Statnamic testing of piles in clay 1

Statnamic testing of piles in clay

Trang 1

STATNAMIC TESTING OF PILES IN CLAY

BY

DUC HANH NGUYEN

A THESIS SUBMITTED FOR THE DEGREE OF DOCTOR OF PHILOSOPHY DEPARTMENT OF CIVIL AND STRUCTURAL ENGINEERING

UNIVERSITY OF SHEFFIELD OCTOBER 2005

Trang 2

A series of pile load tests have been carried out on an instrumented model pile installed in instrumented clay beds prepared in a 1-g calibration chamber under two stages of consolidation, i.e one dimensional and triaxial consolidation A variety of loading techniques (Constant Rate of Penetration at different rates, Maintained Load and Statnamic) have been applied during the model pile tests

On the basis of these tests, in conjunction with data from previous studies, shear rate effects in clay, i.e the enhancement of soil shear resistance under high rates of shearing are highly non-linear The available non-linear power laws for rate effects were applied to the test results to predict the static load-settlement curve from rapid load pile tests It was found that these models can give a good prediction of the ultimate static pile capacity, but they overpredict the settlement at load below the ultimate value Following this, an alternative method of deriving the static load-settlement curve from a rapid load pile test, a non-linear power law incorporating changing damping parameters, has been proposed This method was used for the model pile tests and then it was calibrated for field load tests carried out on a full size instrumented pile installed in a stiff glacial till

A simple theoretical method, which was proposed by Randolph & Wroth (1978) to establish the relationship between the pile load and its settlement for static pile loads, was modified for static pile load tests and then developed for rapid pile load tests

The gradual decrease of the pile shaft resistance after its peak value to a residual pile shaft resistance, which is known as the softening effect, plus the changes of pore water pressures and the inertial behaviour of the soil around the pile were also reported and discussed

Trang 3

The Author would like to express his deepest gratitude to his supervisors Prof Bill Anderson and Dr Adrian F.L Hyde for their advice, encouragement and constant guidance throughout this research programme, and for their valuable time and efforts in shaping the framework of this thesis Also, the Author take this opportunity to thank Prof Bill Anderson and Dr Adrian F.L Hyde for their generosity in helping me when I had a difficulty in finance at the end of the study

The Author would like to thank technical staff at the University of Sheffield, particularly Mr Paul Osborne and Mr Mark Foster for their assistance throughout the experiments

Special thanks are due to Dr Michel Brown for his guidance and advice on the laboratory experimental aspects of this work at the beginning of the research Thanks are also due to academic staff in the Geotechnical Engineering Group for their friendship Thanks are accorded to my friends for their assistance and sharing their experience

The Author would like to express his gratitude to David Lovegrove and his wife, for their support and encouragement throughout the study I feel this country is much more beautiful with their friendship

Finally, grateful thanks are extended to the Vietnamese government for providing a full scholarship that enabled the Author to conduct this research

Trang 4

TABLE OF CONTENTS

Page ABSTRACT ii

1.2 Research objectives………2

1.3 Outline of thesis……… 2

CHAPTER 2 - LITERATURE REVIEW 2.1 Introduction………4

2.2 Static load testing methods……….5

2.2.1 Maintained load test……… 5

2.2.2 Constant rate of penetration test……….6

2.2.3 Osterberg load cell test……… 7

2.3 Rate effects……….8

2.3.1 Rate effect studies using triaxial tests and torsion tests……….9

2.3.2 Rate effect studies using direct shear tests……… 11

2.3.3 Rate effect studies using penetrometer and shear vane tests………… 13

2.3.4 Rate effect using a model instrumented pile in a clay bed……… 15

2.3.5 Results from field studies……….16

2.4 Dynamic pile load tests……….18

2.4.1 The stress wave propagation equation……….19

2.4.2 Pile dynamic resistance………20

Trang 5

2.4.3 Static pile capacity……… 22

2.4.3.1 Case method of analysis……… ……23

2.4.3.2 Signal matching method……… 23

2.4.4 Dynamic load test advantages and disadvantages………26

2.5 Statnamic load test………26

2.6 Statnamic data interpretation………28

2.7 Quake values for shaft and toe resistances and the softening effect………….32

2.8 The changes of pore water pressure during pile installation and the subsequent loading stages…… ……… ………… 37

3.5 Instrumented model pile……… 69

3.5.1 Pile tip component……… 69

3.5.2 Pile shaft sleeve component……… ………… 71

3.5.3 Actuator - Pile connection………72

3.5.4 Pile shaft load cell performance……… ……73

3.6 Servo-hydraulic loading system……… 73

3.7 Logging and control system……….75

3.8 Instrumentation calibration……… 76

3.9 Testing procedure……….78

3.9.1 Constant rate of penetration tests……….78

3.9.2 Statnamic tests……….79

Trang 6

3.9.3 Maintained load tests……… 80

3.10 Bed dismantling……… 80

CHAPTER 4 - TESTING PROGRAMME 4.1 Introduction………101

4.2 Clay bed preparation and transducer locations……… 102

4.3 Constant rate of penetration tests (CRP tests)………103

4.4 Statnamic tests (STN tests)……….104

4.5 Maintained load tests (ML tests) … ………105

CHAPTER 5 - BED PROPERTIES 5.1 Introduction………114

5.2 Clay bed 1-D consolidation………114

5.3 Clay bed isotropic triaxial consolidation ………117

5.4 Performance of the calibration chamber during the pile load tests…………117

5.5 Bed properties after the testing programme………119

CHAPTER 6 – PILE TEST DATA AND DISCUSSION 6.1 Introduction………139

6.2 Typical results of the pile load tests……… 139

6.3 Pile shaft resistance results and models for the pile shaft resistance……… 140

6.3.1 Non-linear models……… 141

6.3.2 A new non-linear model for pile shaft rate effects……….145

6.3.3 Pile shaft softening effect……… 150

6.3.4 Repeatability of the static pile shaft resistances……….152

6.4 Pile tip resistance results……….153

6.5 Application of the proportional exponent model to the pile total load…… 157

6.6 A simple theoretical approach for the load transfer mechanism………158

6.6.1 Available models for load transfer……….158

Trang 7

6.6.2 Modifications to the existing models for load transfer for static

pile load tests and a new model for rapid load pile tests… … 160

6.6.3 Application of the models to static pile load tests.………167

6.6.4 Application of the models to rapid load pile tests… ………168

6.6.5 Quake value for the pile shaft resistance of a rapid load test………….170

6.7 A comparison between maintained load tests and CRP tests……….172

6.8 Pore water pressures around the pile during pile load tests………173

6.8.1 Pore water pressures during CRP tests at a rate of 0.01mm/s…………174

6.8.1.1 Pore water pressures at the pile shaft………174

6.8.1.2 Pore water pressures around the pile shaft………175

6.8.1.3 Pore water pressures at the pile tip………176

6.8.1.4 Pore water pressures below the pile tip……….176

6.8.2 Pore water pressures during maintained pile load tests……….177

6.8.3 Pore water pressure regime during rapid load pile tests………178

6.8.3.1 Pore water pressures at the pile shaft………178

6.8.3.2 Pore water pressures around the pile shaft………178

6.8.3.3 Pore water pressures at the pile tip………179

6.8.3.4 Pore water pressures below of the pile tip……….179

6.9 Clay bed inertial behavior……… 179

CHAPTER 7 - FIELD LOAD TESTS 7.1 Introduction……… …… 254

7.2 Ground conditions……… ………… ……… 254

7.3 Pile tests……… ………255

7.4 Prediction of the pile static capacity using the Unloading Point Method… 255

7.5 Application of the analyses to field tests…… ……….…… 257

CHAPTER 8 - CONCLUSIONS AND RECOMMENDATIONS FOR FURTHER WORK 8.1 Introduction……… ……… 269

Trang 8

8.2 Main conclusions……… ……… 269 8.3 Recommendations for further studies………… ……… 273

Trang 9

Table 3.5 Material properties

Table 4.1 Testing programme for Bed 1 Table 4.2 Testing programme for Bed 2 Table 4.3 Testing programme for Bed 3 Table 4.4 Testing programme for Bed 4 Table 4.5 Testing programme for Bed 5

Table 5.1 Volume of water expelled during 1-D consolidation Table 5.2 3-D consolidation degrees of Beds 1 to 5

Table 5.3 Undrained shear strengths of Bed 1 determined by hand vane tests Table 5.4 Undrained shear strengths of Bed 2 determined by hand vane tests Table 5.5 Undrained shear strengths of Bed 3 determined by hand vane tests Table 5.6 Undrained shear strengths of Bed 4 determined by hand vane tests Table 5.7 Undrained shear strengths of Bed 5 determined by hand vane tests Table 5.8 Moisture contents of Bed 1

Table 5.9 Moisture contents of Bed 2

Trang 10

Table 5.10 Moisture contents of Bed 3 Table 5.11 Moisture contents of Bed 4 Table 5.12 Moisture contents of Bed 5

Table 5.13 Shear strengths from undrained triaxial tests

Table 6.1 Static pile shaft resistance of Beds 2 to 5 Table 6.2 Pile tip loads for tests in Bed 1

Table 6.3 Pile tip loads for tests in Bed 2 Table 6.4 Pile tip loads for tests in Bed 3 Table 6.5 Pile tip loads for tests in Bed 4

Table 6.6 The influence of initial effects in the calculation for the pile settlement

Table 7.1 Soil properties from laboratory tests for Grimsby clay Table 7.2 Grimsby soil description

Trang 11

Figure 2.8 (a) Schematic of the test arrangement (b) geometry of penetrometer for side friction tests

Figure 2.9 Undrained peak strength measured from vane tests Figure 2.10 Slow and quick-penetration tests

Figure 2.11 Shaft resistances and pile movements

Figure 2.12 Wave propagation in a bar produced by an impact load

Figure 2.13 Idealization of a pile as an elastic rod with soil interaction at discrete nodes

Figure 2.14 Model of downward and upward waves due to soil interaction Figure 2.15 Smith Model for pile and soil

Figure 2.16 Randolph & Deeks model for pile shaft and soil Figure 2.17 Randolph & Deeks model for pile tip and soil Figure 2.18 A typical statnamic loading-time relationship Figure 2.19 Statnamic device

Figure 2.20 Forces acting on a pile during statnamic loading Figure 2.21 Unloading point method

Trang 12

Figure 2.22 Load Settlement response

Figure 2.23 Shaft quake values compared with the pile diameter

Figure 2.24 Ramberg-Osgood model for the relationship of shaft resistance and displacement

Figure 2.25 Idealised softening behaviour for a pile in clay Figure 2.26 Chandler & Martins’ test apparatus

Figure 2.27 Strain Path Method to deep penetration in clays

Figure 3.1 Strain distributions during pile installation according to the strain path method

Figure 3.2 Pumping slurry to the consolidometer Figure 3.3 The consolidometer

Figure 3.4 Miniature Druck Transducer

Figure 3.5 Transducer arrangement in the calibration chamber Figure 3.6 Hole arrangement at the bottom plate

Figure 3.7 The accelerometer and its protection Figure 3.8 1-D consolidation in the laboratory Figure 3.9 Schematic diagram of 1-D consolidation Figure 3.10 Loading plate and its o-rings in the laboratory Figure 3.11 Calibration chamber volume change units

Figure 3.12 Removing the consolidometer after the finish of 1-D consolidation Figure 3.13 The calibration chamber sand retaining ring and its arrangement

Figure 3.14 The triaxial calibration chamber membrane and drainage sand layer at the top of the clay bed

Figure 3.15 The calibration chamber top plate and its attached membrane Figure 3.16 Top plate arrangement during 3-D consolidation

Figure 3.17 Schematic diagram of 3-D consolidation

Trang 13

Figure 3.18 Using the casing tube and auger to make a hole in the bed for pile installation

Figure 3.19 Schematic diagram of 3-D consolidation after pile installation Figure 3.20 Schematic diagram of the instrumented model pile

Figure 3.21 The pile tip load cell

Figure 3.22 The pore water transducer at the pile tip Figure 3.23 The pile shaft load cell

Figure 3.24 The pore water transducer at the pile shaft

Figure 3.25 Schematic diagram of the connection between the loading system and the pile for CRP and Statnamic tests

Figure 3.26 The connection between the loading system and the pile for CRP and Statnamic tests

Figure 3.27 Typical calibration results of a pore water pressure transducer Figure 3.28 Input loading pulse and actual loading pulse for a statnamic load test Figure 3.29 Schematic diagram of the connection between the loading system and the pile for maintained load tests

Figure 3.30 The connection between the loading system and the pile for maintained load tests

Figure 3.31 The clay bed when the tests had finished

Figure 3.32 Carrying out hand vane tests and taking the samples for triaxial tests

Figure 4.1 Transducer arrangement for Bed 1 Figure 4.2 Transducer arrangement for Bed 2 Figure 4.3 Transducer arrangement for Bed 3 Figure 4.4 Transducer arrangement for Bed 4 Figure 4.5 Transducer arrangement for Bed 5

Trang 14

Figure 5.1 Bed settlements during 1-D consolidation

Figure 5.2 Pore water pressure distribution during 280 kPa 1-D consolidation of Bed 1

Figure 5.7 The final transducer locations of Bed 1 Figure 5.8 The final transducer locations of Bed 2 Figure 5.9 The final transducer locations of Bed 3 Figure 5.10 The final transducer locations of Bed 4 Figure 5.11 The final transducer locations of Bed 5

Figure 5.12 Fluctuation of top and side cell pressures during a rapid pile load test Figure 5.13 Changes of pore pressures in the clay bed due to the drop of the top cell pressure

Figure 5.14 Changes of pore pressures in the clay bed due to the drop of the top cell pressure over a period of 200ms

Figure 6.1 Load-settlement curves for a CRP test at a rate of 0.01mm/s (B2/7/CRP-0.01)

Figure 6.2 Load-settlement curves and pile penetration and velocity with time for a CRP test at a rate of 200mm/s (B2/6/CRP-200)

Figure 6.3 Load, settlement, pile velocity, and pile acceleration variation with time for a statnamic pile load test (B2/9/STN-35)

Figure 6.4 Load – settlement curve and load and settlement variation with time for a maintained load test (B2/13/MLT)

Figure 6.5 Skin friction load cell values for tests B1/1/CRP-0.01 and B1/2/CRP-100 Figure 6.6 Skin friction load cell values for tests B2/4/CRP-0.01 and B2/7/CRP-100

Trang 15

Figure 6.7 Skin friction load cell values for tests B3/21/CRP-0.01 and B3/18/CRP-100

Figure 6.8 Skin friction load cell values for tests B4/5/CRP-0.01 and B4/2/CRP-100 Figure 6.9 Skin friction load cell values for tests B5/15/CRP-0.01 and

Trang 16

Figure 6.24 Application of Balderas-Meca’s model for pile load tests in Bed 5 (B5/15/CRP-0.01 and B5/14/CRP-100)

Figure 6.25 Application of Equation 6.6 for the ultimate pile shaft resistance (Bed 2)

Figure 6.29 Application of Equation 6.6 for the ultimate pile shaft resistance (Beds 1 to 5)

Figure 6.30 Application of the proportional exponent soil model for pile load tests in Bed 1 (B1/1/CRP-0.01 and B1/2/CRP-100)

Figure 6.31 Application of the proportional exponent soil model for pile load tests in Bed 1 (B1/1/CRP-0.01 and B1/4/STN-15)

Trang 17

Figure 6.42 Application of the proportional exponent soil model for pile load tests in Bed 3 (B3/1/CRP-0.01, B3/6/CRP-0.01 and B3/3/CRP-25)

Figure 6.46 Application of the proportional exponent soil model for pile load tests on Bed 3 (B3/6/CRP-0.01, B3/15/CRP-0.01 and B3/8/STN-30)

Figure 6.48 Application of the proportional exponent soil model for pile load tests in Bed 3 (B3/6/CRP-0.01, B3/15/CRP-0.01 and B3/10/STN-35)

Figure 6.49 Application of the proportional exponent soil model for pile load tests in Bed 3(B3/6/CRP-0.01, B3/15/CRP-0.01 and B3/12/CRP-250)

Figure 6.54 Application of the proportional exponent soil model for pile load tests on Bed 4 (B4/5/CRP-0.01 and B4/2/CRP-100)

Figure 6.57 Application of the proportional exponent soil model for pile load tests in Bed 4 (B4/5/CRP-0.01, B4/8/CRP-0.01 and B4/7/STN-30)

Trang 18

Trang 19

Figure 6.82 Comparison between the new model and Randolph & Deeks model (B4/4/CRP-0.01 and B4/5/CRP-200)

Figure 6.83 Comparison between the new model and Randolph & Deeks model (B4/10/CRP-0.01 and B4/11/STN-35)

Figure 6.84 Post peak softening effects in CRP tests at a penetration rate of 0.01mm/s (Bed 1)

Figure 6.85 Post peak softening effects in CRP tests at high penetration rates (Bed 1) Figure 6.86 Post peak softening effects in CRP tests at a penetration rate of 0.01mm/s (Bed 2)

Trang 20

Figure 6.93 Post peak softening effects in CRP tests at high penetration rates (Bed 5) Figure 6.94 Pile tip load-settlement curves (B4/5/CRP-0.01 and B4/16/CRP-125) Figure 6.95 Pile tip load cell and its working mechanism

Figure 6.96 Application of the proportional exponent soil model for the total pile load (B1/1/CRP-0.01 and B1/4/STN-15)

Figure 6.97 Application of the proportional exponent soil model for the total pile load (B2/12/CRP-0.01 and B2/9/STN-35)

Figure 6.98 Application of the proportional exponent soil model for the total pile load (B3/13/CRP-0.01 and B3/17/CRP-200)

Figure 6.101 Application of the proportional exponent soil model for the total pile load (Brown, 2004 data)

Figure 6.102 A simple model for the vertical soil deformation

Figure 6.103 Application of the linear and non-linear models for pile shaft load transfer mechanism (B2/1/CRP-0.01)

Figure 6.104 Application of the linear and non-linear models for pile shaft load transfer mechanism (B2/7/CRP-0.01)

Figure 6.105 Application of the linear and non-linear models for pile shaft load transfer mechanism (B2/6/CRP-200)

Figure 6.106 Application of the non-linear models for pile shaft load transfer mechanism (B2/11/CRP-300)

Figure 6.107 Application of the non-linear models for pile shaft load transfer mechanism (B4/3/CRP-150)

Figure 6.108 Load-settlement curves for the CRP pile test at the rate of 0.01mm/s and the maintained pile load test (B2/12/CRP-0.01 and B2/13/MLT)

Figure 6.109 Load-settlement curves for the maintained pile load tests (B2/13/MLT and B2/14/MLT)

Trang 21

Figure 6.112 Changes of pore pressure measured the transducer at the pile shaft during CRP tests at the rate of 0.01mm/s

Figure 6.113 Changes of pore pressure measured by the transducers around the pile shaft during CRP tests at the rate of 0.01mm/s

Figure 6.114 Changes of pore pressures measured by the transducer at the pile tip during CRP tests at the rate of 0.01mm/s

Figure 6.115 Changes of pore pressure measured by the transducers below the pile tip during CRP tests at the rate of 0.01mm/s

Figure 6.116 Excess pore pressures during a maintained load test (B2/13/MTL) Figure 6.117 Changes of pore pressure measured by the transducer at the pile shaft during rapid pile load tests

Figure 6.118 Changes of pore pressure measured by the transducers around the pile shaft during rapid pile load tests

Figure 6.119 Changes of pore pressure measured by the transducer at the pile tip during rapid load tests

Figure 6.120 Changes of pore pressure measured by the transducer below the pile tip during a rapid pile load test

Figure 6.121 Measured clay bed accelerations (B2/9/STN-35) Figure 6.122 Measured clay bed accelerations (B3/10/STN-35)

Figure 7.1 Borehole record (T.L.P Ground Investigations) Borehole 1 Figure 7.2 Borehole record (T.L.P Ground Investigations) Borehole 2 Figure 7.3 Information from SPT, CPT and SCPT at Grimsby

Figure 7.4 Statnamic field test result

Figure 7.5 Constant rate of penetration and maintained load test results

Figure 7.6 Comparison between the Unloading Point Method and static pile load test results

Figure 7.7 Comparison between the proportional exponent model and static pile load test results (Equation 7.6)

Trang 22

Figure 7.8 Comparison between the proportional exponent and proportional multiplier models (B4/4/CRP-0.01 and B4/5/CRP-200)

Figure 7.9 Comparison between the proportional exponent and proportional multiplier models (B4/10/CRP-0.01 and B4/11/STN-35)

Figure 7.10 Comparison between the proportional multiplier model and static pile load test results (Equation 7.8)

Figure 7.11 Comparison between the proportional multiplier model and the proportional multiplier model

Figure 7.12 The sensitivity of the model to PSTNUltimate

Trang 23

A Skempton’s pore pressure coefficient [-] Ab Pile tip cross sectional area [m2]

Ap Pile cross sectional area [m2]

B Skempton’s pore pressure coefficient [-]

Cb Dashpot constant in Randolph & Deeks model for a pile tip [kNs/m] D Pile diameter [m]

E Pile elasticity modulus [kN/m2] F Applied load [kN]

Fa Pile inertial force [kN]

Fd Downward force at the pile tip in a dynamic load test or damping resistance in The Unloading Point Method [kN]

Fd1 Downward travelling force before the interaction with pile shaft resistance in a dynamic pile test [kN]

Fd2 Downward travelling force after the interaction with pile shaft resistance in a dynamic pile test [kN]

Fo Initial downward force in a dynamic pile test [kN]

Fr Net force at a time of t = 2L/c later than the time of obtaining Fo in a dynamic test [kN]

Fs Static pile resistance in The Unloading Point Method [kN] Fsoil Total pile load in the Unloading Point Method [kN] FSTN Applied load in a statnamic pile test [kN]

Fu Reflected upward force at the pile tip in a dynamic load test [kN] Fu1 Upward travelling force before the interaction with pile shaft resistance in a dynamic pile test [kN]

Fu2 Upward travelling force after the interaction with pile shaft resistance in a dynamic pile test [kN]

Fur Return upward force at a time t = 2L/c later than the time at which the initial downward force was obtained in a dynamic test [kN] G Soil shear modulus [kN/m2]

Trang 24

Ib Influence coefficient in Randolph and Wroth model to calculate pile base settlement [-]

Ip Plasticity index [-]

Ir Rigidity index of soil [G/cu] [-] Js Smith damping factor [s/m]

JT Gibson & Coyle damping factor [(s/m)N]

Kb Spring stiffness in Randolph & Deeks model for a pile tip [kN/m] Kh Lateral pressure coefficient [-]

KJ Janbu constant [-] L Pile length [m] M Pile mass [tonne]

Mb Lumped mass in Randolph & Deeks model for a pile tip [tonne] N Gibson & Coyle damping coefficient [-]

Nb Pile tip bearing capacity factor [-] Nw Wave number [-]

Qb Pile tip bearing capacity [kN]

Qd Damping component at the pile tip [kN]

Qu1, Qu2 Ultimate pile capacities reached in times of failure t1 and t2 [kN] Rf Stress-strain curve-fitting constant [-]

Rs Static resistance [kN]

Rt Total dynamic resistance [kN]

Su Undrained shear strength of clay [kN/m2]

Suo Undrained shear strength corresponding to the standard peripheral velocity [kN/m2]

T Time for the stress wave to travel from a pile head to its toe and come back the pile head in a dynamic load test [s] Ts Pile shat resistance in a dynamic load test [kN]

V Volume [m3]

Z Pile impedance [m/kNs] a Acceleration [m/s2]

c Stress wave velocity in a dynamic pile test or the damping coefficient in The Unloading Point Method [m/s]

cu Undrained shear strength of soil [kN/m2]

Trang 25

jc Case damping coefficient [(kNs/m) ]

kf Final spring stiffness in Ramberg-Osgood model [kN/m] ko Initial spring stiffness in Ramberg-Osgood model [kN/m] k1 Dayal & Allen damping constant [-]

ms Order of the curve in Ramberg-Osgood model [-] n Briaud & Garland viscous exponent [-]

nJ Janbu constant [-]

pa Atmosphere pressure [kPa]

pf Load corresponding to yield point in Ramberg-Osgood model [kN] qd Dynamic deviator stress in a triaxial test [kN/m2]

qs Assumed static deviator stress in a triaxial test or pile shaft quake [kN/m2]

qb Pile base quake [m]

r Radial distance from a pile [m]

rm Radius of the influence zone around a pile shaft [m]

u Pore water pressure or the particle displacement in a dynamic pile test [kPa]

Δu Excess pore water pressure [kPa] v Velocity [m/s]

Δv Relative velocity between a pile and the adjacent soil [m/s]

vd Downward component of the particle velocity in a dynamic test [m/s] vo Reference velocity [m/s]

vod Original net velocity in a dynamic load test [m/s] vp Peripheral velocity of a shear vane test [m/s]

vpo Standard peripheral velocity of a shear vane test [m/s] vs Assumed static shearing rate [m/s]

vr Net velocity at a time of t = 2L/c later than the time of obtaining vod in a dynamic pile test [m/s]

vu Upward component of the particle velocity in a dynamic pile test [m/s] vx Temporary maximum velocity in Gibson/GRL method for a dynamic pile test [m/s]

w Displacement [m]

Trang 26

α(ε) Balderas-Meca damping coefficient [-] β Damping coefficient for soil [-]

ε Pile strain in a dynamic pile test [-] γ Shear strain [-]

γsoil Soil bulk density [kN/m3] μ Poisson’s ratio [-]

ρ Pile material density [kN/m3] or heterogeneous factor in Randolph and Wroth method to calculate pile settlement [-]

σn Applied normal stress [kN/m2] σz Vertical stress [kN/m2]

σ1 Major principal stress [kN/m2] σ2 Intermediate principal stress [kN/m2] σ3 Minor principal stress [kN/m2] τd Dynamic shear resistance [kN/m2]

τd(ultimate) Ultimate dynamic shear resistance [kN/m2] τs Static shear resistance [kN/m2]

τs(ultimate) Ultimate static shear resistance [kN/m2] τt Total friction [kN/m2]

ξ Parameter in Randolph & Wroth model to calculate pile settlement [-] CRP Constant rate of penetration

KSS Reconstituted clay

LVDT Linear displacement transducer MLT Maintained load test

STN Statnamic

Trang 27

CHAPTER 1 INTRODUCTION

1.1 Background

Normally pile foundations are used where there are no competent strata near the ground surface The pile foundations can transmit applied loads to lower competent strata via shear stress mobilised on the surface of the shaft of the pile, called pile shaft resistance, and via end bearing at the pile tip, called pile tip resistance

If a pile is founded on a firm stratum such as rock the pile load bearing capacity is derived mainly from the pile tip resistance and the pile is called an end bearing pile On the other hand if a pile is embedded in soft soils its bearing capacity is derived mainly from the pile shaft resistance and in this case the pile is called a floating or friction pile

Many attempts have been made to establish the correlations between the pile shaft and pile tip resistances and the soil properties which are determined by laboratory or in situ tests (Poulos, 1989) However, due to great uncertainties involved in predicting the pile capacity, pile load tests are often required to determine or check the pile bearing capacity

Normally three types of pile load tests (static pile load tests, dynamic pile load test and statnamic pile load tests) can be employed The details of these pile load tests will be discussed in Chapter 2

The statnamic pile load test is becoming more widespread The most widely used method to analyse statnamic data, called the Unloading Point Method (UPM), works well for piles in sand but it has been found to overpredict the static capacity for piles in clay by up to 30% (Brown, 2004) Therefore, to get a better prediction of the static pile bearing capacity from statnamic tests for piles in clay further research is necessary

Trang 28

Additionally, since pore water pressures and soil inertial behaviour around a pile during pile testing provide useful insights into the pile/soil interaction, further investigation of these values is necessary

1.2 Research objectives

The objectives of this study are:

♦ To improve and modify an existing model pile test arrangement to carry out statnamic pile tests in addition to maintained load tests and constant rate of penetration tests at different rates in clay beds prepared in a calibration chamber ♦ To investigate soil behaviour in statnamic pile load tests in comparison with that of static pile load tests

♦ To propose a soil model to predict the complete static load-settlement curve from that of a statnamic pile load test

♦ To investigate the pore water pressure and accelerations around the pile during pile testing

1.3 Outline of thesis

Chapter 2 presents a literature review of pile load tests which include the static,

dynamic, and statnamic pile load tests Special attention is given to the statnamic pile load tests and rate effects which are thought to play an important role in statnamic test

data interpretation Chapter 3 concentrates on the testing equipment and test

procedures This includes a description of the experimental setup, details of transducers and their calibration, details of an instrumented pile and data acquisition used, the clay bed preparation method and the testing procedures for the maintained load, statnamic, and constant rate of penetration tests The testing programme in addition to the details of pore water pressure transducer and accelerometer

arrangements are given in Chapter 4 Chapter 5 presents the clay bed properties

These include the behaviour of the clay beds during the one dimensional and triaxial consolidation processes, the soil moisture contents and undrained shear strengths determined by hand vane tests at different locations in the clay beds and by conventional unconsolidated undrained triaxial tests The final locations of transducers during dissection of the clay beds were measured and are reported in this

Trang 29

chapter Chapter 6 focuses on the pile load test data in addition to discussion and

analysis The pile load test data is from the maintained load, constant rate of penetration, and statnamic pile load tests which were carried in the five clay beds Several available soil models are used to check their ability to derive the complete static load-settlement curve from that of a statnamic pile load test Following this a new soil model is proposed A simple available model is also used and developed to establish the relationship between the pile settlement and its shaft resistance for both static and statnamic pile load tests The pore water pressure behaviour during maintained load, constant rate of penetration, and statnamic pile load tests is discussed The application of the new soil model to a field pile load test on a full size

pile is presented in Chapter 7 The most widely used Unloading Point Method is also

applied to the field test data and then comments about this method are given Finally, a summary of main findings from both the experiments and the data analysis are

presented in Chapter 8 Recommendations for further work are also made in this

chapter

Trang 30

Static load tests have been used for a long time They are the most reliable methods and provide a benchmark reference for other types of tests However, these methods are expensive and time consuming due to the cost and time for setting up and dismantling the loading reaction for a test For this reason, dynamic tests have been developed In the beginning, dynamic tests were not reliable and were used as an indication of the pile axial capacity during driving More recently, this method's accuracy has improved significantly by taking stress wave effects into consideration and is widely used in some countries However, it still requires a complex analysis of data The idea of combining the advantages of static and dynamic tests led to the birth of the statnamic test

Although the stress wave effects are reduced significantly in statnamic tests, to obtain the static pile capacities from the statnamic tests their results still need to be analysed Several methods have been proposed and work relatively well for sands Unfortunately, they usually overpredict bearing capacity for clays To obtain better predictions, it is necessary to carry out further research

In this literature review attention will be given to the statnamic pile test and the analysis of the test results However, the statnamic test has a close relationship with

Trang 31

both static and dynamic tests so these tests will also be introduced Therefore this chapter will include a review of the following:

♦ The main types of pile load testing, their applications, advantages and disadvantages, as well as assumptions and parameters required for test data analysis ♦ Statnamic testing, result interpretation and the influence of rate effect which is believed to play an important role in the test data and its analysis

♦ The potential formation of residual shear planes or ‘strain softening’ effects, the pile settlements at which the pile skin friction and pile toe resistance are fully mobilised

♦ The change of pore water pressure occurring during the pile installation and the subsequent loading stages

2.2 Static load testing methods2.2.1 Maintained load test

In the maintained load test (MLT) the pile is subjected to a series of static load increments and decrements Each load is maintained for a minimum specified time or until the rate of the pile settlement meets a specified criterion before applying a further increment or unloading, whichever is required (Tomlinson, 2001)

The increments of load are conventionally chosen as 25% of the working load However, when the pile is loaded to failure the chosen intervals of 25% working load might miss the ultimate capacity In these cases, intervals of 12.5% the working load or less may be chosen However, the testing time will increase when using smaller load increments In most cases the ultimate capacity may be taken as the load at which the pile penetration is equal to 10% of the diameter of the base of the pile (BS 8004:1986), or the load at which settlement continues to increase without further increment in load

A rate of movement of 0.25 mm/hour may be taken as the limiting rate for normal purposes (BS 8004:1986) Sometimes, particularly for clay soils, a limiting rate of 0.1 mm/hour or less might be used

Trang 32

The test load may be applied in one of the following ways: a) by means of a jack which obtains its reaction from kentledge heavier than the required test load; b) by means of a jack which obtains it reaction from tension piles or other suitable anchors Care should be taken to avoid the reaction system influencing the pile settlement In the case of using kentledge reaction the nearest edge of the crib supporting the kentledge stack should not be closer than 1.3 m to the surface of the pile test In the case of using anchor-pile reaction all anchor piles should be at a distance of at least three test pile shaft diameters, centre to centre, and in no case less than 2 m (BS 8004:1986)

The MLT, also called a proof load test, is usually carried out to confirm that the pile settlement under the working load can be tolerated by the superstructure Therefore, in the MLT the pile is not subjected to the pile ultimate capacity or the failure load, which causes a significant settlement without a further loading Due to the natural variation in ground conditions a test at working load is insufficient The settlement of the test pile under its working load might satisfy the settlement criterion, whereas that same load applies to the next untested pile could cause excessive settlement For this reason, the test load must be carried to some multiple of the working load (i.e 11/2 or 2 times) (Tomlinson, 2001)

The MLT results are the most reliable as the test procedure replicates well the working load under the superstructure However, the test is expensive and time consuming The loading and unloading procedure needs at least 19 h If setting up and dismantling the test are taken into account it normally takes about 5 days to one week to finish a test

2.2.2 Constant rate of penetration test

In the constant rate of penetration test (CRP), the pile is loaded in such a way that the rate of settlement is maintained at a specified value The British Standard (BS 8004:1986) recommends that the penetration rate should be about 0.75 mm/min for friction piles in clay or 1.5 mm/min for end-bearing piles in sand or gravel Some other specifications suggest slightly different penetration rates such as the ICE Piling

Trang 33

Specification (ICE, 1997) requiring a rate of 0.01 mm/s for predominantly cohesive soils and a rate of 0.02 for predominantly cohesionless soils

Contrary to the MLT, in the CRP the test only ceases when the pile is subjected to the ultimate pile capacity In most cases the ultimate capacity may be taken as the load at which the penetration is equal to 10% of the diameter of the base of the pile (BS 8004:1986) When the tested pile is very long or has a large diameter the following factors should be taken into consideration in determination of the ultimate capacity: a) the elastic shortening of the pile which for very long piles might itself reach 10% of the pile base diameter; b) the practical difficulty of loading a large pile to a settlement as great as 10% of the base diameter For such piles an assessment of the ultimate capacity may be made from the characteristics of the load-settlement relationship

The method of load application is similar to that of the MLT However, due to the pile being subjected to the failure load care should be taken with the reaction to make sure the test does not have to be aborted prematurely Therefore, this test is not suitable for a large diameter pile as very heavy kentledge loads or high-capacity anchors are required

The CRP test is relatively short compared with the MLT For instance, a friction pile in clay with a base diameter of 600 mm needs only about 1.5 hours to finish the test However, with a large pile diameter the requirement of heavy kentledge loads or high capacity anchor piles is a drawback of the test

2.2.3 Osterberg load cell test

Due to the development of drilled shaft piles with large diameters, an innovative static pile testing technique known as the Osterberg load cell test or the bi-directional static load test has been developed (Osterberg & Pepper, 1984)

The test is performed by a load cell, commonly called the Osterberg cell or the cell’ The O-cell is a hydraulically driven, high-capacity sacrificial jack-like device that is installed at the bottom of the reinforcement cage of a drilled shaft or at the tip of a driven pile (Figure 2.1) Unlike the conventional static and dynamic tests that

Trang 34

‘O-apply a compression load at the top of the pile, the O-cell loads the pile from the bottom by inflating the O-cell, which is pressurized using a hydraulic pump with the oil pressure monitored by a pressure transducer attached to the hydraulic conduit By doing this the end bearing support provides reaction for the shaft friction along the pile, and vice versa Thus the test ceases when one of the following occurs: a) the applied load reaches the ultimate side capacity; b) the applied load reaches the ultimate end bearing capacity; c) the maximum O-cell capacity is reached

During the test the movements of the top and bottom cell are measured by telltales, which are attached the top and bottom cell and extend to the pile top (Figure 2.2) The upward movement of the pile top during the test is measured by digital gauges mounted on a reference beam, which is placed over the pile head and supported by two posts driven into the ground a sufficient distance apart (i.e 10 feet or two shaft diameters, whichever is larger) to eliminate the influence of the shaft movement during the test

The O-cell test offers a number of advantages over the conventional load tests, including economy, high load capacity (by up to 150000 tons), safety, reduced working area, and the ability to separate the end-bearing and side friction components Disadvantages include the need for advance installation, sacrificial load cells, the test capacity is limited to twice the lower of the ultimate pile shaft capacity and the ultimate pile end bearing capacity, the test is not also suitable for some types of piles (Poulos, 2000) The pile shaft resistance is also assumed not to be affected by the pile directional movements, upward or downward However, Wood (2003) found that the direction of the pile movement affected both stiffness and ultimate pile behaviour

2.3 Rate effects

It has been well known that soil strength under a dynamic load is larger than that under a static one This phenomenon is usually referred as the rate effect and it is exhibited by both sands and clays However, in clays the rate-dependent response is far more complicated than that in sands (Hyde et al 2000) It is believed that three elements in clay contribute to the rate dependency of engineering properties: the pore

Trang 35

water, the particle contacts, and the water/soil-skeleton interaction (Mitchell, 1976; Pike, 1981)

The pore water in clays is a viscous fluid Therefore the higher water content the higher the viscosity of the clay

The particle contacts also exhibit viscous behaviour The contacts are formed by penetration of the mineral particle and its adsorbed water layer into the adsorbed water layer of the adjacent particle The viscosity of the adsorbed water is larger than that of the free water Deep penetration of the contact, as in the case of over-consolidated clays, would therefore lead to higher viscosity of the clay Also, clays with thick adsorbed layers, such as high plasticity clays, would exhibit higher viscosity

The water/soil-skeleton interaction will vary with varying shearing rates At low rates of shearing, the particles in the soil skeleton will have enough time to re-arrange and the shearing will occur along the path of least resistance, whereas at high rates of shearing, this re-arrangement will not happen At higher rates, the soil response is accompanied by increased dilatancy or decreased contraction, along with more negative, or less positive, pore water pressures, respectively (Muir Wood, 1990; Bjerrum et al 1958; Crawford, 1959; Whitman, 1957)

2.3.1 Rate effect studies using triaxial tests and torsion tests

Gibson and Coyle (1968) carried out triaxial tests on both sand and clay specimens at various rates of strain to compare the total dynamic resistance with the static value The sands varied in grain size and grain shape, and the clays varies in plasticity and moisture content Gibson and Coyle concluded that:

Rt = Rs + RsJTvN (2.1) where Rt is the total dynamic resistance

Rs is the static resistance JT is the damping factor v is the velocity of shearing

Trang 36

N is the parameter drawn from the test results, which was 0.18 for clays and 0.2 for sands

Balderas-Meca (2004) used triaxial tests at rates varying from 0.001mm/s to 200mm/s to investigate the dynamic response of clays The testing arrangement is shown in Figure 2.3 Two types of clay were used for the research The first type was reconstituted clay (KSS) with liquid limit (wL = 37%) and plastic limit (wP = 17%), which was created by consolidating a slurry mixture of Speswhite kaolin (50% in mass), silt (25%), and sand (25%) The second type was undisturbed glacial clay with liquid limit (wL = 20-36%) and plastic limit (wP = 12-18%), which was obtained from boreholes at a site near Grimsby, England Specimen dimensions of 100mm diameter and 200mm height were used for both clays From the study, Balderas-Meca concluded that:

1 α(ε)( )β α(ε)(0.001)β

= (2.2) where qd is the dynamic deviator stress [kPa]

qs is the assumed static deviator stress determined at a axial displacement rate of 0.001mm/s [kPa]

Δv is the axial rate of displacement [mm/s]

vo = 1000mm/s, which was used to normalized Δv β is the parameter deduced from the test (β = 0.2)

α(ε) is the damping coefficient, which was dependent on strains as shown in (Figure 2.4)

Cheng (1981) carried out torsion tests on hollow cylinder clay specimens at different strain rates The specimens were 76.2mm high with inside and outside radii of 76.2mm and 101.6mm, respectively No confining pressure was applied to the specimens and a state of pure shear was induced in the specimen by rotating one end of the specimen at a constant rate The maximum strain rate of the study was up to 15rad/s From the study, he proposed the relationship between the ultimate dynamic strength and the ultimate static strength in the form of:

⎝⎛ −+

=ss α 1 e−βγ⋅sd

Trang 37

where sd is the ultimate dynamic shearing strength [kN/m2] ss is the ultimate static shearing strength [kN/m2] α is the dynamic component [kN/m2]

β is the damping coefficient ⋅

γ is the strain rate

Both α and β were found to be dependent upon soil moisture content The dynamic component, α, was 77, 57, and 36 when the soil moisture content, w, was 30 ≤ w < 31, 31 ≤ w <32, and 32 ≤ w < 34 respectively Coefficient, β, was 0.15, 0.20, and 0.30 when the soil moisture content, w, was 30 ≤ w < 31, 31 ≤ w <32, and 32 ≤ w < 34 respectively

There were some limitations in these studies as follows:

♦ Apart from Balderas-Meca (2004) they only focused on rate effects at the ultimate resistance

♦ Due to rate effects only being considered at the ultimate resistance these rate effect models are inadequate as in statnamic tests the pile resistance develops gradually to the ultimate and the values below the ultimate are important for determining the pile working load

♦ Balderas-Meca (2004) considered rate effects with the development of shear resistance and proposed a relationship between the damping parameter and strain Different soils have different stress-strain relationships Therefore this proposal cannot be applied for a wide range of soils

♦ Further discussions about Balderas-Meca’s model are made in Section 6.3.1 where it is examined to check its capability for statnamic test result analysis

2.3.2 Rate effect studies using direct shear tests

Heerema (1979) used a flat metal plate in contact with a soil sample to simulate the response at the pile-soil interface (Figure 2.5) He used undisturbed overconsolidated clays from North sea and Belgium with undrained shear strengths ranging from 55-620kPa The normal stress between the sample and the metal plate could be varied over a wide range (about from 50kPa to 240kPa) During the test, the steel plate was moved up and down against the surface of the sample The velocities could be varied

Trang 38

from near-zero to about 1.0m/s The test arrangement allowed determination of the dynamic resistance, velocity, and normal stress From this research, Heerema concluded that:

τt = σn0.7[(-0.0041Su + 4.44)v0.2 + (0.0029Su – 0.32)] (2.3) where τt is the total friction [kPa]

v is the velocity of shearing

Su is the undrained shear strength of clay [kPa]

σn is the applied normal stress on the sheet plane [kPa]

Tika-Vassilikos (1991) carried out direct ring shear tests to investigate the shear resistance between London clay (liquid limit = 71%, plastic limit = 26%, and clay fraction (% < 2μm) = 53%) and steel at different rates (1.8mm/s, 18mm/s, and 43mm/s) The clay samples were prepared with a water content of 30.5% and an overconsolidation ratio of 2 for the experiments The ring shear apparatus used for the tests allowed shear at large displacements to be investigated It was reported that the shearing resistance between soil and steel was significantly dependent on the rate of shearing A drop of shearing resistance to a residual value after a large displacement was also observed This was reported to be attributed to the orientation of the clay particles parallel to the shear direction

Chin and Seidel (2004) used a novel test device (Figure 2.6), which was developed at Monash University, to investigate the dynamic response of the pile-soil interface A clay sample with dimensions of 555mm x 160 mm in plane and 40mm in thickness accommodated by a shear box was placed in contact with a pile section The device included two main actuators, the vertical one applying the normal stress for the sample-pile contact, the horizontal one applying the shearing stress at the pile-soil interface Chin and Seidel carried out a series of tests with three types of clay, which had low (Ip = 8%), medium (Ip = 20%), and high (Ip = 37%) plasticity To take into account the pile roughness effect, a smooth concrete pile, rough concrete pile, and a smooth steel pile were used The device allowed to be conducted tests at velocities up to 1600m/s From the study, they concluded that:

e β

= (2.4)

Trang 39

where τd is the total dynamic friction [kPa] τs is the static friction [kPa]

α is the dimensionless parameter, which is between 0.51 and 1.7 β is the parameter [s/m], which is between 2.5s/m and 3.5s/m v is the shearing velocity [m/s]

Some limitations in these studies are:

♦ The model proposed by Heerema (1979) is difficult to apply for statnamic tests as it requires undrained shear strength and applied normal stress

♦ The model proposed by Chin and Seidel (2004) seems not to work when the shearing velocity equals zero as τd/τs is zero rather than 1

♦ Chin and Seidel (2004) investigated the shearing behaviour between soil and both concrete and steel to simulate shearing between a pile and soil However, the authors only considered rate effects at the ultimate resistance

2.3.3 Rate effect studies using penetrometer and shear vane tests

Dayal and Allen (1975) investigated the dynamic resistance using a cone penetrometer which had a maximum stroke length of 610mm and the following geometry:

1 outer diameter of 35.6mm 2 cone angle of 60o

3 cone base area of 1000mm24 friction sleeve area of 15000mm25 length of 457mm

The penetrometer was instrumented to allow it to measure the interface friction and the tip resistance independently The tests were carried out with clay (liquid limit = 37%, plastic limit = 21%, Plasticity index = 16) and the arrangement of the test is shown in (Figure 2.7) The penetration speeds were 1.3mm/s up to 0.9m/s From the tests, Dayal and Allen concluded that:

τ (2.5) where τt is the total friction [kPa]

Trang 40

τs is the assumed ultimate static friction [kPa] determined at a penetration velocity of 0.1mm/s [kPa]

v is the penetration velocity [mm/s]

vs is the reference velocity assumed static velocity of 0.1mm/s k1 is the damping constant and raging from 0.17 to 0.93 (Table 2.1)

Similarly, Litkouhi and Poskitt (1980) used a penetrometer to investigate the dynamic response of a pile-soil interface at different velocities in remolded London clay (liquid limit = 70%, plastic limit = 27%, Plasticity index = 43) and Magnus clay (liquid limit = 31%, plastic limit = 17%, Plasticity index = 14) and Forties clay (liquid limit = 38%, plastic limit = 20%, Plasticity index = 18) The apparatus arrangement is shown in (Figure 2.8) Litkouhi and Poskitt used a penetrometer which had a stroke of 30cm and had the following geometry

1 outer diameter of 10mm 2 length of 260mm

3 cone angle of 120o

The instrumented penetrometer only measured the tip resistance using a load cell mounted at the penetrometer and the skin friction was derived by subtracting the tip load from the total load From their test, they concluded that:

τt = τs (1 + JvN) (2.6) where τt is the total friction [kPa]

τs is the reference velocity assumed static velocity of 0.3mm/s v is the penetration velocity [m/s]

J is the damping constant [(s/m)N]

N is the damping parameter with a value of about 0.2

Biscontin and Pestana (2001) published the results of vane shear tests which were carried out at different peripheral velocities ranging approximately from 1mm/min to 1500mm/min An artificial clay with a mixture of 72% kaolinite, 24% bentonite, and 4% type C fly ash was used for the tests The clay had a liquid limit of 115%, plastic limit 40%, plasticity index of 75% A single standard 55 mm diameter field vane was used for the tests with an aspect ratio H/D of 2, blade thickness less than 2 mm and

Định dạng
Số trang	164
Dung lượng	1,96 MB