this trend.
When a pile is ioaded, the effective stresses inside the soi1
mass range fiom low values,
co~responding to the initial overburden stress, to very high
ones; at high confining
pressures, dilatancy vanishes and crushing becomes the only
mechanism of deformation
in addition to simple slip. These combined effects lead to a
decrease of the sheax-ing
strength of sand. This explariation was supported by the fact
that the cnishing sound was
heard during the experimental work as the dtimate load was
reached, at confining
pressure values greater than 60 kPa At one stage during the pile
loading at a 100 kPa
confinhg pressure, two soil pressure transducers with a capacity
of 700 kPa were failed.
That meant the stress in the soil exceeded the tramducers
capacity, which resulted in the
sand crashing.
-
4.5 SUMhlARY
An experimental investigation of the axial response of tbree
steel piles with
different taper angle installed in sand was presented and
discussed in this chapter. The
analysis of the results showed consistent agreement with the
results obtained by other
researchers. Based on the results presented and the cornparisons
between the response of
tapered piles and straight-sided wall pile, the following
conclusions were made:
1. The pile axial capacity increased with an increase in
confinulg pressure for d l piles
examined in this study.
2. The resdts indicated a beneficial effect of the pile taper in
ternis of an increase in the
axial capacity and niffness.
3. The analysis of ihe test results indicates that there was a
limiting fiction value that
depends on the angle of shearing resistance of sand and the pile
friction angle.
4. As the effect of the taper was much less significant for
higher confixing pressure values
(Le. greater embedment depth), it may be recomrnended that the
taper be confied to the
topmost part of longer piles.
5. The pile tip contribution 10 the total pile capacity was less
for tapered piles than for
straight-sided wall pile.
-
Tcrbte 4-1 Axial Compression Results ( first group of tests)
Applied confining pressure O kPa 20 kPo 40 kPa , 60 kPa t3canng
Cnpacity h QT2 1.70 7.10 1 1.50 14.50
2 .OQ 9 .O0 17.00
Table 4-2 Axial Compression Results ( second group of tests)
Table 4-3 Pile Tip Resistance at Qu ( second group of tests)
Applicd Confininp Pressure Rearing Cepacity
Qu (W QS QT 1
KQ=QT I /QS Capacit)lNolume 1 ( Q W s
Applied Confininp Pressure l'ip Kcsisionce (kN)
Shaîl Resistance (kN)
Perceniage of Shaft Resistancc Perccniaee of Tip
Rctsisiancc.
TI S
T I S
T I S
Tl S
O kPa 4.54 5.00
O kPa 2.89 2.96 2.1 1 1.58
42.20 34.80 57. 80 65.20
40 kPa 15.36 2 1 . 0
20 kPa 1 1.50 14.50
1.10 1 .O3
60 kl'a 1.82 2.29 2 1.78 17.2 1 92.29 88.26 7.7 1 1 1.74
1.37 3.50
1,26 2.62
20 W u 5.86 6.08 8.64 5.42 59.59 47.13 40.4 1 52.U7
60 kPa 19.50 23.60
40 kPa 2.32 2.64 1 N.68 12.72 88.95 82.8 1 1 1 .O5 17.19
80 kPa 0.71 1.4
24.58 22.60 97.19 94.17 2.81 5.83
1.21 4.44
100 kPa I .58 1.8 1
26.42 24.19 94.36 93.04 5.64 6.96
80 kPn 24.00 25.29
100 P a 26.00 28.00
1 ,O5 5.47
1 .O8 5.93
-
Figure 4- 1
confining pressure
10 1 S 20 Settlement (mm)
Load-seulement curves of pile T 1 wirh different codining
pressures (firsr group of tests)
1s 20
Setticment (mm)
Figure 4-2 Load-settlement curves of pile T2 with different
confining pressures (first group of tests)
-
Figure 4-3 Constmction of the offset limit Ioad (after Canadian
Foundation Engineering Manuai, 1992 )
-
15 20 Settlemmt (mm)
Figure 4 4 Load-senlement curves of pile Tl with dEerent
confhing pressures (second group of tests)
Figure 4-5 Load-seulement curves of pile S with different
confining pressures (second group of tests)
-
-1.4 - a: O kPa confining pressure
c: JO kPa confining pressure
b: 20 kPri confining pressure
Load (kN)
- - * - -5.OkN (Tl) - - * - -1O.OW~l)
-1.4 1 d: 60 kPa confining prrsswe
Figure 4 6 Load distribution along the pile with different
values of load applied at pile heads of T 1 and T2 (first group of
tests)
-
Load (kN) O 10 20 30
a: 0 kPa cotirinirig pressure
c: JO kPa confirring pressure
b: 20 kPa coiifiiiitig prcssurc
Load (kN) O 20 40 60
d: 60 kPa conlining pressure
Figure 4-7 Load distribution dong the pile with different values
of load appiied at pile heads of T 1 and S ( second group of
tests)
-
Load (kN) O 1 O 20 30 40
- - * - -2û.O kN(T1) - - * - -30.0kN(T1)
-1.4 -
e: 80 kPa coiifinirig prcssiirc
Figure 4-7 (continued)
Load (kN) O 20 40 60
-
Unit load transfer (kPa)
O 2 4 6 8 10 12 14
a: O kPa c o ~ i i i n g pressure
Unit luad transfer (kPa)
O 20 40 60 80
- 1
c: 40 kPa confining pressure
Unit load transler (kPa)
O I O 20 30 40 50 60
J
b: 20 k f a conliniiig pressure
Unit load transfer (kPa)
-1 J
d: 60 kPa confining pressure
Figure 4-8 Unit load transfer to the soi1 when ultimate load was
reached at piles T 1 and T2 (first group of tests)
-
Unit load transfer(kPa)
O 5 1 O 15 M
Unit load transfer (kPa)
O M 40 60
c: 40 kfa confiiiing pressure
Unit load transfer (kPa)
O 10 20 30 40
Unit load transfer (kPa)
0 2 0 4 0 6 0 8 0 1 0 0
d: 60 kPa confining pressure
Figure 4-9 Unit load transfer to the soi1 when ultimate load was
reached at piles Tl and S (second group of tests)
-
Unit load transfer (kPa)
O 20 40 6 0 8 0 1 0 0 1 2 0 1 4 0
C
e: 80 );Pa confining prcssurc
Fisure 4-9 (continued)
Unit k d transfer (kPa)
O 20 40 60 80 1 0 0 1 2 0 1 4 0
-
Unît ioad transfu ( kPa)
10 20 30 40 50
Unit ioad transfer (kPa)
1 O 20 30
b: Tapcred pile. T2
Figure 4- 10 Variation of unit load transfer to the 5011 curves
with piles Tl and T2 at different confining pressure applied (first
group of tests)
-
Unit load tnnsfer (kPa) O 10 20 30 40 50 60
-1 J
a: Tapercd pilc. Tl
Unit load transfer (kPa) O 10 20 30 40 50 60 70
-1 J
b: Straiglitaided wall pile. S
Figure 4- 1 1 Variation of unit load transfer to the soi1 curves
with piles TI and S at different confining pressure applied (second
group of tests)
-
O 1 O 20 30 40 50 60 70 80 90 100 Confining pressure (kPa)
Figure 4- 12 The distribution of the ultimate load between the
pile point and the pile shafl for various applied confining
pressures (second group of tests)
-
CHAFTER 5
UPLIFT BEKAVIOUR OF TAPERED PILE
5.1 INTRODUCTION
In practice, a working pile is not aiways subjected to a
compressive load. Piles
supporting transmission towers and jetty structures have to
resist uplifi loads.
Tapered piles have a substantial advantage with regard to their
load-carrying
capacity in the downward frictional mode. The uplift performance
of tapered piles has not
been fully unùerstood. This chapter describes the results of the
experimental investigation
into the characteristics of the uplift performance of tapered
piles. The observations
include the load-displacement behaviour, ultimate uplifi load,
ratio of uplift to
compressive !oad and load -fer patterns.
5.2 TESTING PROCEDURE
The pile axial compressive loading test started after the
installation procedure was
completed as described in Chapters 3 and 4. The pile was fîrst
loaded ciownward with
zero applied confinhg pressure. After the downward axial loading
was completed, a
pullhg jack was set and al1 the instruments were reset to zero,
the uplifi test was
conducted. The testing procedure and readings for the axial
pullout tests were similar to
those for axial compression tests described in Chapter 4 except
that the load was applied
in tension. Each loading increment was about 10% of the expected
pile uplift capacity
until 15 mm upward pile movement was attahed or the failure
(significant change in
displacement due to a small load increment) occurred first.
-
5.3 TESTING RESULTS
5.3.1 Upiift Load-Disphcement
The load applied at the pile head and the displacement of the
pile head were
measured during the loading tests and plotted in Figures 5-1 and
5-2 for the first and
second group of tests, respectively. Figures 5-1 (a) and 5-2 (a)
show the load-
displacement cuves at different values of confining pressure for
tapered pile, Tl , in the
fkst and second sets of tests, respectively. Figure 5-1 @) shows
the load-displacement
curves for tapered pile, T2, and Figure 5-2 (b) shows the
load-displacement curves for
straight-sided wall pile, S. It can be noted fiom the figures
that the pile's uplift capacity
increased due to the increase in the confining pressure. It may
also be noted that the piles
with larger taper angle, Tl, displayed a softer response
manifested by larger
displacements at the sarne load level, except for initial
loading which was afTected by the
res idd stresses as discussed later.
5.3.2 UItimate Uplitt Load
The ultimate pullout Ioad for each pile was detemiined fiom the
load
displacement Cumes. The fdure load of a pile was considered to
be the load that resulted
in 6.25 mm upward movement. The results were compared in Table
5-1 based on the pile
uplift capacity ratio, KP, and the net uplift capacity ratio,
KF$J. The ratio KP was defined
as the ratio of the uplifi capacity of the two investigated
piles, while the ratio KPN was
defmed as the ratio of the net uplift capacity (pile upiift
capacity subtracted by pile-self
weight) of the two piles. The ratio of the net uplift to push
down shaft capacity for the
same pile under the same confining pressure was also obtained.
The results of the fkst
-
group of tests, piles Tl and T2 in loose sand, and the resulu of
the second group of tests,
piles T 1 and S in medium dense sand, are given in Tables 5- 1
and 5-2, respectively.
5.3.2.1 Loose sand status
It can be observed fiom Table 5-1 that pile Tl displayed lower
uplift capacity
manifested in values of KP and KPN lower than 1, for dl values
of confining pressure.
The uplift capacity of pile Tl is 7 to 12% lower than the uplifi
capacity of pile T2 for the
given values of confining pressure, with the lowest capacity at
confining pressure e q d to
20 and 40 kPa. The same trend was observed in KPN. The ratio of
net uplift capacity to
push down shaft capacity for Tl varied between 41% at zero
confining pressure to 33% at
40 kPa, while it varied fiom 66% at zero confhing pressure to
46% at 40 kPa for T2.
These values suggested that this ratio was less for piles with a
larger taper angle and
higher confming pressure.
5.3.2.2 Relatively medium dense sand status
It can be seen fiom Table 5-2 that the axial uplift capacity of
the tapered p lile was
lower than the axial uplift capacity of the straight-sided wall
pile for al1 values of
confining pressure. However, the difference was insignificant,
especially at higher
confining pressure. The ratio KP varied between 0.86 to 0.98,
with the maximum value
occurring at a confining pressure of 20- 40 kPa The ratio KPN
varied between 0.83 to
0.99, with the maximum value occuning at a confining pressure of
20 kPa. The
cornparison between the renùts of the two sets of tests
suggested that the eEect of the
taper angle on the uplifi capacity of prototype piles installed
in dense sand wodd be
-
small, especially for longer piles (as the confining pressure
increases with depth). The
variation of the ratio of net uplift capacity to push down shaft
capacity was small in the
case of the straight-sided wall pile, S, mtween 59% and 70%)
while a larger variation
was calculated for the tapered pile, Tl, (between 37% and 58%).
A cornparison with the
results of the first set of tests suggested that this ratio was
higher for piles installed in
dense sand.
5.3.2.3 Effect of pile placement method
The ratios of net uplift capacity to push down sh& capacity
in this study were
lower than the results referred by Nicola and Randolph (1 993).
In their study, the ratio of
tende and compressive shaft capacities varied with an average of
about 0.7 for piles
driven into sand. In the current study, the pile was placed in
the centre of the soi1
chamber and the sand was then poured around it, resulting in no
densification due to the
method of placement. In the case of pile driving, the soi1 is
displaced and the sand
becomes denser in the close vicinity of the pile, and
consequently, the ratio of uplifi
capaci~j to pushdown capacity becomes higher. Levacher and
Sieffert (1984) investigated
the axial performance of piles installed in sand. They concluded
that the placement
method had a significant effect on the axial performance of
piles.
5.3.3 Pile Head Stiffness
The eflect of confinhg pressure on the pile head e e s s is
illustrated in Figure
5-3. Severai observations can be made fiom this figure. FirstIy,
as the connning pressure
increased both the initiai and secant stifniess of ail piles
increased, as expected. However,
-
the inmease in the secant *ess (at ultimate load) was much more
significant.
Secondly, piles with smaller taper angle, Tï, in the first set
of tests and S in the second
set, had smaller values of initial stiffhess and higher values
of secant &ess. The higher
initial stifhess values of piles with larger taper angle may be
attributed to lower rrsidual
stresses developed during the pushdown loading tests. However,
as the pullout loading
continued, the residual stresses were dissipated and piles with
smaller taper angle
displayed higher secant stifniess values. This behaviour was
more evident in the second
group of tests because more significant residual stresses were
developed due to
application of higher loads to piles in dense sand. Thirdly, the
secant stifhess was 15-
20% of the initial stiffhess for al1 piles, which represented a
highly nonlinear behaviour in
this loading mode. This nonlinearîty was more pronounced in
loose sand and piles with a
larger taper angle.
5.3.4 Load Distribution
The forces transmitted at different locations were calculated
from strain gauge
readings as
4, =E, E A , (5-1)
where q, is the pile axial load at the location of main gauge i,
E, is the main mesurement
of strain gauge i, E is the elastic modulus of the pile material
and Ai is the pile cross-
sectional area at the location of strain gauge i.
Figures 5 4 (a) (b) and 5-5 (a) (b) show the load distribution
dong the piles under
various load incrernents (given as ratios of the ultimate uplift
load, P.) for the first and
second groups of tests, respectively. It may be observed from
both figures that the
-
general trend of the load distribution was the same for ail
three piles at ail loading
increments. It can aIso be noted that the load was transferred
to the soi1 gradually except
for a distinct change close to the pile tip. This change was the
result of residual stresses
developed during the downward loading test that was performed
before the uplift loading
test. The presence of the residual stresses was evident fiom the
compressive stresses
shown near the pile tip. It may also be noted that the residual
stresses were more
pronounced in piles installed in the dense sand.
5.3.5 Shaft Friction
The shaft friction of the pile was calculated from the strain
measurements during
the pile test. The readings of the strain gauges recorded when
the ultimate uplift load was
applied were used to calculate the load distribution dong the
pile. The difference
between the force calculated fiom any two sets of strain gauges
dong the pile wall minus
the corresponding weight of the pile represented the total load
transferred to the
surrounding soi1 between the two points. Dividing this value by
the conespondhg
surface area. the average of shaft fiction was obtained as
wheref, is the average shaft friction between stations i and j ,
S, is the surface area of the
pile between stations i and j, and Wg Ys the pile weight between
stations i and j. The shaft
friction curve was obtained fiom the values of shaft fiction
dong the pile.
Figure 5-6 shows the comparison of the shaft fiction for piles
Tl and T2 in the
first group of tests, while Figure 5-7 shows the comparison
between Tl and S in the
-
second group. Both figures illustrate that the shaft friction
was slightly lower for piles
with larger taper angle. Comparing the shaft fiction for the
tapered pile Tl in the two
sets of tests, the effect of initiai sand density on the shaft
fiction could be observed. As
expected, the pile in the dense sand had a higher shaft fiction,
however, this effect was
Iess significant at higher confining pressure.
The shafl friction curves at different values of confuiing
pressure are shown in
Figures 5-8 and 5-9 for piles Tl and T2 in the first group of
tests, and Tl and S in the
second group of tests, respectively. Both figures show that the
intensity of the shaft
fnction increased with an increase in the confining pressure.
However, the increase in the
shaft friction was less in the higher confuiing pressure range
(greater than 60 kPa). This
suggested that there was a limiting fiiction value in this mode
of loading.
5.3.6 Downward and Upiift Shaft Friction
The cornparison of shaft friction at ultimate tensile and
compressive capacity are
s h o w in Figures 5-10 (a) (b) and 5-1 1 (a) (b) for the first
and second groups of tests,
respectively. The compressive results were extracted fiom
Chapter 4. The general trend
was that for most of the pile Iength the tensile shaft fiction
was lower than the
compressive shaft fiction, but close to the pile tip the
compressive shaft fiction
decreased and the tensile fnction increased. This trend was
similar to that reported in
Nicola and Randolph (1993), where the theoretical basis for a
consistent difference in
tensile and compressive sh& capacity of straight-sided wall
piles in sand was explored.
Their work showed that there are sound reasons for expecting the
tensile shaft capacity to
be significantly lower than the compressive shaft capacity for
straight-sided wall piles in
-
fiee draining soils. The increase in the tende niction close to
the pile tip could have been
fictitious and codd be ateibuted to the presence of the residual
stresses which affected
the strain measurements used to calculate the shaft fiction
along the pile.
5.4 DISCUSSION: RESIDUAL STRESSES
The existence of residual stress has k e n known and
investigated by other
researchers such as Stewart and Kulhawy (1 98 11, Briaud and
Tucker ( 1 984) and Poulos
(1987). In the field, these stresses develop during the driving
of piles, where their value
could be significant, or as a result of the load testing of
bored piles.
During driving or downward axial loading tests, a pile moves
downward, and the
pile-soi1 friction along the shaft and the point soi1 resistance
acts upward on the pile to
resist the pile's peneûation. Mer dnving and during the
unloading that follows, the soi1
under the pile tip pushes the pile back and stresses dissipate.
However, a significant
residuai point load c m exist in the pile toe, especially with
lower confining pressure
applied since point capacity is larger and a large rnovement is
needed to unload the pile
tip, wMe little movement is needed to unload the pile shaft.
Poulos (1987) emphasised the importance of considering the
residual stresses in
the interpretation of btrumented pile loading tests. He noted
that if zero r e s i d d
stresses were assumed, only the incremental stresses and loads
were measured, and a false
picture of the shaft and toe resistance was obtained. Hence,
substantial ciifferences
appeared to exist between the skin Wction values in compression
and tension, whereas
the values were the sarne (in his opinion). He also pointed out
that the effect of residual
stresses was more signifiûant on the initial uplifi stiffness of
piles dnven in sand.
-
In the current study, the pile installation method did not
result in any residual
stresses. However, the piles were tested in compression prior to
the uplift test which
might have renilted in sorne residual stresses developing dong
the lower part of the pile.
The load transfer c w e s obtained in this study varied
considerably as s h o w in Figures 5-
6 and 5-7. This variation could be amibuted partially to the
residual stresses. However,
it is the author' opinion that the soi1 reaction to the pile
motion was inherently different in
the two loading modes, especially for tapered piles.
5.5 SUMMARY
An experimentai investigation of the axial uplifi response of
three steel piles with
different taper angles installed in sand was presented and
discussed in this chapter. The
q l i f i performance characteristics of the piles were
investigated and the following
conclusions were drawn:
1. The pile axial uplift capacity increased with an increase in
the confinuig pressure for al1
piles examined in this study;
2. The ratios of uphft to compressive load and load transfer
patterns for straight-sided
wall piles were similar to those obtained by other researchers.
These ratios were less for
tapered piles than saaight-sided wall piles since tapered piles
possessed much higher
bearing capacity and slightly less uplifi capacity;
3. The uplifi capacity of tapered piles was comparable to that
of straight-sided wall piles
at higher confining pressure values, suggesting that the
performance of actual tapered
piles (with greater length) would be comparable to that of
straight sided wall piles;
-
4. Residual stresses developed during the pushdown loading phase
and their effect were
more significant on the initial uplift capacity of piles. This
effect was more pronounced in
the case of straight-sided wail piles in dense sand.
-
Confining pressure
O S 10 15 Upward movement (nm)
O 5 10 15
Upward movement (mm)
Figure 5- 1 Load- upward movement curves of piles at different
confining pressure values in first group of tests a: Tl b: T2
Confining pressure
O kPa - o . - M kPa -_-- 40 kPa ; - * . o . - 60 kPa ---
L 80 kPa -1WkPa j
O 5 10 1s O 5 1 O 15 Upward movement (nm) Upward movcmmt
(mm)
Figure 5-2 Load- upward movement curves of piles at different
confinhg pressure values in second group of tests a: Tl b: S
-
1 lntial stiffness I
O 20 40 60
Confining pressure (kPa)
1 Secant stiffness at Pu (
Confining pressure (kPa)
a: First group of Tests
lntial stiffness I
O M 40 60 80 100 Confining pressure (kPa)
- - --
Secant stiffness at Pu
O 20 40 60 80 100
Confining pressure (kPa)
a: Second group of Tests
Figure 5-3 The effect of cunfining pressure on the uplifi pile
head st if iess a: First group of tests b: Second group of
tests
-
Load (kN)
O 0.2 0.4 0.6 0.8 Load (kN)
O 1 2 3
Load (kN)
-2 O 2 4 6
Figure 5-4 (a) Load distribution dong the pile at different load
incrernents applied at pile head of T 1 in first group of tests
-
Uplift Ioad applied at pile liead (w I
Load (kN)
O 0.2 0.4 0.6 0.8
Load (kN)
-2 O 2 4 6 8
Figure 5-4 @) Load distribution dong the pile at different load
incrernents applied at pile head of Tl in first group of tests
-
Uplifi load appIied at pile lmd (kN)
Load (kN)
-5 (W
-1 O 1 2 O S
a. O i ù ? ~
Load (kN)
I .