F LIBRARY
Michigan State

University
~ A

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on Of before date due.

DATE DUE DATE DUE DATE DUE ll

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

____Ji _
‘—*n —_
I ‘I!

MSU In An Affirmative Action/Equal Opportunity Institution
chS—od

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

CYCLIC LOAD EFFECTS ON MODEL PILE BEHAVIOR IN FROZEN SAND

By

David L. Stelzer

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requiresments
for the degree of

DOCTOR OF PHILOSOPHY

Department of Civil and Environmental Engineering

1989

ABSTRACT
CYCLIC LOAD EFFECTS ON MODEL PILE BEHAVIOR IN FROZEN SAND
BY
David L. Stelzer

Small cyclic loads, 3 to 5 percent of the long-term
sustained load, superimposed on a static load will increase
settlement (displacement) rates of friction piles embedded
in frozen ground. Determination of the basic mechanisms
responsible for this increase in pile settlement rates and
a suitable theory for their prediction were the objectives
of this research.

A series of model steel piles with controlled surface
geometries (lug size, shape, and spacing) were frozen in
sand and loaded while carefully controlling test variables.
The data collected provided information on the displacement
mechanisms operating at the pile/frozen soil interface.
Pile settlement rates due to static and incremental static
loading were analyzed and compared to rate increases
resulting from superimposed dynamic loads. Variables
controlled during these tests included cyclic load
amplitude, frequency of load application, static load
magnitude, and temperature. Sand density and ice volume
fractions were controlled during sample preparation.

Experimental results showed that cyclic loads

significantly increased model pile displacement rates over

those observed for the sustained load, but were smaller
than rates inferred from published results. Typically,
superimposing a small cyclic load (with amplitude X)
produced a displacement rate increase approximately
equivalent to the rate increase produced by adding a small
static load (magnitude close to 0.6 times X) to a similar
pile. The measured displacement (creep) rates appeared to
be independent of frequency in the range of 0.1 Hz to 10 Hz
for small cyclic loads superimposed on a sustained load.
Data corresponding to frequencies which induced resonant
response in parts of the loading system were excluded from
this study.

Larger lugs (more than 2 times maximum particle size)
compressed frozen sand at their leading edge and densified
the adjacent frozen soil. Particle crushing along with
formation of a zone of material flow and/or slippage
occurred in front of and around the lugs. Ice melting at
particle contact points and water movement to adjacent
locations with lower stress accompanied this slip process.
The onset of increasing pile displacement rates (tertiary
creep) was shown to be a function of lug height and

spacing.

To my family

iv

ACKNOWLEDGEMENT

The author wishes to express appreciation to the National
Science Foundation for supporting this work under grant No. GEE-8412275.
Financial support from the Department of Civil and Environmental
Engineering at Michigan State University, arranged by Dr. Mark B.

Snyder and Dr. William B. Taylor, was appreciated.

The author is grateful to Prof. 0.3. Andersland for his excellent
guidance throughout the course of these studies. Prof. Andersland's
advice was thoughtful and helpful. Drs. Robert K. Wen, Gary L. Cloud,
and Grahame J. Larson deserve special thanks for serving on the author's
guidance committee and reviewing this dissertation.

My wife Karen and daughter Gabriel deserve special recognition

for providing their emotional support and love.

TABLE OF CONTENTS

LIST OF TABLES .......................................... X
LIST OF FIGURES ......................................... xi
LIST OF SYMBOLS ......................................... xxi
CHAPTER I. INTRODUCTION ............................... l
1.1. Frozen Soil Pile Foundations ................... l

1.2 Cyclic Pile Loading: Research Problem and

Objectives ..................................... 5
CHAPTER II. LITERATURE REVIEW ......................... 7
2.1. Mechanical Properties of Frozen Soils .......... 7

2.1.1. Effect of Frozen Soil Composition on

Behavior ............................. 7
2.1.2. Frozen Soil Creep Strength ........... 10
2.1.3. Temperature Dependence of Frozen Soil
Strength ............................. 20
2.1.4. Constitutive Equations for Frozen
Soils ................................ 23
2.1.5. Frozen Soil Strength Testing ......... 31
2.2. Response of Frozen Ground to Dynamic Loading... 32

2.2.1 Dynamically Loaded Foundations in
Unfrozen Soils ....................... 32

2.2.2. Stress-Strain and Energy Absorbing
Properties ........................... 33

2.2.3. Deformation Influenced by Cyclic
Loading .............................. 35

2.3. Adhesion and Friction Between Pile and Soil.... 39

vi

2.3.1. Mechanical Interaction Between Frozen

Soil and Pile ........................ 42

2.3.2. Pile/Soil Adfreeze Bond Estimation... 42

2.4. Review of Pile Design Methods ................. 43
2.5. Pile Settlement Estimation Equations .......... 52

2.5.1. Shear Deformation of Soil Along Pile

Shaft ............................... 52
2.5.2. Penetration Deformation of Soil at
Pile Base ........................... 56
2.5.3. Expansion of a Cylindrical Cavity by
Pile Settlement ..................... 63
2.6. Cyclic Loading of Piles and Pile Driving ...... 65

CHAPTER III MATERIALS, SAMPLE PREPARATION, EQUIPMENT,

AND TESTING PROCEDURES ................... 71
3.1. Materials ..................................... 71
3.2. Sample Preparation ........................... 72
3.3. Temperature Control Equipment ................. 78
3.4. Displacement Measurement Equipment ............ 79
3.5. Pile Loading and Load Measurement Equipment... 80
3.6. Testing Procedures ............................ 81
3.7. Model Pile Loading Procedures ................. 82
3.8. Displacement-Time Data Plotting Techniques. .. 85
3.9. Determination of Pile Settlement Equation
Parameters .................................... 87
CHAPTER IV EXPERIMENTAL RESULTS ..................... 90
4.1. Sample Inspection After Testing ............... 90
4.2. Full Duration Loads on Lugged Model Piles ..... 92

vii

4.

4.

4.

CHAPTER V

5.

.3.

4.

5.

6.

l.

.10.

Step Loads on Lugged Model Piles ..............
4.3.1. Step Load Increase Magnitudes ........
4.3.2. Repeated Step Load Increases .........

4.3.3. Rapid Adjustment to a Nearly Steady
State ................................

Short Duration Loads on Model Lugged Piles....

Multiple Short Duration Loads on Lugged Model
Piles .........................................

Threaded Rods as Model Piles ..................

DISCUSSION AND APPLICATIONS ..............
Creep Equation Parameters from Static Loads...
Pile Capacity From Straight Shaft .............
Load Capacity From Pile Lugs ..................
Temperature Dependance of Pile Capacity .......
Dynamic Effects on Model Pile Load Capacities.
Limits of Model Pile Displacement .............
Comparison to Uniaxial Compression Tests ......

Comparison to Ladanyi's (1976) Cone
Penetration Tests .............................

Large n Values from Cyclic Loads on Model
Piles .........................................

Applications to Field Piles ...................
5.10.1. Analysis of a Pile Failure ...........

5.10.2. Another Perspective on a Proposed
ASTM Standard ........................

5.10.3. Additional Pile Capacity from Driven
Piles ................................

viii

104
114

121

126

133

148
151
169
169
173
182
185
187
194

201

205

206
212

212

216

217

5.10.4. Additional Research on Static and

Cyclic Pile Loads .................... 219

CHAPTER VI SUMMARY AND CONCLUSIONS ................. 221
6.1 Summary ....................................... 221
6.2 Conclusions ................................... 222
6.2.1. Creep Settlement Parameters ........... 223

6.2.2. Dynamic Load Effect on Settlement

Rates ................................. 224
6.2.3. Pile Surface Roughness and Load
Capacity .............................. 225
APPENDIX A FIGURES FOR DETERMINATION OF TABLE 4.1
CREEP PARAMETERS ........................ 227
APPENDIX B PILE TEST DATA .......................... 267

REFERENCELIST 0.0000000000000000000000000.00.00.00.00. 315

ix

Table

Table

Table

Table

Table

Table

Table

Table

Table

Table

LIST OF TABLES

Relative Importance of Material and Field- and/or Test-
Condition Parameters on Dynamic Properties of Frozen Soils.
(after Vinson, 1978).

Creep Equation Parameters from Full Duration Tests of
Lugged Piles.

Creep Equation Parameters for Lugged Piles from Step Loads.

Displacement Rate Increases from Short Duration Load
Increases on Lugged Piles.

Threaded Pile Test Results

Displacement Rate Increases from Multiple Short Duration
Loads on Lugged Piles.

Creep Equation Parameters from Full Duration Tests of
Lugged Piles With and (Estimated) Without Resistance on
Shaft Behind Lug.

Determination of Temperature Dependent Creep Equation
Exponents. (Data From Table 4.2)

Lugged Pile Test Data at for Model Steel Piles Frozen in
Sand at -3 deg C with Particle Sizes 0.42 to 0.59 mm.

Temperature Adjustments for Parameswaren's (1985) Cyclic
Uniaxial Compression Tests.

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

.10.

.11.

.12.

LIST OF FIGURES

Temperature profile in the ground, permafrost area
(from Andersland, 1987).

Frozen sand strength mechanisms (after Ting et a1., 1983).

Compression tests with frozen Ottawa Sand (after Sayles,
1973). (a) Stress-Strain Curves; (b) Mohr Envelopes.

Schematic representation of the failure envelope for
frozen Ottawa Sand (after Ladanyi, 1981).

Mohr envelopes for creep strength of Ottawa Sand
(T - -3.85 deg C) (after Ladanyi, 1981).

Failure strain versus strain-rate for a frozen sand
(after Ladanyi, 1981).

Compressive strength vs. strain rate (after Bragg and
Andersland, 1980).

Temperature dependence of uniaxial compression strength
for various frozen materials (after Ladanyi, 1972).

Constant-stress creep tests (after Andersland,et a1. 1978).
(a) Creep curve variations.

(b) Basic creep curve.

(c) True strain rate versus time.

Creep curve for step loading (after Ladanyi, 1972).

Damping ratio of frozen soil (after Vinson, 1978).
(a) Versus frequency. (b) Versus temperature.

Effect of cyclic stress on temperature of frozen
sand (after Parameswaran, 1985).

Typical strain relations for frozen sand
(after Trimble and Mitchell, 1982).

xi

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure
Figure

Figure

Figure

.13.

.14.

.15.

.16.

.17.

.18.

.19.

.20.

.21.

.22.

.23.

Idealized plate in contact with ice and frozen soil

(after Weaver and Morgenstern, 1981a).

(a) Air trapped during sample preparation allows slip at
ice/plate interface.

(b) Interaction between plate protrusions and soil
grains.

(c) Shear plane developes in ice.

Vyalov relationship between failure stress and time
(after Neukirchner, 1988).

Load test of steel pipe pile
(after Linell and Lobacz, 1980).

Long-term cohesion of frozen soils
(after Weaver and Morgenstern, 1981b).

Design chart for friction piles in frozen Ottawa Sand
(after Weaver and Morgenstern, 1981b).

Design chart for friction end-bearing piles in frozen
Ottawa Sand (after Weaver and Morgenstern, 1981b).

Uncoupling of effects due to pile shaft and base
(after Weaver and Morgenstern, 1981b).

Notation for the theory of tapered piles
(after Ladanyi and Guichaoua 1985).

Vibratory frequency versus force amplitudes for model
piles embedded in unfrozen sand (after Schmid 1969).

Wave form of model piles at various resonance
frequencies (after Schmid 1969).

Typical creep curve showing the displacement of wood pile
in frozen clay with time (after Parameswaren 1984).

Model pile embedded in frozen sand.
Lug provides controlled surface roughness.

Model pile testing equipment.
Typical model pile displacement curves.

Typical log(time)-log(disp1acement) curves
for model pile movement in frozen sand.

Primary-creep parameters from time-creep-strain
data. (after Andersland, et a1. 1978).

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

.2b.

.2b.

.3b.

.3c.

.3d.

.3e.

.6b.

.7b.

Time-displacement curves for model steel piles
with 3.18 mm lug heights in frozen sand.

Log(time)-1og(disp1acement) curves for a pile test using
various C-type starting points. (a) Test #105.

Log(time)-log(displacement) curves for Test #114
with various starting points.

Log(time)-log(displacement) curves for Test #111
with various starting points.

Time-displacement curves for model piles with 3.18 mm lug
heights in frozen sand @ -3 ° C and 25.4 mm displacement
limits. (a) With 0 mm C-type starting points.

With 1 mm C-type starting points.

With 3 mm C-type starting points.

With 4.4 mm C-type starting points.

With 6 mm C-type starting points.

Time-displacement behavior of similar piles subjected to
similar loads in frozen sand using a 1 mm C-type starting
point.

Time-displacement behavior of two similar model piles in
frozen sand subjected to similar load increases. Plots
used 5.75 mm C-type starting points.

Time-displacement behavior of a model pile before and
after a load increase. (a) Load increased after a
relatively small pile displacement.

Load increased after a relatively large displacement.
Time-displacement curves of a single pile test in frozen
sand after a load increase. (a) Using C-type starting
points.

Using B-type starting points.

Time-displacement behavior of a tapered-lug model pile
(Test #60) loaded by several steps including a
superimposed dynamic load.

Time-displacement of a tapered-lug model pile before and

after loading. (a) Static load increase at 9.73 mm
displacement of Test #60.

xiii

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

.9b.

.9c.

.10.

.ll.

.12.

.12b.

.13.

.14.

.14b.

.14c.

.15.

.16.

.17.

.18.

.19.

.20.

Static load increase at 12.96 mm displacement of Test
#60.

Static load increase at 19.76 mm displacement of Test
#60.

Comparison of load increases on a model pile (Test #60)
using C-type starting points.

Time-displacement behavior of a model steel pile (Test
#29) loaded by several steps.

Time-displacement behavior of a model steel pile before
and after loading. (a) First step load during Test #29.

Second step load during Test #29.

Time-displacement behavior of a model pile (Test #33)
frozen in a dense sand and loaded by steps including a
superimposed dynamic load.

Time-displacement behavior of a model pile (Test #33) in
dense sand before and after a load increase. (a) Step
load 2.38 mm after beginning of Test #33.

Step load added at 7.46 mm after beginning of Test #33.

Temperature increased from - 3 deg C to -2 deg C at 9.51
mm after the beginning of Test #33.

Time-displacement behavior of a lugged pile (Test #35)
shown with several superimposed dynamic loads and one
static load increase.

Time-displacement behavior of a lugged pile (Test #51) in
polycrystalline ice with a constant load.

Time-displacement curves for a model pile in frozen sand
at -3 deg C with periods of unloading.

Time-displacement behavior of a single pile (Test #92)
shown 1 mm after restoration of the static pile load.

Time-displacement curve for a model pile in frozen sand
showing the relationship for short duration load
increases VIII through XI of Test #92.

Short duration time-displacement curves for a model pile
in frozen sand, load increases VIII Through XI of Sample
#92.

xiv

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

.21.

.21b.

.21c.

.21d.

.22.

.23.

.24.

.24b.

.24c.

.25.

.26.

.27.

.28.

.29.

.29b.

.30.

Time-displacement behavior of model piles subjected to
superimposed dynamic loads. (a) Curve of tapered lug pile
(Test #60) in loose sand at -3 deg C.

Curve of 0.79 mm lug in dense sand (Test #33) at -3 deg
C.

Curve of 0.79 mm lug in polycrystalline ice (Test #35) at
-3 deg C.

Curve of 0.79 mm lug in loose sand (Test #88) at -3 deg
C.

Time-displacement behavior of a single pile (Test #123)
loaded by successively larger loads during a short total
duration.

Time-displacement of threaded rods in frozen sand.

Time-displacement behavior of threaded rods in frozen
sand. (a) Threaded rod time-displacement curves started
0.14 mm after initial loading.

Threaded rod time-displacement curves started 0.3 mm
after initial loading.

Threaded rod time-displacement curves started 0.7 mm
after initial loading.

Time-displacement behavior for 16 thread per inch piles
embedded in frozen sand.

Time-displacement behavior for 72 thread per inch piles
embedded in frozen sand.

Time-displacement behavior of a 72 thread per inch pile
embedded in polycrystalline ice and sand.

Time-displacement behavior for a 24 thread per inch pile
embedded in polycrystalline ice.

Time-displacement behavior for a 24 thread per inch pile
embedded in polycrystalline ice. (a) Static load increase
of 52%.

Static load increase of 27%.
Enlarged time-displacement behavior for a 24 threaded per

inch pile (Test #64) in polycrystalline ice showing a
displacement rate increase from a cyclic load.

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

.31.

.32.

.32b.

.l.

.10.

.ll.

.12.

.13.

.14.

Time-displacement behavior for a 24 thread per inch pile
(Test #86) in loose frozen sand with periods of
superimposed cyclic loads.

Enlarged due to a superimposed cyclic load. (a) Time-
displacement behavior after 0.65 mm displacement.

Time-displacement behavior after 0.95 mm Displacement.

Files with different lug sizes approach steady state (b-l)
at various rates.

Idealized comparison of compaction ahead of
pile lugs having different lug heights.

(8)
(b) Region

Comparison between penetration tip and pile lug.
Cavity expansion model (after Ladanyi, 1976).
of increased pressure around pile lug.

Difference in creep equation proof load results from
difference in pile lug size.

Difference in creep equation proof load shown for lugged
pile with different embedments along their straight shaft

portions.

Comparison of straight-shaft pile behaviors at various
constant displacement rates.

Best fit lines are shown to generalize peak capacity and
residual capacity of straight-shaft piles (data from
Figure 5.6).

Creep parameter n increases with decreasing lug size.

Load capacity of lugged model piles based on step—load test.

Dynamic loading effect on displacement rates (normalized)
versus frequency.

Displacement rate ratio versus displacement rate
before dynamic (1) and static (+) loadings.

Displacement rate increases due to
short duration load increases, static and dynamic.

Relationship between displacement limit and displacement
at failure.

Comparison of thread spacing to pile displacement at
failure.

xvi.

Figure

Figure

Figure

Figure

Figure
Figure

Figure

Figure

Figure
Figure
Figure

Figure

Figure

Figure

Figure

Figure

Figure

.15.

.16.

.2b.

.2c.

.4b.

.4c.

.4d.

.6b.

.6c.

.6d.

.6e.

Displacement-time curve for concrete pile in frozen soil
(Parameswaren 1982).

Pile test results-Shaw Creek site (Luscher et a1. 1983).

Time-displacement curves for model steel piles in frozen
sand at -3 deg C with 3.18 mm lug heights and 19.0 mm
(average) displacement limits.
Time-displacement curves for model piles in frozen sand at
-3 deg C with 3.18 mm lug heights, 19.0 mm displacement
limits. (a) 0 mm C-type plot starting points.

1 mm C-type plot starting points.

3 mm C-Type plot starting points.
Time-displacement curves for model steel piles in frozen
sand at -3 deg C with 3.18 mm lug heights and 76 mm
(average) displacement limits.
Time-displacement curves for model piles in frozen sand at
-3 deg C with 3.18 mm lug heights, 76 mm displacement
limits. (a) 0 mm C-type plot starting points.

1 mm C-type plot starting points.

3 mm C-Type plot starting points.

5 mm C-Type plot starting points.
Time-displacement curves for model steel piles in frozen
sand at -3 deg C with 1.59 mm lug heights and 25.4 mm
(average) displacement limits.
Time-displacement curves for model piles in frozen sand at
-3 deg C with 1.59 mm lug heights. (a) 0 mm C-type

starting points and 25.4 mm displacement limits.

1 mm C~type starting points and 25.4 mm displacement
limits.

3 mm C-type starting points and 25.4 mm displacement
limits.

6 mm C-type starting points and 25.4 mm displacement
limits.

0 mm C-type starting points and 19.0 mm displacement
limits.

xvii

Figure

Figure

Figure

Figure

Figure
Figure
Figure

Figure

Figure

Figure
Figure
Figure
Figure

Figure

Figure

Figure

Figure

Figure

.6f. 1 mm C-type starting points and 19.0 mm displacement

limits.

.6g. 3 mm C-type starting points and 19.0 mm displacement

limits.

.7. Time-displacement curves for model steel piles in frozen

sand at -3 deg C with 1.59 mm lug heights and 76 mm
(average) displacement limits.

.8. Time-displacement curves for model piles in frozen sand at

-3 deg C with 1.59 mm lug heights, 76 mm displacement
limits. (a) 0 mm C-type plot starting points.

.8b. 1 mm C-type plot starting points.
.8c. 3 mm C-Type plot starting points.
.8d. 5 mm C-Type plot starting points.

.9. Time-displacement curves for model steel piles in frozen

sand at -3 deg C with 0.79 mm lug heights and 24.1 mm
(average) displacement limits.

.10. Time-displacement curves for model piles in frozen sand

at -3 deg C with 0.79 mm lug heights, 24.1 mm
displacement limits. (a) 0 mm C-type plot starting
points.

.lOb. 1 mm C-type plot starting points.
.lOc. 3 mm C-Type plot starting points.
.lOd. 6 mm C-Type plot starting points.
.lOe. 12 mm C-Type plot starting points.

.11. Time-displacement curves for model piles in frozen sand

(0.74 to 0.84 mm grain size) at -3 deg C with 0.79 mm lug
heights and 76 mm (avg.) displacement limits.

.12. Time-displacement curves for model piles in frozen sand

at -3 deg C with 0.79 mm lug heights, 76 mm displacement
limits. (a) 0 mm C-type plot starting points.

.12b. 1 mm C-type plot starting points.
.12c. 3 mm C-Type plot starting points.

.l2d. 6 mm C-Type plot starting points.

xviii

Figure

Figure

Figure

Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure

Figure

.13.

.14.

.15.

.le.

.15c.

.15d.

.1.

.10.

.ll.

.12.

.13.

.14.

.15.

.16.

.17.

Time-displacement curves for model steel piles in frozen
sand at -3 deg C with 0.79 mm lug heights and 76 mm
(average) displacement limits.
Time-displacement curves for model steel piles in frozen
sand (0.74 to 0.84 mm) at -2 deg C with 0.79 mm lug
heights and 76 mm (average) displacement limits.
Time-displacement curves for model piles in frozen sand
at -2 deg C with 0.79 mm lug heights, 76 mm displacement
limits. (a) 0 mm C-type plot starting points.
1 mm C-type plot starting points.
3 mm C-Type plot starting points.
6 mm C-Type plot starting points.
Pile Test #20.
Pile Test #22.
Pile Test #23.
Pile Test #24.
Pile Test #25.
Pile Test #26.
Pile Test #27.
Pile Test #28.
Pile Test #30.
Pile Test #31.
File Test #32.
File Test #34.
Pile Test #38.
Pile Test #39.
Pile Test #40.

Pile Test #41.

Pile Test #42.

xix

Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure
Figure

Figure

.18.

.19.

.20.

.21.

.22.

.23.

.24.

.25.

.26.

.27.

.28.

.29.

.30.

.31.

.32.

.33.

.34.

.35.

.36.

.37.

.38.

.39.

.40.

.41.

.42.

.43.

Pile

Pile

Pile

Pile

Pile

Pile

File

File

Pile

Pile

Pile

Pile

Pile

Pile

Pile

Pile

Pile

Pile

Pile

Pile

Pile

Pile

Pile

Pile

Pile

Pile

Test

Test

Test

Test

Test

Test

Test

Test

Test

Test

Test

Test

Test

Test

Test

Test

Test

Test

Test

Test

Test

Test

Test

Test

Test

Test

#43.

#44.

#45.

#46.

#48.

#49.

#52.

#54.

#61.

#62.

#65.

#67.

#69.

#70.

#71.

#73.

#74.

#75.

#76.

#78.

#80.

#81.

#82.

#85.

LIST OF SYMBOLS

A, actual area

a, pile radius

AF’ area of shear plane

b, a creep parameter

c, material dependent constants

c, system damping coefficient

clt' long term cohesion

D, damping ratio of a soil

EO(T), a fictitious Young's modulus at a constant temperature, T

F, a function

f, a function of the type of materials and the conditions
at the pile/frozen soil interface

f(0), a function of temperature

C, shear modulus

6*, complex shear modulus

G, a function

R, system spring coefficient

k, material dependent constants

L, length of pile

Lo, initial length

M, dynamic magnification factor

m, a constant based on the pile’s surface roughness

xxi

N, normal pressure between soil and pile
n, a creep parameter

P, constant pile load

Pc’ pile proof load

PC', adjusted pile proof load

PE, pile end bearing

p., pressure induced by soil expansion

p , initial earth pressure

5 , hydrostatic pressure

P , pile capacity

Q, pile capacity due to cylinder expansion
R', universal gas constant

r, distance from pile’s center

ri, expanded cavity radius

rio’ initial cavity radius

5, a constant displacement rate

31, a fictitious instantaneous displacement

s(t), total displacement in terms of time

T, temperature

t, time
tf, time to failure
tF, time at start of tertiary creep

to, material and temperature dependant constant
which correspond to short term tests at a
given temperature and stress level

TR, transmissibility of applied force

tang, friction coefficient alonf pile surface

xxii

U, apparent activation energy
6, velocity of a soil element at distance r
ua, pile displacement

V the volume displaced by punch penetration
(if penetration, less than the punch radius)

Vi the volume of the expanded cavity

Vio’ initial cavity volume

ui, displacement of the cavity in the radial direction
a, slope of tapered pile surface from pile axis

6, ratio of applied load frequency to natural frequency of system
9, shear distortion rate

6, pile displacement

6, pile displacement rate

6c, arbitrary (reference) pile displacement rate

2, axial strain rate

0 - absolute value of temperature

0c, (arbitrarily) 1 deg C

f, damping ratio

a, true stress

a, lateral stress on the pile

a uniaxial compressive strength for frozen soils

C,

ac, temperature-dependant proof stress

000, proof stress at 0 deg C

acuo, creep proof stress at a temperature close to
the freezing point

a a temperature-dependant creep proof stress

cufi’

ace, proof stress at temperature T

xxiii

pressure at the pile base

a , effective stress

e
of, long term allowable stress
oi, initial instantaneous stress

ak(T), temperature dependant proof stress

am, mean stress

00, material and temperature dependant constant
which correspond to short term tests at a
given temperature and stress level

or, radial stress

a , radial stress at radius r

re e

a uniaxial tensile strength for frozen soils

T!
at, circumferential stress

a], 03, correspond to axial and confining pressures in the
triaxial cell

1, shear force between unfrozen soil and the pile
r, shear stress at a distance

Ta’ long term adfreeze strength between pile and frozen soils

TF’ stress along shear plane at failure
Tlt’ long term shear strength
¢lt’ friction angle of the Mohr envelope for this frozen soil

d, friction angle of the Mohr failure envelope for this soil
w, temperature-dependant exponent

w, natural frequency of system

2(c), creep strain rate

a, true strain

i o u
C , instantaneous strain

xxiv

0
cc, an arbitrary creep strain rate

(C)

c , creep strain
0
:6, equivalent strain rate

Ce” effective strain

ie . . .
c , instantaneous elastic strain

1 . .
s p, instantaneous plastic strain

a an arbitrary small strain (for calculation purposes)

k,

INTRODUCTION

1.1. Frozen Soil Pile Foundations

In permafrost regions, pile foundations can be used
advantageously for several reasons (Linell and Lobacz,
1980). They allow the structure to be raised above ground
level so that an air space remains below the floor. This
isolates the building from the annual frost heave and thaw
subsidence of the uppermost ground layer. The air space
insulates permafrost from heated buildings and helps
maintain the frozen ground temperature. Because strength
of permafrost is temperature-dependent, a foundation pile’s
load capacity will be higher for colder permafrost. Piles
transfer loads deep into the ground to higher strength
permafrost layers which are less susceptible to ground
surface temperature fluctuations. When local building
materials are scarce, piles can be the most economical way
to provide stable foundations.

Pile installation in permafrost generally depends on
the type and temperature of frozen soil encountered
(Heydinger, 1987). Piles may be driven, as in unfrozen
' soils, into fine-grained relatively-warm permafrost. In
harder permafrost, oversize holes may be drilled or augered
so that the pile can be set in and eventually frozen into a
slurry of natural soil, sand, or silt. A combination of

these two methods (Manikian, 1983) involves drilling an

1

2

undersize hole before driving a steel pile. Warm water is
sometimes added to an undersize hole to lower driving
resistance.

Reinforced concrete, steel, and timber are commonly
used for foundation piles placed in permafrost (Linell and
Lobacz, 1980). In some areas timber is more readily
available, can be relatively inexpensive, has versatile
lengths (up to 70 feet), and (if properly preserved) is
long lasting. Steel H sections and pipe piles are often
chosen because of their high load capacity, suitability to
driving, and resistance to weathering if protected.
Concrete piles have many of a steel pile’s good points but
are less acceptable because of their weakness to tension
forces that may develop during annual freezing at the
ground surface. Cast-in-place concrete piles are even less
acceptable because cement hydration adds heat to the
permafrost and because the concrete may freeze and weaken
before complete curing has occured.

The time dependance of frozen soil stress-strain
relationships leads to one of the major problems with
piles in permafrost. Prediction of long-term load capacity
and settlement must, by necessity, be based on short-term
testing. Extrapolation of short-term pile data to a long-
term load capacity prediction is complicated by many
factors, including soil temperature, soil type, pile type,

soil/pile adhesion, and pile installation method. Some

3

examples of these complicating factors include:

a) Seasonal soil temperature changes due to climatic
changes (Lachenbruch, 1988) and/or physical ground
surface changes (Brown et a1, 1969) which adversely
effect pile load capacity (Figure 1.1).

b) Ice formed in frozen soils, often randomly, during
deposition and/or freezing can greatly effect pile
settlement and load capacity (Nixon and McRoberts,
1975)

c) Salts or other contaminants in the frozen pore water
may not be obvious without testing, yet may
significantly reduce a pile’s load capacity (Nixon,
1987).

d) Pile geometry has been shown (Ladanyi and Guichoua,
1985) to have a large effect on pile capacity and
allowable settlement.

e) Pile/soil adhesion varies with the pile material and
soil type (Linell and Lobacz, 1980).

f) Piles embedded in frozen slurries have the highest
pile/soil adhesion values (Linell and Lobacz 1980).

g) Pile testing for load capacity must wait until the
slurry has frozen and any excessive lateral
pressures induced by freezing have returned to
equilibrium values (Neukirchner, 1988).

Due to the high cost of full-scale pile testing, many

of the problems associated with piles and other foundations

 

  
       
    
 
 

 

 

 

_Tmin Tm 0°C T max
"' ‘:. r.=-- r. " Active
- 'Range of annual . layer
-ground temperature;
2' variations ; ‘. -
Depth of zero .- . : -\; , . .
annual amplitude _- .
(approx. 10 to 20 m)
Permafrost
layer
Geothermal gradient = dT/dZ
7
l
Unfrozen

 

Depth, 2

Figure 1.1. Temperature profile in the ground, permafrost area
(from Andersland, 1987),.

5
have been investigated in a controlled laboratory

environment using frozen soil samples or models. Once the
problem has been thoughly studied in the laboratory (and
warrents further consideration) full-scale field tests may

be conducted more economically.

1.2. Cyclic Pile Loading: Research Problem and Objectives

Piles supporting structures in permafrost areas will
settle under imposed loads over a period of years. Load
capacity is normally calculated on the basis of adfreeze
bond at the pile/frozen soil interface. This adfreeze bond
is determined from long- and short-term creep tests under
static loads. In field situations, a cyclic load is often
superimposed on the static load. These cyclic loads may
include vibrating machinery (turbines, power generators,
and compressors) or travelling loads (cranes, fork lift
trucks, ect). Shock and transient vibrations are also
caused by forges and pile drivers. Field observations have
shown that the steady-state pile settlement rates (creep)
in frozen soils are enhanced by the superimposed cyclic
loads. _Larger settlement rates can reduce the service life
of existing piles or create a need, during design stages,
for more piles. This laboratory research project was
initiated to quantify these displacement rate increases and
identify contributing factors.

Pile settlement in frozen ground is typically

6

predicted on the basis of a creep equation relating shear
stresses at the soil/pile interface to pile displacement
rates. Creep parameters are used to characterize soil
type, soil/ice structure, temperature, and loading
conditions. This study used model steel piles in a
laboratory setting to carefully change and control test
variables in order to increase our understanding of the
processes responsible for pile settlement. Variables
included cyclic load amplitude, frequency of load
application, static load magnitude, and temperature. Sand
density and ice volume fraction were controlled during
sample preparation.

Load capacity of friction piles in frozen ground is
known to increase significantly when the pile surface is
roughened by the addition of protrusions or lugs. A series
of model steel piles with controlled surface geometries
(lug size, shape, and spacing) were frozen in sand and
loaded to provide information on the displacement
mechanisms operating at the pile/frozen soil interface.
Criteria were sought which would help define the pile
surface roughness geometry that most effectively controls
pile settlement and increases pile capacity. Details are

presented in subsquent chapters.

LITERATURE REVIEW

2.1. Mechanical Properties of Frozen Soils

Frozen soil is a complex mixture of mineral and/or
organic particles, unfrozen water, polycrystalline ice, and
air (Vyalov, 1965). The strength of a frozen granular soil
results from ice strength, soil strength (interparticle
friction, dilatency, and particle interlocking), and
soil/ice interaction when subjected to a given strain rate
at a specified temperature and stress state (Ting et a1.,
1983). Ice is responsible for creep, the time-dependant
deformation of a frozen soil under sustained loading. For
long-term loads creep deformation is important to
engineering design and prediction of its magnitude and rate
(with respect to stress, stress history, temperature, and

soil composition) becomes necessary.

2.1.1. Effect of Frozen Soil Composition on Behavior

Weaver (1979) classified frozen soil into four
significant catagories: dirty ice (0.9-1.0 Mg/m3), very
dirty ice (1.0-1.7 Mg/m3), ice-poor frozen soil (1.7-2.0
Mg/m3), and ice-rich frozen soil containing inclusions of
ice. Ladanyi (1981) added the category of unsaturated
frozen soil to these frozen soil types.

The behavior of a frozen sand is described in Figure

2.1. Starting with the peak uniaxial strength of ice, ice

Ana: . .ae us mag. nouuev «3352.02. Sum-3.3: pawn :ououm ZEN can»:

.>\.> oz<m .6 20:05.... ”.533
66 #6 NS 0.0
_ _ _ *_ . _ o

 

2053.0 .
J<o.hthoa>I IhOZUZhw U0.

- f _

 

 

 

 

 

 

 

. v
oz_zur»wzu5m mo. l . [Ollllulln
. .
-mozgozi . N ands:
35552.; + a
rhezumhm
46m . Amooa .pcmamuopfiw Ihxo<zwwdmhm
r 3256 new unocnwsoo Bout . n
#6..» oupouaxu 0&4... n .p
32355 + . 7.6.033 ¢ .. .v
:wozucbm 4.0m onlON oz<m <>><kho
r _ . _ _ _ _ v

00. on O
33.6

9

strength is shown to slowly increase due to an ice
strengthing effect of the scattered sand grains. Kaplar
(1971) and Goughnour and Andersland (1968) stated that
particle contact was established at 40% and 42% sand
volume, respectively. Peak strength can be seen to sharply
increase above these sand volume fractions. The soil
grains, when unfrozen, will develop strength from friction,
dilatency, and particle interference, all of which are
dependant on confining pressure and void ratio (Ting et
al., 1983). The ice is thought to stabilize the soil
skeleton (structural hinderence in Figure 2.1) due to
strong ice to unfrozen-water film to sand-grain bonds
between sand particles. At relative densities greater than
50%, ice/sand adhesion is thought to enhance dilatency ,
effects and increase peak strength.

The behavior of ice-rich sand is dependant on the
amount of ice and its orientation to soil stresses. For
conservative design, some authors (Nixon and McRoberts,
1976) consider ice-rich soils to have only the strength of
ice. Kaplar (1971), Alkire (1972), and Baker (1979) have
shown that strength decreases rapidly when ice saturation,
within frozen soils, is decreased.

The size of soil grains will also effect strength by
altering how water is frozen within the soil. Water
molecules bond to the surface of clay particles and remain

unfrozen at normal freezing temperatures (Dillon and

10
Andersland, 1966; Anderson and Tice, 1972). The nature of

some clays will prevent freezing of significant amounts of
water down to -30 deg C. (Tsytovich, 1960). Scott (1969)
reports little unfrozen water in sand at any temperature
below freezing. Capillary flow in some silts can cause ice
lenses to form (Anderson et al., 1978) parallel to the
freezing front.

Salt dissolved in pore water of a clay will reduce
frozen soil strength at a given temperature by increasing
unfrozen water contents (Nixon and Lem, 1984). Ruedrich
and Perkins (1974) showed that the maximum uniaxial
strength of a sand was reduced by the addition of salt to
pore water. Alwahab (1983) observed a significant
reduction in model pile load capacities when pore water_of
frozen sand was changed from distilled water to relatively

hard tap water.

2.1.2. Frozen Soil Creep Strength

Tsytovich (1975) described the process of creep in
frozen soils. When stress is applied to a frozen soil, ice
will melt at pressure points within the soil mass. The
melt water will then move to locations of lower (confining)
pressure and refreeze. Soil grains encountering less
resistance will deform and/or break bonds within the
remaining ice matrix and rearrange themselves in a

generally denser configuration. Breaking the cohesive bond

11
tends to weaken a soil while particle rearrangement tends
to strenghten the soil due to better particle interlock.
If the strengthing process dominates, the deformation rate
will continue to slow down and may never reach a constant
deformation rate. If the weakening process dominates, the -
deformation rate will reach a minimum value then increase.
Failure is often considered to start when the rate
increases. The rate will increase until rupture or when
the soil has little or no resistance to stress.

Creep within the ice matrix requires consideration of
the applied stress, time to failure, minimum creep rate,
temperature, and deformation at failure. In order to
quantify the strength of a frozen soil, a series of
triaxial tests are generally used to define the friction
angle, ¢ and the time-and-temperature dependent cohesion,
c(t,0) (Andersland et al., 1978). Ladanyi (1972) notes
that constant-temperature compression-creep tests (Vyalov,
1962; Sayles and Epanchin, 1966) determined that the onset
of failure will occur at similar magnitudes of strain. For
frozen fine-grained soils (Andersland et al., 1978), large
strains can develop during secondary creep and failure is
assumed to occur at a strain of 20% (design consideration
may limit failure to a lower strain).

In general, longer load durations correspond to lower
allowable stresses. Vyalov (1959) described long term

frozen soil strength by

12

0
0f = *0 (2.1)
ln (tf + t ) - 1n (to)

 

where

t = to exp(ao / 01) (2.2)

and of is the long term strength, ai is the initial
instantaneous strength, tf is the time to failure, and the
material and temperature dependant constants are
represented by ac and to' Vyalov (1963) said that t* could
be neglected for long term strengths. Sayles (1973) found,
for practical load durations, that the long term allowable
stress for triaxial tests of frozen Ottawa sand approached

a minimum level, thus

a
O
of = (a1 - 03) z (2.3)

In (tf / to)

where 01 and a3 correspond to axial and confining

 

t

pressures in the triaxial cell and the parameters 00, 0

correspond to short term tests at a given temperature and
stress level.

As mentioned before, at a sand concentration close to
40% and greater, frozen sand begins to develop strength
from the ice matrix, interparticle friction, and dilatency.
In Figure 2.2(a), Sayles (1973) explained that the ice
matrix is relatively stiff and provides its peak strength
at about 1 % strain while friction and dilatency develop

more slowly and provide a second peak at about 10% strain.

Figure 2.2.

13

 

 

 

 

 

 

 
  

 

 

A
E
:m24’_ (E5: OMMPiWPa
a J\
éflél 515]
.3 2.76
E a '39 _
o L 1 1 l 1 1 ’
O 2 4 6 8 IO 12
AXIAL STRAIN, 61: PERCENT
(a)
albi-
I
. second peak
*:12 .
E 3 first peak
g ice
4 -/
o 1 L 1 l 1

 

 

0 4 8 12 16 2O
NORMAL 513555. 0'. MP3

(0)

Compression tests with frozen Ottawa Sand (after Sayles,
1973). (a) Stress-Strain Curves; (b) Mohr Envelopes.

14

The stress conditions at the first peaks of these tests are
plotted as a Mohr envelope in Figure 2.2(b) and labeled
"first peak". This is similar to the Mohr envelope
developed from tests of ice samples. After the first peak,
frictibn and dilatency continue to slowly develop and
produce a second peak in Figure 2.2(a) at the higher
confining pressures of 6.89 MPa and 5.51 MPa. The stress
conditions at 8.9% axial strain for the 5.51 MPa test and
at 10% for the other triaxial confining tests are plotted
as a Mohr envelope and labeled "second peak" in Figure
2.2(b). This is shown at an angle of 31 deg from the
horizontal and is (approximately) the friction angle for
this sand when tested unfrozen.

Ladanyi (1981) used information from Sayles (1973) and
Chamberlain, et al. (1972) to produce Figure 2.3. The
strength of ice dominates region A and peak strengths were
determined at about 1% strain. Region B represents peak
stress conditions found near 1% strain and at approximately
10% strain (where dilatency and friction added strength to
the frozen sand). In region C, confining pressure partly
melted the ice matrix and its contribution to strength was
reduced. The high confining pressures of region D crushed
sand grains and melted the pore ice so that the resulting
behavior is similar to unfrozen sand under undrained
conditions. Ladanyi (1981) notes that most permafrost

stress conditions would fall into regions A and B at

15

.Adoaa .uhcnpea nouuev vcom oaauuo cououu
you 2.382.; 25.33 on... no souueusoaoudou cause—030m .m.~ ouawum

wfifimfit
u-pgcflznfiz mflfia. xsfiw z- uZ—zwwaz mz_zwrxm
QZFHduawfixflmEm — Qfihuflauflmflwcm ZZDFm zzzflm

 

  

 

$83815 BVBHS

 

l6

confining pressures below 55 MPa.

Sayles (1973) also showed Mohr envelopes for creep
strength from triaxial creep tests at various times to
failure (Figure 2.4). The envelope for the longest time
(to failure) can be seen at 29 deg from horizontal, which
is close to the apparent angle of internal friction (30
deg) of the same sand when unfrozen. Andersland and
AlNouri (1970) showed that the apparent angle of internal
friction was similar for frozen and unfrozen sands and also
showed this angle to be close to zero for frozen clays.

The ratio of aC/aT (uniaxial compressive
strength/tensile strength) for frozen soils is usually
lower than 5 (Ladanyi, 1981) which contrasts with the value
of ac/aT = 8 for an ideal elastic—brittle material
(Griffith, 1924). At higher temperatures (Ladanyi, 1981)
or at slower strain rates (Haynes et al., 1975), oC/aT
approaches unity and shows plastic behavior. Bragg and
Andersland (1980) showed, in Figure 2.5, an abrupt change
from a plastic behavior (with a failure strain greater than
4%) at strain rates less than 0.00025/sec. to a more
brittle behavior (with a failure strain less than 1%) at
strain rates greater than 0.00025/sec.

In Figure 2.6, a break in the -2 deg C curve at the
strain rate of 0.00001/sec is thought to be associated with
pressure melting at sand grain contacts. If refreezing did

not occur immediately after melting, the unfrozen water

l7

.xsaas .aseueaa nonmav Au moo ma.n- 1 av
pawn oswuuo mo numcouuu doouo uom aedo~o>co use:

mm: swmuahm 4<t¢oz

 

m . v 0
AT _ . a .
c 00
:Nd
”95:3 2 6:5. 1

 

 

.«.~ ousmaa

'SSBBIS BVBHS

Ed“

l8

.Aaaaa .aseaeua “sumac

 

 

pace Cancun a How ou¢u-:«uuua useuo> canyon ousauem .n.~ ounmam
Tm ~uh<¢ z—SBM a<~x<
to. to. to. “.2 to. hm. .
. _ _ _ .o
Uon—u a
UooT a. w.
.0 I
I‘lllle U00... 0 m...
r .reIu Uou- o .. _ u
as... .....
O,“ 4 m
. [Pelflifl CICIMIOICODIOIC In 1N
z
_ L _ Sumo-Q _ O—

 

 

l9

.Acoa. .6euawuoec<
use amen: nouuev ouch :«euua .u> newcouue e>qamoudaoo .o.« shaman

.7903 0.9. c3...“ .25.

 

 

 

 

 

N.o. 1o. 79 so. to. to.
..... . . . 114.. . . —....q. . . |1.l... . . . 1:... . . N
.. oh. .. .0 ..

I UeC— I IO 1 V
I. UGO ' lo I.
H UeN .I .0 U0
n ) o lo
|<
fil . . . . C O h m . C O C . . . . o #0“
e a . l a e el e e. u e
.u. n~ .. ¢-.-1u-qn-uu.-Jnu.- .
.I C ION
r- I.
rt»..- . p —:.-~ - - —-p-b- - b i—PPPPPbL - run-pp . - 0V

 

(gun/N w) utduuts ugssaadwog

20

content could be increased and thereby decrease the
strength of the sample. This strength decrease could alter
the stress versus strain-rate curve. Samples tested at
strain rates higher than 0.00001/sec may fracture some

bonds in the ice matrix of frozen soils (Ladanyi, 1981).

2.1.3. Temperature Dependence of Frozen Soil Strength

The basic relationship of (increased) frozen soil
strength with (decreased) temperature is shown in Figure
2.7 from Sayles (1966). The theory of rate process has
been used to describe chemical processes (Glasstone et a1,
1941) and deformation of unfrozen soils (Mitchell et a1,
1968). It has also been used by other investigators
(Andersland and Akili, 1967; Goughnour and Andersland,
1968; Andersland and ANouri, 1970; Andersland and Douglas,
1970) to describe the temperature-dependant behavior of
frozen soils. According to this theory, deformation occurs
when energy (thermal or mechanical) is added to "flow
units" (of molecular size within the considered solid) so
that the energy barriers preventing relative movement
between these flow units are overcome. Andersland and

(C)

Alnouri (1970) expressed creep rate 2 as

2‘3) = A expi (-U / R') / (T) 1 (2.4)

where R' is the universal gas constant, T is absolute

temperature, and U is apparent activation energy. Ladanyi

21

 

f—SANOY SILT -

i

r- OTTAWA SAND

  

UNCONHNED COMPRESflON STRENGTH kg/cm2

 

 

 

’
’
’
Luce _
CLAY _
'0 -so -no «so -wo

TEMPERATURE , °C

Figure 2.7. Temperature dependence of uniaxial compression strength
for various frozen materials (after Ladanyi, 1972).

22
(1972) transformed this equation to

[ (9) (“U / R’) ] (2 5)
a = 0 exp .
C9 0° (273 n) (273 - a)

 

with

a = 273 - T (2.6)

where ”co is the proof stress at 0 deg C and OCH is the
proof stress at temperature T.

There has been disagreement over the usefulness of the
rate process theory. Hoeskstra (1969) stated that this
theory shouldn’t be used with substances that change phase
with temperature as frozen clays do. Ladanyi (1972)
pointed out that the activation energies obtained by
plotting the natural logarithms of creep rates versus
reciprocals of absolute temperature are very high (389
kJ/mol in one case) compared to the value of 84 kJ/mol
found by Gold (1970) for polycrystalline ice. To explain
this discrepency, Andersland and Douglas (1970) stated that
the activation energy determined during frozen soil
deformation may be effected by mechanical changes occuring
in the soil. Recent research by Orth (1985) supports the
rate process theory for frozen sands and also notes that
nonunique activation energies for ice (80 kJ/mol at below -
8 deg C and 120 kJ/mole at above -8 deg C) found by Frost
and Ashby (1982) support his and earlier findings.

Ladanyi (1972, 1981, 1985) has stated that the most

—\.

23

convenient means of describing the temperature-dependent

frozen soil strength, a is by empirical relationships

c9
such as Vyalov’s (1962) power law which was normalized by

Assur (1963) and shown.by Ladanyi (1972) as

a =

C0 aco [1 + (a / 9c)]” (2.7)

where so is (arbitrarily) 1 deg C and 0 is defined by
equation 2.6. Ladanyi (1972) states that w = 1 can be used

to approximate change over limited temperature intervals.

2.1.4. Constitutive Equations for Frozen Soils

Some typical creep curves for frozen soils under
uniaxial compression are shown in Figure 2.8. (Andersland
et al., 1978) Figure 2.8(b) shows the three general stages
of constant temperature deformation behavior for frozen‘
soils and other materials that exibit creep. The first
stage is primary creep where the deformation rate initially
increases rapidly, then slowly decreases. The behavior of
some silts and sands do not leave this stage except when
subjected to high stresses and/or high temperatures. The
second stage is characterized by a constant deformation
rate. Ice and ice-rich soils may have an extended period
of secdndary creep. The third stage of creep is indicated
by an increasing deformation rate which leads to failure of
the sample. Figure 2.8(c) shows that the second stage of
creep contains the minimum deformation rate.

Because of the large strains obtained for creep tests

24

Figure 2.8. Constant-stress creep tests
(after Andersland, et a1. 1978).
(a) Creep curve variations.
(b) Basic creep curve
(c) True strain rate versus time.

25

icerich soils

 

Tertiary creep.
Yield strength exceeded
Secondary creep dominant.

  

Primary creep dominant.

 

iceooor soils

   

 

 

a 525

Time r

 

(a!

 

  
 

---4-------——-—--

I
I
l
l
l

I

 

 

"---------------------

e 526

Time t

‘1'

(hi

.u 22 ESE

III

1L

II

 

Time t

‘r

 

(cl

8.

Figure 2

26

on frozen soils, true stress, a is used as
a = P/A (2.8)

where A is the actual area and P is the constant load.

True strain, a can be shown as
e = 1n (L / Lo) (2-9)

Primary creep has been described by Hult (1966) as

. . . i . c
hav1ng an instantaneous strain, a and a creep strain £( ).

The instantaneous strain contains an elastic portion, cie,
and a plastic portion, sip. The instantaneous plastic
strain has been written (Marin et al., 1951; Ladanyi, 1972)
as

. a '
51p = ck [————]k (2.10)

0k(T)
where ck is an arbitrary small strain (for calculation
purposes), ak(T) is the temperature dependant proof stress,
and k is a constant which is somewhat dependant on
temperature. The elastic instantaneous strain, ale can be
shown as
. 0'
ale = —— (2.11)
£O(T)

where EO(T) is a fictitious Young’s modulus at a constant

temperature, T. It has been suggested by Hult that 21p may

be neglected because it is relatively small. Vyalov (1959)

27
showed that the entire instantaneous creep strain may be
neglected when loading exceeds one day because 5i becomes
relatively small (less than 10%) as compared to the creep

(a),

strain, a

(C)

The primary creep strain a can be described by

.(C) = f( a, t, T ) (2.12)

where a is true stress, t is time, and T is temperature.
Vyalov (1962) proposed an equation of the type that has
been used by later researchers (Ladanyi and Johnston, 1974;
Eckardt, 1982) for constant-temperature constant-stress

conditions as

 

 

l t (2.13)

where b and n are creep parameters, EC is an arbitrary
strain rate, and do is the temperature-dependant proof
stress (determined by the choice of EC). This equation can
also represent secondary creep when b equals 1.
Differentiation of this equation yields the creep rate,

5
2(C) = p—ilJb [ 1“ b tb’l (2.14)

 

which may be used to form a ratio of two creep rates

(with similar test conditions except load) as

28

 

] (2.15)

when b = 1 and/or t2 = t1.
Secondary creep may be described (Hult, 1966) by a
method that approximates the accumulated total strain,

a e e a e e a i
as a combination of a fictitious instantaneous strain, a( )

(i)

E

F( a, T) (2.16)

plus a period of constant creep rate,

°(C)

C

G( a, T ) (2.17)
in a total strain equation form as
a = 5(1) + €‘(C) t (2.18)

where t is total time, 0 is a constant stress, and T is a
constant temperature. A series of creep tests, with each
test at a constant temperature and a constant stress level,
can then be used to calculate functions F and C so that
Equation 2.18 can approximate total strain of this material
while it is in secondary creep. Ladanyi (1972) showed a
summation procedure to obtain total strain when a frozen
soil is loaded in steps and each step reaches secondary
creep (Figure 2.9).

Vyalov (1962) described creep of ice-poor soils by

29

al

 

 

I'"°-'-‘j . j'T'

 

 

 

 

 

 

 

Figure 2.9. Creep curve for step loading (after Ladanyi, 1972).

1

 

] a t (2.19)

a = [
w (a + 1)k
where e is strain, a is constant stress, t is time, 9 is
the number of degrees below freezing, and w,k,c,b are

material dependent constants. Ladanyi (1972) added the

effect of confining stress to this equation so that

 

 

., = D to. - w.) 1“ t” (2.20)
with
( l + sind )
j = (2.21)
( 1 - sin¢ )
and
1
D = [ k 1° (2.22)
w (a + 1)

where d is the friction angle of the Mohr failure envelope
for this soil. Ladanyi and Johnston believed it reasonable
to assume j = 1 when 01/ ja3 is less than 1.25 so that

Equation 2.19 becomes

:1 = D (a, - 03)C tb (2.23)

for confined compression creep tests. Weaver and
Morgenstern (1981b), interested in predicting pile
settlement, reviewed existing uniaxial and triaxial
compression tests to select the constitutive relationship
that best fit data for ice-poor frozen soil creep behavior.

An equation from Ladanyi (1972) for multiaxial stress

31

conditions was selected
j + 2

8 = D { [————;——00e + (1 ' jlam l

C tb (2.24)

where ‘e is the effective strain, 0e is the effective
stress, and am is the mean stress. When j = 1, this

equation becomes

t (2.25)

2.1.5. Frozen Soil Strength Testing

Constant-strain-rate tests, relaxation tests, and
creep tests are commonly performed on frozen soils to
determine time dependent strength properties. During a
constant-strain-rate test, the samples are strained at a
constant rate and the resulting range of stresses are
recorded. Scott (1969) states that this test avoids the
problems of sudden loading found in the creep test and
relaxation test. A relaxation test will load a sample
quickly to a given strain then hold that strain while
recording the resulting stress levels. Vyalov, et a1.
(1969) used this type of test to determine long term
strengths of frozen soils. A creep test will load a sample
quickly to a given stress then maintain this stress for the
remainder of the test. Sometimes this test is altered by
increasing the creep load incrementally after given periods
of time.

Under similar conditions, similar values of (é/c)max

32

were found by constant-strain-rate tests and constant-
stress creep tests (Mellor, 1979). Goughnour and
Andersland (1968) have shown how data from constant-stress
creep tests may be visualized as data from constant-strain-
rate tests (and vice versa).

Some conditions of testing have been found to alter
test results. Baker (1978a, 1978b) found that sample ends
need to be carefully prepared. Andersland and AlNouri
(1970) found temperature history to be important.
Additional information on frozen soil test conditions can
be found in Sayles, et a1. (1987) which recommended

standard testing methods.

2.2. Response of Frozen Ground to Dynamic Loading
The literature mentioned in this section includes
topics on unfrozen soils due to the lack of frozen soil

information in this subject area.

2.2.1. Dynamically Loaded Foundations in Unfrozen Soils
Richart et al. (1970) has shown methods which use
measured or estimated dynamic stress-strain properties to
estimate the response of shallow foundations on elastic
half-spaces to dynamic loading. These methods include
geometrical damping, the spreading of wave energy from a
source on top of a half—space, but neglect the internal

damping of the soil. Richart et al. (1970) noted that the

33

Eassumption of no internal or material damping, when the
(damping ratio, D, of a soil was less than 3% (as many
‘unfrozen soils have), would alter the computed shear
tmodulus, G by less than 6% from the complex shear modulus,
6*, computed with the influence of damping. Assuming soil
to be elastic (no internal damping) made computation of a
foundation’s response to dynamic loading easier and nearly
as accurate. Figure 2.10 shows damping ratios for frozen
soils which are considerably higher than 3% for soil
temperatures higher than -10 deg C. Use of elastic theory

methods for dynamic response of frozen soils may produce

unacceptable errors.

2.2.2. Stress-Strain and Energy Absorbing Properties
Frequency, amplitude, and load duration are important
to solving dynamic loading problems (Vinson, 1978). Two
common problems encountered in geotechnical engineering
are: (1) high frequency, low amplitude, and sustained
loading from vibrating machinery or (2) low frequency,
higher amplitude, transient or irregular loading from
explosions or earthquakes. The properties generally tested
for are included in two groups: (1) dynamic stress—strain
properties which include various wave types and velocities,
Young’s moduli, and shear moduli or (2) energy absorption'
properties which include various stress-strain, phase-lag,

and damping measurements.

A
(130 l'I’l'l'l l'l'l‘l'r r' I'VTTF Tfi'rrr'

r--4°c

 

 

 

 

 

 

 

   

 

 

 

 

 

 

 

 

_ Frequency f. H:
(a)
030 T l l _
Clay: OMC-V, OC-V, 86-8, 66-8
Silt: AS-V, HS-V, MS-S
Till: TT-S (non plastic)
: OS-V, OS-S, OSL-V
Q20- ‘
.<
":3
.§' AW
3 HS-V
to 030V
:3
0.10 - '7
OMC-V OS L-V
0C4!
8C4;
6C4;
1 "'5
k6— ms
1 L i ~——<M$S
0 -5 -l 0 -l 5 -20

(b) Temperature 7', °C

Figure 2.10. Damping ratio of frozen soil (after Vinson, 1978).
(a) Versus frequency. (b) Versus temperature.

35
Cyclic loading superimposed on a statically loaded

pile normally includes a low frequency, low amplitude,
sustained cyclic load with a much larger static load.
Although many dynamic triaxial, dynamic uniaxial
compression, and resonant column tests (Kaplar, 1969:
Stevens, 1975; Chaichanavong, 1976; Vinson et al., 1977)
consider only a cyclic load or superimpose a cyclic load on
a hydrostatic load, they provide some insight on how
frequency and amplitude of loading effect various soils.
Table 2.1 was compiled by Vinson (1978) to show the
relative importance of the frozen physical conditions and
testing conditions on the dynamic stress-strain and energy

absorbing properties.

2.2.3. Deformation Influenced by Cyclic Loading

Some triaxial and uniaxial compression tests on frozen
soils show how dynamic loading can effect total strain.
Strain rates appear to be relevant to the problem of
increased pile displacement rates due to cyclic loading.
Parameswaren (1985), in Figure 2.11, showed large strain
rate increases after a cyclic load (dashed lines) was
superimposed on a static load (solid lines). A temperature
increase was measured within the sample during the dynamic
loading from which Parameswaren (1985) calculated a volume
of unfrozen water and speculated how this might accelerate

strain rates. In uniaxial compression tests of frozen sand

Table 2.1.

36

Relative Importatce of Various Parameters on Dynamic

Properties 0! Frozen Soils (after Vinson, 1978).

 

Material parameters:
Material type and
composition

Material density or void

ratio (for fully saturated -

soils)

leeoontemordegreeol'ioe
ssturationpsrusl’roeen
Walnut

Field- and/or tast-
rondltt‘on powers:
Temperature

Strain (or stress)
amplitude of loading

Frequency of loading
Conhning pressure

Very important: dynamic stress.
strain properties for coarse-
grained soils an be nearly an
order of magnitude greater than
(or “gained soils

Important; over a reap ol’ void
rsu‘os (tom 0.3 to ice. dynamic
stuns-strain properties can
decrease by a factor at 5; over
the range ol’ void ratios
modeled with s y'ven soil type
the influence of dynamic stras-
strain propertia will not he as
great; dynamic stresspstrm's
properties of iadeu'case

' substantially with dementia
deuity

Very important; however.
cannons: o! mas-pained
soils with ”7. ice saturation is
rare in nature; deuce oi is
saturation for line-pained soils
'3 relatd to unfrozen water
cooled and the influence in this
one is refleaed by temperature

Very important for frozen soils.
particularly in the range 0 to
- S‘C

L'nimportant for axial strain
amplitudes less than 10‘ "x. for
from soils and ice: important
(or axial strain amplitudes
between 10” and I0“'/.l'or
frown soils

Relatively unimportant

lmportsnt for coarse-grained
soils and ice: unimportant for
fine-grained soils

Very important: clergy-absorbing
properties can dill’cr by an order
of magnitude or yester for
dill’erent soil type

Prohshly unimpttant; there
may he a slight decrease in
damping ratio with decreasing
density over the rang
assoa'sted with the inn soil
1’93

lmportntuesoonunentsunda-
Dynamiestmstrain propnsiu"

Very important for frozen soils
in the range 0 to - S’C;
relatively unimportant in the
range - 5 to - 20C

L'nlmportant for anal strain
amplitudes less than l0' “4:
Important for axial strain
amplitudes between IO' ’ and
l0' ' '/.l'or most frozen soils

Probably important
l'nimportant

37

 

-2.o ‘ r .. r
-2.5 -
-3.0
-3.5 ‘ 1

°C

 

TEMP,

 

 

 

 

 

 

 

10 r l
9 - Cyclic
3‘ Stressing —___
s 8 _ / l pl.
2 _ _J
“ 7 r 2——
< a’
a'
E 6 i—- ’I
00 5 Temperature Measured by a
Thermocouple Embedded in the Sample
L 1 ‘
4

0 50 100
TLME. h

Figure 2.11. Effect of cyclic stress on temperature of frozen
sand (after Parameswaran, 1985).

38

Ame .3233: one again. .333

mean conch—m you assuage» cannon and??? .w~.~ 0.3m;

 

 

[ i/o/ ~
[/

u was 2. 2:2 :5
coca coo.
‘ d

 

 

J

320 2:: o
n... u x
95. 3:93.. 4

 

 

 

39
at 11 deg C, Trimble and Mitchell (1982) compared a

constant stress sample to a sample that was repeatedly: 1)
loaded to the same constant load for 0.25 seconds and then
2) completely unloaded for 0.25 seconds. Figure 2.12 shows
that volumetric strains, total strains, and strain rates of

these tests are similar.

2.3. Adhesion and Friction Between Pile and Soil

Adhesion has been extensively studied by Jellinek
(1957a, 1957b, 1960a, 1960b, 1960c, 1967) with results
reported for tensile and shear tests on ice adhesion to
stainless steel, lucite, fused quartz, and polystyrene.
Jellinek (1957b) proposed that a transition layer of
unfrozen water lies between the ice and the other
materials. Strong tensile and lesser shear strengths were
found between the ice and the other materials (Jellinek,
1967). Experimental control of adhesion is reported
(Jellinek, 1957a; Alwahab, 1983) to be difficult so that a
lot of scatter exists in experimental results. Weaver and
Morgenstern (1981a, Figure 2.13) suggest that entrapped air
could weaken bond at the ice/pile interface.

Friction between a pile and soil grains can be thought
of as occuring in unfrozen soil. The shear force between

unfrozen soil and the pile, can be described as

T = N f (2.26)

Figure 2.13.

40

Idealized plate in contact with ice and frozen soil
(after Weaver and Morgenstern, 1981a).

(a) Air trapped during sample preparation allows slip
ice/plate interface.

(b) Interaction between plate protrusions and soil
‘ grains.

(e) Shear plane developes in ice.

 

(a)

 

 

Figure 2.13.

 

 

 

 

 

42

where N is the normal pressure between soil and pile and f
is a function of the type of materials and the conditions
at the pile/frozen soil interface. Normal pressure can
vary considerably due to pile installation methods (Ladanyi
and Guichaoua, 1986). Friction can be virtually eliminated
if a very cold pile forms an ice coating as it freezes its

surrounding sand slurry (Linell and Lobacz, 1980).

2.3.1. Mechanical Interaction Between Frozen Soil and Pile

Mechanical interaction may evolve as a physical
rearrangement of the soil particles with respect to the
pile. Alwahab (1983) added oversize protrusions (lugs) to
model piles which created, when loaded, regions of
increased pressure ahead of these lugs. These pressures
are similar to those pressures attributed to cavity
expansion in frozen soils (Ladanyi, 1976). Increasing the
diameters of these lugs clearly increased the model pile
load capacities. Weaver and Morgenstern (1981a) showed
indications that shear behavior at the interface between
metal plates and frozen soils was dependent on plate

roughness and soil type.

2.3.2. Pile/Soil Adfreeze Bond Estimation
Adfreeze bond at the pile/frozen soil interface, used
for estimating pile load capacity, has often been

considered (Linell and Lobacz, 1980; Weaver and

43

Morgenstern, 1981b) to be some fraction of the shear
strength of the frozen soil surrounding the pile. Weaver
and Morgenstern (1981b) considered the long term adfreeze

strength between pile and frozen soils, ra to be

a Tlt (2.27)
where m is a constant based on the pile’s surface

roughness. Long term shear strength was then defined as

flt = c1t + a tan (dlt) (2-23)

where c1t is long term cohesion, a is the lateral stress on
the pile, and ¢lt is the friction angle of the Mohr
envelope for this frozen soil. Weaver and Morgenstern
(1981b) considered a tan(¢lt) to be small compared to clt’
hence they assumed the long-term adfreeze bond to be

(2.29)

Values of m were described as ranging from 0.6 for steel to

1.0 for corrugated steel.

2.4. Review of Pile Design Methods
The best information for pile design in permafrost can

be obtained from full-scale piles (Linell and Lobacz, 1980)

44

built at the project site by the anticipated construction
method and load tested during the surrounding permafrost’s
warmest condition. Factors such as soil type, soil
temperature, pile installation, and pile type are thereby
combined in the right proportions so that loading the test
pile is similar to loading a design pile. However, pile
test information is not generally available during the
early planning of a project or may be considered
unnecessary for small projects (which could be overdesigned
and verified later). Preliminary designs may be estimated
from knowledge of the project’s soil types, soil
temperatures, confined triaxial compression test
parameters, pile size, and anticipated pile loads (Weaver
and Morgenstern, 1981b).

A proposed standard method by the American Society for
Testing and Materials (ASTM) entitled "Standard Method of
Testing Individual Piles in Permafrost Under Static Axial
Compressive Load" has recently been outlined by Neukirchner
(1988). While this standard has not yet been approved, it
uses many procedures accepted in "Piles Under Static Axial
Compressive Load" (ASTM D-1143), the existing standard for
unfrozen soils. The biggest difference between the
"unfrozen" and "frozen" methods concerns loading
procedures. The "unfrozen" test increases the pile load in
several steps (25% of design load) and generally holds each

incremental load for less than 2 hours. The "frozen" test

45

procedure uses two test piles, each one loaded in two steps
with the load maintained for each step from 6 hours to 5
days.

If a test program were to use the proposed standard
method, two piles would be prepared in a manner similar to
that for unfrozen piles (ASTM D-1143) and temperatures
would be measured along their lengths. A 100% of design
load (adjusted for temperature) would be added to the first
pile while a 200% of design load (adjusted for temperature)
would be added to the second pile. Both loads would be
maintained until a steady creep rate is obtained or for
three days if a steady rate does not develop. Additional
loads would then be added in order to fail the first pile
in 6-12 hours and the second pile in 3-5 days. The results
from these tests would be compared to a diagram similar to
Figure 2.14 to determine if the design pile can sustain its
design load throughout its service life.

If the Linell and Lobacz (1980) test loading procedure
were used, one pile would be prepared in a manner similar
to ASTM D-1143 and temperatures would be measured along its
length. Incremental loads of 10,000 pounds would be added
and maintained for 3 days (the recommended minimum).
Incremental loading would continue until 1.5 inches of
settlement (failure has occured) or a total load of 2.5
times the design load is obtained. A load determined to

initiate pile failure could then be divided by a factor of

46

.Amama .uossouuxSoz Houuev
new» one nmouua euaaueu monsoon augusouundon >o~e>> .#H.N unawam

Anymorv mu .ouaaunm ou mafia

 

mucnuncou I cu .oe
unnaueh cu oEaH l mu
unouum ouaaumm l we nouns:

Itlslltttlltl
||I|I||I|IIIIII\llltttttttls
Islltltttllllllllmmmmflmmmwwwwwmmmwuttttttttttlltttisttttttttttttt.

 

 

me A «a A an A 0 mos o

 

 

 

 

 

 

ssaaas eanIreg

Jo

(----)

,-eaq

I

I

47
saftey of at least 2.5, for dead load plus normal live
load, or by a factor of saftey of at least 2.0, for dead
load plus maximum live load. Although the loading periods
are shorter than 3 days, Figure 2.15 illustrates how a
failure load would be determined by this procedure.

Pile capacities based on pile tests or other
preliminary work may be estimated by using previously
established stress-strain-time relationships for frozen
soils. Linell and Lobacz (1980) roughly estimated pile
capacity by using soil temperature, soil type, and pile
type to estimate sustainable adfreeze bonds on the pile’s
surface. In their examination of pile capacity estimation,
Weaver and Morgenstern (1981b) considered compression piles
to fail when the adhesive portion of the adfreeze bond .
between pile and soil was broken (resulting in large rapid
settlements for relatively smooth straight piles) or when
the frozen soils surrounding the pile deformed excessively
(along the sides and at the end) during the design life of
the pile.

Weaver and Morgenstern (1981b) used various estimates
of soil cohesion (which they converted to_pile/soil
adhesion values as detailed in section 2.3.4) to create
Figure 2.16. Constitutive equations that best fit a
collection of uniaxial compression data and triaxial
compression data (see section 2.1.4 for details) were

incorporated into pile settlement equations based on creep

48

 

 

 

1
I45.
Minuet Thaw Zone err f' Elastic deflection of pin:

Pl 1,0 Va 5’
8'
Permafrost 1,153

 

 

 

 

 

Assumed Lend Distribution

Average Adfreeze Bond Stress in Permafrost. psi

 

 

      
 
  

 

 

 

 

o 2 I; 2.0 22 39 3:
Load. hips
0° 20 so 63 up too :20 up too neg zoo
A/Assumed failure
. . . --*-------
OJOr Computed shortemng.8.ef pile assuming ‘.--- 4
distribution of teed as shown
i 0.20. Determination 1
2 of failure tend
g 0.30» 4
3
5
5 0.4m *
0.50)- 4
(16¢:o 5 ‘ i 3 t0 .2 1‘4 f6 :8
Time. doys
Pile type: 8-in. pipe, 36 mm Soil profile: O-l ft peat. 1-20.4 ft (bottom of pile) silt
Pile length: 20.9 ft Backfill around pile: silt-water slurry
Length below surface: 20.4 ft Avg temp of frozen soil: 29.2°F
Embedment in frozen soil: 16.1 ft Test performed: July 1958

Loading schedule: 10-kip increments applied at 24-hr intervals. The deflection shown for an incre-
ment is that observed just prior to application of next increment.

Note: Pile not isolated from soil in thaw zone.

Figure 2.15. Load test of steel pipe pile
(after Linell and Lobacz, 1980).

49

 

500

§

§

§

 

Long Term Cohesion. C" (kPa)

too - \‘

 

 

 

Temperature (‘0)

Figure 2.16. Long-term cohesion of frozen soils
(after Weaver and Morgenstern, 1981b).

50

 

to“ . .
A a -

a. -:
a: “a -

 

I I I I l '1'
Settlement
eslgn Range

I I
I

to”

r l [nnl

 

Normalized Pile Displacement

 

l

'_.._T ______

r‘o

t I I liq I
Pile Radius .
Time in Hours (h) 7
Pile Displacement Rate ‘

l lllllL

     
     
 
   
  

Allowable Adfreeze
Strengths for
Steel Plies

1 s lssul

 

[LLLUI t tthtn

 

to" '
10

102 103

Applied Shaft Stress. -

Figure 2. 17.

:- (kPa)

Design chart for friction piles in frozen Ottawa Sand

(after Weaver and Morgenstern. 1981b) .

51

 

      

10" I I 'lriirl I I'll!

A l'

g :a - Pile Radius //
6' .t - Time in Hours (h)

.I: u - Pile Displacement Rate
v a

. Si '
= e.

a to"

IIIIIIII
l l lssul

     

l

to"

Normalized Pile Displacement

 

 

 

 

'0-4 1 tittttl 1 111111
to2 103 to‘
Applied End Bearing Stress,
GE (kPa)

Figure 2.18. Design chart for friction end-bearing piles in frozen
Ottawa Sand (after Weaver and Morgenstern, 1981b) .

52

deformation theory (see section 2.5) to predict long term
settlements. Normalized design diagrams were created to
predict pile settlement from stress around the pile’s
shaft (Figure 2.17) and to predict pile settlement due to
stress at the end of the pile (Figure 2.18) and each figure
must be used with equal pile displacement values, Ua‘ The
figures shown are for an Ottawa sand. Design diagrams for
ice, Hanover clay, and Hanover silt are included in their

paper but are not shown.

2.5. Pile Settlement Estimation Equations

2.5.1. Shear Deformation of Soil Along Pile Shaft
Johnston and Ladanyi (1972) found that rod anchors

grouted into permafrost displayed typical creep behavior.

The anchor behaved like a short pile and it displacement

was described as
s(t) = si + s(t) (2.30)

where s(t) is total displacement in terms of time, si is a
fictitious instantaneous displacement, s is a constant
displacement rate, and t is time. This equation is similar
in form to the constitutive equation for secondary creep of
frozen soils found in Equation 2.18. The assumptions made
for this model include: a) Pile is cylindrical, b) Uniform
shear stress along the pile surface, c) Uniform soil type,

uniform temperature, and d) Neglect pile weight. End

53

bearing was nonexistent because the anchor was pulled
upward.

It was assumed that shear stress at any distance from
the pile’s surface could be described by an equation from

Nadai (1963) as

r = ra (a / r) (2.31)
where ra is shear stress at the pile surface, a is the pile
radius, r is the distance from pile’s center, and r is
shear stress at a distance. An expression for shear
distortion rate, é was used

-d(h)

7 = -——————— (2.32)
dr ‘

where u is the velocity of a soil element at distance r.
An equation from Odquist (1966) was used by Johnston
and Ladanyi (1972) to describe an incompressible von Mises

material in a general state of stress as

 

1 (2.33)

where 0e is equivalent stress, Ce is equivalent strain

rate, 2c is a strain rate chosen for computation purposes,

n is a creep parameter, and a is a temperature-dependant

cue
creep proof stress. This equation is a variation of

Equation 2.13 (with b = 1) which best fit Nixon and

54

McRoberts (1976) uniaxial-creep—stress and strain-rate data
for ice and ice-rich soil.

As a pile settles, a given soil particle will only
move vertically and/or normally to the pile’s surface, so
that plane strain can be assumed. The stress [ 01,

02 (= aL/Z + 03/2), 03 ] and strain rate [ 21, £2 (= O),
23 (= -E1) conditions for plane strain and constant

volume were used to derive

J3 (0 - 0 )
E. = film“) .‘ [ 1 3 1“ (2.34)
U

 

A horizontal normal force (lateral soil pressure) and a
vertical shear force (resistance of soil to axial pile
load) were thought (Johnston and Ladanyi, 1972) to develop
simple shear in the frozen soil surrounding the pile. The
stress, 1 (= aL/Z - 03/2) and strain rate, 3 (= 2:1)
conditions of simple shear are included in the plane

strain equation so that

 

3(“+1)/2 EC [ 1“ (2.35)

a
CUB

Nixon and McRoberts (1976) suggest that pile behavior
in ice and ice-rich soils are better described by a two term

form of the previous equation so that

. = 3(n1+1)/2 31 n1 + 3(n2+1)/2 32 ,“2 (2.36)

After substituting Equations 2.31 and 2.32 into this

55

equation, integration of du/dr between the boundry
conditions (u = ha and i = ra at r = a and
u = 0 at r = o) yields

3(n1+1)/2 3;

 

5e
ll

“1 ' 1 3(n2+1)/2 82

+ (2.37)

 

This equation is valid only for the temperature used
to evaluate its creep parameters. Other temperatures would
each require a series of compression creep tests to
determine additional sets of creep parameters.

Weaver and Morgenstern (1981b) assumed that a single

term form of Equation 2.37

3(n+1)/2 B Tn

__ = (2.38)
a n - 1

a
a

 

was better able to describe pile behavior in ice or ice-
rich soils. This equation was then used to determine pile

capacity, Ps’ thus

P = 21raL 1'

1.1
ZnaL [-EL](1/n) [
a

n - 1
3(n+1)/2

 

1 .
1(1/n) [—13—] um (2.39)

Weaver and Morgenstern (1981b) described an equation

that best fit uniaxial compression creep data for ice-poor

56

soils (Equation 2.25). This equation was similar in form
to Equation 2.33 (for ice and ice-rich soils), hence the
derivations used for pile displacement in ice and ice-rich
soils (Equations 2.34 through 2.38) were used by Weaver and
Morgenstern (1981b) with Equation 2.25 to give the

displacement (ua) for a pile in ice-poor soil, thus

3(c+1)/2 D Tc

b c - 1

 

and pile capacity PS, based on shaft shear stress ra, is

 

given by
PS . = 2 1r a L 1a
u C‘]. l
= _£L_ (1/C) (l/C) ___ (1/C)
27raL [atb] [ 3(C+l)/2] [D ] (2°41)

The relationship between pile capacity (applied stress),
pile displacement, and temperature as determined by Weaver
and Morgenstern for a pile frozen into Ottawa sand is

shown in Figure 2.17.

2.5.2. Penetration Deformation of Soil at Pile Base

The theory of cavity expansion was first used by
Bishop (1945) as a simple yet reasonably accurate model of
deep indentation. Bishop, et al. (1945) and Hill (1950)
believed the pressure that indents a deep hole in an

elastic-plastic material to be related to the pressure that

57

expands a cavity of equal volume in a similar material.
Johnston (1970) and Ladanyi (1966) approximated the volume
(Vs) displaced by punch penetration (if penetration is less

than the punch radius) as

2

V = n (r ) s (2.42)

io
where rio is the punch radius and s is punch penetration.

The volume (Vic) of the cavity before expansion is

2

Vi0 = - n (r
3

3
i0) (2.43)

By this theory, the volume (Vi) of the expanded cavity,

vi = vi0 + vS (2.44)

and a volume change ratio can be determined as

Vi 3 s

___ = 1 + 1—————— = l + 3 (2.45)

m In

io io
Ladanyi (1972) described the axial steady state creep

rate by an equation similar to Equation 2.33, thus

51 = a [ ] (2.46)
a f(0)

 

where 2c is a creep rate selected for calculation purposes,
n is a creep parameter, f(6) is a function of temperature,
and acuo is the creep proof stress at a temperature close
to the freezing point. If the instantaneous creep is

58

small, then this equation can be converted to a creep

strain equation

 

£1 .= [ ] (2.47)

0 15(9) t]('1/n) (2.48)

S C110

o
ll

[ c

is used for brevity.

For ice and ice-rich soils in a prefailure state
(according to the von Mises criterion) that behave as
described by the previous equation, the conditions of
equilibrium are shown by Ladanyi and Johnston (1974) as

d(ar) 20 -a

+ = O (2.49)
dr r '

 

where or is the radial stress and at is the circumferential

stress. The solutions for this equation are given by

Odquist and Hult (1962) and Hult (1966) as

r
“r = (are-p0) [—91‘3/m + 10,, (2.50)
r
and
r
at = (are-p0) [1 -(3/2n)1 [jg—1‘3”) + p0 (2.51)

where radial stress are at radius re can be found in the

prefailure zone and po is the hydrostatic pressure.

Ladanyi and Johnston (1974) describe the resulting

59

displacement (at radius, r) equation as

u 1 30 - 3p e
___ = ___ [ re ° 3“ [___]3 (2.52)

r 2 2 n as r

 

Ladanyi and Johnston (1974) also approximated the
relationship between the expanded cavity volume Vi and the

initial cavity volume Vio as

V.
1

 

3 ](‘3)

a [l + u./r. ] a [l - ui/ri

1 lo (2.53)

V.
10

where rio is the initial cavity radius, ri is the expanded

cavity radius, and ui is the displacement of the cavity in

the radial direction. A combination of Equations 2.45 and

2.53 yields
3s

B

[1 - ui/ri](’3) - 1 (2.54)

Combining this equation with Equation 2.52 yields

-——+1 (1-—[ 1 °1

B 2 2 n as

 

“ )('3) (2.55)

when ( or = pi) and (re = r.).

e 1

According to Ladanyi and Johnston (1974), Hoff’s
elastic analogue allows the strain relationship of
Equation 2.47 to be replaced by the steady state creep rate
relationship of Equation 2.46. A derivation similar to

those shown in Equations 2.49 through 2.51 yields a

6O

displacement rate, hi equation for ice and ice-rich soils
as

1 3p. - 3p
= _ [ 1 ° 1“ (2.56)

2 21105

 

H l C'
1"“ P'

Weaver and Morgenstern (1981b) carried this equation

further and showed a settlement rate equation,

{1 2 B 30
._a__ = —.{1 - _ [.__§_]n}(-3) - — (2.57)
a 3 2 2n

where 0E is the pressure at the pile base and

N

La)

1 a
B = = c (2.58)

W) 1“

 

 

s [ acuo

Equation 2.57 is a rearrangement of equation 2.48 and is
used for brevity. Equation 2.57 was simplified by Weaver

and Morgenstern (1981b) to
—- = B [——1 (2.59)
This simplification allows easier (and reasonably accurate)

estimation of pile end bearing in ice and ice-rich soils as

1.1 2 1r n a2
P = [———"5‘——](1/n> (2.60)
D a 3

 

For ice-poor soils, described by Ladanyi (1972) in

Equation 2.23, Weaver and Morgenstern (1981b) reported that

61

the effect of confining pressure was limited (Ladanyi and
Johnson, 1974) for pile end bearing and used Equation 2.25
to describe soil behavior. Equation 2.25 is similar to
Equation 2.47, for ice and ice-rich soils, hence derivation
of pile end bearing capacity was similar for both soil
types. Weaver and Morgenstern (1981b) showed the pile

displacement equation for iceepoor soils as
= D [___—___] (2.61)

where b and c are the creep parameters and D is shown in
Equation 2.22. This equation was then converted to a pile
end bearing capacity,

u 2 n n a2

P = [ a.__.__b](1/n) (2.62)
D at 3

 

Weaver and Morgenstern (1981b) created a design chart to
show the relationships of pile end bearing capacity
(applied and bearing stress), pile settlement, and
temperature for a given pile frozen into Ottawa sand as
shown in Figure 2.18. Weaver and Morgenstern (1981b) noted
that adding a pile end bearing estimation to a pile (shear
on side) capacity estimation is only approximate because
peak end bearing capacity develops at a larger displacement

than peak friction pile capacity, as shown in Figure 2.19.

62

ones use UMMMMQMH Manounsomnoz use uo>nos you n
H“ ou use nuoouuo Ho msuaasowcw .ma.~ w
on: «m

 

 

 

63
2.5.3. Expansion of a Cylindrical Cavity by Pile Settlement

As the pile shown in Figure 2.20 settles, it will push
outward to expand the cavity in which the pile rests.
Lateral pressure will increase along the pile surface,
additional friction forces will develop, and the pile
capacity will increase.

An equation for lateral pressure increase was used by
Ladanyi and Johnston (1973) for analysis of pressuremeter
data. It combined an equation for primary creep with
Odquist’s (1966) rate equation for creep cavity expansion.
Ladanyi and Guichaoua (1985) rewrote this pressuremeter
equation to describe the pressure created by movement of

the pile (Figure 2.20) as

pi - po = c [“1 /riJ‘1/“’ (2.63)
and
n _ 2 b
C =0” (—) t‘ w“) { (——) [—Jb}(1/n) (2.64)
J3 /3 2
C

where pi is the pressure induced by soil expansion, po is

the initial earth pressure, b and n are creep parameters, t

is time, a is the proof stress, and 5c is the strain rate

c6
choosen for computation purposes. This expression for

lateral pressure was then used to estimate frictional

forces over the surface of the pile with

L/COSa

2n 6 = 2n] (pi - p0) (sina + tana COSa) ri dx (2.65)
0

using notations described in Figure 2.20. Pile settlement,

64

 

 

Figure 2.20. Notation for the theory of tapered piles
(after Ladanyi and Guichaoua 1985).

65
5 may be described as

‘u. = s tana (2.66)

where ui is the pile—cavity radial displacement, thus

Equation 2.65 can be integrated to give

(1 + tan§)/(tana) D
ZflQ = 21rC [—12 a:
2 - (1/n) 2

 

1 - d/D](l/n)

L

(1 - [d/D](2 ’ lfl”) s(l/n) (2.67)

This equation may be shown more simply as
s = K Q t (2.68)

which can be rearranged and differentiated to become the

pile settlement rate,

6 = b K Qn t(b'1) (2.69)

2.6. Cyclic Loading of Piles and Pile Driving

Due to a lack of existing data for cyclic loading
superimposed on statically loaded piles in frozen soils,
information on pile driving has been reviewed. Schmid’s
(1969) study of model piles vibrated into unfrozen
uniformly graded sand, showed several modes of vibration
for the pile (Figure 2.22). The lowest mode, described as

if the pile vibrated or moved as a rigid body with the

.Amoma cannon nouuev ones deepens: sq convenes madam
Hopes you nopsuuaded eonom nanuo> accesvouu huounnnu> .H~.~ decade

66

 

 

 

.a: . a
0000 0000 0000 000a 000“ 000.
.. 8. o. 9 o
. . (kw _ . t . t(4 P .
are 69¢ are .Lf._ .8

I/(‘\ Tu
— .l'

. ..._.. .
r0.
8... .. 833... 83o: :23 .35 .g

:8 52s as $6 3 «to sis 256 .
£93 5 I to 3 a: 82.. f.

2.... s33 .38.-a I 8.2 .

 

( SONflOd ) J

67

.Aaeaa eaaaom saunas nanoeasuaue

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

codename» users; on n63.» apnea mo and.“ 263 .NN.~ and»:
. . anhigstde
:0 00. 00.. 00: {K
‘
Q I . . 00008810,;
.uzca
on: 8.. e3 8.. 6:93....
83:82.
am

.iiirmius

 

 

 

 

 

68

soil, is shown as a wide flat peak in Figure 2.21. The
maximum penetration velocity occured at this frequency of
vibration. Schmid (1969) observed that most soil
penetration resistance occured at the pile tip during the
rapid tip impacts created by resonant vibration.

The lowered resistance along the pile shaft occurs
(Kovacs and Michitti, 1970) when the relatively stiff pile
moves in unison with the cyclic force, the soil/pile bond
breaks, and individual soil grains tend to rearrange
themselves. The pile/soil interface remains in a condition
of reduced friction by the continued application of the
cyclic force. This "fluidized" condition could not be
achieved in a very dense sand and it was proposed that the
lack of void space would not allow grain rearrangement
(Kovacs and Michitti, 1970).

Pile driving in frozen soils has been observed to
cause melting around the pile (Linell and Lobacz, 1980).
Dufour et a1. (1983) thought that the frozen soil yielded
plastically at the driven pile tip and melting followed.
The melting and rapid impacts associated with driving piles
in permafrost may create conditions quite different from
those associated with small cyclic loads.

Direct information on cyclic loading superimposed on
piles in frozen soils is limited to data presented by
Parameswaren (1982, 1984). A typical pile displacement—

time curve (Parameswaren, 1984 shown in Figure 2.23)

69

.Agma Ganesha-Bush ~3on 25» nu“:

heao coneum nu

 

 

 

 

 

 

 

 

 

 

 

03; too: we uncapped—mac on» 9:32? 2550 moouo 13.3.3. .n~.~ snow:
3 .22:
cm~_ con. omN. coo. ens com om~
. q . q - q ltil
. as. 22.6 a. titul--ttt
\‘Islls'- I‘D! 0‘2 “NOAH h.
(tnfifouh \ti/
T 6
1.
\tii/ To. .. 25 a.
X e n e. .. 3.. :
a o. .. 3... o.
t \t . n-
s“ n.2 .. em a e
\ o. 72 .. 2.6 o
n o. .. no.- u
2 via. s opts o
rt s
\s 0 use N at I h 2:; 50:23
:2 >30 3.5 20:20:. sue: 02:42-32 2902.
Z. a; .U.. {one 6:: 352305.35
— P _ p _ P

o

3 .c

:9.—

'1N3W33V1d5l0

llllll

70

represents a peak to peak cyclic load of nearly 5% of the
static pile load superimposed on the model pile. The
displacement rate for region 7 of this figure was more than
doubled by the added cyclic loading. This increase in pile
displacement (settlement) rate must be accounted for in
determining the design load per pile. Suitable
constitutive relations for predicting the increase in
displacement rate as a function of cyclic loading
(amplitude, frequency, etc) have not been available for

frozen soils.

MATERIALS, SAMPLE PREPARATION, EQUIPMENT,
AND TESTING PROCEDURES

3.1. Materials

Uniform soil samples were needed to examine the
displacement behavior of model piles embedded in the frozen
soil. Local differences within a frozen soil sample could,
for a period of time, change a pile’s displacement rate and
distort data collected during that period. Comparisons of
individual tests were more meaningful when samples were
closely similar. Sand was used to produce uniform samples
because it was not prone to develop ice lens during
freezing, its unfrozen water content varied little, and its
gradation could be controlled. Sand was also a good choice
because it is commonly used in slurries that are frozen.
around field piles. A commercial washed silica sand (from
Unimin Corporation, Oregon, Illinois) with subangular
particles of specific gravity at 2.65, a gradation between
0.075 mm and 0.840 mm, and a D50 of 0.40 mm was readily
available. Some samples used only that sand portion
between 0.420 mm and 0.590 mm to improve local gradation
control throughout each sample. Distilled water was used
for all saturated frozen sand samples and all ice samples
to reduce adfreeze bond variability due to water impurities
(Alwahab, 1983).

The model piles needed to be stiff, strong, and have a
constant surface roughness. The roughness of corrugated

field piles (Thomas and Luscher, 1980) was simulated by

71

72

adding oversize protrusions, called lugs, to relatively
smooth 9.52 mm diameter steel rods as shown in Figure 3.1.
Lug heights ranged from 0.76 mm to 3.18 mm. Another
surface roughness type tested used U.S. standard screw
threads, fine (24 per inch) and coarse (16 per inch),
machined into a 9.52 mm diameter steel rod. A variation of
72 threads per inch was specially cut into a third 9.52 mm

diameter steel rod.

3.2. Sample Preparation

Density, freezing method, lateral stress on the pile,
boundary interference with the pile, and pile alignment all
influenced loaded pile behavior. Sample preparation
methods which controlled these factors were developed and
refined to provide closely similar samples.

A split mold was first clamped around the soil
restraint plate. The lugged piles were cleaned using
longitudinal strokes with a #400 (for metal) sandpaper
followed by wiping the pile surface to a dull shine with a
paper towel. A wire brush was used to dislodge any
clinging materials from the threaded piles before wiping.
Each model pile was then placed through a hole in the
restraint plate, braced at its ends in a position normal to
the plate, and sealed into this hole with a pliable putty.
In some cases, plastic tubing was used to cover part of the

pile. After the sample was frozen, this tubing was removed

73

P

Dynamic Load, d

1

Static Load, P T
f‘

Pile Covered
with Plastic
Tube, then

Removed after

 

. 7 Freez ing to
Pile Form a Void
Displacement Space for

Simulation of
Lug Spacing

we.

I

 

>1: lair.

'. -'Soil: .

'j Frozen-.1

 

r-E--

'SThermistor '
. Q ‘

. .t-r——-q

A .

\L e

:14.

_J

 

:89)-
le.fi L Lug:.-.'
-. 2"Height:

 

Displacement"‘

 

,Limit " :'. .

-_':

.' - tip-'3 '

w

.
' a

pie Immersed

Protected Sam

 

*Jd.f.Diameter

 

 

 

g2

 

 

Medel Pile

Silicon Grease Applied—
Behind Lug to Reduce
Load Transfer on

Selected Piles

Figure 3.1.

in Liquid Coolant

Model pile embedded in frozen sand.
Lug provides controlled surface roughness.

74

 

8i

 

 

i

 

 

 

 

\\\X\\\\\\T

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

l”
\\\\\\\\\

 

 

 

 

 

 

 

r

 

 

 

 

 

.ucoaduaao acuunou dado «one: .~.n shaman

nouoaoneaeoo<
neuemaouuo use noon
uoueneceo couuocam
neuuuume< meson
sexesm eunuchconuoedm
enemas: nausea

uo>aa acumen can
one“: so nemcm euucx
ue>oA

emoueoduuuno unencuw
unneeded

neuuueo deepen

 

 

 

 

 

he

a m

 

 

izodds69553£

 

 

 

 

Let. i
.p 3 > O
JMI _ _1 r.)

dado coca
nouoauuasz wsuwmoq
boundam com
Houzdaoo adconuom
momma; coma

and ea>o

uoxonun ocunmunom
antennae: Henna:
ounam usueuunom Auom
teem Cancun
acunuenonh

edum aoeum deco:

 

 

 

 

étmtilmtdfdcszirtH:Z-J

 

 

 

 

 

 

 

 

 

75
to leave a void (with a diameter equal to that of the lug)

ahead of the lug as shown in Figure 3.1. A thermistor was
braced at a position within the mold that would avoid
interference with sand displacement mechanisms involved
with movement of the model pile, yet provide representative
temperatures along the embedded model pile. The mold edges
were sealed with pliable putty before placing the sample
mold assembly in a top-opening freezer (Sears Kenmore 15)
to chill at -14 deg C.

A layer of water was then placed into the mold. Sand
was slowly poured from a small funnel through the standing
water to settle lightly on the surface of the accumulating
particles. As the sand’s top surface increased within the
mold, water was added as needed to maintain a level above
the sand. Periodic tapping on the sides of the mold served
to increase the density of some samples. The average solid
volume fraction for loose sand was 0.58 and for dense sand
samples was 0.64. The average ice matrix density for these
samples was 0.89 Mg/m3. These values are shown for each
individual sample in Appendix B.

A different procedure was used to fill the mold for
preparation of polycrystalline ice samples. Pans of
distilled water were first frozen, then broken into pieces,
and passed thru a chilled food processor (Cusinart). The
finely chopped ice (average particle size approximately 1

mm) was sprinkled into a 1 cm layer of chilled water (near

76
0 deg C). After this water layer was saturated with ice
particles, the mixture was stirred slowly to evenly
distribute the ice particles and release any trapped air
bubbles. Another 1 cm layer of chilled water was then
slowly added to repeat this process until the mold was
sufficiently filled. After insulating the mold (described
in the next paragraph), chilled water would freeze around
the ice particles to form a polycrystalline ice sample.
The average ice density of these samples was 0.90 Mg/m3.
These procedures were followed for both ice and sand/ice
samples.

To encourage longitudinal freezing along the pile axis
and minimize lateral pile pressures due to freezing of
trapped water, insulation was added around the mold so that
the top was more heavily insulated than the sides. The
bottom was uninsulated and left open to the -14 deg C
freezer temperature. After 24 hours, the excess ice or
frozen sand was scraped away from the frozen sample and the
split mold halves were removed. The frozen sand was
suspended below the reaction plate by adfreeze bonds with
the model pile. Triple rubber membranes were placed around
the sample to protect it from the ethylene glycol-water
coolant and to help minimize effects of small temperature
changes of the surrounding coolant. Open ends of these
membranes were sealed with gaskets and hose clamps to the

soil restraint plate and a lower circular plate. This

77

effectively sealed the bottom section of the pile and
sample from the coolant. A small plastic tube, sealed into
a hole thru the lower circular plate, allowed a thermistor
lead wire to pass outside of the coolant tank.

The model pile, solidly frozen into the sample,
extended through a hole in the soil restraint plate. A
rubber hose, placed over this end of the model pile, was
clamped to the soil restraint plate. The open end of this
hose will later rise above the liquid coolant to allow the
protruding pile end to be attached to the loading system.
The hose protects the upper section of the pile (and
sample) from the coolant while offering no restriction to
pile movement.

The putty that sealed the opening between the 9.52me
diameter pile and the 17.5 mm hole in the soil restraint
plate during sample preparation was left in position
because it remained pliable at test temperatures, offered
little resistance to pile movement, and protected the
frozen sample from small coolant leaks. Earlier tests with
frozen sand samples under similar conditions showed no
appreciable influence from the restraint plate (Alwahab,
1983) for a lugged pile, centered in a 152 mm by 152 mm
sand sample. Preliminary tests during this study, using
threaded piles in ice samples, showed no apparent change in
pile behavior when additional putty was used to create a

larger unsupported section at the restraint plate.

78
3.3. Temperature Control Equipment

The prepared sample was then transferred to a well
insulated 58 liter coolant tank and attached to a restraint
frame that could pivot in all directions (within limits) on
a 31 kN capacity bearing (a Spherco rod end). The ethylene
glycol-water coolant, colder than the test temperature, was
then supplied by a circulating coolant bath (Forma
Scientific 2425 Bath and Circulator) to fill the tank and
quickly chill the testing equipment within the tank. The
coolant bath was then adjusted to the warmer test
temperature. Inflow from the coolant bath, colder than the
outflow from the tank, was directed onto the top of the
soil restraint plate to produce, as shown by preliminary
tests (based on up to 6 internal sample temperature
measurements), a nearly uniform temperature gradient along
the pile’s length.

The wire lead to the sample thermistor (Fenwal
Electronics UUA35J84) was connected to a digital data
logger (Hewlett-Packard 3467A Logging Multimeter). The
internal sample temperature was checked to determine when
coolant bath adjustments were needed. The thermistor had
been calibrated earlier in an ice bath and had an accuracy
of i 0.1 deg C. The laboratory room temperature was
maintained near constant to assist the coolant bath’s

temperature control.

79

3.4. Displacement Measurement Equipment

With the frozen soil and model pile in position and
the coolant flow established, supports for the Linear
Variable Displacement Transducers (Schaevitz 1000 HR-DC and
3000 HR-DC) were attached, through stiff connections, to
the model pile. One or two LVDT’s were used in earlier
tests, but three LVDT’s were used in later tests for
greater measurement precision. These stiff connections
minimized strains or displacements not associated with pile
movement through the soil. Displacement was measured
between the LVDT supports and the soil restraint plate.

The LVDT’s were remotely powered by a DC source (Sola
Electric 24VDC1.2A) and their resulting DC output was read
and digitized by a data logger (Doric Minitrend 205).
Earlier, each LVDT and the data logger had been carefully
calibrated to i 0.003 mm by a LVDT micrometer (Schavitz),
Because of the relatively slow data logger response
(approximately 1 second) measurements of cyclic
displacements due to dynamic loading could not be directly
recorded. The data recorded during these periods
represents a sampling of the displacement data that could
later be averaged. The digitized signals and their
respective times were recorded by a personal computer
(Hewlett-Packard 85B) on magnetic tapes. The recorded
signals would later be run through a short program and

plotted on a pen plotter (Hewlett-Packard 7470A).

80

Since automated data recording allowed a great number
of data points to be collected, a simple data plotting
program was used to connect consecutively recorded data
points. Some graphs will appear as "stair steps" because
the data, plotted point to point, was recorded at the limit
of displacement transducer sensitivity. Gaps may appear in
some curves between data files or when data was lost.
Electrical interference caused some scatter of the plotted
data recorded during precise measurements of threaded pile
movement. Continuous recording at equal time intervals
would have facilitated best fit curve plots and is

recommended for future work.

3.5. Pile Loading and Load Measurement Equipment

Stiff connections, above the LVDT supports, were added
to connect the 4.45 kN capacity load cell (Strainsert FLlU-
ZSGKT). This load cell was remotely powered (Sanborn 8805A
Preamp Carrier) and its DC output was read from a strip
chart recorder (Hewlett-Packard 7702B Recorder) or from the
screen of a storage oscilloscope (Techtronix 564). Cyclic
loading at 10 Hz was more accurately measured by the
oscilloscope. The load cell, preamp carrier, recorder, and
oscilloscope were calibrated together on a separate load
frame using known weights.

A second pivoting connection was added above the load

cell to connect the lever loading arm and to minimize

81
eccentric loading from this lever (Figure 3.2). Knife-edge

type connections were used at the ends and fulcrum of this
lever to minimize friction and maintain a nearly constant
mechanical advantage of 4 times the static weight on the
model pile. The static load weights and holder (suspended
on a knife-edge connection) were temporarily supported by a
hydraulic jack until testing began.

The electrodynamic shaker (APS 113) was connected to
the lever arm so that 2 times the shaker force was added to
the pile’s static load. During some tests, the shaker was
connected directly over the model pile in order to provide
better-formed force waves. The pin-type connection between
the shaker armature and the lever was rigid yet could
rotate as the lever moved. This small amount of lever arm
movement did not alter the dynamic force magnitude
significantly during pile tests. A function generator
(Zenith SG-1271) signaled waveform, frequency, and
amplitude to the shaker’s power amplifier (APS 114) which
in turn supplied controlled power to the shaker. During
some tests, an accelerometer (PCB 308B) was used to check
the frequency and magnitude of accelerations at various

locations on the loading system.

3.6. Testing Procedures
Once all connections were made and coolant flow was

established, the coolant tank top was heavily insulated. A

82

very small load was applied (lever self weight) to bring
all system connections into contact and check the load
system alignment. Pivoting connections above and below the
sample appeared to remove almost all eccentric loads from
the pile. Force, temperature and displacement measurements
were started and rechecked during a 18 hour waiting period.
Consistant use of this waiting period, before adding test
loads to the pile, was thought to stabilize temperatures
and allow lateral pile pressures to diminish to similar
levels. After checking the recording systems, the reaction
force from static weights and their holder were slowly
lowered onto the lever arm by the hydraulic jack to begin a

pile test.

3.7. Model Pile Loading Procedures

Three different pile loading sequences were used to
study the model pile’s behavior. For the first case, load
was maintained constant throughout the test as shown for a
typical full duration load sequence labeled P3, P,, and P5
in Fig. 3.3. Small variations in temperature, sand density,
pile alignment, and other physical conditions altered pile
displacement rates and made direct comparisons between full
duration load tests less certain. Case P2’ and P1" (before
load increase) were considered as full duration loads up
until additional loads were added.

For the second sequence, loads were increased after

83

nut—=0 ucosoonadnac e39 dove... Hanan—ha. .n.n ouawum

0

— v
t L 9

v; n;

0!

 

(II) 3 ’aumttdlto 'Ud

<9

84
periods of displacement as in the typical step loads labeled

P,’, and P5’ in Fig. 3.3. Slight physical variations within
samples effected comparisons between step loads. Step loads
added after pile displacement rates reached nearly constant
values appeared to yield more consistent results.

For the third sequence, a constant load was maintained
until a nearly constant pile displacement rate was achieved.
The load was next increased for a short duration, then
reduced to its initial value. A nearly constant pile
displacement rate was allowed to return before the next
load increase. A short duration load is shown as P," in
Fig 3.3. The displacement rate resulting from a short
duration load increase was compared to the displacement
rate immediately before the short duration loading. This
comparison normalizes the effect of physical variables and
total pile displacement on pile displacement rates.
Parameswaren (1984) used a similar technique to measure the
effect of dynamic loading on a constantly-displacing model
pile frozen in soil.

In a variation of the short duration loading sequence,
more than one short duration load was added in succession
(as shown by P2" and P3" in Fig. 3.3) before the load was
reduced to its initial value. These multiple-short
duration loads have the same advantages of the single loads

and provide additional data.

85
3.8. Displacement-Time Data Plotting Techniques

To examine and emphasize data obtained during periods
of nearly constant displacement rates, log (displacement) -
log (time) plots were started at different points along the
pile displacement. When similar (displacement) starting
points were used to plot data for different tests, the creep
curves could be compared. Three types of starting points
will be described.

Point A on the linear displacement - linear time insert
of Fig. 3.4 represents the initial time and initial
displacement for a load increase on a model pile. The
selection of an A-type starting point will affect the
linearity of the initial portion of a log (displacement) -
log (time) plot. Small data inconsistencies, such as
elastic deformations and system seating errors, distort the
initial portions of these plots.

Point C (Figure 3.4, insert) represents a data point at
displacement 86C after a load increase. Pile displacement
data for different tests may be more closely compared when
their data plots are started after a given displacement with
a C-type starting point. As the curves approached nearly

steady state conditions (after 86 displacement), the

C
variation of etc between curves becomes less important.
Typical log (displacement) - log (time) curves P3, P4, and
P5 (Fig. 3.4) are full duration loads (also shown in Fig.

3.3 by curves P3, P4, & P5) with C-type initial points after

86

.ccen couonm ca ucoao>oa swam demos new
no>uso Ausoaoooadmucvmoa-AoauuvmoH Heoumha

A- v . sea...

.¢.m ousmfim

7

wt
I
0

rm

0

o

a.

S

m

 

 

IO

' dItp

0!
<-

87

a given displacement. Curves P,’ and P5’ of Fig. 3.4 (also
shown in Fig. 3.3 as P,’ and P5’) represent step loads plotted
with C-type initial points.

Point B (Fig. 3.4, insert) remains at the initial time of
load increase but is ASE (a given displacement) after the
increase. To compare load increases with a B-type starting

point, a value of A5 was arbitrarily chosen to produce the

B
most linear, most parallel log (displacement) - log (time)
curves as idealized by P2", P3", and P," of Fig. 3.4 (also
shown in Fig. 3.3 as P2", P3", and P,"). The 2 curves labeled
P," (Fig. 3.4) use a C-type beginning at a time and
displacement just before the short duration loads were

added. The ratio of short duration load, P," and the

initial load, P " may be compared with other short—

1b
duration/initial-load ratios or it may be used in equation
3.4. The multiple-short duration loads P2" and P3" may be

directly compared using equation 3.4.
3.9. Determination of Pile Settlement Equation Parameters
Pile displacement - time equations, similar to the

primary creep equations given by Ladanyi and Johntson

(1974), can be written as

5 = [-——] [-—-] t (3°1)

where 6 is pile displacement, P is load, 50 is an arbitrary

88

displacement rate corresponding to a proof load Pc’ t is
time, n and b are exponents in the creep equation. These
equation parameters may be determined graphically as in
Figure 3.5.

Differentiating Eq. 3.1 with respect to time gives

C b “ bt(b-l) (3.2)

One
II
P"!
6...]
t—I
I—J

where 3 is the pile displacement rate. Taking a ratio of 2

displacement rates, as described by equation 3.2, gives

0”

P t
_ = [_LJn [41001)
Pl t1

(3.3)

One
H

For t2 = t1 and/or b = 1, equation 3.3 reduces to

' P
—2 = [-2—]n (3.4)

3.

When displacement rates are nearly constant, b = 1 can be
assumed. Using n determined by equation 3.4 and assuming

b = 1, a value for 5c can be arbitrarily chosen and Pc may

be determined.

89

Stress difference a, - o3, MN/m2

 

   
 

10-3 r r r rrttri i ii rurrrrT T7 r rrT ‘10-;
l l f / 3
+ I / ..
f- l A/ .4
....... __.____ _ __
i-rm . '1, , 4
--# 1-b'tana ..
Ii-tunp

104 10-4

 

 

 

7: a. _ lbli,/bl'] "" ‘3
3:. c, j
E :
E 1.0 llllN/m2 - a, - a, "
g r- / ..
i /« 0.517 MN/m’
O- 104 F/ a'in‘ '03l 310-6
: :
C :
b a, - a, - 0.26 MN/m2
10" 1 t, L 11.1Lll 1 1 1 1 (,1417 1 ,7 1 13,..- 10.5
1 10 100 1000

Time t,h

Figure 3.5. Primary-creep parameters from time-creep-strain
data. (after Andersland, et a1. 1978).

Factor 0

EXPERIMENTAL RESULTS

Prediction of pile settlement due to superimposed
cyclic loads first required information on pile behavior.
under constant loads and after static load increases. The
results from static load tests on various model piles
provided examples of typical model pile performance. Use
of new interpretation techniques with time versus
displacement data allowed detailed analysis of this pile
behavior. These techniques also provided a better
understanding of pile displacement rate increases from
static load increases. With this knowledge of static pile
behavior, changes due to cyclic loads for various load
magnitudes and frequencies could be studied. Comparisons
between static load increases and superimposed cyclic loads
allowed the effects of cyclic load variables (frequency and

magnitude) to be quantified.

4.1 Sample Inspection After Testing
The frozen soil and model piles were examined after
each test for contamination, sample imperfections and
information on soil-pile interaction. Shallow grooves cut
parallel to the model pile on opposite sides of the frozen
soil allowed the sample to be split cleanly into 2 parts.
Changes in color and the presence of crushed particles

in front of the lug were noted in the otherwise uniformly-

90

91

colored, uniformly-graded sand. The compacted sand and ice
formed a wedge with close to a 45 deg angle in front of the
lug. An open void always formed behind the lug after pile
displacement thru frozen sand and ice. This void, although
impossible to measure precisely because of sample
disturbance during equipment disassembly, was nearly equal
to the pile displacement measured during the test. Almost
all pile displacement resulted from relative movement
between the pile and the surrounding frozen soil.

When a void, created ahead of a lug, was used to
simulate one lug in a series of lugs (Figure 3.1), a
cylinder of frozen sand was sheared off and pushed ahead of
this lug before the lug reached the void. Pile
displacement rates increased just before this failure in
shear. The outside diameter of this cylinder of sand
equalled the lug diameter, the inside diameter equalled the
pile shaft diameter, and the length was observed to be
slightly less than the displacement remaining at the start
of pile failure. For example, all 0.79 mm height lugs
tested in this manner were observed with 4 to 6 mm frozen
soil pushed ahead of their lugs while 8 to 17 mm was pushed
ahead of the 3.18 mm height lugs. These cylinder lengths
are given for each sample in Appendix B.

A void was also observed after pile displacement
through polycrystalline ice samples. A white, possibly

crushed ice zone (vs clear for undisturbed ice) was found

92

ahead of the lug. In regions where rapid lug displacement
had occured, the white zones extended radially outward into
the sample suggesting a larger fracture zone.

Changes in sand color and the presence of some crushed
particles were found along the length of threaded rods
frozen in sand. Some particles remained wedged between the
threads but between many threads only bits of ice were
found. The slip surface occured at the outer edge of the
pile threads. A number of loose sand grains were found
between the pile and the adjacent discolored sand/ice
solid.

The threaded-rod/polycrystalline-ice samples showed a
white zone along the length of the pile at the pile/ice
interface. The slip surface occured between the inner edge
of this zone and the outer edge of the pile threads. It
was common to find ice, initially frozen between two
threads, had been displaced so that only a fraction of it
remained wedged between ice slip surface and the top of the
trailing lug.

It appears that pressure melting of ice, movement of
unfrozen water to adjacent areas with lower pressures, and
refreezing describes part of the process responsible for

pile creep displacements.

4.2 Full Duration Loads on Lugged Model Piles

Full duration loads on lugged model piles frozen in

93
loose sand are shown in Figure 4.1. A void space,
beginning at 25.4 mm ahead of the pile lug’s leading edge
and equal in diameter to the lug, limited the amount of
direct pile to frozen sand contact area and served to
simulate the condition of one lug in a series of lugs.
Limiting the lug displacement in this way caused these
piles to increase their displacement rates and begin to
fail after 8 to 9 mm of total displacement. Little or no
secondary creep (constant displacement) was observed during
these pile load tests. Plotting the test data with
log(displacement)-log(time) coordinates usually transformed
these results into relatively straight lines that were
overly effected by small aborations during initial pile
displacements. Starting each pile test plot after a given
displacement reduced plot variations and provided better
comparisons between pile test results.

Various C-type starting points (Figure 3.4) have been
used in plots of a single pile test in Figures 4.2a, 4.2b,
and 4.2c (Pile Tests #105, #114, and #111 respectively).
The differences between the 0 mm and the 1 mm C-type
starting point plots in these figures are apparent.
Changes are smaller between larger C-type starting point
plots (such as 4.75 mm and the 6.00 mm C-type starting
point plots) shown in figures 4.2b and 4.2c. The
displacement rates of these piles are clearly displacement

dependent.

94

.ocee senoum cu eufiuuos and as a~.n new:
sound aeeue demon new nuance usoaoueamuucioaua .a.¢ ouswqm

Aug. u .oaae

 

 

ow mm om mm am my 0" m o
J‘4 1 11 . ‘ d ‘ 1 q d J 1 I q i d 1“ d J G ‘ 1 q d 4 4 d 1‘ 1 J )1 4 q 4 d d 1

as a.nu as on.n can» 2: HN.~ was
an inn 3 SN o:
5. «.3 28 eo.~ o:
as {on 23 36 :23 23 an.“ 2:
ea e.n~ zx o~.m gas
so £33 a .33 .3.— .oz use...
udofleoeamean eaum
\

\ s

\\
mo: \ \
\ ‘\ s
t . e 7///
unnamed page
no uueum r
\
eggs «gas page -Hs .

 

 

(mm) 9 '3uama02tdSIo attd

 

95

.ncat uaos any .uucuoa unuuuaun amhuno a:0«uu>
«can: and» caua a you u¢>uao AucoaooaamauvvmoH-AoauquoA

 

.N.¢ ouawum
3565 u .32.
2: S o g 1o
d \ \ \ due
\
\ \ \\
muafiom wcfiuumum «guano \\\\ \

 

.2. ¢ . e \
\
\\..\
\

:2: m

33..“

:2: o

 

 

Cd

(mm) 9 ‘3uauaoatd31q 3114

96

.nuauon usuuuauu a=o«u¢> nun:

 

   

«gut uuoh How no>u=u nusoloooaaauvvmoa-noauuvmoa .nu.q ouamum
oo~ Cu Anusozv u .oauh o.~ ~.o
ousaficm oHHm mo unaum nu
.o.~

mucuom wcw

as m~.c
as c.c

 

unhanu

     

unaum

 

cg

 

(mm) 9 ‘3uamaoctds1q 3115

97

cod

.uucuoa nuauuaua nsouuu> nu“:
and; unoh wow uo>hso nuaoacouamaquon-noauuvmoa

cg Anuaosv u .anua o.~

 

 

d

     

wusafiam uawm mo uuaum nu

ucfiom mafiuuaum wa>auo as

as mn.c
33 «.¢

~.o
g.

.o~.¢ ouswum

o.~

 

 

o~

0

(mm) 9 ‘3uamaow1dstq aIIJ

98
Pile Tests #105, #114, and #111, plotted with equal

C-type starting points, are shown in Figures 4.3a through
4.3e. The beginning portions of these curves,
(particularly the 0 mm starting point curves in Figure
4.3a) are sometimes erratic as shown by the high
sensitivity of the log coordinate system at its lower end.
This portion of each curve (the initial 1 mm or less) was
neglected when fitting parallel straight-lines for each

group of curves shown in Figures 4.3a through 4.3e.

Parallel straight-line-approximations can then be
treated, as Andersland et al. (1978) did with unconfined
compression of frozen soils shown in Figure 3.5, to
determine the material creep parameter b (dimensionless), n
(dimensionless), and ac (proof compression stress at an"
arbitrarily chosen strain rate 2C). Since pile loads will
be analyzed instead of compression stresses, PC (model pile

proof load) is substituted for ac and 6c (arbitrary
reference pile displacement rate) for 2c as shown on
Figures 4.3a through 4.3c.

These parameters can then be used in the pile creep

equation (3.1)

a = (5c / b)b (P / PC)“ tb (3.1)

to predict pile displacement. Note that creep equation
parameters derived from curves with C-type starting points

other than 0 mm, can only predict pile displacement beyond

99

.3.—«om msuuuaua 2;»..0 a! o 59.; A3 .33: usoaoonammwv as «.mu can u . n- w

and. Banana 3.. 3530: ms.— II 3.." :33 no.3." dance now «25.5 usoaoouaaaavész.

0A:

 

 

 

       

 

.n.¢ ouamum
cu A9505 u .33. a; 70
. 1 #5

0:3 .30: a a
23 30033.5 \ ..T...
2630 we a
cauuaauxouaa< m
25..— unwuauum _U a
I
.04 m
ousawcm mafia mo unaum O m
.m
OJ
m

3: £33.90“ w a. on; u 0..

z“: 3 a... as u co

Tia. u c 3.0 u m

nuouoluuom scauasam dunno

.:~‘
A:

 

100

oo~

o~

.musuoa msuuunun onhu-o as H soda .bm.q ousmum

Anusosv u .oluh

o.~

 

 

mofia

¢-*

-*

d

  

GM

0
u:\aas.on «.w zxom.summ
zxs an as s~o.c n o
e.« a c me.o u a

ouoaduam acquazom cacao

 

OH

 

(mm) 9 ‘3uauaoatd31q 311d

101

OOH

.nuSHOQ mauuunun onhu.0 as m sud: .on.¢ ouswum

A.u=o;c u .onsa

 

 

 

e-&

fiam‘

0
pie... to: a o 525an
a: an a... :85 u o

- was; 36:;

nuouoldunm caduusam Queue

 

 

—.o

~.o

M
Q

Ca

(mu) 9 ‘3uamaoatd31q 3135

102

oo~

.ausuoa msuuuauu onzu-o as ¢.c nus: .om.¢ ousmum

Aauaozv u .o-«a

 

«afi‘

O 0
as}... to- w e saws-om
H-a 23H an a: c~oc.o I o
- mé .. a g... u a

nuouoaduam noducaom noouo

 

 

 

~.o

m.o

0“

(mm) 9 ‘nuamaoaIdSIQ atya

 

103

 

.musuoa mcuuuaum oQAu-o as m no“: .mm.¢ muswum
OOH Ow Aauaosv u .osuh fl ~.o
~.c
m.o
~
«fid‘ “as
o u
“axes s.cn « e zxom.~uoa

c.¢ I s

-

 

23H um 35 c~co.o I o

am.o I A.

uuouoaduam coauasvm dunno

 

 

o~

(mm) 9 “nuauaoutdsxa 3113

104

their respective starting points. However, the parameters
obtained from curves started at C-type points other than 0
mm can be used to compare pile types, soil types, and other
test variables.

Full duration load tests for 6 model pile types are.
plotted with various C-type starting points in Appendix A.
The creep parameters obtained from these figures are
included with those from Figures 4.3a through 4.3e and are
included in Table 4.1. At this time note that most model
piles with C-type starting points of 3 mm and greater, have
b values nearly equal to 1.0. Straight line approximations
of these curves represent nearly constant displacement
rates.

Full duration pile load tests demonstrate pile
behavior and failure displacement in the most reliable
manner, but are time consuming since more than one pile
test is needed to describe general pile behavior. Adding
incremental loads to a pile’s initial load allows

comparisons to be made during a single pile test.

4.3 Step Loads on Lugged Model Piles

Displacements for 2 similar piles, shown in Figure
4.1, are very similar to those in Figure 4.4. After a pile
displacement of 2.75 mm, the pile load was increased during
Pile Test #107. This same load increase was applied,after

a pile displacement of 4.75 mm, to Pile Test #105 . In

105

nun—anon 32233.26
0 now a. no can. :- on.o a» sc.o
u out n- on can. I: co.o oa o~o.o .-
u a... a- a. v...- I 36 3 {are .-

uI o.

 

 

 

 

 

 

 

 

 

 

 

 

as}... done! a... on 15 3583... o.— ..a. a a. 33...... u a
-----_------_-----_-----_-----+------+-----+-----_-----+------+-----+-----.-----+------+-----+-----_-----+------+-----+-----_-------------+-----. .....
. _ . _ _ _ _ _«u.« _onoo.e_ ~.n _ c.. _ss.« . «no.9. o.. _no.o _o~.~ . o~.a_ ..~ _s..o _ u «a to.a~ _ ~.< .
-----_--Ju--_-----_-----_-----_------_-----_-----_-----_------_-----_-----_-----.------_-----.-----_-----_------_-----_-----_-------------.-----_
on.« _ssoo.o. a.n .oo.o _ _ . . .n~.« .cnoo.o_ «.4 _so.o _u..u . ano.o_ o.n _no.o _on.. _ n..o_ o.u _on.o . o «a \¢.nu . «.4 _
-----_------.-----_-----_-----_------_-----_-----_-----.------_-----_-----_-----_------.-----.-----_-----_------_-----_-----_-------------_-----_

. _ . _on.« _noco.o_ o.c _oo.o _~«.~ _uooo.o. o.n _o..o _o~.— . osa.o_ 4.x .no.o .on.~ . o«.o_ ~.« _a..o . c on e as _ o.< _o~.n
IIIII_-InnuI-uu-nu_nn-uu_onnuu+IIIII 4 4 _ 4 -Iun4uuuuu4unaun_-onuu4uuuu-o+uunou4ncu-a_-c---4---u--4--uu-+u----_u-------o-un-4----:4n--I-
so.“ _ «ao.o. ~.o _ a." . _ _ . _«u.~ . N«.3 a.» .~..e _oo.e _ a..o_ n.« _no.o _oo.o _ o.~_ o.~ _so.o _ o no \..~. . o.< .
-----_------_-----_-----.-----.------_-----_-----_-----_------.-----_-----_-----_------_-----_-----_-----_------_-----_-----.-------------_-----. .
«c.u _ «no.9. -.o _uo.o _ _ _ . _ua.~ . sua.a_ a.» _~..o _on.n _ nno.o_ o.n _n~.o _.«.n _ o~.o. 4.. .on.o _ c on to.n~._ o.< _

. _ _ _ - . __._____.___..___

_ _ . .«~.~ _osoo.o_ a.~ _ a." _oo.~ .ouoo.e_ o.~ _uo.a _o¢.« _ o—a.o_ n.n _n~.o _on.~ _ n~.o_ o.n .cn.o . o no 4 a. _ ..< _an.~

_ _ _ _ 4 III+IIIII+IIIII_n-nnn+uunucu4ouuua4n-uun- 4 4 uu4unuuu_IIIII4IIIIII4IIIII4IIIII_IIIIIIIIIIIII4II-I-+I-I--
oo.c . «u.o. 4.“ . 9.“ _ . . . _...o _ ou.o. «.5 _no.e _oa.o _ ...e_ n.o _o..e _o..e . ...a_ n.n _n..o _ u so as..« _as.< _

_ . . _ . _-----.-----_-----.------.-----_-----_-----.------.-----_-----_-----_------_-----_-----_-------------_-----_
an.~ _«sooo._~ an _ o.~ . _ _ . _nn.~ .«woo.o_~ nu _co.e _nn.n _ coo.o_ ... .so.o _co.. _ on.o_ a.“ _on.o _ .- no ~ as _u~.< .

_ _ . _ _------_ - _- _ . -_-----.-----. - _ _ _-----_-----_------_-----_----r_-------------_-----_
co.“ _ o~._ o.- . a." _ _ . . _ao.e _ .«.o. n.. _uo.e _no.o . e«.o. u.~ _u..o _...o _ nu.o_ «.. _co.o . +. no 4 a. .n~.< .a~.o

u u u _ u u u _ u 0 u _ u u u _ u o o _ u 0
5.1.5... _ amid... _ .515... _ 7"...“- _ salon... _ _ I2... _I.

..u.._.._._u..._...._u..._..... 3...... "1...... .3..

4 4 4 _ 4 4:1 4 _«naua4IIIIII4nuuacocu---_ r 4 4 .IIIII4IIIIII4IIIII4IIIII_ usage-\vooa<_ .os.ou.n

I. o I .au .33.. 33-2 I n .. :t 2.3. 85-0. I. a a .8 33.. 15-0. I n I .2. :3- oauaé. .l o u :8 an...» é»-o_u30oiu..><_ .3.: 9...

 

- - _ . _ - - _ --- - ----_-------------.-----.-----

.oouuu woods; «0 canon cauuausa Adah Iona nucuoluunm coda-awn Ao-uu .~.c can-h

106

cod

Anusonv u .0luh
On

‘

.usaon mauuuaua omhu-o as A a mafia: econ sououm sq mecca
unauaua ou vouoowasa moans uumuaua mo Mea>a£on usoaoonanmav-oaaa

.q.q «Hausa

a.

 

 

unaum amok
HOUH< n! H UOUHdum 0:30
2x o~.« @ hose uaoa cash

uuuum ouch
uouu< a! a vouuaum o>uso
23 cN.N w ncat uuoh «Hum

       
    

.o mac n. a scam a: mm.o-«<.o
awn—«A usoaooaanmuo as :43
uamsoz was as as."

F‘o

 

 

CH

(mm) 9 ‘3uam9091d81a atxa

107

.ausuon wsuuuauu onhu-o as n~.n can: auosm .aomuouosa oaoa.uaafisum
on vouuonbsu vane canopy an undue doves unauaua an» no uo«>a£ua usosooaaemfiv-osuh .m.¢ ouswfim

muse .0!
cod as A .c u «a s s.o
fl

 

C

.o

     
 

unaum uaoa wouu<
ea nn.n .oauouosu smog
uouu< as oo.a vouuuum o>uso

zx mo.n @ moat uuoa mafia

H
(mm) 9 ‘nuaueontdsxq 3114

uncum uuoe
uouu< as nn.n .oaaouosn coca
uoumd as co.n vouuuum o>uso
2: ac.« w sass unoa «sum

 

 

Ca

 

108

Figure 4.5, displacement data from #107 is shown using a C-
type starting point of 5.75 mm (3 mm after after the load
increase). ’Based on our knowledge of full duration pile
behavior after a 3 mm displacement, Pile Test #107 can be
assumed to be close to the steady state creep rate
condition. Pile Test #105, plotted in Figure 4.5 using a
5.75 mm C-type starting point (1 mm after its load
increase) nearly matches Pile Test #107’s behavior. A
comparison of plots in Figure 4.5 with Pile test #111 (a
full duration test plotted with a 6 mm C-type starting
point) in Figure 4.3e would show all three to be similar
and would appear to represent general pile behavior.

The log (time) - log (displacement) curves for a model
pile, started 1 mm before and 1 mm after a load change, are
shown in Figure 4.6a and 4.6b. Pile Test #107’s load
increase at 2.75 mm produced plots that are less similar
and more unsteady than the plots produced by the load
increase on Pile Test #105 at 4.75 mm. Straight line
approximations are more easily made for Pile Test #105 in
Figure 4.6b and these approximations are similar to those
from full duration loads (with 6 mm C-type starting points)
shown in Figure 4.3e. The load increase at 4.75 mm
produced better and more consistent results.

To closely observe pile displacement after a load-
increase, data from Pile Test #105 has been plotted (Figure

4.7a) using various C-type starting points after the load

109

.usoaoonHaaHo oHHA Hquu hHo>HudHou I noun. ooaaouosH band an.
.ouaouosH vooH a hound vsa owwuon oHHn Hove! a no uoH>u£on usoaoouHmaHo.oaHH

 

   
     

.o.6 «human
2: . oH H9365 u on: H .33 93m: H.o
+4 4 s
use: 0:0.
u. uo>uso you
acoHuanxouqu
ocHH ustauu
onuouosH cued ouomom as H
can unoaooanuHa Haves
an ms.H us vouuuum o>u=o
.zx cN.~ u: hoHt unofi oHHm.
. H

usouoocHAnHa an mm.~ uouu<
zx ao.n ou mummuosH
fined 23 o~.~ HaHuHsH

      

ouuouusH cued ououon as H

can usoaooaHauHa Hauoa
as m~.m on couuuum o>uso
.23 on.“ an noHt aqua oHH

0 O

us\aa H.oI m @ zxnm.suom
sz um as eco.o I o

a.H I s o¢.o I A

auouoaaucm :oHuaavm nacho

 

 

 

OH

.0

(um) 9 ‘3uauaoetdsxq 3114

110

ooH

.ucoaoonHmaHo omuqH hHo>HuaHou,a wound tomaouusH ouoH .no.c oustm

Anusosc u .oass
OH H H.

 

 

1

   
        

usoaooanuHa as mm.¢ uouu<
28 oc.n ou oaaaouocH
wood 23 o~.« HuHuHcH

oncouocH cued
noum< ll H van
usoaoouHmnHa Hayes
a! nn.n an vouuuum
o>uso .zx mo.n

an tht uaoh OHH

oaaouosH fined Houu< as H can
unannouHaaHa Hooch a: n~.m an vouuuum o>uso
.zx oo.n an mcHt amok oHHm

6)....1963 a. «3. u
as mmoo.o u
o.s u : ma.o u

auouoaduum soHuasvm Queue

0
a

co
a

 

 

oH

(mm) 9 ‘nuamaoatdsxq GIIJ

111
increase. A pile at a constant displacement rate would be

parallel to the dashed line in Figure 4.7a. Nearly
constant displacement rates can be observed between the
asterisk marks of the curves started at 1.0 mm and 0.065 mm
after the load increase. Constant displacement rates can
be estimated for those portions between asterisk marks.
The displacement rate estimated from the 0.065 mm C-type
curve does not represent general pile behavior but can be
compared to other load increases plotted at 0.065mm C-type
starting points. The relatively-short total displacement
(0.65 mm) of the 0.065 mm C-type curve is an advantage that
will be used later.

Various B-type starting points (also shown in Figure
3.4) are used for plots of Pile Test #105 in Figure 4.7b.
The plot, started at 0.2 mm more than the displacement at
the time of load increase, is similar to the C-type plot in
Figure 4.7a started 1.0 mm after the load increase. The
displacement rates between asterisk marks of these two
curves are approximately the same. The 1.0 mm B-type
starting point distorts the log (time) - log (displacement)
curve to a point where it is no longer useful. The 0.065
mm C-type and the 0.065 mm B-type curves of Figure 4.7a and
4.7b, respectively, are nearly the same at this
displacement transducer sensitivity. In general, B-type
starting point plots would allow displacement rate

estimation after shorter total pile displacements.

.ausHoa msHuuaua oahu-o msHmD any .oaaouosH ouOH a uouuo
taco sonouu :H uaou oHHa onsHa a no uo>uao usoaooanaHo-oaHH .~.¢ ousmHm

Aauaonc u .Q-aa

 

112

      

oH H H.o Ho.o
. mo.HV
oaaauosH one; .
uouu< as oo.H omaouosH oaoH
oouuaum o>u:o. uouu< as moo.c
. couuaum o>uau
d
to
TL
a
6333.. anon. H3: 0
ll ms.¢v omaouosH onoH mm
:34 as 8.0 60385 . m .o I
o>usu a
m
9
u
.3
9
m
:aosm soauuom zx ac." sane
usoaooaHmaHo a! ms.¢ uouu<
zx ao.n cu zx o~.« scum
condouosw coed moHt amok oHHm.
m

 

 

 

 

113

CH

.uusHoa maHuuaun anhu-n msHmD .95.: ouamHm

Aauaonv u .oaHH

 

 

     
 

 

 

   

H H.o Ho.o
. .1 no.0
ouaouosH
coed mo
oaHH an uaoa oaaouosH cued ammouocH
-ooaHmuHa sash mo oaHH us uaoa vaoH mo oaHe
one: as H -oocHauHa cash no usoaouaHamHo
an vouuaum one: as «.o an sash one: as nmo.o
o>uao vouudum o>uao us couuaum o>u=o
.\
Av% mu
.:\\\\ .\\\\ onuouosH vaoH no
Kym”)? uaHom 25.: a co. c . m .o

         

U. “OUH.UW 0:30
0.0
.r» o)\ .\
O
>. a
so

%0
.\.\

.aaoSm :oHuuom
.zx ac.m xHao usoaooaHauHa aa ms.¢

Hound .zx oc.n ca 23 cw.~ aoum
vomuouoan owed ncHt uaoh oHHm

 

 

(mm) 9 ‘3uameoaIdsxa 311a

114
4.3.1 Step Load Increase Magnitudes

These comparisons involve model piles with a 0.76 mm
high tapered lug embedded in frozen low density (loose)
sand. This lug pushes the frozen sand outward instead of
forming a compacted sand wedge to displace the surrounding
frozen sand, as described in Section 4.2 for a square-edged
lug. The pile in Figure 4.8 moved very slowly for the
first 4 hours. This behavior was not unique to this lug and
was sometimes seen at the beginning of tests of pile with
square-edge lugs. Piles that showed this behavior had
small lugs (0.76 to 0.79 mm) and in at least one case (Pile
Test #25) was found to be slightly out of alignment. After
a few millimeters displacement, typical pile movement
resumed so that the step load at 9.73 mm can be considered
to be uneffected by this initial failure.

Three step loading sequences are shown in Figure 4.8.
A static load has a small cyclic load superimposed which is
followed by a static load increase. The displacement rates
of the static load before cyclic loading and the
displacement rates of the static load increases are shown
for each step load in Figures 4.9a through 4.9c. The creep
equation parameters determined from these step loads are
included in Table 4.2. The cyclic loads were relatively
small and applied for relatively short periods of time.
They were assumed to have little or no effect on the

succeeding load increase and will be considered in the

115

.vqu auadshv vowonaauoasa a msupsaosu amoua cho>om an voccoH

Acct yucky oaua dono- w:~-vouonau a no nou>¢£oa usoaooaaauuv-oauh

A335 u .25.

Haw mm— .u— my

.o.c ousmsm

 

 

sao¢.-.l

2H01 3 NH IEO'O + [Z'I‘

 

asemmMHWHV\i

\

I—

   
 

NX_£Z'I'
'1‘
N1 £1I°

2H01 3 NH 620'0 + 81

zxoo.~ u
too; chuwcH

_

T. EEmo.-
0

nu 77;:r1u3;1;
_+ A
0

0

z

7

x.

N as

a 8.:

I

0

n“

2

 

 

 

0 won m- an beam as om.o ou u¢.o
uuawa usoaooaaamua as on

 

 

 

o~

0m

cc

(mm) 9 ‘3uaueow1dsxq 9115

116

. .co¢ uaoH mo usoaooaanauo as nh.m u. ouaouosu coca ouuaum Any
mauvaou uouuu vs. ouomoa came Hobo! ma~.vouonau a mo usoaooaamauc-oaua .m.< ouswum

 

oo~ ed Aouaosv u .OI«H a ~.c
«sud use: A \\\
oSu uuooauousm
o>uau no .
.caonm uoz mu scuuaauxouno<

cowuuom popcoq zaaaoweacho
.ea m~.¢ an 2x mg.“

on comoouochlcoza .es um.¢ an
~55 2x «85 + co; 3 2x 8.“
scum commouocH coo; mafia

ecu; uguunuum
/

  
 

 

0%
0
0e
mv
.abw
.2
0 0).? . O 0
an? a) 2}... a an M o 535-...“
2.0 00 as: us. aa 36 I o
oeoo aoe t n.» n a 84 .. a
0.) 00¢? nuouoaauom souuaaum noouo

 

 

 

~.o

Ca

(mm) 9 ‘3uauao91dsgq atxa

117

cog

 

.oot amok mo usoaooaagaav as oo.- an coaches“ coca ouuaum .ba.v ou=Mum

Aauaoav u .oauh
OH fl

l+

.nzonm uoz cauuuom venue;
kHHaouaasha .aa om.- an 2: n~.~ ou
coaaouosu cosh .aa.n~.wa an a: OH
9 23 aNc.QH a~.H on 23 m~.~

. aouh voanouosm cued oaum

 

~no

 

)9
o5 are Exes ~6qu @ Edd-mm
.«a a. 2“: on a... 36 .. o
no oz » ca .. : cad .. a
O 0
00.0 000+ queue-anon souuaavm guano
é

 

 

d
(mm) 9 ‘auemaoatdsrq 311a

~.o

o~

118

.oo% anus mo usoaooaamauv as os.oa ua oaaouosu vaoa cuuaum .om.c ouawam

1.“:oac u ..aaa
ea ~

1.

 

~o.o

 

”=305m uoz scuuuom covaoa
zaaaouadaan .aa cm.- an 23 mw.H ou
comuouosm cosh .aa ow.e~ an n: ed
w as Hmo.ou -.~ cu 2x -.a

BOHh fiGfldOHOp—H UCOA Odwm

0 0

nexus ~.on « w zxmm.ouom
23“ an as «m.o u o

: m.e u : ca.c n a

ououoaauam souuasou nacho

 

 

 

 

~.c

ca

(mm) 9 ‘3uanaoatdsta 911a

119

Table 6.2 Creep Equation Parameters for Lugged Piles from Step Loads.

----+ ----- + ------- + ------ +----+ ------ + ------ + ----- + ------ [ ---------------
Pilelrest [ Sand [Solid [Pi1elznbed-[Dia- [In- [Load [Creep Equation
Teat|Ten- [ Size [VolunelLug [ lent |p1ace-[itia1[In- [ Parameters

 

No.[per- | Range [Frac- [SizelAhead/[nent [Load [creaae|----+----+-----
[aturel [ tion [ [Behindlat In-| | | b | n [ 2c
I I l ‘ I ll“: lcrunl l I I I
Ides cI III-III l 3 I II I u/ml In I “II 3 : I mm
22 [ -2 |. 07-. 86| 58.1 [0.79[ 76/57] 31. 52[1. 31 [ 13.6 | 1 | 7 7|1.00
----+ ----- + ------- + ------ +----+ ------ + ------ + ----- + ------ [----+----+----
23 [ -3 |. 07- .86[ 58.2 [0.79[ 76/62[ 8. 16|1. 38 | 32 2 [ 1 | 6 1|1.26
----+ ----- +~ ------ + ------ +----+ ------ + ------ + ----- + ------ [----+----+----
24 | -3 [.07-.84[ 59.0 [0.79[ 76/66l 2.97[0.885| 30.7 | 1 [ 6.7[1.08
[ | [ | | | 2S.72[1.16*[ 12.9 | 1 | 8.2[1.17
[ [ [ [ [ | 33.12|1.31*[ 18.9 | 1 [ 6.3|1. 16
----+ ----- + ------- + ------ +----+ ------ + ------ + ----- + ------ [----+----+----
25*a[ -3 |.07-.86| 58.3 [0.79[ 76/61[ 2.16|1.56 | 28.6 [ 1 | 9.3[1.59
----+ ----- + ------- + ------ +----+ ------ + ------ + ----- + ------ [----+----+----
26*1[ -3 [.07- 84| 58 0 [0 79[ 76/62| 26. 29|1. 56*[ 8 0 [ 1 [ 6.0|0.82
----+ ----- + ------- + ------ +--.-+ ------ + ------ + ----- + ------ [----+----+----
27 [ -3 [.07-.84| 57.6 [0.79[ 76/60[ 11. 54[1. 38*[ 17.2 | 1 [ 9 1|1.20
----+ ----- + ------- + ------ +----+ ------ + ------ + ----- + ------ |----+----+----
28 [ -3 |.07-.84[ 58.1 [0.79[ 76/67[ 2.38[1.11 [ 40.0 [ 1 | $.5[1.26
[ | | [ | | 11. 25|1. 56*[ 16.0 | 1 [ 6.5[1.27
----+ ----- + ------- + ------ +----+ ------ +- ----- + ----- + ------ [----+----+ ......
29 [ -3 |.07-.86| 58.0 [0.79[ 76/60[ 2.27[1.12 | 31.9 [ 1 | 5.3|1.22
[ | | [ [ [ 10. 86[1. 38 | 24.2 [ 1 [ 5.3[1.10
----+ ----- + ------- + ------ +----+ ------ + ------ + ----- + ------ |----+----+----
32 [ .3 |.07-.86[ 63.6 [0.79[ 76/62| 9.97[1.72*| 9.9 [ 1 | 3.9[1.45
[ | [ [ | [ 11.76[1.90*| 22.6 | 1 [ 8.3|1.69
[ | | | | [ 18. 74|2. 37*[ 17.1 | 1 [ 6.3[1.S3
----+ ----- + ------- + ------ +.---+ ------ + ------ + ----- + ------ [----+----+----
33 [ -3 [.07-.84[ 64.5 [0.79[ 76/62[ 2.38|1.63 | 6 8 | 1 [ 6.2|1.56
| .3 | | | | | 7.40[1.74 [ 30.9 [ 1 [10 [1.76
[-3.-2[ | [ | [ 9.51[2.28 [ 14 *n[ 1 [10*n|---
[ -2 | [ | | [ 32. 90|2. 28 | 15. 2 [ 1 | 9.2|1.49
----+ ----- + ------- + ------ +----+ ------ + ------ + ----- + ------ [----+----+----
34 | -3 [.07-.84| 63.8 [0.79[ 76/64[ 1.1S|1.29 [ 27. 6 | 1 [ 7.9[1.70
[-3,-2| [ | | | 3.57[1.65*| 25 *n[ 1 | 8*n[----
| -2 | | | [ | 16. 55[1. 65*| 26.2 | 1 | 8.7|1.38
----+ ----- + ------- + ------ +----+ ------ + ------ + ----- + ------ |----+----+----
35 [ -3 lice. 0. 90g/cc[0. 79| 76/76| 32. 85[0. S6 [ 32 0 [ 1 [ 4 6|0.22
----+ ----- + -------------- +----+ ------ + ------ + ----- + ------ [----+----+----
38*g[ -3 [ice, 0.89g/cc[0.79[ 76/76| 15.83|0.44*[ 40.0 [ 1 | 4 4|0.18
[ | [ [ | 28. 88[0. 62*[ 12.1 [ 1 [ 4 O|0.16
----+ ----- + -------------- +----+ ------ + ------ + ----- + ------ l----+--9-+----
39*gl -3 lice, 0.90g/cc[0.79[ 76/76| 16.88|0. 44*[ 26.3 | 1 | 3 9|0.20
[ [ | | [ 27.10[0.56*[ 43.2 [ 1 [ 4 9|0.24
----+ ----- + -------------- +----+ ------ + ------ + ----- + ------ |----+----+----
44 [ -3 lice. 0.903/cc|O.79[ 76/76| 20.15[0.56*| 31.2 [ 1 | 6 2[0.33

Table 6.2 (cont.)

3120

----+ ----- + -------------- +----+ ------ + ------ + ----- + ------
66 I .3 lice. 0.89g/cc[0.79[ 76/76| 28.96[0.56*[ 3.0
I I I l 41.24I0.50*I 5.2
----+ ----- + -------------- +----+ ------ + ------ + ----- + ------
68 [ -2 lice. 0.90g/ccl0.79[ 76/76[ 29.3S[0.56*| 7.1
----+ ----- + -------------- +----+ ------ + ------ + ----- + ------
49 I -3 [ice. 0.903/cho.79I 76/76| 7.57|0.56*| 32.0
----+ ----- + ------- + ------ +----+ ------ + ------ + ----- + ------
55 I -3 [.62-.59| s7 *v[0.79[ 76/66| 3.88[1.55 I 5.8
----+ ----- + ------- + ------ +----+ ------ + ------ + ----- + ------
60*t| -3 [.62-.59[ 57 *v[0.76[ 76/62[ 9.73[1.00 | 18.2
I I I I I 12.96|l.l8 I 7.5
I I I I I l9.76[l.27 I 28.3
----+ ----- + ------- + ------ +----+ ------ +----e-+ ----- + ------
78*g| -3 [.62-.59[ 57 *v[3.18[ 76/32| 2.51[1.56 | 96.3
----+ ----- + ------- + ------ +----+ ------ + ------ + ----- + ------
82*g[ -3 I.42-.59I s7 *v|1.59[ 76/63I 2.26[0.93 I 68.8
----+ ----- + ------- + ------ +----+ ------ + ------ + ----- + ------
105 I -3 I.42-.59I 57 *v|3.18[ 25/63I 4-75I2-20 I 40.0
----+ ----- + ------- + ------ +----+ ------ + ------ + ----- 6 ------
107 I -3 I.42-.59I s7 *v[3.18[ 25/6l[ 2.75I2.20 I 40.0
----+ ----- + ------- + ------ +--4-+ ------ + ------ + ----- + ------

a Pc determined at 6 -0.1 mm/hr
* Estimated at [static load + 0.5(dynamic load amplitude)[

*a Pile out of alignment

*f Test appears to be “fast“ and have a low P value

*g silicon grease applied to pile and lug surface to reduce bond

*n In Pile Test #33. if n - 10 is assumed, the -3 to -2 deg 0 increase
is equivalent to a 168 {-3 deg C constant temperature) increase

In Pile Test #36, if n - 8 is assumed, the -3 to -2 deg C increase

is equivalent to a 258 (.3 deg C constant temperature).increase

*t Leading edge of lug is tapered instead of square-edged

*v Estimated from the average of similar tests

121

short duration load section (4.4) of this chapter.

Log (time) - log (displacement) plots for load
increases are started at the time of increase and 1 mm
after the increase are shown in Figure 4.10. The
difference between curves for the 7.5% load increase is
notably smaller than the difference between curves for the
28.3% load increase. The 7.5% curve, started at the time
of load increase, was nearly straight. After the smaller
load increase, the pile quickly regained its

load/displacement equilibrium.

4.3.2 Repeated Step Load Increases

A model pile with three step loads is shown in Figure
4.11. Log (time) - log (displacement) plots for this load
increase (Figure 4.12b) were started 2.0 mm before the load
increase, 1.0 mm before the load increase, at the time of
load increase, and 1.0 mm after the load increase. The
plot started at the 24.2% load increase (at 10.86 mm), is
similar to the plot started 1.0 mm after this load increase
(at 11.86 mm).

In contrast to the relatively steady creep rate
observed around 10.86 mm, Figure 4.12a shows the 31.8% load
increase at 2.31 mm to cause a larger readjustment in pile
displacement rates. While the slow initial pile
displacement (similar to the initial displacement in Figure

4.8) may have worsened this situation, unsteady conditions

122

 

10

 
 
 

Time, t (hours)

 

 

 

0.1
0.01

(mm) 9 ‘3uemaostdsxq 9115

pile (Test #60)

Comparison of load increases on a model

using C-type starting points.

Figure 6.10.

123

.nmoun Heuo>om an nausea Amwa umohv
swam Hoouu Hopes e we mou>e£on usoaoooamaum-oaufi .~H.< ousmwm

Amazonv u .eauh

 

ow mm om me 0v mm om mm ow m" o. m o
dddl-luluqfiJIdJ-dldl-dd.dqllflql-dIdlI-du4dti1dd-idIIH‘I‘dd-‘I
55m.~\
I.
.0 20. 2; n
I. Umoa HmfiuwcH
X.
N“ l
5.3.2!
_
I
r.
C...
X.
N
SE
IeN.mN
_
z
“I
8
x. I
N

.0 m8 n- a 23m 5. 36585
uuaua usoaoooaneuo as as
230: m3 5. 2.8

 

 

 

OH

ON

Om

Os

(mm) 9 “auamaoatdsyq at};

124

.mu% unOH msqusv mood noun unuam Adv

.msavooa nouns use daemon dawn Hooua demos a mo mou>a£on usoaoomaamuo-oauh .NH.e unawam

Anusonv u .esuh

 

 

       
 

    

         

 

use eeeeuosn coed daemon aa c.~ um 23 ~N.H
scum pesuauouoa euouoaouem sequesvm noose

OOH o~ I H ~.o
- . \ To
ss~.o no moo: g be mo>u=o sou
:ofiumswxoudd< was 2393mm...
5 x
\\m).
a»
.0
0.89
.~
2%...
0
65
.(0
so? aw
to 01 9 u o
O .
a are... a) up}... do: a a 32.7%
as 3.7 20: 0a .a god .. o
4). ea «6 I a o; .. a
‘90 e06 . .. u .......I. a ower - ,.
see ease 09 eases 5.. too.— uoumd a... or" us zu— he;

 

 

a~

(mm) 9 ‘3uemeostdsrq 3118

125

.aua amok mauusv coca noun ccoomm .n-.q omswum

Anusosv u .oauh

          

 

oo~ ed
0 0
as}... do: a a 33.7%
a: 2. a... Sod .. o
in I c o; .. n

.
oueouosu omQu sound as o.H um 23 wo.a

vac endomosn wood shaman El O.H us 2: n:.~.

scum vosuaueuoa euouoamuom scuueaum dunno

 

o“

 

 

(mm) 9 ‘auemaostdsrq aIIJ

126

are generally found at pile displacements of less than 3

mm .

4.4.3. Rapid Adjustment to a Nearly Steady State

The time-displacement behavior for a model pile frozen
in dense sand is illustrated in Figure 4.13. Step loads
are shown at 2.38 mm and 7.40 mm of displacement and a
temperature increase was started at 9.51 mm of
displacement. Note the relatively sharp displacement rate
change at 7.40 mm. The displacement rate change at 2.38 mm
is equally sharp but difficult to see in Figure 4.13. The
log (time) - log (displacement) plots in Figure 4.14a and
4.14b are relatively straight and parallel to each other
even though the increased—load curves were started at the
time of load increase. The ratio of displacement rates
shown in Figure 4.13 is almost the same as the ratio
determined from straight line approximations in Figure
4.14b.

The data plot shown in Figure 4.14c started 1.0 mm
before the coolant bath temperature was increased and the
second plot was started after the pile displacement had
stabilized in the higher temperature frozen soil.
Temperature stabilization required more than one hour and
4.43 mm of displacement. The curve at -2 deg C was started
4.43 mm after the temperature bath surrounding the pile and

the sample had been increased. Despite the large

.omoa euasshv vomoasuuoasa o maucsaosa menus as boomed use uses onset

om

a su sosoum Anna ueoav oaum doves a mo nou>oson usosoooammwp-os«a

mm om

Assad—3 u . clan.

.mm

.ma.e «Hesse

 

we ow mm om om ma OH m o
1‘1..‘fl‘l‘dqddd-dqdd.ded-IdddI-dd‘d.1‘Id-quJ-q‘dd-Ilddd-14“
\u:\4s ssmm.fll
\ .wm u
\ TL
.seoq.sv\ _ _ _ _ W.
IIIIIIII .I. T. T. T.
H; as — e e e e x6
e 0.8 I w. w. w. w. M I
.u _. _. _. gamma.“
nu + .U n. no u .u H .u. H
l O O O
45: .299 o ,o. w m
z e E I. E
rmmuizl. ”w x. x. ya
8 00 I. N N N L
m0e e
Au. nunu X. as as so
. nu7u1. “N
Cats .I. T. T.
Tic. as no nu AU
u 3.x. "a an "n
OXINN .1 z z 2
IN 0
n 000 m amassed
S 1...... m.c~ u u 5:. .
”WWW 9: mm a L
:sz
. z
c.
”a m. 48.8 I caduceus masso> essom
mu . .o nae n- a scam as ee.o-e~o.c
7o uusud usoaoooadmua as on
m. m 233. m3§26

 

   

  
 

   

 

 

 

OM

OM

Om

cc

(mm) 9 ‘3uemeoa1dsrq 3113

128

.nm% week we msassumon segue as an.~ ueoa meow Aev .oeeouosu ueoH e soume

 

 
     
 

 

 

use enemas uses oesou sq Anni ueohv odds douoa e no so«>e:on usoaooeaomuu-osaa .e~.e «mamas
as: .0!
Cu A osv nu «a _.o
‘4 , ~.o
osug use: A
as» muooauousu
o>uso mo
souueaaxoudd< an. I a a
05..— usmueuum D H M \ M
.I m.o

 

 

0

u;\aa H.cIom @ zaen.HIoa
zas us as «coo.c I o
«.e I : c.~ I a

euoueaeuem saucesvm moose

b

 

 

(mm) 9 ‘3uemeoatdsxq 911a

129

2:

.mn% duos no masseuse; "use. as oe.h an coves cues swam

a:

1.05053 u ...19

 

.AQH.¢ shaman

 

 

J

 

O U

u:\:: H.cI M @ zxea.sIoa
zss u. as «mooc.o I o
~.os I a o.s I a

eueuoaeuem scuueaam noose

 

 

A:

(mm) 9 ‘3uameoatdsjq alga

130

.mn% amok no msussuwoa on» noume as Hm.m me

o mou «I 0» o wou n I scum uoeeouosu ensueuoasoh .oc~.e ousmum

oo~ Om Amusosv u .eauh a

 

 

u mac m- an

«

 

o woe N. u.

 

 

«

 

 

~o’

ca

(mm) 9 ‘auamaoatdsxq 3115

131

difference between starting points of the plots (5.43 mm),
a good comparison may still be obtained if the pile's
behavior is relatively steady within this range of
displacement.

The behavior of a lugged pile frozen in
polycrystalline ice is shown in Figure 4.15. The
relatively large initial load (1.53 kN) was removed after
approximately 14 mm of displacement to prevent further
rapid failure. The initial rupture of ice adhesion is
shown in Figure 4.15’s insert. After the ice adhesion
rupture and removal of the large initial load, the pile
continued to displace with a smaller static load. Dynamic
loads were later superimposed on this small static load. A
relatively sharp change in displacement rate can be seen
after a static load increase at 32.85 mm. The ratio of the
straight line estimates of displacement rates before and
after 32.85 mm is shown on Figure 4.15.

The log (time) - log (displacement) plots are not
shown for this type of sample because the effect of the
initial rapid displacement on the pile’s general behavior
is not known., The displacement rate ratio, determined from
the relatively straight lines of the time-displacement
plot, can be used to compare the rate changes due to
various load increases.

Parameters determined from various step loads are

listed in Table 4.2. The creep equation parameters show

132

.oeeouusu ueoH cuueue use use mueou oaseshu uoeodsuuoqse ~euo>oe

so“: ssose Anna uses» saws uomwsH e we sou>eson usosooeadeqqusuH

Anusesv u .eaua
m N

.na.e assess

 

 

o meu n. so can esudaeuehuohaom
asses acoaoouHaaea a: on
used»: was as a~.o

mu 0
u .u
S.
.m _m 3 . ”Hum
m m 05.8 8
Z T. 0
m. mm“;
a a C...
m m m 05...... «N
\.
m_ J. _

 

 

ZHOIQ N3 8IO'O+ 9S°O

ZHDIaNX £20°0+ 95'0

zH018 NH EIO'O+ 9S'0

 

 

d

  

ZHDIB Ni 600'0+ 9S'0

   

170

 

 

I

1

p901 IBIIIUI‘
l

 

 

ZHOIQ N3 fiOO'O? 95'

    
     
     

       

 

N>I £51
9901 IBIJIUI

A

ll

 

 

o~

ON

on

Co

(mm) 9 ‘nuamaostdsrq 311g

133

considerable variability even when derived from a single
sample. Sample material variability and slight temperature
changes during a step load may have contributed to
variations of creep equation parameters. The short
duration loads shown next will shorten the periods of test
data collection to reduce displacement rate variablility

and allow more comparisons during one pile test.

4.4 Short Duration Loads on Model Lugged Piles

Data from a model pile test with a slowly increasing
displacement rate is shown in Figure 4.16. A group of
similar piles with very dissimilar displacement behavior is
shown in Figure A.13 (Appendix A). Creep equation
parameters determined from these plots for full duration
loads (or from load increases on them) would be
questionable. However, if a very short section from any
one of these curves is considered, a nearly constant
displacement rate would be observed. A short duration load
takes advantage of this short term consistency.

The short duration loading was convenient for
applications of cyclic loads to model piles. Superimposed
cyclic loading could be carefully monitored during the
entire short duration period and a constant loading
condition could be maintained. Performing many short
duration loads on one sample allowed a range of frequencies

and amplitudes to be tested. Short duration static load

134

.ueoH useuesoo e saws ecu osuaaeumhuOAHoa su

Aams ueokv eaum uommsa e no Heu>e£ea usosoueaaequoaah .o~.q ousmam

assuage u .oase

 

 

e m ~ fl 0
soxoun we: uson e>ueosu< sound
23 mm.o cu uomeouoon
2a s~.~ no 6664 adds guesses no
«t
x.
N

o mou m- we can osuaaeumhuozaom
salad usoaooeaaeua as on
news»: was as aa.o

 

 

Ca

om

Om

cc

(mm) 9 ‘3uemeostdsxq 911g

135

increases were also inclUded to allow comparisons with
dynamic short duration load increases.

The time-displacement plot of Pile Test #92 is shown
in Figure 4.17. The three periods for lever arm adjustment
are shown. The log (time) - log (displacement) behavior
after reloading is shown in Figure 4.18. The typical
decreasing displacement rate was disturbed by the arm
adjustment process so that these three displacement rates,
determined after different total displacements, are nearly
the same. After each arm readjustment, the pile was
allowed to reach a nearly constant rate before short
duration loads were again added.

A series of short duration loads (static load
increases and superimposed cyclic loads) are shown in
Figure 4.19. After each load increase, the initial pile
load was reapplied and maintained until a nearly constant
displacement rate was re-established. At this point,
another short duration load was added. Slopes of the
initial displacement rates, illustrated in Figure 4.20,
vary slightly between short duration loads. Slight
variations were normalized when displacement rate ratios
were used for comparisons.

With each short duration load increase, the time
displacement curves often show small (0.02 mm for Test VIII
in Figure 4.20) periods of rapid displacement (believed due

to material and/or loading system characteristics). These

136

.msuueouss no euouuem saw: u mou n. us

use. sououu su edun Aeuos a new Je>hso usosooeadequoauh .na.e ousmum

$5 a is...
on as an on as on cu as

 

 

‘ 4 - ‘ ‘ 1 4

unmsoz was as am.a

Ann. mane

 

 

 

 

 

 

as m.-:
Hzaflfi92§~xfi< s
mflRH—nZHER—SOEBEBQ

ES m.MNI.
.«..eum
\\ use.

 

 

¢~.o .UHh 2H m‘ «ZOHBHOZOU Emma mum s

_ _ _ _ _ _

A 1 1 A 1 _L l l

l A

A

l J L A 1 l

 

_F

 

_ a _ 2 en" _ I _ 2 «as _ a _ . 2 «as

 

Cu

ma

ON

(IN) 3 ‘mmanasm 311.!

mm

on

AZSA

137

.ueoa dawn ouueue as» no souueuoueou Houue as H saosu

Amos ueoav suds eamsue e no sou>esea usoaooeaaequoaga .ma.e ouswum

cod o~

d‘

Anusosv u .oaua ~ ~.

 

 

 

 

 

cg

I—IO

(mm) 9 ‘3uemaostdsta GIIJ

138

.lec ueou. no ax swsousu ~HH> someouosu use” sequeusu 323 you musesoqueaou
can msuaose uses soeouu s« 0.3.— Houos .— you 25:0 usosooeadequosE.

on

an

05 u .85.
a...

am

On

 

 

4

d

 

zu— c~.~

 

QV

.a~.e «Hausa

¢~

C
N

F"
N
(W) S ‘mmona 310 am

NN.

139

.NG% Ofifism HO HN rum—JOH—PH H~H> NOQQGHOQd ‘QOH .fiflfla CONOHH and

sung Heuos e you ee>uso usoseoeadequesuu souueusu uuosm

 

 

 

AHINN

3
s

‘0
N

3.
s 8:
(II) 3 ‘mmuamsm ma

 

 

 

 

 

 

”H Hunk

€60.85.
.4 .4 s. a.
s m w. _w
K — q u q
and... x I
as ou.
\\ .38." I
33
OOI¢VN \\
\
ne.~ I .« x «a
L
unease
€50.85.
C. C. C. C. C.
v. v. v. w. w.
m ,7... M n m.
4 - _ - _ I _ - 4
ea .o.w II
N «.Ou +
seas \ .39...“ I
\\ n¢aa
.22 \\
\
G.~I.«\«« I
L

 

 

00.0“

noION

65.0«

“5.0"

(u) g ‘monasm rm

 

 

 

 

 

 

 

    

.o~.c assess
one u .85.
w. a w... a a
m «u M w. M
KW,- I _ I .l. I . . a
a: 6.2 .e. I
u «.3 +
2.. SJ
2.. 2.... I
\ .23
.22” \ \
\ I1
34 I .m \ «M
l
x .53
2.5 u is...
. m m M m m
flu]. _ . _ . _
a: 64% I
u «.2 +
as o«.«. as c~.~

 

 

    

O<Od
en.H I am \ ««

HHH> Emma

IL

 

 

n¢.-

Q¢.-

INIHIDVIJSIG 311d

nn.-

an.n~

(m) 3

uh.au

-.o~

.LNENSDV’IJ 510 3'1 Id

~o.a~I

no.a~

(W) 3

140

are followed by slowly decreasing displacement rates, (seen
as fairly steady in Figure 4.20) which could often be
approximated by straight lines. The displacement rates
before short duration loading can be estimated in this way.

Note that the displacement rate during short duration
loading can be considered as relatively steady over a small
range of displacement. This was shown earlier for a
portion of the 0.065 mm C-type curve in Figure 4.7a between
asterisks. The displacement rate during a short duration
loading does not represent general pile behavior. However,
the relatively fast displacement rates from short duration
loads can be compared with each other.

For better comparisons, the displacement rate
estimated during short duration loading can be divided by
the displacement rate estimated before the load increase to
provide a displacement rate ratio for each test. These
ratios are shown in each part of Figure 4.20. Displacement
rate ratios tend to normalize any differences preceeding
the short duration load increases and clearly provide the
magnitude of change (displacement rate increase) produced
by adding a given load variable. The effect of a dynamic
load increase can be compared with the effect of a static
load increase

Data from tests where cyclic loads were superimposed
on lugged piles with approximately the same lug heights and

tested at the same temperature are shown in Figures 4.21a

141

.o.mou n- as used omooa s« Acme unoav dawn mam uouoneu mo o>uso Aev .mueoa

 

 

oaseshu ueeonsauomse ou ueuooeAse mosun Houos mo soa>eson usosooedaeaqusuH .H~.e ousmum
~m Om Aeusosv u .elua mN mm
I I. c~
\d
T..
T.
\ a
\ m
e
\ p
:
I l
r.
I.
X.
"N
a Q."
.u-.c .mam sw saonm aw _
conflueosoo museum 4. .I.
.esoHuHusou >ueoums= _ r.
383:: S as \ I
one: uses—00333 u a. \ _+
\\ nu
\. . I
.l- \ w
0, \.\x TI eaueuon mad ueuonea esoSm m.e .mam
c.
\\ x. o mou n. no usem as an.o-«<.c
«Ne N awn—«A usosooedaga as on
. W
0
H
LP

       
 

 

 

ON

(mm) 9 ‘3uameostds1q 31:5

.o mou n- ue Aunt assay used oesou s« was as o~.c mo o>uso .bau.e ousmah

Aussosv u .oaaa mv De

 

142

 

4 '1 J 4 11 d J J

 

8901 311d

T. T.
1. .2
.V .7
_+ x.
0 N
0 lCeN
7y
I.
x
N 1
as
I
n. a.

L
m 6.»

ew.c I sequoeum oaaao> uuuom A
6 non n- on econ as ee.o-eso.c
swag; usoaooeadmaa as us

Z
Z
9
X.
N.

 

 

 

m.~

(mm) 9 ‘3uamaostdsxq 91;;

143

RN

.0 mou n- we Anna ueoav sou osuaseuehuohaod s« and as ah.c mo o>uso .oa~.e omsmum

 

 

3 385 u .25. 2 1......
1 . I I . mm
m.
Th4
a
q
\. O
.\I a
\\ D.
00..» .\ .. :m
a» .\\
90 .\ 0
0%. \\ .
Awncv \\.I\ M”
‘0. \.\
9319 \\.. MM
8 .\ a
09 \\ 0
%% \.\. .
e \\. M”
as ma.NnI \\ _+
.
0 0 1mm
1. .0
t. 76
I. 12
N N do} cad I huumsen cum I
as o mou m- as con osuaaeuehuohaom
1. salad usoaeoeaomua as ms
nu
n“
z

 

 

mm

(mm) 9 ‘nuemaostdsxq 3115

144

.o meu n- no Aunt uuohv uses oeooH su mad as ah.c mo o>usu .u-.e ousmdm

 

    

 

 

2... 5.55 u .25 2... 3...
I mm.o
.J oo.c
u:\ss e¢.o
o~.~ u IIIIIIIIII
ucxsa m~.~
”I I 8.6
S
9
+
0
0
8
o I 2.6
x.
I. N
v. a o as. m- 3 6:3 as 3.6.8...
9 M 3a...— usoaooeamma as on
x. H
N z

 

(mm) 9 ‘auemaoatdstq 311g

145

Table 0.3 010151th late Increases (rm Short Duration Load Increases on Lugged Piles.
Pilelfeet I Send [So-lid [Pilelued-[laitial I Percent IDiaplaee-Iniaplace-I Percent [Creep
Teatqu-I Size IVelI-elLu. I eat I Lead I Lead In- I not at [meat letelDisplece-I Par-
lo.lpere-l Inge Il’rec- ISiselAheed/I letore I crease [the Time I letore [lent Ratel er

 

 

 

 

 

 

 

 

 

 

Iture I [tin I Ilehindllnoreeeel 0d Pre- I of head I Loed Ilncreaee I eter
I I l l “-00 l I qua-07 Ibex-000 liner-«0| l :0
IdegCIr-I! I-I-I-Ih IIIISI I [ml 2 I
___‘ 5 5 5 5 [ 5 [ c [—
55 [ -5 |.07-.00[ 00.5[0.70[ 70102[ 1.27 [31.5110 [ 7.27 I 0.000 I 7 [ -
5 5 54 5 4 | 5 [ 5 5 —-[
55 [ -5 [100. 0.00.1“ [0.70[ 70170[ 0.50 [30.0110 I 51.00 [ 0.7 | 7 I -
-- 5 5 5 5 5 I 5 | 5 5 — [ -----
00014-5 [.02-.50[ 57 ~[0.70[ 70102[ 1.70 [32.0110 I 10.00 I 1.57 I 7 | -
5 5 5 5 5 | 5 I 5 ¢ I -----
70-.[ -5 [.02-.50[ 57 ~[5.10I 70152[ 1.50 [3 5.11.2~[ 2.05 | 0.002 [ 05 [7.2
5 5 5 5 5 [ 5 [ 5 5— [
010d -5 [.02-.50[ 57.0 [0.70[ 70102[ 0.05 [3 0.71.1~[ 5.05 I 0.100 I ~25 [2.7
5 5 5 5r 5 | 5- [ 5 5 5|
025d -5 [.02-.50[ 57 ~[1.50[ 70105[ 1.50 [35.1110 I 5.70 I 0.112 [ 5 [1.0
I I [ I I I 1.50 [31.7110 I 7.00 I 0.102 I 5 [1.0
[ I [ [ I 1.50 [+3.6etaticl 0.25 I 0.005 I 10 [5.2
I I l l l I 1.50 [+5.0000010I 0.55 I 0.100 I 10 [0.0
I I I I I I 1.50 [00.000.00.14 0.07 [ 0.000 I 50 [0.0
5 5 5 5 5 [ 5 [ 5 5 --|
07 I -5 [.02-.50[ 50.0 [1.50I 70100[ 1.55 I: 5.21 1 I 2.00 [ 0.050 [ 52 [0.5
l I I l I 1.55 I: 5.211 [ 5.21 I 0.050 [ 50 [0.0
I l l I I I 1.55 I: 5.211 I 5.00 [ 0.027 [ 50 [0.0
5 5 5 5 5 [ 5 [ 5 5 I -'
00 l -5 [.02-.50[ 50.0 [0.70[ 70101[ 1.50 [3 5.11 1 [ 0.01 [ 0.00 I 20 [ 5.7
I I I I I | 1.50 [35.111 [ 7.05 [ 1.10 [ 25 [0.5
l I I I I [ 1.50 [35.111 [ 11.20 [ 0.70 [ 20 [0.5
I I I I I I 1.50 [31.711 I 12.70 [ 1.00 [ 0 [5.5
5 5 5 5 5 [ 5 [ 5 5 In"-
00 [ -5 [.02- 50[ 50.7 [1.50[ 70100[ 5.00 [3 2.01 1 [ 0.50 [ 5.0 I 11 [ 0.1
I I [ I I [ 5.00 [35.211 [ 11.12 I 5.5 [ 20 [0.0
[ I l I [ [ 5.00 [35.210.1I 10.02 | 0.0 [ 0 [1.2
5 5 5r 5 5 [ 5 | 5 -5- |
02 I -5 [.02- 50[ 57.0 [1.50[ 70100[ 2.20 [3 2.51 1 [ 0.70 [ 0.00 | 10 [ 5.0
l I l I I I 2.20 [32.510.1I 7.71 I 0.00 I 5 [2.0
l l I I I I 2.20 [32.5110 [ 0.57 [ 0.75 [ 0 [5.1
l I l l I I 2.20 [+1.0000010I 0.00 I 0.07 l 10 [0.2
l I I I I [ 2.20 [3 5.111 I 15.55 [ 0.07 [ 20 [0.0
I [ l I I [ 2.20 [35.110.1l 10.57 [ 0.00 I 10 [5.5
l I I [ [ [ 2.20 [35.1110 I 15.50 [ 0.00 I 20 [5.7
l I l I I [ 2.20 [310.111 [ 10.70 [ 0.00 [ 50 [0.5
I I I I [ [ 2.20 [310.110.1[ 20.00 I 0.57 [ 51 [0.5
l l I I l I 2.20 [310.1110 [ 21.07 [ 0.00 [ 00 [0.1
I I I I I I 2.20 [+6.50tatic[ 22.15 [ 0.50 [ 05 [0.0
I I I [ I [ 2.20 [052000014 20.02 [ 0.00 | 20 [0.0
l I l I I I 2.20 [3 0.01 0.5[ 27.00 [ 0.50 [ 12 [2.0
l l I l l I 2.20 [3 0.710.5[ 20.50 [ 0.02 [ 10 [5.0
l I I I I I 2.20 [310.110.5I 20.10 [ 0.05 [ 51 [0.5
I l I l l I 2.20 [35.115 [ 50.01 [ 0.07 [ 10 [5.5

Table 0.3 (cont.)

146

 

 

5 5 5 5 5 I 5 I 5 5 I

05 I -5 I.02-.50I 57.0 [1.50I 70102I 2.20 [310.1110 I 0.00 I 0.00 I 02 I 5.0
I I l I I I 2.20 [310.11 1 I 7.52 I 0.00 I 07 I 5.5
I I I I I I 2.20 [310 11 5 I 0.10 I 0.07 I 05 I 5.2
l I I I I I 2.20 [310.11 2 I 0.71 I 0.75 I 07 I 0.0
I I I I l I 2.20 [310.11 0.0I 12.00 | 0.52 I 55 I 0.0
I I I I I I 2.20 [310.11 0.5[ 15.05 I 0.52 I 51 I 0.5
I l I I l I 2.20 [310.11 5 I 15.00 I 0.07 I 51 I 0.5
I l I I I I 2.20 [310.11 1 I 15.00 I 0.01 I 75 I 5.7
I I l I I I 2.20 [310.11 1.5I 10.20 I 0.57 | 55 [ 0.0
I I I I I I 2.20 [3 5.11 0.5I 10.50 I 0.50 I 25 I 0.5
I I I I I I 2.20 [3 5.11 5 I 10.00 I 0.05 I 20 I 0.5
l I l I I I 2.20 [3 5.11 1 I 17.55 I 0.02 I 52 I 5.0
I I I I I I 2.20 [3 2.51 0.5I 17.00 I 0.01 I 10 I 5.0
l I I I I I 2.20 [3 2.51 5 | 10.10 I 0.00 I 10 I 5.5
I I I I I I 2.20 [3 2.51 1 I 10.51 I 0.05 I 20 I 7.0
I I I I I I 2.20 [310.11 0.5I 21.55 I 0.00 I 50 I 0.0
I l I I I I 2.20 [310.11 5 I 21.72 I 0.00 I 50 I 0.0
I I I I I I 2.20 [310.11 1 I 22.00 I 0.00 I 75 I 5.7
I I I I I I 2.20 [310.11 1.5[ 22.55 I 0.50 I 07 I 5.5
I I I I I I 2.20 [310.11 0.2I 25.00 I 0.50 I . 00 I 0.1
I I I I I I 2.20 [310.11 5 I 25.55 I 0.57 I 05 I 5.1
I I I I I I 2.20 [310.1110 I 25.00 I 0.51 I 00 | 5.0
I l I I I I 2.20 [310.11 1 I 25.00 I 0.00 I 05 I 5.2
I l I I I I 2.20 [310.11 0.5I 25.50 I 0.05 I 00 I 5.1
I I I I I I 2.20 [310.11 1 | 25.00 I 0.05 | 01 I 0.7
I I I I I I 2.20 [310.11 5 I 20.17 I 0.57 I 00 I 5.1-
I I I I I I 2.20 [310.11 0 I 20.02 I 0.55 I 70 I 5.0
I I I I I I 2.20 [310.1110 I 20.00 I 0.55 I 71 l 5.0
5 5 5 5— 55 I 4 I ¢ ¢ I

00 I -5 I.02-.50I 57.0 [1.50[ 70100I 1.05 [3 2.51 0.5I 5.01 I 0.22 I 10 I 5.7
I I I I I I 1.05 [+1.0000010I 0.05 I 0.20 I 10 [12.0
I I l I I I 1.05 [3 5.01 0.5I 0.00 I 0.100 I 20 I 0.7
I I I I I I 1.05 [+2.7000010I 0.05 I 0.21 I 25 I 7.0
I l I I I I 1.05 [310.01 0.5I 0.00 I 0.171 I 05 I 5.2
I I I I I I 1.00 [310.01 5 I 7.10 I 0.105 I 00 I 5.0
I I I I I I 1.05 [310.01 0.1[ 7.07 I 0.20 I 07 | 0.0
I I I I [ I 1.05 [02.700.01cl 7.72 I 0.20 I 51 I 10.0
I I I I I I 2.00 [+2.7000015I 7.77 I 0.27 I 15 I 0.5
I I I l I I 1.05 [310.01 0.1[ 10.20 I 0.100 I 01 I 5.0
l I I l I I 1.05 [310.01 5 I 10.50 I 0.107 I 52 I 0.0
I I I I I I 1.05 [50.0500010I 10.02 I 0.155 [ 55 I 0.5
I I I I I I 1.05 [+2.5000010I 11.05 I 0.101 I 20 [11.0
I I I I I I 1.05 [3 5.01 0.1[ 11.20 I 0.100 I 50 I 0.7
I I I I I I 1.05 [3 2.51 0.1[ 11.00 I 0.100 I 0 I 2.0
I I I I I I 1.05 [+1.1000010I 11.05 I 0.100 I 11 I 0.0
I I I I I I 1.05 [015.05.010I 10.55 I 0.172 I 102 I 0.5
I I I I I [ 2.20 [3 2.21 0.1[ 15.50 I 0.50 I 0 I 0.1
l l I I I I 2.20 [3 2.51 5 I 15.02 I 0.50 I 10 I 5.0
l I I I I I 2.20 [3 2.51 0.5I 10.55 I 0.07 I 11 I 0.1
l I l I I I 2.20 [01.2500010I 10.00 I 0.57 I 15 [10.1
I I I I I I 2.20 [3 5.01 0.5I 17.02 I 0.01 I 20 I 0.0
5 5 5 5 5 I 5 5 I

 

A
Y

1'3bl3 6.3 (33313.)

147

 

A
‘T

A
V’

 

¢ % ¢ 3 | 3 ¢ I
I I I I I I 2.23 I32.3333213I 17.32 I 3.33 I 13 I 7.3
I l I I I I 2.23 I;13.1/ 3.3I 17.73 I 3.33 I 33 I 3.1
I I l I I I 1.33 I+13.33331cI 23.32 I 3.171 I 173 I 3.3
I I I I I I 2.23 I313.1/ 3.3I 21.33 I 3.32 I 33 I 3.2
I I I I I I 2.23 I;13.1I 3 I 21.33 I 3.31 I 33 I 3.3
I I I I I I 2.23 I313.1/13 I 21.33 I 3.33 I 33 I 3.3
I I I I I I 2.23 I33.3333313I 21.32 I 3.33 I 33 I13 3
I I I I I I 2.23 I313.1I 3.1I 22.17 I 3.33 I 73 I 3.3
I I I I I I 2.23' I+3.3333313I 22.33 I 3.33 I 23 I 3.7
I I I I I I 2.23 I+3.3333313I 23.23 I 3.33 I 23 I 3.3
I I I I I I 2.23 I:13.1/13 I 23.37 I 3.33 I 33 I 3.1
I I I I I I 2.23 I: 3.3/13 I 23.33 I 3.33 I 27 I 3.3
I I I I I I 2.23 I; 3.3/13 I 23.13 I 3.33 I 32 I 3.3
I I I I I I 2.23 I+2.3333313I 23.33 I 3.33 I 13 I 3.3
I I I I I I 1.33 I+23.333313I 27.33 I 3.133 I 333 I 7.3
l I I I I I 2.33 I+3.3333313I 23.23 I 1.13 I 33 I 7.3
I I I I I I 2.33 I; 3.3/13 I 23.33 I 1.33 I 33 I 3.3
I I I I I I 2.33 I313.1I 3.1I 23.33 I 1.31 I 33 I 3.7
I I I I I I 2.33 I: 3.37 3.3I 23.31 I 3.33 I 13 I 3.3
I I I I I I 2.33 I313.1/ 3.3I 23.72 I 1.33 I 33 I 3.3
I I I I I I 2.33 I313.1/13 I 33.13 I 1.13 I 37 I 3.3
I I I I I I 2.33 I+1.3333313I 33.33 I 1.13 I 13 I 7.7
I I I I I I 2.33 I; 3.1/ 3.1I 33.33 I 1.33 I 12 I 2.3
I I I I I I 2.33 I3_3.17 3.3I 31.23 I 1.33 I 13 I 2.7
I I I I I I 2.33 I; 3.1113 I 31.73 I 1.33 I 22 I 3.1
I I I I I I 2.33 I33.3393313I 32.13 I 1.33 I 3 I 3.3
I I I I I I 2.33 I31.3313313I 32.33 I 1.33 | 23 I 3.3
I I I I I I 2.33 I+3.3333313I 32.77 I 3.33 I 32 I11.3
I I I I I I 1.33 I313.3/13 I 37.33 I 3.33 I 33 I 3.3
I I I I I I 1.33 I;13.3/13 I 37.33 I 3.27 I 33 I 3.2
I I I I I I 1.33 I33.3333:13I 33.37 I 3.33 I 32 I 3.3
I I I I I I 1.33 I313.3/13 I 33.33 I 3.27 I 33 I 3.1
I I I I I I 1.33 I+3.3333313I 33.33 I 3.23 I 33 I 3.3
I I I I I I 2.33 I33.1333313I 33.33 I 3.33 I 33 I 3.7
I I I I I I 2.23 .|+3.3333313I 33.22 I 3.71 I 23 I 3.3
I I I I I I 2.23 I;13.1/13 I 33.32 I 3.73 I 31 I 3.3
I I I I I I 2.23 I:13.1/13 I 33.33 I 3.73 I 33 I 3.1
I I I I I I 2.23 I33.331.e13I 33.21 I 3.72 I 23 I 3.3
I l I I I I 2.23 I37.3333313I 33.23 I 3.31 I 32 I 3.3
I I I I I I 2.33 I33.3333313I 33.33 I 1.37 I 33 I 3.3
I I I I I I 2.33 I:13.1/13 I 31.33 | 1.33 I 33 I 3.3
I I I I I I 2.33 I;13.1/13 I 32.31 I 1.23 I 33 I 3.3
I | I I I I 2.33 I313.1/13 I 32.33 I 1.23 I 33 I 3.1
I I I I I I 2.33 I33.3333113I 33.33 I 1.31 I 23 I 3.3
3 ¢ ¢ 3 3 | ¢ | 3 ¢ | -
? 01353133an 233.3 1113233333 3331's In 333331234 33332233317 but. 333. 33 b3 3331.].

1'12 .3

81113331 323333 31791133! 33 p113 3nd 111; 31323333 33 23331333 bond

1.334111; 33123 at In; 13 3393233! 11333331! 3! 3qu32323d¢3d

233133333 {23- 3213 33732333 32 31.1132 3333.3

83111323 37m £32- 12333

148
through 4.21d. The major difference involves the material

in which each pile was frozen. The larger displacement
rate increases are obtained for loose frozen sand as shown
in Figures 4.21a and 4.21d. The dense sand, Figure 4.21b,
and the polycrystalline ice, Figure 4.21c, show little or
no displacement rate increases. These and other results

from short duration loads are summarized in Table 4.3.

4.5. Multiple Short Duration Loads on Lugged Model Piles

The short duration load technique can be repeated on
the same frozen sand sample in order to provide much more
data. The time-displacement behavior for this type of
loading sequence is shown in Figure A.3 (Appendix A) on
Pile Test #123. After an extended period of displacement
at the initial pile load of 3.10 kN, the load was increased
0.12 kN (5.2%) for a 90 minute period. Two additional 0.16
kN pile load increases were added during subsequent 90
minute intervals. The last 0.16 kN increase (3.74 kN total
pile load) remained for more 90 minutes. An alternative
loading sequence might have restored the load to 3.10 kN
after the last addition. The log(time)-log(displacement)
curves from these load increases, using 0.1 mm B-type
starting points, are shown in Figure 4.22.

Curves B, C, D, and E in Figure 4.22 show a relatively
faster displacement during periods of (primary behavior)

compared to the nearly constant, late primary-secondary

149

.:o«u3u:o Hauou uuona 3 acquav upaoa homuua hH3>wmmooosm an

 

 

 

 

 

 

     

 

 

 

conned Amuat uaohv mafia causaa a mo uo~>u£oa ucoaooaamauo-3euh .N~.¢ ouswum
Ca Aauaosv u .oluk O.~ ~.o
. ~.o
can: uucqom mauuuaum
25.-.. 3.. H...
o mac n- 33 333m 3: mm.o-~3.o
uuauq ucoaooaunuua :3 ms
3:33.: was aaod.n
2x o~.m
m.eu:
nm.c
2x c~.m
m.mnc
zx Ne.m
m.muc
zx mm.m
m.enc
gulls ...... ELEM
cooauom cocwauouon coo:
«Ham
muouwsaunm noouo

 

 

 

 

(mm) 9 ‘3uamaoatdstq at};

1.5I)

 

 

 

 

 

 

 

3 I «3+ 333333333 _ _ 3333333333 3 33333333 . 333333 +3331+ 333333 + 3333333 + ..... +3---
3... . - . 8.: . ..... 733-3.». . _ . _ . _
......... _ . - I. 3 1.55... _ _ _ _ _ _
2... . - . 8.: . ..... 73333.2 _ . _ . . .
I - .- . . a.” . .......... . . . _ _ _ .
3.3 . - . 3.3 . ..... 733-2.». . . . . . .
. . - . 2 . .......... . _ . _ _ _ _
8.9 _ - . 8.: . ..... 73333.2 . . . . _ .
_ _ . ... . .......... _ _ . _ . _ _

- . 8.3 . 3 . ..... . 3 . 3.3 .3.: .3.". ”.3 72.2. a- 73
u u u n . u u u u u a
5:33.32... 33. .Bsu. 33.I.I..l. u.l..l.u2o.
33333332 33333332 333333 . I . i _ _ 03 - _ . _ _
v33.— _ Q33 _ ‘33.— uo _ ~33. 33: SI 73333332313: _ 333-. _ 3.3...

no»: _ 333a.- _ 3.: 34»- 3' . 333333 _ 3383- :33341333. 333:— 33 73333.3:
33.. 3333.33. 3333. a. 3333 . -33... -3 33 . v.3 . 3333 . 33.3333. :3 .1.-32...:
-8193.heads-37333332938. 33.2.3.- . 133—-3372... 338. 38 . .3.-«7:3

.83.— ..3-35 33 3.38 83.33 333.8 8333. 383 .3382: 3:. 333381333 .3... 33:

151
behavior of curve A. Curves B, C, D, and B would move

closer to curve A (by slowing down) if each curve was
extended, but any comparisons between them would extend
over a larger range of total displacement and thus
complicate these comparisons. Curves B through E can not
be readily compared to the nearly-steady Curve A, but can
be compared with each other since they were drawn during
the same early stage and are within a relatively small
range of total displacement. Data comparisons are also
improved because no pile unloading period was allowed to
complicate conditions during the test.

With parameter b = 1, equation 3.4 written in the form

. _ n

was used to compute n values shown in Figure 4.22. These n
values, computed between curves B, C, D, and E, are similar
to the value of n = 4.0 computed from the full duration

loads (at C-type starting points of 5 mm) in Table 4.1.

4.6 Threaded Rods as Model Piles

Each thread of a threaded rod can be considered as a
lug and the entire pile as a series of closely spaced lugs.
Pile test data for a single size thread (24 per inch with a
1.06 mm peak to peak distance) in a frozen loose sand is
shown in Figure 4.23. Note that pile failures (Tests #91

aand #86) began at approximately 1.0 mm which is nearly

1.522

 

 

 

 

 

 

 

 

 

 

 

 

 

Table 3.4. Mould 2113 foot. Results
typo IrnssaolP11sIc-rypcI3anoIrcst I3c11cI1n- I Loan I01s- I01s- I erscp .
at I to Irutlseut-I or I!-- IVol- Iluoll Inez-u. I placo-Ipllco-I Pun-ton
Loan Imssoluc. I to; Ilcollpon-I ass I LcsoI I ssnt I not I----+---+---
Docs-ILanstnI Irc1nt I Itnac Insc-I I In LoacInsts I n I n I:
steal I I I I I maI I luv In. u-I I I °
I I I l I I l I I cr-aI-Icr-aOOI I I
I I I I - I I“! CI 3 I H’II 3 I I: I - III/h: I I IMP.
3 3 c 3 . . I c I 3 I 3__.
rn11 I 1.00 I 01 I 0.10 Im «I -3 I30.3 I 2.03I constant I - I - I0.00I3.2I2.10
I I 00 I 0.13 | I I57.0 I 2.02I constant I I I I I.
I I 00 I 0.10 I I I 57* I 1.3bl constant I I I I I
I I 01 I 0.10 I I I 37¢ I 0.30I constant I I I I I
3 3 3 3 3 3— I 3 I 3 I 3 #
0011 I 1.00 I 01 I 0.30 IN «I- I30.3 I 2.03I constant I - I - I0. 03I0. 3I1.3s
I I 00 I 0.30 I I Is7.3 I 2.02I constant I I I I I
I I 00 I 0.30 I I I 37' I 1.30I constant I I I I I
I I 07 I 0.30 I I I 570 I 0.30I constant I I I I I
3 3 3- 3- 3 3 I 3 | 3 I 3 3
Fall I 1.30 I12 I 0.0 Is mI— Iso.s I 2.00I constant I - I - I0. 0 I2. 0I
I I I I I I53.1 I 1.71| con-tan: I I I I I
3 3 3 3 3 3 I 3 I 3 | 3 3
3011 I 1.50 I I 0.10 Im uI- I30.s I 2.0sI constant I - I - I0 30I2.3I1.30
I l l I I I53 1 l 1.71I con-sans I I I I I
3 3 3 3 3 3 I 3 I 3 | 3 3
3011 I 1.50 I I 0.3 Im mI- I30.3 I 2.0bI constant I - I - I0.00I3.3I1.33
I I l I I ISM I 1.71I meant I I I I I
3 3 3 3 3 3 I 3 | 3 I ----+---+----
3011 I 1.50 I I 0.7 Is mI- I30.s I 2.0tI constant I - I - I1.0 Is.2I1.05
I I I I I Is3.1 I 1.71I constant I I I I I
3 3 3 3 3 3 I 3 I3 3 I 3 3 -
atop I 1.00 I 07 I 0.1 Is MI- I 37* I 0.30I3100stst1cl 0.103 I0.0033I1.0 I3.2I1.05
3 3 3 3 3 3 I 3 I 3 | 3 3
stop I 1.00 I 00 I 0.1 Im al- I 57* I 1.30|+ 70stst1cl 0.00 I0.0077I1.0 Ib.£I2.60
3 3 3 3 3 3 | 3 | 3 | 3 3
Stop I 1.00 I 30 I 0.1 IsanoI -2 Is1.1 I 0.07I+ A23t3elcl 0.23 I0.072 I1.0 I2.0I1.10
3 3 3 3- 3- 3 I 3 I 3 - |----+—--3----
snortI 1.00 I 70 I 0.02 IsanoI -3 Is1.1 I 1.11I+ 03stat1cI 0.123 I0.037 I - I3.2I -
3 3 3 3 3 33 | 3 I 3 I----+---+----
shostl 1.00 I 30 I 0.01 IsanoI -3 Is7.3 I 2.02I:2.0I 1 I 0.227 I0.2t I - I2.5I -
I I I I I I I 2.02Igs.2/ 1 I 0.~50 I0.1at I - I2.sI -
I I I I I I I 2.02I35.2/ 1 I 0.050 I0.152 I - I3.5I -
I I I I I I I 2.02I:3.2/ 1 I 0.050 IO.lh8 I - I3 oI -
3 3 3 3 3 3r I—- 3 I- 3 I----+---+----
shoztl 1.00 I 03 I 0.01 I1cc I-3,0.00./ch 0.00I+ 32stst1cI 2 I0.0031I - |t.1I -
— 3 3 3 3 3F 3- I 3 I ------- 3 ------ I----+---+----
snottI 1.00 I 03 I 0.01 I1cc I-3.0.00./ch 0.01I+ 27stst1cI ? I0.0050I - I3 3I -
I I I I I I 0.31I:s.0/ 10 I ? I0.0007| - I3.0I -
5 PC dot-wand or. Z; - 0.1 all/h:
0 Sand 10 31:0de 0.32 to 0.59 II. 100 10 polycrystalline

'v hum-cod from the worn” of similar cuts

? Displacement. tacos mounted utter largo Jump: in total dlaplncmont

.923 «33.5 cu moo» topoousu .uo ucoaoowadmuoos... .33 0.3:»:

Ana—.55 u . al.:

 

mm on mm 0N m. o. m o
_

 

153

 

    

    

   

  

 

1 1 1 - 1 I 1 1‘ - 1 1 1 1 — 1 1 1 1 1 1 1 I1 ~ 1 1 1 1 q 1 1 1 1 N 1 1 D
T;
usoaouaadox ”Eon—0097.30
Z? . mafia—5a coo—3:58.39 13.5005 1 m .o
.«IIIIII. L I n \
o.» rt 9 .
V 7 .3
m .+ O .
c .0 w on: and S
o c. on: .34 S .
w m 3 no: So on
a an: 8.~ S .
m a I o. a
a... M u .3.: :0 .oz anon.
M m naouum 133.; . 33m .
H
2
1
0 non n- on 9.3. no 3.6-3.0 .
uouofinua I: and downed-Boone souuusvm
15.5.an :2: you canon—E. «a noouo ca 60952: uoz out
351 aooum fiction—fl. nauqun you Hmé .wuh mom 3. flax.
m . fl

 

 

(mu) 9 ‘nuamaovtdszq 312M

154

ooooounh Adv .osoa cououu :« noon voodousu mo uo«>ozon usoaoooadmwo-osua

OOH

.wcaoooa Amanda“ Hound la «H.c vouuouu oo>hso ucoaoooadmwo-oauu ton

Om

J

Anusogv u .0aua

In

0.

fl

.o~.3 cscwan

~.o

 

 

«on

$3\

q

cm‘

~31

IIq

 

U 0

n:\:: ~.ou M o cmza~.~ucn
23H on cannao.c I o

«.n I : on.o I n

uuouoaouom saunasvm dunno

 

.A..o

~.o

 

 

(mm) 9 “nuamaoatdsyq OIIJ

155

.msuoooa Hounds“ nouns as n.c vouuoua no>uso ucoaooounauv-oauu to» condense .nq~.¢ ouswum

oo~

Cu

.9355 u .333. o;

~o.c

 

 

+

   

0 0

taxes ~.oI « o cmzmo.aucc
2:” on so nnncc.o I o
o.o I : mc.o I a

aucuofiduom scuuosvm doouo

 

 

(mm) 9 ‘nuamaootdsrq 3133

156

.mcuvooa Hands.“ uouuu .5: 5o noun—3m noise usoaoouamuuvoauu Uou cocoon—fl. ASN.3 0.5m:

on:

A:

A9335 u .093. o. a a .o

 

 

q

4

of.
H3.

 

 

(am) 9 “:uamaoetdsuI OIId

157
equal to the peak to peak distance.

Figures 4.24a through 4.24c show log (time) - log
(displacement) behavior for the pile tests shown in Figure
4.23. While measurements for these relatively small
displacements are more subject to system errors as compared
with the lugged pile data, it is clear that these piles
achieved nearly steady rates at much smaller displacements
than the lugged piles.

A time-displacement curve for a model pile with larger
threads (16 threads per inch and a peak to peak distance of
1.59 mm) in frozen loose sand is shown in Figure 4.25.

Pile failure for Test #90 appears to begin, with some
initial oscillation, at 1.5 mm which is nearly equal to the
thread to thread distance.

The behavior of a model pile with small threads (72
per inch and 0.35 mm peak to peak distance) in frozen loose
sand (particle size of 0.42 to 0.59 mm size) is shown in
Figure 4.26. In this case failure occured at about 0.14 mm
displacement. In polycrystalline ice this same pile failed
at 0.17 mm (Figure 4.27), compared to an apparent pile
failure of 0.14 mm in frozen loose sand. Any interaction
between sand and threads may have been prevented by the
small size threads which limited failure to ice at the
soil/pile interface.

Time - displacement data for the 24-threads per inch

piles, frozen and loaded in polycrystalline ice, are shown

158

.vsuu

 

 

sououm :« cocoonso madam send you cooks» wH you uo~>o£on usosooaaanuo-osfie

o.

d 1- 11 4

A9505 9 . 03:. m

nn.c I ocaocsnn cacac> ndficm .m><
0 non n- no scan as on.c-«o.o
scocncua on ~n.n

Acao.n~c non” non cocoons on

7 atom Hooum cocooush.

.m~.o cannon

 

 

 

(mu) 9 ‘nuauaavtdSIa 8112

159

.ocaa conoum cu ooooonao madam son“ you poops» as you Hou>a£on usoaooodduup-osuh

0..

 

A3305 u .03.: m.c

1

 

0 non n- as coca ca cm.o-~o.c L
uouoaoua ea «n.m

Anso.n~c scan non uncanny «a
«com Hooum topooush

 

 

.c~.3 cocmnn

c.o

(mu) 9 ‘nuamaoetdsgq 3114

160

.vcum can and asuaununhuo>aom :« popponeo
oaum coca non cooks» an a mo uo~>n£0n usoaoowamnuo-32uh .nu.c onswuh

om av om Amazes. u .0339 o. o

 

 

11.-.111-141.1-11111--‘~111111111—111-.4111—11-1-o14

. . 3.0
an on o u
- a noun oaam
o moo a u a us on: 0.09

and 19: ww.

  
  

. anaconda as «n.c
Aan3.n~c scan non cocoons «a
icon dooum covnoush

 

 

(mu) 9 ‘3uamaoatdsgq aIIJ

161

in Figure 4.28. Initial failure begins at about 0.19 mm.
This is close to the failure displacement shown in Figure
4.27 for the 72 thread per inch pile. Additional tests are
needed to show if the failure displacement in-
polycrystalline ice is totally independent of thread size.

Pile displacement data, summarized in Figures 4.29a
and 4.29b, was obtained using 2 displacement transducers
positioned on opposite sides of the model pile. This
arrangement recorded large displacements at the start of
each load increase. Later, a third displacement transducer
was added and all three displacement transducers were
repositioned evenly around the model to provide more
reliable results. Although the total displacements shown
in Figure 4.29a and 4.29b are questionable, the
displacement rates shown after each displacement jump
appear to be reasonable. Straight lines were used to
approximate the displacement rates of these piles during
given periods. Ratios for these displacement rate changes
are shown on each figure.

An enlarged portion of the data from Figure 4.29b is
shown in Figure 4.30. Pile Test #86, reported earlier in
Figures 4.23 and 4.24, is shown in more detail in Figure
4.31 and Figure 4.32 with the displacement rate increase

approximated by straight lines.

162

cm

ON

Aouzonv 9

.oo« annuaounhuo%aon :« convonEo
cage 50:“ non enounu «N a you uoq>¢£on usoaoouanuuo-uauh .w~.c unawam

.oluh

D. m

 

 

1

q

q d u I n 1 a - u a -

 

o moo n- on 00H ocuauauuAHUAHom
2.32.33 .5. and

.5338 :2: non 2335. 5
atom Hooum tenuoush

m.o

 

 

m.o

(mm) 9 “nuamaoutdsxq SIIJ

163

.aun mo ouoouosa coca ouuuum adv .oou osuaaoumhuoaaod :1
tectonic oawa soon you nouns» «a a you u0u>i£oa usoaoooamn«o-oauh .o~.¢ ousmum

 

 

 

 

mm om mm ON .3305 u .033. o. m o
q 1 1 1‘ 1 - 1‘ 1 1II1.~ 1 1‘ 1 1 ‘ 1 1 I‘J - 1 1I 1 1 q I1I 1 1 1 11 1 1 1‘ 1 o
.9254 N
no lash Ian:
“0 no. «Och .
TL
.3 mu
nu n,
W Aoaoouosn wood. u£\§l aged 7
9 ouuaum paw .- noun—v w.n I W
m on}... n86 .0 :6
o,
e no
m m
m
. . .
nu nu
Issa‘ \II I 9
5:55 MW 0
386 N _+
m. .0 Lia
nu
5:3... mm.
mmo.o\ m u not n. u. can 253329853.—
\\ uouoaouo as ~n.a
\ w 3568 :2: non 2.3.35. om .1
\ H atom Hooum cocoon—F
2

 

 

 

e...

(mm) 9 ‘nuamaoatdsxq 3114

164

.osu mo oaoouonu vaoH cuuaum .nm~.q gunman

om mN ON 23:05 u .033. S

 

 

 

 

 

 

m o
q d a d J a d u a J d W q 1 + q 4 d 4 J 1 1‘1 1 d a d a J 4 N00
.9254 N
task Cain
AoanOHOCH coca u:\al Onoo.o '0. such mu
033m .3 a :83 «.N n W; «6
u£\la duo.c "N
T d
9
_ by
0 T.
7 _+
T. ..
"N nu
H E . .
i a 28 o\ \ w m
nu AU
9 m
MM 8 i m.AV
d w
a nu
H“
7.
0 new n- on can osuaaauuhuoxaom A
uouoadun as «n.o
3363 :2: you «335. «a
«com Hooum covuaush
m.o

(mm) 9 ‘nuauaaatdsxa 9114

165

Quad ouaohu a scam oaaouoau ounu ucoaoouaauuo a mcuzosa ooa oauaaauahuohaoa :«
Acwt uaoav «Hug Sosa non vouaounu «N a Mom uoq>unoa ucoaooaeaauv-cauu conunacm .om.c ousmum

 

 

 

 

A: m o Ann—.55 u .25. c N
. . . . . _ o.m
Aomuouocu vac: o«~o%o u£\aa smoo.c .
«m.¢fl a aouhv o«.a I
u:\33 dac.c
n ~.m
_
n ¢.m

o mac m. an ccum as mm.o-~<.c
uouoauun aa.~n.o J e.m

Aaaq.n~v nucH “an unauuna «m
«com Acoum vuvmoush

zH010 96H OZO'O + IV'O -—

 

 

 

 

(mu) 9 “nuauaoatdstq SIIJ

166

on

.avoou cuauho voaonauuoasa mo «canyon saw: anon :onoum «mood :«

 

 

  

 

   

    

Aomt uaoav caun 50:1 uon oaaunu «N a you uo~>¢soa ucoaoooannqv-osua .am.¢ ousmum
m 3585 u .32. o
1 1 1 1 q 1 1 1
0 m3 m- a. 2.3. a... 36.3.: 2
nous—33 a... «n5 Z .0
33%...8 :2: no; 335:. a .0 7.
atom dooum 00.335. .7. m
.. a
\ m6 0
O .
. 7v nu
S
.6 _+ 7.
. 7».
79. nu nu ma
20 oz 0
7o. nu7u 1L 8
zwmz_+ w a
.6 oz _+ “0 T. IL
7. Nu 74+.mmnv8 P. ma mm
mu .0 mmAUpor; mm a h”
. 7.. _+ B . W «0 C.
0 I m 1..
z a mu m m a m m
.+ a I a 0 \z) mu
T. I H N
m m m m z m.
m m. m (
a m
au 2
I
m
2 l

 

 

‘2
p...

o.~

(mm) 9 ‘nuamaavtdsjq OIIJ

167

.ucoaoouuaauv a! nc.c uoumo uo~>¢£oa ucoawouaamuv.oauh Adv

 

 

 

.vQOA ouaohu vaaoaluuoaaa a on mac vomuuacm .Nm.c ouswum
o.c m.m Auuaonv u .0319 o.m
_ 1 _
v. 136
_ m
.0 \ 7
q. \ m m
0 \ _ _+
A8835 23 0298 2).... 9:6 m \ \ z 0
3.2” a :85 «NA .... N \ \ .0 .I u
2}... 2:... Q \ \ z w
J\\_\\ L. mm
. a m
o\ o a
. . . z m
we m“ _+ mw 1mm.o
m s .o z
I
v. m m m
0 B I
z u. m
— _+ I 9
w. mu m u
m m m
m z
a m
3
an
I
mfluu . Im®.O
S... .32.. 5 :36 8- a!» 2:5

 

 

(mm) 9 ‘quamaoatdsxq 91:4

‘168

a".

.ucosooouanun as no.c nouua uo~>¢£on ucoaooaaaauv-oauh

.nmm.e ousmwm

 

 

 

m 1 . ‘ Q; :32: u .3: 1.
4 . . q
z
0
Z
N
d
9
l
u:\aa m¢~.o
:4 a
u;\aa m-.o
\“
\\ z 1
\ w
u;\aa a¢~.p\\ .+
\\\ .U
\\\ p.
\ is... m:.c W
\ n 1
d
8
WY 9 86 .33.. c. :38 8. an.» 2.35
.40 m.
.3. H
2

O¢.o

m¢.o

oo.~

 

 

(mm) 9 ‘auamaaatdsya atxa

DISCUSSION AND APPLICATIONS

5.1. Creep Equation Parameters from Static Loads

The results from full duration model pile tests,
summarized in Table 4.1, may be analyzed to better
illustrate general pile behavior. The parameters n, b, and
PC are shown for 3 model pile types in Figures 5.1, 5.4,
and 5.8. The piles represented in these figures are
similar in all ways except for their lug size (3.18, 1.59,
and 0.79 mm).

In figure 5.1, the b values for each lug size approach
a value of 1, which represents a constant displacement
rate. The b values for the smallest size lug approached
this constant value more quickly than the larger lug siZes.
Values for the 1.59 mm lug at 3 mm and 6 mm (Figure 5.1)
appear to be low because other b values for this lug height
are higher in Table 4.1. As a smaller lug advances through
the frozen soil, fewer particles are moved, compacted, and
displaced around the front of the advancing lug. The
smaller volume of sand grains and ice ahead of the smaller
lug should accumulate with less pile displacement. A two-
dimensional approximation of what these zones of compacted
sand might look like is shown in Figure 5.2.

Ladanyi (1976) showed a compacted soil condition for a
a flat cylindrical punch (Figure 5.3a) after it had

penetrated a frozen soil and reached a relatively constant

169

170

.aouou waouua> us Aalav ouauo honoua sooouooo nouns was usououuuo and: noaum .~.m ousmum

o

Aalv usuom msuuuoum onhh.o
¢

_

N

To

 

r.
Ego: 0:... EE mfn Olo
.2961 93 55 mm; film
29o: 9: EE and I

ausuom msuuuoum ooha-o
saw can n on 361.0; 0» uncoo<
.naa asan.~ you ao=~a> a

 

r n:
‘9.
o

\‘

u
q zonaUUJBJ daeza

I
“2
o

 

IO;

.uusmwos and usououuuo,ms«>an «mad mafia

 

 

Mo boos. coauooaaoo no coauuooaoo oouaaooon ~.m ouswqm

 

 

 

r a

\

 

171

~
.
.

 

 

 

TJ aw “soaoooqouaa.vw A .m 1 .w .usosoooaoawo
_

 

 

 

 

 

 

 

A .7 ,1.

 

 

 

c 00000
c 00000
000000
000000
-oooooo

 

172

.msa mafia poncho ousnuoun comaouosa no common Apv .Awmaa .qhsoooq noummv

Hovoa sownsoaxo Auu>ao any .ma~ mafia use nu» coauauuosom sooauon somwuodsoo .n,n ousmHm

ucoso>oz oawm
no noduoouuo Adv any
9 . >uw>ao Houfiuozmm _

onanuh >uosson cuunoumnu«umufim .

. OH hadhfi
... uses and»
.. cananoum

ucoum Owumaam
msqoco>o<

    

use no acouh :« .
anon ounn:uo..
vso vouoanlou.\\
\\..

   

  

 

 

u ..xouno4.
usosooonn.«o

bl :58. .
van osuuco>o<
neuron uso>

KQEAPD I'D!JPUFIKO

pox uoyaoxaauad
'szunoa orusotd-orasuta

 

 

 

 

 

 

Thouosuao IL

173
displacement rate. The region just ahead of the flat tip

(half sphere of radius a) would correspond to the region of
compressed sand shown in Figure 5.2. In Figure 5.3a, the
region labeled as "advancing plastic front" illustrates
where frozen soil plastically deforms around the advancing
lug. A compressed zone of ice and sand ahead of the
advancing pile lug and flow of sand and ice around the pile
lug are shown in Figure 5.3b. After the lug has passed by,
a void remains. Figure 3.1 shows a void ahead of the pile
lug, created during sample preparation in order to simulate
lug spacing on a multiple lugged pile. As shown later,
piles with these displacement limits reach a critical
displacement when each pile begins to accelerate. This
pile failure involves shearing of frozen soil ahead of the
lug.

The proof load PC in Figure 5.4 represents the pile
load at an arbitrary pile displacement rate of 0.1 mm/hr.
This load tended to approach a constant value for each lug
size as the plot starting point was increased. At this
displacement rate, the resistance along each piles smooth
shaft should be constant (for equal lengths) so that the
difference between PC values can be attributed to variation

in lug size.

5.2. Pile Capacity For A Straight Shaft

A more complete picture of pile lug behavior in frozen

174

.ouuo was undo :« oosououuav souu auasmou oooa uoouo souuoavo doouo sq oesououuuo .<.m ouswum

Ally usuom msuuuoum oohh.o

 

 

 

 

 

 

0 ¢ N o
F _ L Augu
£90: 03 EE mfin olo
29»: 9: EE 3.. mum
3301 03 EE and I
Ind
P‘I||.||||||||Aw .d
1» «41 4 1
4 O
10.. m.
LP\\\\\$ m
3 E .9
m7 W04 3...
m
TON

 

10:0.

175

sand is obtained by considering resistance to pile
displacement along the straight shaft of a pile. With this
knowledge, model pile capacity can be separated into a
shaft resistance component and a pile lug resistance
component.

To determine that portion of the total load
attributed to a pile’s straight shaft, silicon grease was
applied on some piles to eliminate or greatly reduce
adhesion, mechanical interaction, and friction at this
interface. A direct comparison between partly greased and
ungreased piles is shown im Figure A.1. Note that the
behavior for Pile Test 99 (partly greased and loaded to
2.21 kN) is very similar to Pile Test 112 (an ungreased
pile loaded to 2.66 kN). The difference between these loads
is 0.45 kN (shear stress close to 240 kPa) and appears to
be representative of the adfreeze strength at the section
of pile behind the advancing lug. The pile sections in
front of the lug were influenced by pressures developed on
the lug and have, most likely, a higher level of adfreeze
strength.

Another comparison can be made using creep parameters
for the 1.59 mm lug height piles in Figure 5.5. Piles with
an average embedment ahead and behind the lug of 76 mm and
63 mm (139 mm total) can be compared to piles with 25.4 mm
and 64 mm embedments (89.4 mm total). The difference in

total embedment (49.6 mm) corresponds to the difference in

176

.asouuuoo uuona unmuwuua Hausa moodu ausoaooaao usououm«v no“:
oawn commaa you saona wood moose souuasao noouo cu oocouomuqo .n.m ouswam

Ally usuom msuuuoum ooha-o
o w N o
:5: .ama 6.: on I

es: .330 EE mm min

I
‘0.

 

Imé

d 79901 JOOIJ

0

(N31)

177
proof load values, Pc’ shown in Figure 5.5. This
difference, about 0.3 kN for the 49.6 mm embedment
difference along this portion of pile can be converted to
approximately 200 kPa.

Earlier attempts to measure this resistance (Alwahab,
1983) using straight shaft piles and constant loads, lead
to sudden failures after very small initial displacements.
The load capacity of the ice adhesion bond is much greater
than the residual capacity remaining after this break. The
constant rate tests by Alwahab (1983) probably better
represent long term conditions for straight shaft piles
since they did not allow rapid pile displacements to occur.

A plot of data from 4 of Alwahab’s (1983) constant
rate pile tests is shown in Figure 5.6. The initial peaks
vary greatly and are dependent on the displacement rates
applied. A plot of peak pile capacity versus displacement
rate (similar to those shown by Alwahab, 1983) gives an n
parameter of approximately 5 in Figure 5.5. This parameter
is typical for frozen soil creep processes at this
temperature. After the short initial displacement required
to rupture the pile/frozen sand adhesion, all residual pile
capacity is derived from mechanical interaction and/or
friction. The estimated residual pile capacities, after
approximately 0.56 mm of displacement, are plotted against
their respective displacement rates in Figure 5.7. Note

that a descriptive creep equation (equation 3.4) for a best

178

.oouou usoaoooaoouv acoumsoo «sowuo> uo
ouoq>o£on cage uuoza-u£m«ouuo uo soaquodsoo .w.m ousmqm

Aaiv vac-confined: oaum

 

 

 

mp0 who 30 who 0.0
.E\EE 2.0 I o
.E\EE 5.0 film

E\EE To «In
.E\EE .00 I
O ®/®\®\mloo_
4
9/[07 . room
I. n 1 1\
./l “
.9eloon
6 e
100v

 

I 00m

(Bax) 311a no 933135 zaaqs

.Ao.n ousmdm scum uuuov nudge uuazm-u:w«uuum mo zuuomdmo

 

guacamou can auuoooao soon ouuaouosom ou :Bosa ouo «mafia yum umon .~.m ousmwm
Au;\alv ouom usoloooaoaun
00— on o.“ “.0
_ _ _
38230 «_E .3393. o . oo.
\
xgooooo 2E xood x \ Q

179

nuns no you anon

nouoaum among uooaouaou «coda "ouoz

      

ICON

loom

100?

 

1 com

(Ian) at}; no 883133 Ivoqs

180
fit straight line through this data has a relatively large

n parameter power term (n = 16). The dashed line in this
figure is best fit to this data excluding the 0.91 mm/hr
displacement rate test (which is different from the other
data) to produce another relatively large n value of 12.
Alwahab’s (1983) tests at -6 deg C did not show the
relatively small peaks (especially in the 0.13 mm/hr test)
observed during the tests at -2 deg C. Experimental
limitations prevented Alwahab (1983) from testing at slower
rates which may have produced tests with relatively small
peaks at -6 deg C.

At warm temperatures and slow constant displacement
rates, little or no peak may result as the ice adhesion
bond shears at the smooth pile surface. Gradual changes
during adhesion failure would allow sand grains to remain
intact in the frozen sand matrix. If the pile were smooth,
relative to the sand particle size, very little
rearrangement of sand grains adjacent to the pile surface
will occur and the pile’s capacity due to this mechanical
interaction will be small. Dilation of sand grains, at
very small pile surface irregularities, could increase the
pile to sand grain contact pressures and thereby increase
friction along the interface. Pressures at the interface
would eventually approach equilibrium with lateral soil
pressures around the pile. Residual pile capacity derived

from frictional forces would in turn be dependent on the

181

creep process of frozen sand surrounding the pile. Load
capacities for these piles should approach and remain at a
constant value for these circumstances.

Initial displacement behavior of a straight shaft
pile, at the time it is first loaded, is influenced by
creep behavior of the ice which carries much of the pile’s
load. A creep equation describing this initial behavior
contains the n parameter characteristic of the creep
process. Once the ice bond at the pile surface is broken,
other materials at this interface (steel and sand) are less
effected by time. A small increase in pile load will be
accompanied by a large displacement rate increase. A creep
equation descriptive of this behavior should have a larger
n parameter. However, the load-displacement behavior
following the adhesion break should retain some time
dependance because pressures at the pile and sand grain
interface are dependant on creep processes of the
surrounding frozen sand. The point of this discussion is
that creep behavior and its n parameter can change with
increasing displacement rate when certain pile types, soil
types, and loading conditions are combined.

The sequence of events described requires a slow
displacement rate as the ice adhesion bond at the pile
surface is gradually sheared. Constant load tests with
straight shaft piles generally rupture the adhesion bonds

which permits a rapid pile displacement. The residual pile

182

capacity following this break is usually very small. As
the adhesion bond breaks on a lugged pile, a portion of the
pile load will be transfered to the pile lug. This
transfer helps maintain a residual load on the straight
portion of the pile. The mechanisms involved are not fully
understood.

Values of 200 kPa and 240 kPa shear resistance
observed at -3 deg C for the straight shaft portion of
lugged piles will be averaged to 220 kPa. This value is
approximate, since it was based on limited data and was
determined indirectly. It will allow comparisons between

pile capacity components in subsequent sections.

5.3. Load Capacity From Pile Lugs

The n parameters for 3 lug sizes, summarized in Table
4.1, are plotted relative to their respective C-type
starting points in Figure 5.8. The n values for each pile
type approach a constant value and increase with decreasing
lug size. However, this graph does change when pile shaft
resistance is accounted for.

Creep equation parameters in Table 5.1 are listed as
they appeared in Table 4.1 and again after each pile’s
resistance was estimated (220 kPa behind the advancing lug)
and subtracted from the parameter calculations. The 25.4
mm straight pile portion ahead of the lug may develop a

higher resistance due to higher lateral pressures around

183

.onuo msa wsquoouooo £u«3 oomoouosu s Houosouoo doouo .m.m ouswgm

A35 usuom msuuuaum 05-0

0 to N
_ _

_
.5301 03.. EE mfin olo \
EBoI 03 EE an; mid . . Q\
5901 03.. EE and To

 

 

Tl

 

 

 

 

do canon. son..— «on:
~.m canoe noun anon vouosnv< oooA.ll II. II.

[0.m

u zaaamazaa daeig

184

 

2: 6:33 8568.8 8.. «6 6.6.36: 6 666 n- ... 65.. I. 3.6 3 «6.6

o cocoa-66656566636666 6

ail 6. 6. on 65 u 8: 6.5.83.6 6.. ..6. 6 6. 686:... 66 a

 

 

 

 

 

 

 

 

4 4 - 4 - _ 4 4 4 _u--+---+--au+---.uu--+---+---+---_---4---4---+---_-------+---+ .....
_ . _ .666 .3666. 6.6 .666 .86 . 86.6. 66 .36 .666 . 2.6. 66 .36 .666 . B6. 66 .666 . -6 6 :66 _ .
_ . . .666 .2666. 66 .86 .666 .3666. 6.6 .26 .666 . 26.6. 66 .36 .86 . 2.6. .6 .666 . 6 «6 :66 . 6... .666
- 4 4 4 . 4 4 4 _ 4 4 4 _ 4 4 4 --_----+no--+---+---_---n----4u--+----
_ _ _ .666 . 6666. 6.... .666 .666 . 66.6. 6.6 .666 .86 . 26. 6.. .26 .26 . 66.6. 66 .36 . -6 6 :66 _ _
. . . _«66 . «666. 66 .666 .36 . 866. 66 .666 .666 _ 3.66. 66 .26 .666 _ 666. 6.. .666 . 6 3 266 . 66 .666
4 4 4 . 4 4 4 . 4 4 --- 4 . 4 4 - 4 -_---4onu--+--u-+---_-------4---4---
66.6 . 66. 6.. . 66 .666 . 66. 66 . 66 .36 . .6. 6.. .86 .36 . n6. «.6 .666 .26 _ «6. 66 .26 . -6 6 2.8 . .
666 . 86. 6.. . 66 .666 . «66. .6 . 66 .86 . 86. «6 .86 .36 . :6. 66 .666 .666 . 8.6. ....n .26 . 6 66 :66 .666 .26
u u u _ u o _ u . u n u u u _ u u 4 _ 0 o
69.6“. . avian. . ......IJ... _ 7...: .. _ suit"... _ _ ....l. _l
a. 6.6.6.... 6.6.6.... "1...... 6.6.6.236... .3..
4 4 4 _- 4 4 4 . 4 p ----4-----. 4 4 4 . 4 4 --+---_ voanonxouos<. .o:_ou6m
I «6 .. .8 868 686-6. I o n .8 868 8.3-0. I. a u .8 868 686.6. I 6 I .8 868 66.3.6. ..I 6 .. .8 868 836.662.663.65. .68. 66..
_ . _ _ -- _- .
.63 6638 868 8 66563.8 635:. 2631-38. 65 5:.
.88 6.33 66 3.8 8383 668 88 6666.368 832.8 68.6 6.... .38

185

the lug. This effect was not estimated or accounted for in
the calculations. Experimental limitations prevented
determination of pressures on the lug. The recalculated n
values for the lug and the pile portion just ahead of the
lug are all approximately equal to 5 (see Figure 5.8) and
appear to be independant of lug size once steady state

creep (b = 1) conditions are approached.

5.4. Temperature Dependance of Pile Capacity

Proof load PC values are used to illustrate the effect
of temperature on pile behavior in frozen sand. Values of
Pc from Tables 4.1 and Table 4.2 are summarized in Figure
5.9. A variation of the temperature dependance equation
2.7 can be used with the proof load Pca (at temperature 6)
values obtained from Figure 5.9 as

6

_ __ (.0
Pee — Pco ( 1 + a ) (5.1)

C

where 9c = 1 deg C, 6 = 0 deg C - T(deg C) the absolute
value of the test temperature, and Pco is the proof load at
a temperature near 0 deg C. Rearrangement of this

relationship yields:

"U
ll
"U
H
+
I
VI

8

co co (5'2)

which is a constant. Equating these relationships at

temperatures of -2 deg C and -3 deg C yields:

186

.uaou vuoa-mouu so nouns madam Hobos comm:~.uo huuoamuu boon .m.m gunman

mo
_

Huom conoum no souuoaum oa=Ho> ouaom

...o
_

Nd
.

0.0

 

.8.. 2.... o 666 a: I
.68 6E o 666 n: I

0.0

1.¢.O

Imd

IN.—

 

lo;

a ‘peoq 3001.1 uopenba (139.13

3

(N3!)

187

3 _ 2 _
(1 + ') w P (1 + -)
1 0(3) 1

w

Pc(2) (5.3)

These proof loads, the sample solid volume fractions, and
the computed w values are summarized in Table 5.2. These w
values are similar to the value (w = 0.64) determined by
Alwahab (1983) for model piles in frozen sand. The step
temperature increases from -3 to -2 deg C appears to allow

determination of the w parameter.

5.5. Dynamic Effects on Model Pile Load Capacities
Experimental data, summarized in Figure 5.10, shows
that model pile displacement rates in frozen sand were
essentially independent of frequency in the range of 0.1 Hz
to 10 Hz. This behavior is in agreement with work reported
by Schmid (1969) for model brass piles embedded in dry,
uniformly graded, unfrozen quartz sand with the same
particle size range. Schmid (1969) reported a pile-soil
resonance at about 50 Hz with a small peak in his load
versus frequency curve (Figure 2.21. This curve was
relatively flat at frequencies above and below resonance.
The smaller model steel pile and stiff frozen sand used in
this study should have a higher natural frequency and pile
displacement should be relatively frequency independent for
0.1 Hz to 10 Hz. Observed resonant frequencies close to
1.0 Hz and 6.2 Hz were clearly due to the loading system

since they could be altered by changing the loading arm

188

«at 0:0 ant names
.«.0 Quack 0H onoouosu ousuuuomfloh-moum 0:» float oosuauouoo i

00.0 0.4 .......An.n .avuv 3 .usosomuu 0000:0000 ounuouoaaoa

i 23 nn.n zu 00.4 ......0 000 N- u. 0.06 uooum saunasdu-moou0 owouo><.
as «0.4 as na.u ......0 0.0 n- u. 0.04 uooum sauuandu 00000 omnuo><
00.0 00.0 ......................=0«uonuh oaaao> vudom ounuo><

.6.. .66.9 scum .u.60
..ucosomuu nodalavfl nacho 0600:0000 ouauouoaloa no soda-:«Iuouoa .~.n ounce

189

.husoavouu aanuo> A00N44aauocv couch 000300040046 :0 uoouuo ms40004 04adsaa .04.n ousm4h

30 m .5885...

 

 

 

 

 

 

.l L
nil
. < la .
T. VA’| L
r
O! 0
1| . I.
>
E

 

r W\\\8 u 6.6. H a. 664 u 646..

04.0 ousu4m a4 .0 .04».. an.» I 0:000u00u 040035. ~044a4m

 

 

 

 

0.4

...;
O
H

N
o
H

m
H

~‘t
.4

In
H

\O
....

IN
0
H

‘ouvu 3m manaavmsm

OIHVNAG 380838

‘3
2g ' omvma Hllh

 

190
set-up.

An increase in dynamic load amplitude significantly
increased pile displacement rates over rates observed for
sustained loads (Figure 5.10). Comparisons show rate
increases roughly in proportion to the increase in load
amplitudes, i.e., the rate was approximately doubled when
load amplitude was doubled. Since several tests were run
on one frozen sample, a creep rate measured just prior to
application of the dynamic load might differ from rates for
subsequent tests on the same sample. To compare tests
performed at different initial creep rates, the ratio 31/ 51
was plotted against 31 as shown in Figure 5.11. Data for
each of the three best fit lines corresponds to a given
dynamic load amplitude. The small negative slopes may show
the effect of starting tests at various displacement rates
before secondary creep was fully attained. Points on the
left in Fig. 5.11 should be closer to the minimum or
secondary creep rate. In addition, 3 shaded data points
(indicated by a slash) represent displacement rate increases
for a small static load increase. For a dynamic load
amplitude equal toa static load increase, the dynamic rate
increase always appeared as some fraction of the static rate
increase.

The displacement rate ratios for static and dynamic
short duration load increases are plotted against the

percent of load increase in Figure 5.12. The cyclic load

191

0.0

.umc40004 A+v 0400». can AHV 04adshv «woman

can» 0:0300040040 asuuo> 04uuu 0004 000300040040 .44.n ousm4h

$5.50 W .68.. 042.22 58...... 6.6.5. 0.28.8588

 

 

 
       

66 .26 66 6.6 6.6. 6.6

. . _ . _
I I L

If > s 66 + z. 666
l \r mm. o<04 I 4.4
. a 64+ 21. 664 k 0 1.
.. + .. N4
. Av s .6 H z. 666 .
.. 4 N6 + z. 666 3.6.. 1 m 4
I q l 4V..—
G

.. X B .. n4
. u «6. H z. 666 .
.. 68.. ...64
. .6 6.6 + 26. 64.6 ‘ O D
l NV 5.4
1 I

04.0 ouam4h :4 .0 00440. 000» 0 uncoouaou 04obahu ~044a4m

 

 

 

‘onvu 3m mmovmsm

‘ S - DNIGVO'I :mom
"- g " sumvm usuv

 

 

192

.04l0shv 1:0 040000 .0000000:4 0004 00400000 00050
on 030 000000004 000» 000000040040 .44.n ouzw4m

A0004 040000 uo 00 00000064 0004

 

 

O— m w 0 N O
. . . . . 0
000000.... 000.. 0.005 +
003.38.. 000.. 0:26 0
low
c 10..
gfl¢‘\0..‘o
W ¢0$\
0.
\\\
\ 0
w no.
a
<
«
+ low
....6 no 66. 6.60
0000500 00004 uco0ounoa 00:44 .0002

 

[OO—

aseazoux 0383 nuamaoatdsxa

(t)

193
amplitude is expressed as a perdent of static load.
Equations for the best fit lines to each group (Figure
5.12) illustrate that a superimposed cyclic load (amplitude
X) would produce a short duration displacement increase
roughly equal the increase produced by a static load
increase (magnitude of 0.6 times X). The best fit line for
static loading could be approximated by an n parameter of 7
while the best fit line for dynamic loading could be
approximated (after adjusting the origin) by an n parameter
of 5. The mechanism causing the shift of the best fit line
for cyclic loading data is not known.

Available information on dynamic loading of lugged
piles in frozen dense sand and ice are summarized in
Figures 4.21(b) and 4.21(c), respectively. These early:
tests did not allow accurate measurement of displacement
rate changes, but appear to show that cyclic loads have
only a small effect on pile displacement rates in dense
sand and ice.

A few cyclic load tests were applied to threaded piles
in sand and ice. These tests show the same general
relationship between cyclic loading and static loading for
these piles. Results for Pile Test #86 (in sand), shown in
Table 4.4, provides n parameter values between 2.5 and 3.5
with an average of about 2.9. Static loads added to
similar piles as step loads or short duration loads during

Pile Tests #67, #69, #80, and #79 indicate n parameter

194

values between 2.6 and 4.4 with an average of about 3.4.
While the small number of tests performed do not provide a
statistically good representation of this behavior, the
implication of this comparison is that cyclic loading (at
nonresonant frequencies) does not produce a large
displacement rate increases in these piles. The effect,
however, appears to be larger than that observed in lugged

piles.

5.6. Limits of Model Pile Displacement

Differences in displacement limits, 6 had little

Ml
effect on the initial displacement-time behavior of the
model piles, but directly influenced the magnitude of the

failure displacement, 6 Model piles with different

F'
displacement limits, curves 106 and 102 in Figure A.5, show
a very similar settlement behavior up to tertiary creep,

8 mm for curve 102. Under similar conditions, a model pile
with a larger displacment limit should begin tertiary creep
after a greater displacement.

Piles with very short displacement limits, curves 103
and 101 of Fig. A.1, behaved differently from piles with
longer displacement limits, curves 99 and 97. Inspection
of split samples revealed that the shorter displacement
limit samples appeared to have a smaller zone of compacted

sand in front of the lug. Upon loading a pile with a very

short displacement limit, shear quickly developed across

195

the observed failure surface, increasing the pile
displacement rate, and producing failure before full
development of the densified zone of sand. A graph of
displacement limits versus failure displacements is shown
in Figure 5.13. With more data, relationships similar to
Figure 5.13 could be prepared to show the influence of pile
diameter (to account for scale effects), pile lug size, lug
spacing, load per lug, time to failure, different soils,
and different temperatures on failure displacements.

For comparison purposes, experimental results have
been summarized in Table 5.3. The load on each model pile
was adjusted (to P’) by subtracting the resistance (0.220
kPa) due to the pile section behind the lug. The area
resisting load at failure (AF) would then be the failure
surface ahead of the lug and through the frozen sand plus
the lug’s outer surface (12.7 mm length). The adjusted
load (P’) was then divided by the shear surface (AF) to

provide the shear stress at failure (1 All shear

F)'
stresses at failure (850 to 1990 kPa) were shown to
decrease with decreased time to failure tF (tF greater than
4 hours). If a model pile was loaded with a small load for
a very long time, the shear stress at failure may continue
to decrease asymptotically to approach the long term shear
strength for the frozen sand at -3 deg C.

Displacement of threaded piles was limited by the

pile’s surface geometry. The relationship between thread

196

.000440u 00 000000040040 000 04044 000000040040 0003000 040000400400 .m4.m 000m40

.50 :6 .0403 060000.488

 

 

 

mm ON 9 o—
. p . L
fin.
\\o
\\0\
\
\\Q\
m \\ {In
\\
\
m \
104
n. .0 .O.nl n .H. #0 620m ...Guouh
0 06040.. 03 .6. 666 IIII. 1m—
D D 00040.. 03 .6. 60.4
4 000400 03 .... 2.6
D Q 543 .5. 66. 044.. 40000 4066:

 

 

ON

(um) J9 'quamaontdsra azntreg

197
Lugged Pile Test Data at for Model Steel Piles Frozen in

Sand at .3 deg C with Particle Sizes 0.42 to 0.59 mm.

Table 5.3.

e ...
M . .548923135 39307306.0.
B 1 up.) 1.46.1398“11L8w.5.0~.5..3.nb11.31
mtitr 1 22 1 25.6.0.
.188 h
T 0.. I\
s n 0000 00000000 MuIIIIIIIIIIIIIIIIIIu
\l 0 0000000000
ts u a 0321.09543297477636034365
.6081 P” 60855.59689197935005827728
Muadr( 1111 2111121211122222111
.SS 0.
I'll Il"l"llll"'ll"llll|ll|lI
f 0
or u MI 97608 4.5.50.34.4.18986859J6m6.£~0.
O 0. 00000000
an 1 444/4.3....3210.0.889883876m7.6.,moo.
00810.1.“ 1111 111.1111
Mhll (
S F
t“ 'll'll'lll
ha“ ) 1644.86688008024941115..J7.O.
.081 Pm 6060860.44432227638889854744
sani‘ 111 1111
11Mfl I\
DP 4..
'l'l IVI'lll'I'II'IIIIII'I"||II
.
.«MR H) 4.4/44400.027224448888271111
3.6“ Cu 555559995200555777752444I.»
my...“ I\ 2222211111112221111112222
ll|I||AV||'|'|IIIIIIII'I'II|II|ll
9 7
.w 6 6
ted \I 8265229417114858797116478
ufla W 60.22%36221222733771283196
“PM ( 2822732222222111112211100
A 1 1
Illl'll"ll'||"ll|l|l|lll|'l|l'l
0 0
1 1 60600060 6060
. . 0.06.... ......
a.“ ) 0365320617116177117117341
n W 1L661r71621226277221227531
PM ( 3222233222222211222221111
2 2 .
t
w \I 8888888888889999999999999
“.60.... fl 1111111111115555555557777
Nahum I\ 3333333333331111111110000
qllll'l"llll'll'l'lljlll'l|l'l'|
s . 1746580297313675420858912
eo 1011001199001011000991122
TN 11111111 111111111 1111

 

*g pile shaft greased behind lug
*f possible faulty test, too fast.

198

spacing and pile displacement at the beginning of failure
(tertiary creep) was illustrated by several figures in
Section 4.7. While this data is limited, a review will
allow several deductions to be made.

Thread spacing is plotted versus observed pile
failure displacement for threaded piles frozen in uniform
sized sand and polycrystalline ice in Figure 5.14. All
threaded piles in polycrystalline ice began to fail after
displacements approximately equal 0.15 to 0.20 mm.
Examination of threaded piles after testing to failure in
ice, showed a loss of some ice between the threads. This
lost ice appeared to have been pushed past the shear
surface, developed at the outermost edges of the threads,
and into the surrounding ice matrix. This failure
displacement approximately 0.15 to 0.20 mm) is not unlike
the displacements (at peak pile capacity in Figure 5.6)
associated with adhesion failure of smooth shaft piles.
Tests of piles with very small threads (approaching the
roughness of "smooth" shaft piles) may approach the
behavior of smooth shaft piles.

Piles with the large size threads in frozen sand began
to fail after a displacement nearly equal to the thread
spacing. The explanation for these observations appear to
be related to grain size (0.42 to 0.59 mm ) of the frozen
sand. These grains would fit between 1.06 mm and 1.58 mm

thread spacings. Examination of these piles after failure

199

.ouaaaom us ucoaoooadmav cane ou wcuoomu cooks» mo somuquEoo .c~.m shaman

Aalv msuoomm voousfi oaum scum ou seam

Whoa Uan 00.0
— _ h

J]

@

aa~¢.or \AHHW\\\ _I

eonum :«ouo vcom mo owned

 

 

 

asanqow _ III .III III .Ilu III. III I.

.c moo m. on mummy H~<

scam aaan.c-~e.o cu noses nu

 

ooH ocuaaouahuohaom :« omega now

«any

l.ninv

IIAv.u

r.AvAN

 

(m) aznureg 39 nuemaostdsrq 311a

200
in sand often showed only loose sand grains or empty spaces

between threads. Before or during failure, ice and sand
(now absent from these threads) must have been pushed out
into the surrounding frozen sand. A displacement equal to
the thread spacing may have been necessary to push all the
interfering grains away from the cylindrical failure
surface outside the outer thread edges before failure could
begin. Some sand grains, although loosened, could have
remained tightly interlocked with the threads and produced
additional shear resistance.

The 0.42 to 0.59 mm sized sand grains could not fit
between the smallest threads and did not develop a good
interlock with this type of pile. Examination of these
piles after testing to failure showed some striations along
the shear surface where sand grains or pile roughness
features had passed, but many areas showed only a smooth
shear surface similar to the failure surface found after
testing polycrystalline ice samples. Without a good
interlock between the smallest threads and sand grains, the
behavior of ice appears to have dominated.

Weaver and Morgenstern (1981a) report that an aluminum
plate of given roughness developed sufficient load transfer
to induce measureable shear deformation during shear tests
of a frozen silt (sample with 30% of grain size less than
the average plate roughness). This same plate showed an

initial displacement at a frozen sand interface (average

201

grain size 200 times larger than the average plate
roughness) but gradually developed sufficient load transfer
to induce shear deformation within the sample. The frozen
silt appeared to interlock well with the plate roughness
compared to little interlock with the larger sand

particles.

5.7. Comparison to Uniaxial Compression Tests

The adjusted n values in Table 5.1 can be compared to
n values determined by Bragg and Andersland (1980) for
unconfined compression tests of frozen sand with a solid
volume faction of approximately 0.64. Bragg and Andersland
(1980) estimated n values of 3.3 at -2 deg C and 8.7 at -6
deg C for strain rates less than lo-fi/sec. Interpolation
between these temperatures yields n = 4.6 at -3 deg C. At
strain rates greater than lo-fi/sec, very large n values
would be obtained because of strength versus strain rate
independence. The similarity between these n values and
the n values observed for lugged piles is not surprising
since the mechanical interaction developed during
compression tests has similarities to the mechanical
interaction with frozen soil ahead of a pile lug.

Bragg and Andersland's (1980) illustration of a
ductile to brittle transition at strain rates greater than
10-‘/sec. may be related to the soil deformation process

associated with a pile undergoing cyclic loading. It

202

appears that very small frozen soil zones, which are
stressed rapidly during cyclic loading, would fail at the
relatively small failure strain (less than 1%) associated
with brittle fracture instead of the larger (greater than
4%) failure strain associated with ductile type creep
deformation. Adjacent areas of frozen soil would then
_carry additional load and the overall pile displacement
could increase.

Parameswaren (1979) performed a series of unconfined
compression tests on frozen sand with a moisture content of
approximately 20%. At strain rates below 3*10-3/sec, n
values of 5.2 at -2 deg C and 8.2 at -6 deg C can be
estimated from his data. An interpolation of these values
yields an n value of 6.0 (at -3 deg C). At faster strain
rates, these relatively high n values represent
independence between strength and strain rate.

Parameswaren (1985) also superimposed cyclic loads of
16.3% on larger static loads during creep compression tests
of the same frozen sand (Figure 2.11). Table 5.4 includes
this data and uses it with a variation of equation 3.4 in
the form

(___—)n = .62 / .51 (5.4)

where a is initial stress, A0 is the amplitude of the
cyclic load, 22 is the strain rate after load increase,

and E, is the strain rate before the load increase. The

2053

owoum Auoluum an on as: sued aeuuucu .ouueuasom uoz aw onoouooa ousm cuouum «

 

 

 

 

m.a - . so.n - ma.~ o~\ m.o+ ace.~
e.HH - em.e - he.“ oax n.m+ acs.~
s - « «.NH - a~.e aux n.e+ aoe.~
- H.6H - No.5 no.~ oax n.o+ aoo.~
- a.- - so.o he.“ 63\ n.o+ aoo.~
- n.o - n.63 au.e aux m.e+ mom.~
Aoo.\.:o~c .uoa\.ucsc Aoau\.uosc A": \ sc Amaze
.«u «u a»

a u a 6 30a amouum museum—nu Aouoouoan b0. 6
-n -u eeuaanv< uuouum Houu< ououonv sued oasouocn umouum
.«w «m ease-neaaea cued cashew cannon Heuuusn enouum aeauusn

.eu-eh caduceuaaoo deansusa
oaaoho Anaodv e.:eue>ueneuem you ausoaueanv< ensueueaaoa .<.m oases

204

large displacement rate increases observed in Figure 2.11
result in much larger n values as compared to values of 5.6
to 6.1 interpreted from Parameswaren’s (1979) data
involving only static loading and temperatures from -2.5 to
-3.2 deg C.

During cyclic loading of Parameswaren’s (1985)
uniaxial compression samples, temperature increases did not
quickly dissipate and the entire sample temperature
increased as shown in Figure 2.11. The strain rate increase
related to these temperature increases of -3.2 to -2.5 deg
C can be estimated by first using equation 5.2 with w =
0.64 to determine a relationship between creep loads which
would produce equal strain rates at different temperatures.
Thus

1.12 (5.5)

Pc(2.5) = Pc(3.2)
This relationship indicates that the load at -3.2 deg C
must be 12% larger than the load at -2.5 deg C to produce
equal strain rates. Taken one step further, a temperature
increase from -3.2 to -2.5 deg C for a sample under a
constant load would increase its strain rate to the rate
resulting from a 12% load increase on a sample held at a
constant temperature of -3.2 deg C. Equation 3.4 with n =
6.1 (interpolated for 3.2 deg C from Parameswaren’s 1979

paper) can be rearranged such that

205

 

) - = 22/ g = 2.0 (5.6)

This means that the strain rate should approximately double
for a temperature increase of -3.2 to -2.5 deg C
temperature increase. This adjustment is listed in Table
5.4 and the resulting n parameters are similar to those
values found by static loading (Parameswaren, 1979) at
constant temperatures. During static loading, measured
temperature increases near a model pile on this project
were found to be very small and quickly dissipated into the
surrounding sample. Conditions for rapid heat dissipation
are also likely to be observed around a full scale pile in

frozen soil.

5.8 Comparison to Ladanyi’s (1982) Cone Penetration Tests

From penetration tests conducted in varved silt at
-0.3 deg C, Ladanyi (1982) estimated n values of 4 to 5 at
penetration rates less than 18 mm/hr. Larger n values, 10
to 20, were estimated at penetration rates greater than 18
mm/hr. This behavior was associated with uniaxial
compression tests (large n values) at strain rates above
about 10-5/sec.

Ladanyi (1976) has also compared a frozen soil element
just ahead of a penetration rod with an element of soil
within a compression test sample (at failure). He

explained that the soil element, ahead of the advancing

206
penetration rod would change from the elastic to the
plastic state in the same manner as the element in the
compression test sample tested to failure. Ladanyi (1982)
later proposed that it was more logical to compare the
frozen soil behavior ahead of the penetration tip to an
expanding hemispherical cavity with a diameter equal to the
cone’s diameter. This representation is shown in Figure
5.3a. Ladanyi stated that the creep equation parameters
for ice poor or dense granular frozen soils should be
obtained by static, stage-loaded cone penetration tests.
These tests are not unlike the step loaded pile tests
completed for this project.

After testing a series of model tapered piles, Ladanyi
and Guichaoua (1985) proposed that an expanding cylindrical
cavity theory, using a primary creep law (Ladanyi, 1972),
was appropriate for tapered pile displacement in a frozen
sand. Limited data did not fully support this theory.
Movement of frozen sand around this study’s model pile lug
(which leaves a void behind the lug) is similar to

cylindrical expansion.

5.9. Large n Values from Cyclic Loads on Model Piles
Parameswaren (1982) described how superimposed cyclic

loads of 11.5% of the initial static load caused pile

displacement rates to double. Figure 5.15 is from his

paper shows the region of pile displacement (above 0.1 mm)

207

.Amaau couuauoaouomv aaou cououu cu caum ououocoo now o>hao oauu-ucoaoooaam«a .n~.m shaman

 

 

 

.. .3:—
cc. cc. coo eo~ o
a a q c
at: a: .6

h e\\7 6
‘0‘ II
I. g «0.0 «H
N
s
‘
1
V.
«a
a
...: .2 .e \\ m
u: S .. 9:33 \\ ....
T 0398 no 50:26on . \ 1 3.9 I
\ \Ie\\\ [L U

\ mood ouuuum

.3 .....zuu‘. full /«° and” u. 9:33

“V\\h‘ \\. cuaoao venomauuoasm

.dJHrIIIILlII||lllllllllllllllllll|u mo mooauom aouoaoo

. .. - a-.o

 

208

where a cyclic load of $1.54% caused the displacement rate
to approximately double, as indicated by the given
displacement rates.

For a number of reasons, caution should be used when
interpreting these results. Based on tests completed
during this study, it is known that the displacement rate
determined from a short duration test (section 2 in Figure
5.15) is relatively fast if compared to the displacement
rate found at the end of a longer duration test (section 1
in Figure 5.15). It has also been shown that when a pile’s
load is reduced (sections 3, 5, and 7 in figure 5.15), a
pile’s displacement rate will be relatively slow for a
period of time before regaining a displacement rate typical
for this reduced pile load. From the information given, it
is not clear if pile failure has started (although this
seems unlikely at a displacement of less than 0.15 mm).

The displacement rate increases in sections 2, 4, and 6 of
Figure 5.15 would have been put into perspective if a small
static load had been added (perhaps at section 4 in place
of the cyclic loading) to produce a displacement rate
similar to the rates produced by the cyclic loads.

To further examine this pile test, a variation of
equation 3.4 can be used as:

P + AP n . .
(‘—;—) = 52 / 51 (507)

where P is the initial load and AP is the amplitude of the

209
cyclic load. A i1.5% cyclic load that doubles the

displacement rate can be shown as:

(1.015 )n = 2.0 (5.8)
which yields n = 46, a value considerably higher than the
value of n = 8 given by Parameswaren (1982) for similar
piles under static load increases.

Cyclic loads applied to various piles during this
project gave n values equal to or less than the values
found by static loading (see Table 4.3). Parameswaren's
(1985) cyclic loading of uniaxial compression samples also
appear to show, after being adjusted for temperature, n
values approximately equal to static load n values.
Assuming that heat generated at the pile/soil interface is
quickly dissipated (as noted in this study), what could:
account for the large difference in n values between cyclic
loading and dynamic loading?

This difference may be partly accounted for by
considering basic harmonic loading theory. From Clough and
Penzien (1975), a ratio of the maximum base force (force at
the pile/frozen soil interface) to the applied force
amplitude (force measured above the pile and frozen soil by
a load cell) is called the transmissibility (TR) and is

represented as

TR - M [1 + (253)210'5 (5.9)

where M is the dynamic magnification factor

210

M - [ (14?)? + (253)? 1‘” (5.10)

with 8 defined as the ratio of applied frequency (5) to the
natural frequency (w), thus
Z?
,3 ... -— (5.11)
w
The damping ratio (5) is written as :
c w
e = -——-—— (5.12)
2 k
where c (system damping coefficient), k (system
spring constant), and w (system’s natural frequency of
vibration) are the basic damped system parameters.

The TR values for the lugged and threaded piles used
on this project are all close to 1 because the applied
frequencies were all well below the natural frequencies of
these pile/frozen soil systems. For this project, certain
resonant frequencies developed due to parts of the test system
which increased pile displacement rates. At 1 Hz, the
static weights were induced to rotate clockwise and then
counterclockwise which produced larger pile displacement
rate increases. At 6.2 Hz, a long metal rod supporting the
static weights would begin to vibrate and also produced
displacement rate increases. These results were not
included in the figures for this study because they

represented a characteristic of the test system and not the

behavior of model piles embedded in frozen soil

211

Although no information was given regarding damping of
the pile/soil system, a conservative damping ratio of E =
0.16 (cyclic triaxial tests at 0.3 Hz on Ottawa sand at -
2.5 deg C, Figure 2.10b) might be assumed. A second
estimate of g = 0.03 (cyclic triaxial tests interpolated to
10 Hz for Ottawa sand at -4 deg C, Figure 2.10a) will also
be considered. Parameswaren (1982) found that the
amplitude of displacement between the pile and surrounding
soil shown in Figure 5.15 was a maximum at a frequency of
8.72 Hz. Clough and Penzien (1975) explain that this peak

response frequency would correspond to a value of B as

flpeak = [ 1 - 2 s? 10.5 (5.13)
Using this equation, natural frequencies at 8.95 and 8.73
may be determined for g values of 0.16 and 0.03
respectively.

If values for 5 = 0.16 and w = 8.95 Hz are assumed, w
= 10 Hz can be used in Equation 5.10 to determine a TR =
2.4. A cyclic load of i1-5% of static load would become
i1.5*(2.4) at the pile/soil interface. Using load
increases of :3.6% in equation 5.7 yields an n value of 20.
Assuming 6 = 0.03, w = 8.73 Hz, and w = 10 Hz would produce
an n values of 15. These n values for cyclic loads are
similar to the n values found from static loading only.

Adjustments for temperature increases from cyclic loading

212

could decrease these n values to magnitudes associated with

static loading.

5.10. Applications to Field Piles
Situations involving field piles will now be
considered in light of the model pile test results

contained in earlier parts of this dissertation.

5.10.1. Analysis of a Pile Failure

Lusher, et al. (1983) described a series of tests
using 450 mm diameter field piles with 18 mm corrugations
spaced at 300 mm intervals along the pile. For three soil
types, these pipe piles were placed into 600 mm holes with
a sand slurry. Once the sand slurry was frozen and the_
test temperatures had stabilized to -0.3 deg C (a coolant
circulation system within the pile helped control
temperatures), a series of incremental loads were added to
each pile with each load maintained for a period of 3 days.

During test loading, pile displacements were measured
at the top of the pile with reference to some fixed point
away from the pile. As the pile settled, soil around the
pile was moved or displaced. For those piles tested in
clays and the ice rich silts at this relatively warm
temperature, this deformation could occur at the pile/soil
interface or entirely outside of the relatively stiff

cylinder of frozen sand slurry. Analyzing their behavior

213

would require additional information.

Although creep parameters were not reported, the dense
alluvial sands and gravels at Shaw Creek Flats in south-
central Alaska would normally display a creep behavior
similar to that of the frozen sand slurry. The frozen
gravels should be stiffer than the sand slurry. Assuming
that the surrounding natural soils are as stiff or stiffer
than the frozen sand slurry, then pile movement and pile
failure can be largely associated with deformation within
the sand slurry.

Figure 5.16 summarizes displacement data for 3 pile
tests conducted at Shaw Creek. The displacement-time curve
for Pile #1 is shown at the top of this figure. All three
curves are shown on the bottom part of this figure. To
analyze the behavior of these piles, consider first the
behavior of a combination of two model pile types: a small
lugged pile and a threaded pile. A physical model of these
combined characteristics might be analogous to a perforated
lug threaded onto a rod which would then be frozen into a
sand sample. The threaded portion of this composite model
pile would quickly develop capacity but would start to fail
after a small displacement (approximately 1 mm for a 1.06
mm peak to peak thread for the 9.52 mm diameter rods used
in this study). As failure continued along the pile/sand
interface, part of the pile capacity (held by these

threads) is transferred to the lug.

 

  
     

mud: #3

.AanH .AI us nosousgv ouua Josue saga

-ouasnou uuou oaum .oa.n ouswuh

214

£13533 .... 9.: .. 3:
.3.-II

3.8 nausea
a. .3.. . ......

....e 2. .o .3.—.05 :06...» ea. .0 8.3-:3 3::-
‘euo.au.eu .5 to. scaveeea too. see: Selina—ea; 13.63 cm:
.39.

no.
:30: . HI;

2 . Else—Ala
a"

    
   

53:91:33 .2, a3:

        

2...... 3... .3. v3.5... 0

'0'0‘

'- ' Imauana

215 .

At 1 mm displacement, the lugged model piles frozen in
sand were shown to have decreasing displacement rates. A
transfer of pile load to the lug of the composite pile, due
to failure at the pile/sand interface, would probably cause
the displacement rate to temporarily increase while the
frozen soil is fully compacted ahead of the lug. The
displacement rate of this composite pile would then
gradually decrease as displacement of frozen sand around
the lug became established. Some friction and mechanical
interaction along the pile/sand interface would probably
remain and add a small amount to the overall pile capacity.
The displacement rate would become nearly steady until the
lug approached the void left by the preceeding lug. Rapid
pile displacement would then follow.

If the composite pile theory is applied to the piles
tested at Shaw Creek Flats, two types of behavior can be
identified. Piles #1 and #3 appear to be in the stage
before adhesion and mechanical interaction at the pile
surface ("threads" of a threaded rod) begin to fail. In
Pile #2 failure at the pile’s surface may have occured at
approximately 7.5 mm displacement and caused a load
transfer to the pile’s projections (lugs). The pile
displacement rate increased after 7.5 mm displacement but
the pile did not totally fail because the pile's
projections continued to carry the pile load. If this pile

had more closely spaced or larger projections, a slower

216

displacement rate would have been observed past the 7.5 mm

displacement.

5.10.2. Comments on the Proposed ASTM Pile Standard

A proposed American Society for Testing and Materials
(ASTM) "Standard Method of Testing Individual Piles in
Permafrost Under Static Axial Compressive Load"
(Neukirchner, 1988), was discussed earlier in Section 2.4.
This method proposes that the test include two piles with
two loads on each pile. The first load would be near 100%
of the pile design load and should displace a relatively
small amount until a steady creep rate is attained or until
3 days have passed. The second load, 200% of the design
load, on each pile should cause failure with displacements
reaching the point of accelerating rates at less than 25
mm. This process may take about 6-12 hours for the smaller
pile load and about 3-5 days for the larger load. The two
failure loads are used as a set of information, and the two
creep loads are used together as a set in the proposed ASTM
method.

Based on tests performed during this study, the
approach proposed by this ASTM standard appears to be the
right one. Loads causing failure are compared as a group
while loads applied for shorter displacements are compared
as another separate group. Directly comparing displacement

rates determined at failure with displacement rates

217

determined after short displacements (as in short duration
loads during this study) can give misleading information.
Neukirchner (1988) explains (correctly) that this method
should provide more useful pile loading information in a
relatively economical and expedient manner, but cautions
that tests from additional piles may be necessary.

Establishing a load-displacement relationship based on
two data points can be risky if the two piles tested are
not very similar. The "similar" piles in Figure 5.16 show
significant variationsu Displacement rate variations
between model piles and even during a single model pile
test was frequently encountered during this project. More
data could be obtained from one pile if short duration type
tests were applied at the end of a pile test (defined as
when the pile displacement rate begins to increase).

It appears that a series of incremental-load type
short-duration tests could be performed at the end of the
standard pile test when the pile displacement rate is
slowly increasing. This series of small loads (3 to 5%, 4
to 10%, and 9 to 15% of the initial load) could each be
applied to the pile for short periods (10 to 30 minutes)
before the initial pile load is restored. If the
displacement rate before this series of loads were similar
to the rate (including a short stabilization period) after
the series, then a comparison between the three load

increases could be considered. The relationship between

218
the three load increases and the n parameter determined

between them would provide test data useful as an
indication of agreement between the two piles tested. The
expense of gaining this information would be minimal since
the test piles and equipment are already in place.
5.10.3. Additional Pile Capacity from Driven Piles

Manikian (1983) discussed driving pipe piles into
pilot holes filled with warm water (to decrease driving
resistance). Driving piles with lugs would be difficult
and could leave large voids behind each lug that reduce
pile capacities. The pile corrugations discussed by
Lusher, et al. (1983) could be added after driving but
would require additional field time which is very
expensive. Black and Thomas (1979) indicated that the 8
pile tests (distributed among three sites in Alaska)
reported by Lusher, Black and McPhail (1983) cost over
$5,000,000 in 1975. Pile modifications made during pile
manufacture would be less expensive.

Before modifications to field piles can be considered,
a method to install them is needed. Manikian (1983)
decribed in detail how warm water (of a given temperature)
had been added to pilot holes (of given diameter) and
allowed to warm the surrounding frozen soil (for a given
duration) to ease pile driving resistances. The relatively
small volume of soil warmed by the water and by the action

of pile driving was able to freeze again quickly. This

219
method could be taken one step further to ease the
resistance of piles with "rougher" surfaces by including
some combination of warmer water, larger pilot holes, or
longer soil warming periods that would also keep freeze-
back time to a minimum.

Roughness could be added to a pile surface so that the
outside diameter was not increased but included grooves or
indentations that, filled with the warmed soil during pile
driving, would freeze solidly later. These roughness
features should be equal to or larger than the grain size
of the natural surrounding soil to produce a good interlock
between soil and pile. Frozen soil within grooves could be
considered similar to frozen soil (limited to 25 mm and
less) ahead of model pile lugs. Pile capacity increases
could be based on the length, depth, and shape of the pile

indentations.

5.10.4. Needed Research on Pile Behavior
Several topics related to pile behavior under static
and cyclic loads, described below, need further study.
a) Static and cyclic load effects on the behavior of
frozen soil in shear at the pile/soil interface.
b) The long-term pile/soil interaction mechanism
dependance on surface geometry needs to be described
in greater detail.

c) The incremental-load type short-duration test sequence

d)

220
would appear to have application to field pile load
tests. Test procedures for field use need to be
developed.
The pile displacement rate increase shown by
Parameswaren (1982, 1984) should be considered during
pile foundation design in permafrost. A field check

of a pile’s natural frequency could be developed.

SUMMARY AND CONCLUSIONS

6.1. Summary

Small cyclic loads superimposed on a static pile load
will, in some frozen ground situations, increase pile
settlement rates and significantly lower their design
capacity. This situation may arise for piles in
perennially frozen ground which are used to support
vibrating machinery (turbines, power generators, or
compressors) or traveling loads (cranes or fork lift
trucks). Small cyclic loads, 3 to 5 percent of the long-
term sustained load, were reported (Parameswaran, 1982,
1984) to accelerate settlement (displacement) rates of
model piles embedded in frozen ground. The basic
mechanisms by which small cyclic loads increase pile
settlement rates and a suitable theory for estimation of
this increase were objectives of this research.

This study used model steel piles in a laboratory
setting to carefully change and control variables in order
to increase our understanding of the processes responsible
for friction pile settlement. Pile settlement rates due to
static and incremental static loading were analyzed and
compared to pile settlement rate increases due to
superimposed dynamic loads. Variables controlled during
these tests included cyclic load amplitude, frequency of

load application, static load magnitude, and temperature.

221

222

Sand density and ice volume fractions were controlled
during sample preparation. Model pile settlement in frozen
ground is typically predicted on the basis of a creep
Wequation relating pile load or pile-surface shear stress to
pile displacement rates. Creep parameters were used to
characterize soil type, soil/ice structure, temperature,
and loading conditions. Experimental results showed that
cyclic loads significantly increased model pile
displacement rates over those observed for the sustained
load, and that creep theory with the appropriate parameters
will predict the effects of small dynamic loads on
settlement rates.

Load capacity of friction piles in frozen ground is
known to increase significantly when the pile surface is
roughened by the addition of protrusions or lugs. A series
of model steel piles with controlled surface geometries
(lug size, shape, and spacing) were frozen in sand and
loaded to provide additional information on the
displacement mechanisms operating at the pile/frozen soil
interface. The onset of increasing pile displacement rates
(tertiary creep) was shown to be a function of lug height
and spacing. A number of conclusions are presented in the

next section.

6.2. Conclusions

The conclusions resulting fron this research project

223
will be presented under three headings: creep settlement
parameters, dynamic load effects on settlement rates, and

pile surface roughness and load capacity.

6.2.1. Creep Settlement Parameters

Displacement rates for most lugged piles decreased
slowly with time throughout a given test and seldom reached
a constant displacement rate. For these conditions, direct
comparison of displacement rates or creep parameters
between piles was difficult. To improve comparisons
between tests, log(time)-log(displacement) plots were
prepared using data obtained after a given displacement.
This technique gave creep parameters which could be readily
compared with like plots of similar pile test data.

Incremental loading for relatively short time periods
eliminated some of the long-term pile displacement
variability caused by small temperature changes, frozen
soil heterogeneity, and decreasing displacement rates.
Although these short-term tests may not accurately predict
long term creep parameters, they did permit comparison of
load variables such as dynamic-load amplitude and frequency
with each other and with small static-load increases.
Increasing a pile load by small steps, with each load held
constant for a short duration, also decreased the effect of
temperature, materials, and displacement variability, while

it provided data that better predicted creep parameter (n)

224

values for long-term pile behavior.

6.2.2. Dynamic Load Effect on Settlement Rates

The technique developed to measure displacement rate
changes for incremental static pile loads was used to
measure displacement rate changes caused by dynamic loads
superimposed on statically loaded model piles. Cyclic
loads significantly increased pile displacement rates over
those observed for a sustained load. Model pile
displacement rates in frozen sand were observed to be
essentially independent of frequency for small dynamic
loads superimposed on a large static load in the range of
0.1 Hz to 10 Hz. The magnitude of rate increase was shown
to be dependent on amplitude of the cyclic load.
Frequencies that induced resonant responses in parts of the
loading system were excluded from reported data.

Creep theory appears to be suitable for prediction of
the change in pile displacement rates due to a small cyclic
load. A superimposed cyclic load (amplitude X) would
produce a short duration displacement increase roughly
equal to the increase produced by a static load increase
(magnitude of 0.6 times X). Compared to small static load
increases, laboratory measurements indicated a reduction of
the dynamic-load n to about 5/7’s of the static-load n
value when small dynamic loads were superimposed on larger

static loads. This reduction in n appears to be

225

appropriate for evaluating the increase in settlement rates
when nonresonant cyclic loading is superimposed on a static
load. Pile capacity must then be adjusted for a given

service life.

6.2.3. Pile Surface Roughness and Load Capacity

Load capacity of all model piles in frozen sand was
dependent on a load transfer component acting at the
pile/frozen soil interface. Mechanical interaction between
frozen soil and model pile surface protrusions (lugs)
provided the larger and more reliable long term load
capacity.

A minimum lug spacing (greater than 10 mm for a 3.18
mm lug height) was needed to develop fully the mechanical
interaction between frozen sand and the pile lug.
Additional data are needed to completely define the
relationship between lug spacing and lug size along with
the influence of particle size, soil type, temperature,
pile size, and loading conditions.

At lug spacings that fully developed mechanical
interaction (greater than the minimum described above), the
pile displacement-time behavior was relatively unaffected
by lug space variations until it reached a critical
displacement (start of pile failure). The relationship
between critical failure displacement and lug spacing

(displacement limit) was shown for 2 lug heights. It was

226

observed that shear stress on the remaining pile/soil
contact area appeared to asymptotically approach the long
term shear strength of the frozen soil. The mechanisms
that initiated pile failure for these short term pile tests
should be similar to those which operate during long term
periods.

Limited information from tests with threaded-rod model
piles indicated that thread heights and spacings larger
than the sand particles fully developed pile/soil
mechanical interaction. In comparison, tests with thread
spacings less than the sand grain size did not fully
develop mechanical interaction and showed a displacement
behavior more like a threaded pile in polycrystalline ice.
In polycrystalline ice, failure occurred at similar
displacements for all threaded piles.

Based on the above observations, piles with surface
irregularities much smaller than the adjacent sand grains
will develop little pile/soil mechanical interaction during
rupture of the adhesive bond component and during
subsequent pile displacement. Frictional forces appeared
to provide the major portion of any remaining pile capacity
following a gradual loss of the adhesion bond along a

relatively smooth pile.

APPENDIX A

FIGURES FOR DETERMINATION 01" TABLE 4.1 CREEP PARAMETERS

227

.uuasau usoaooeamuuv Aeusuo>ev an o.on use sesame: mad as an.n and:
c not a- no use. conouu an sedan «ecu. novol you ee>uso usoaeoeuneuvuosua .~.< enema“.

Ante u .oswe

 

 

 

 

 

 

 

 

 

m~ o“ m o
....32- -1. a
main so cowfiqa< omoouu soowfiwm « .o om:
as ~.os .zx sN.~ so“ a scam assm o - No 6
as N.o~ «2x -.~ mod \
as ~.- .2: u~.~ he .
as ~.m~ «z: ~N.~ ma
ss o.o~ zx.eo.~ -~
as o.as z; c_.m obs as .
as o.¢~ z: -.m cog \
x
z .ussus a .eooq .oz 6
acosuooadafin owed boos \\\\\\ x. 1
. \
\ \s
\ .
x
0 \
3 1m
as he mob o
ousaaom mafia
. no vuoum 1
~11 o~_ ..H

 

 

 

(mm) 9 '2uamaoatd810 atrd

228

.ausaom «sauna». uoaa enhu-o II c any ..uuflaa unease-uduuv as o.a« .eunuuon use

 

II o~.n and: c «on a- as use. neuouu :« sedan demo! uou eo>u=o nauseoeamnuo-oiqa .N.< ouowum
Ages a .maaa
cod o~ a ~.c.
4 J N ~.o
\
\
usom one as o>uso mo
:ofluoawxou d< ocwq ugmwouum mu
\\.. L
«sesame mass to rumba nu . a
. \

 

Nu a

sacs

cm‘ boos

 

 

Cu

 

(mm) 3 ‘1uamaoetdsrq 911d

229

OOH

cg

.eusuoa uswuueue noun 0950-0 :3 a

Aggy a .oaua

.a~.< ensues

 

 

mmaa

Luca

 

neuu umuH

 

 

cg

(mm) 3 ‘1uamaoetdstq 311d

230

 

.eusuoa «sauna». noun emha-o as n .o~.< ousmum
66s 63 Ages 3 .oaae 1.:
~.o
:
. .d
1 To
.\ m.
nu
.L.
S
.a
T.
9
1 a a
m
a
u
14
a)
m,
m
«ma ‘ unma meg Home

 

 

 

ed

231

as

.353: uses—00:33.. A332,: In on was 3:30: 95 as 36 £33

 

   

 

9

cu

cm

0 mom a. as vs.- cououu Bu sedan «sou. «econ non nebuao unease-uneuv-oawa .n.< enamum
Ange a .mafia
am am ov om ow ad a
. q . . . . . . i q . . .
com: ® beam ssmm.ou~e.o
em

Pt

. mmmu

I

C.

x. A

"N

.\s
. E o
c.. 79

€.. h..9

.1. % Z X 23. :8 4

7 u... m. N Zu mm.“ N-

X. N 20— -..~

N z: ~e.m
2x em.m .
zo— mm.m
z: Nw.m
2x 0 .m
z: o~.m MN~

a .eooq «Hum .oz same

 

 

 

 

am

(am) 9 ‘3uamaoetd51q and

232

.352— ucuuuou- yoga 09596 as 0 new duds: usoaoosaaaup as 2. .3530: as.”
as 36 .3qu new a. 0. was» 630.3 s.— 339 .702. now sagas econ—3393725.“. .e.< 933..—

Anne a .6329

 

25 OH H 70
- . 2.6
.d
T?
T...
3
0
T?
a
uso: use an 695:0 now u.
song—33.394 on“; unmfiouum D o
m
m
x \\ A d. ‘14
9
m
m
26— eoé @.
«a: boon. x caé
w mm: buoy A:

 

 

 

233

 

 

.eusuom «savanna uoaa ombuoo as « .b¢.< ouamum
Anne L .msfia
oo~ o~ ~.o
1
zx eo.~
w «Nd‘ umo 2x o~.m
\. w ma- amok

 

 

 

~.o

(mm) 3 ‘1uamaoetdsrq 3118

2.

234

oo~

.nusuoa «sauna»: nods 0939-0 as n

2:: a as:

.oq.< shaman

 

 

zxoc.~ @
«NH& game
zxo~.m w

\ 8: :3

‘1‘

.\

\O

\\

(mm) 3 ‘1uamaostdsrq 311d

 

Cu

 

235

.uusuom uswuuoua uoaa oah8-0 as n

Aunt 2.6afie
ooH as

.oq.< enemas

~.o

 

zx mzvtm o
«Nu‘ boos

 

 

 

 

ca

~.o

(mm) 3 ‘1uamaoetdsrq 311d

236

.uuusua use-ooeamuuv Annouo>ev
and: u not n- no vsee sonouu an sedan Acou-

ua v.n« as. .uauuos was as an.”

 

 

 

 

(mm) 9 'auamaoatdsto atra

devo- wou nebuso nausea-amuuo-os«a .n.< ousmum.
.uac a .6339
ON m— cu m o
d d d a dl d 1 1 u 1 q q 4 q q d 4 d o
>~c
was cusses .msue :6 .u.n- e scam seam.o-~q.o .
vowadd< ommouo cooflHHma _
es ~.m~.«zx -.~ we
as m.nH «2m 5“.“ bbh _
ss m.s~ 23 -.N Neg
as m.- 23 -.N «ea m
as m.- 2; NR._ mew - 1 ,
as «.mm 23 sn.~ say
as e.n~ z: -.N be“ .
ss_¢.m~ zx be.~ mad : .
z .uusfia a .8863 .62 \\\\. s .
usosoooaamwa mafia. uses 4 o"
1
4
0 L
. m- Ca Ca oo~ we .
1m—
1
/ .
ousuwom mafia. .
no .
cos to u um mss

 

 

 

ON

237

.uuusaa usosooeaaauv as c.n~ use «usuon usuuusun eahu-o as a new .uunwuos use
as un.~ Saw: u not a. an vs.- sououu an «Juan moves you no>uao usosooeamuav-os«s .o.< ousmum

Pi

H.o
~.o

oo~ Cu Anusonv u .oaaa

[1 J

 

usoz use no o>u=o non
sowuosaxouddc as“: u£~qouum nu

ousauom mafia uo buoum nv

x K

(mm) 3 ‘Juamaostdstq atld

had

a Home
Mgas
cog anus amok

 

 

om

 

238

God

.euusua usosooeaAuuv as ¢.ou use guano“ usuuuouu onhu-u as H .bo.< ousmuw

Asst 3.6519

~.o

 

 

saga umoH

cofi‘

umOH

nag
§ anus

N.o

 

ON

 

(um) 9 ‘3uamaoetdsrq BIId

 

239

oofi

.euqsaa usosooeadauv as ¢.n« use nusuoa usuuueue cahu-o as n

Cu

2.: H.

.osfia

.ow.< enamam

 

 

“fiaa game

oo~§ anus

Nada Home

 

 

 

ON

3‘1uamaoetdsrq 311d

(mm)

240

.euulaa usosooeaanuv as ¢.nu as. ousaom usuuueun oahuuo as o .vo.< ousmum

Ange u.oase

 

co“ 0" ~.o
. \ \
1
mfida umoa
boa as
a :8 m.

 

 

 

N.o

ON

(mm) 3 ‘1uamaoetdsrq 311d

241

099

.nuuaad usoaeoea9uuv as o.h~ use suauo9 «cannons o9hu-u as n .wo.< ousuuu

Ausc u.oaaa
S s so

 

 

\\

 

 

 

~.o

09

(mm) 3 ‘1uamaoerdsrq BIId

242

oo~

.0uuaua ua0a000a90u0 as o.h~ 0:0 ousuo9 usuuuaue 0990-0 as H

Ages b.02ue

.ue.< museum

9.0

 

 

 

 

 

~.o

(mm) 9 “ausmaoetdsyq eIIJ

243

009

.00uaua 0:0s000A9000 as 0.9a 090 006009 0:000000 0990-0 as o .00.< 000009

o9

Ange 0.0200
0

u.

 

 

J

 

o
9

 

 

o9

.O

3 ‘1uamaoetdstq 3118

(mm)

244

oo9

.009a99 000s00099090 as 0.59 0:0 000909 0:900000 0990-0 as 0

Anne 0.0a00
S 9

..S .< 3:0:

9.0

 

 

 

9.0

 

 

c9

3‘1uamaoetdsgq BIId

(mm)

245

.009s99 0a0a00099090 900000>0v as on 060 0090909 099 as no.9

 

 

 

£093 0 000 n- 00 0s00 seuouu :9 00999 90000 9000- son 00>usu 0:0s00099090-0a9a .s.< 009099
0:0 a .0090
cm m“ 09 m o
. 114 9 4 . 9 . 4 . J 9 J 11- . 4 O
0 om- w 0000 as9m.o-<.o
N L
cm - m
29 0m. 90
29 m9.9 <9
29 c~.~ Na .
29 c~.N na
29 o9.m oo
3 2
0.0060 0990 .62 same as

 

 

 

 

(mm) 3 ‘Juamaostdsxa 311d

246

.uua9oa u=9uucun u09a oAhuuc II o adv .nu9l99 acosoon9nu9v as mu .uunw9os us9
as on.9 £u93 0 not a. u: can. Banana :9 .099A 90cc: you uo>uao uaoaoou9mu9v-oa9a

u u .we
oo9 O9 9 av 99 9

.o.< ousmum

9.

 

uzom use an nw>usu uou
:o9uaa9xounad 0:99 unw9auum mu

Boa
game

 

¢¢a
game Naa
umoH.\\ mm2
anus ¢wa
game

 

 

 

 

Q9

3 ‘1uamaoetdsgq 9115

(mm)

247

oo9

09

2.: a

.nuaaoa u=9uu¢uu u09a Oahu-o HI 9 .nm.< ounu9h

 

 

92‘
away

«$2
Hume

~$\

away

@a
umwh

¢w§
umma

 

 

9.0

o9

g‘nuamaaetdsxq BIId

(mm)

k

248

009

.uua9oa azuuunun 9095 0228-0 :6 n

Auzv a .osfia

.um.< ou=u9m

9.o

 

 

ea‘
away

3. \

away

¢§
Hams

oma game

0 .

 

o9

 

9.0

g‘nuamaaetdsgq BIId

(mm)

249

.0uc9oa usauu0u0 u09m 0229-0 as n .va.< 0uam9m

2.52 . .oafia

 

cofi o_. _ ~.o
. . \\ _.o
«$4 Hams
.\
D
\\
\\
\\
\
.2
~¢9
«ma
anon game
«m‘
“was
D

 

 

 

09

(mm) 3"3uamaoetds:a aIId

250

 

.0u9a99 0:0300092096 90m0uo>0v as 9.0a 6:0 0u£w90£ «:9 as m~.o
:09: u now a. 00 2:00 s00ouu :9 00999 90000 90cc! now 00>uao usoaoo09209v-0a9a .a.< ouaw9m

902V a .us9a
om mm om m9

u on: @ Ucmm seam.oumc.c

22 99.9 «~90
2: cm.9 9N9‘
29 mm.9 $990
29 55.9 @990

 

2

q

2009 099m .02 name

«~399am «99; so “swam my

09

   
        

9~9a

   

 

 

 

99

m9

ON

9‘3uamaaatdstq GIId

(mm)

251

.00a9on 06900000 0092 0020-0 as c 209 .009s99 0a0a0o090090 as 9.0“ .0030905 059
II 02.0 :09: 0 000 n- 00 0:00 600000 :9 0099& 9000a 000 00>0so 0:0a0009m09v-0a9h

 

.3.< 250:
20:0 0 .oa90 .
oofl ca 0 a o
- . N.o
.9990: mac 9 mw>u990 HON

.xouqa< 0:99 0209000m mu
m
m.
8522 32 0o 0:3 O D a
“a
d
T.
9
3
a
m
a
u
11
9)
m
m

NN9‘
0009 9~90
“may

 

 

 

 

252

099

o9

.000902 0:900000 0092 0220-0 as 9

2.: g

.0298

.ao9.< ou=m0m

9.o

 

 

«N90
0002

 

 

 

N.o

g‘nuamaaatdsrq SIId

(mm)

253

 

009

.000902 0:900000 0092 0229-0 as a

9020

9a
009

.0 mafia

.oo9.< 000000

 

 

 

g‘zuamaaetdstq BIId

(mm)

254

009

.002902 00900000 0092 0229-0 as 0

90:0 .0 0390

.0o9.< 000090

9.0

 

 

~90
009

  
  

0992
0009

 

 

~.0

 

 

0N

3‘3uamaoetdstq SIId

(mm)

255

009

.000902 00900000 0092 0229-0 as «9 .009.< 000092

0040 .u @300

 

 

 

«~90
0009

 

 

 

N.

00

3‘3uamaoatds1q GIId

(mm)

256

.009a99 0:0a00092090 n.0b0v as 02 000 000090: 009 as 09.0 :09) 0 000 n-.00 90090
09000 as 00.0 00 02.0v 0000 000000 09 00992 9000s 000 00>090 0:0a00092090-0s99 .99.< 000092

0 . 05
Q9 9 =0 0 90 m o

 

 

 

 

 

1 1 1 0 9!! q 0|\|-|\IHI-I-I\I\H\\O
09 -
L
L
99 i m
L
0
- Q9
.9
0
0
1m“
20 99.9 mm .
09 20 99.9 90 .
20 99.9 MN
20 99.9 cm .
4
mm.o u =o9uouum aaa9o> 099om .m>< m .0ao9 .oz - cm
0.9- “0 ncam 5290.0-99o.o 0990 game

 

g ‘1uamaaetdstq 3115

(mm)

257

.000902 00900000 0092 0220-0 as 0 20v .009s99 0:0a00092090 as 09 .0050905 009

as 09.0 :09: 0 000 0- 00 0000 000000 09 00992 9000s 000 00>000 0:0a00092090-0a99

0; 0 .0299
009 09 9 0 9

.09.< «00090

9.0

 

1‘ d.

000: 0:0 00 00>0=0 009
.00022< 0099 00090000 . fig

000
0009

:2 . D
0009 000 N0
0009 009

 

      

9.0

 

09

 

g ‘nuamaaatdstq 3115

(mm)

258

009

.000902 00900000 0092 0220-0 as 9 .099.< 000092

9020 .0 0599

 

 

(mm) 3 ‘3uamaaetdsrq GIId

 

 

 

259.

cam

cu

.u0aaoa acquua0n 0009 onha-o as n

00:0 0 .oafia

.ou~.< ousuuh

~.o

 

 

 

 

 

~.o

g ‘3uamaaetdsgq GIId

(mm)

260

.uoH

.a0auom u=u0un0a

.

2:: H.

mafia.

00am omha-o as o

.om0.< «0:000

~.o

 

 

 

~.o

OH

 

 

(umn ‘3‘3uamaaetdsta 311d

261

.000sud 0cosoo¢~nuuv nouauo>¢v Is on can nusmaoz and as on.o
£003 u mac n- 0: van. sououm £0 00009 00.00 doves now nobuao 0soaoou~aauvuosua .n~.< ousmum

.unv 0.oaaa

 

 

mm om m_ o_ r a
Jq“q444l.4.‘44...JJ4.‘..\\\o

.\.1
20 No.0 an Q m

20 No.0 on L

aae.¢~ I

w 20 no.0

can“ 20 mm.0 mm 4
20 eq.0 mm 4 o.

.Oz 1

 

cues ~00; anus «000

1 mu

1 cm

1

  

L

   

mma
mm. o m- @ usam egom.o-~¢.o
a 05000: was sam~.o L mm
00204 unmamomsamso sacs .

l

 

 

 

 

(mm) 3‘3uamaaatdstq SIId

 

262

.000a00 0sosoouaauuv Aouuuobhv as on can 00=uuos wad as on.o #003 u not N- 0-

 

Aas co.o 00 en.ov van. sououu :0 0.009 0000. novel you no>uao usuaoooamn0v-oa0a .e0.< 009000
00:0 0.0300
cm mN om m0 o0 m o
mm.o u
ac00onum 03:00> @000m .w>< .
u out w team ssqm.cnmho.o 4
4
\ 1m

0
00 \ 4
0

1 o0
.4
A
L

i m0
4
L

1 cm
Nu m
.0 m0 0

mm

 

 

 

3‘1uamaoetdstq SIId

(mm)

263

.00:00& ua000000.0000 0050-0 as c any .000a00 0:0!00000000 as 05 .0030005 and

 

  

 

as ah.c A00: u «on «- 00 can. Banana :0 00000 “oval you nopuso usosouaqnaue-oa09 .n0.< 000000
0 00a
80 S 0 5 0 00 0 0.o
0 4 0.3
000: 0:6 00 m0>0=o uou NH
.xouqn< 0:04 0am0000m mg I
l
10
\\
0N0
000a 00a \\\
000a \ «N0
0009 mga 000a

\. o0

 

 

 

3‘3uamaaatdsgq SIId

(mm)

 

 

 

264

OOH

.002009 ua000000 0000 0050-0 as 0 .nm0.< 000000

a

00:0 0 0300

 

 

-a
0008

 

 

 

y‘auamaaetdstq 311d

(mm)

265

oc~

.n0suoa «:000000 0000 onha-o as n

0000 0.0300

 

 

 

0N0
0000

ad‘
0008

.0m0.< ouswum

0.0

 

MN§
0008

 

 

 

 

0.0

3‘3uamaaatdstq 3115

(mm)

266

OO0

.000000 00000000 0600 0008-0 as 0

o0 00:0 0.0e00

.000.< 000000

0.0

 

 

 

 

 

g‘zuamaaetdsrq BIId

(mm)

APPENDIX 3

FILE TEST DAIA

267

Table 3.1. Tabulated Pile Test Data.

-g: 0.074 to 0.84 mm grain size sand
-u: 0.42 to 0.59 mm grain size sand

3
s
i polycrystalline ice

 

Test Fig- Test Sand Solid Ice Embed- Lug Thread

No. ure Tem- Type Volume Den- ment Size Size
No. pera- Frac- sity Ahead/ /Pile
ture tion Behind Embedment
Lug
deg C * % g/cc mm/mm mm mm/mm

 

1 through 17
Notes: Poor quality data due to coolant leaks and an
inconsistent pile-loading system not shown.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

18 A.14 -2 sg 58.3 n/a 76/66.3 0.79 -
Notes: Very small leak found after testing
Mechanism for small displacement jumps unknown.
19 A.14 -2 sg 58.1 n/a 76/62.9 0.79 -
20 8.1 -2 sg 57.8 n/a 76/62.l 0.79 - >
Notes: Large displacement rate increases, cause unknown.
21 A.14 -2 sg 58.8 n/a 76/65.3 0.79 -
22 B 2 -2 sg 58.1 n/a 76/57.4 0.79 -
23 8.3 -3 sg 58.2 n/a 76/58.4 0.79 -
Notes: Lever arm knife edges set on moveable rings for
less static load variation.
24 B 4 -3 sg 59.0 n/a 76/64.5 0.79 -
Notes: Test appeared to be slow. Alignment problems
or slightly higher vol./solids may be a factor.
25 B s -3 sg 58.3 n/a 76/61.1 0.79 -
Notes: Pile allignment estimated at 2 deg from vertical.
26 B 6 -3 sg 58.0 n/a 76/62.l 0.79 -
27 3.7 -3 sg 57.6 0.90 76/59.7 0.79 -
Notes: Coolant bath began to fail after approx. 12mm displ.
28 8.8 -3 sg 58.1 0.92? 76/66.9 0.79 -

 

29 4.11 -3 sg 58.0 0.91 76/59.9 0.79 -

Table 3.1.

White) (A)
UNI" O

34

35

h.)
0‘

37

38

39

4O

41

42

43

44

45

46

268

 

 

 

 

 

 

 

 

 

 

12
failure

 

 

 

 

 

 

 

 

 

 

(cont.)
3.9 -3 sg 58.3 0.91 76/63.7 0.79
Notes: Possible alignment problem during initial displ.
3.10 -3 sg 62.4 0.93? 76/61.3 0.79
3.1l -3 sg 63.4 0.92 76/62.5 0.79
4.13 -3/-2 sg 64.5 0.93 76/62.1 0.79
3.12 -3/-2 sg 63.8 0.92 75/53.7' 0.79
4.15 -3 i 0 0.90 76/76.4 0.79
n/a -3 i O 0.91 76/76.4 0.79
Test Data: minutes 8, 9, 10, 11, 11.32,
micrometers O, 33, 38, 114, 227,
Pile load - 2.22 kN.
n/a -3 i 0 0.90 76/76.4 0.79
Notes: Sudden pile failure at pile load - 2.67 kN.
3.13 -3 i O 0.89 76/76.4 0.79
Notes: Silicon grease applied to pile before sample prep.
3.14 -3 i O 0.90 76/76.4 0.79
Notes: Silicon grease applied to pile before sample prep.
3.15 -3 i O 0.89 76/76.4 0.79
Notes: Sample behind lug broken away during initial displ.
3.16 -3 i 0 0.90 76/76.4 0.79
Notes: Teflon coating applied to pile before sample prep.
Pile was turned to break adhesion before testing.
3.17 -3 i O 0.89 76/76.4 0.79
Notes: Pile was turned to break adhesion before testing.
3.18 -3 i O 0.90 76/76.4 0.79
3.19 -3 i 0 0.90 76/76.4 0.79
3.20 -3 i+sg 32.6 n/a 76/76.4 0.79
Notes: Ice chips mixed with 0.074-0.84 mm sand before
adding chilled water. This sample was not uniform.
3.21 -3 i O 0.89 165.1/- 0.79

Table 3.1 (cont.)

47

48

49

SO

 

 

 

 

 

.....

 

 

000000

269

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

-3 i O 0.89 76/76 4 O -
Pile without lug failed suddenly after 0.55kN added.
-2 i O 0.90 76/76.4 0.79 -
-3 i O 0.89 76/76.4 0.79 -
-3 i 0 0.89 76/76.4 0.79 -
Thermistor located 1.5 mm from pile shaft recorded
approx. temp. increase of 0.06 deg C at time of
rapid pile failure.
-3 i 0 0.89 76/76.4 0.79 -
-2 i O 0.90 76/76.4 0.79 -

-3 su 57.1 0.90 76/63.7 0.79 -
-3 i O 0.89 76/76.4 0.79 -
-3 su ~57 ~O.89 76/64.0 0.79 -
-3 su 57.4 0.90 76/63.7 0.79 -
-3 su 56.1 0.89 76/62.3 0.79 -
-3 su 57 0.91 76/62.l O -

: Pile load of 0.73 kN caused sudden failure.
-3 su 57.1 0.89 76/60.7 0.79 -

: Coolant leak altered frozen sand around pile.
-3 su ~57 t~0.89 76/6l.5 0.79 -

: Tapered lug used.
-3 i O 0.90 - - 1.06/165.1
-3 i O 0.91 - - 1.06/84.9
-3 i 0 0.90 - - 1.06/23.8
-3 i 0 0.90 - - 1.06/80.3
-3 i O 0.90 - - 1.06/81.1
-3 i O 0.90 - - O.35/87.4

270

Table 3.1 (cont.)

 

 

 

 

'27 'Z'ZT'T' '22' 1'37 3.7-"Z? - "" 1.06/87.4
'ZZ' 737'?" '3' :37 13'}? - '7" 0.35/37.-
73' TZE'T “J." 3'7" :23 "" 1.06/25.4
'76' 'E'S'I'T '2? :7" I}? - "7"‘1.o-/33.3

Notes: Coolant leak probably altered last portion of test.

 

 

 

 

 

 

 

 

732"?" '3' '37: '37? - "" 1.0-m.-
Notes: Pile coated with ice then surrounded with sand.
'72" 'ZT'ZE'T 'T '3' 3'3? - 'T" 1.0-,0.-
"73' '37-'73" '22' '27: 7172' - "" 1.05/0.-
Notes: Pile coated with ice then surrounded with sand.
72' 732'"?— "I' "'5' 3'2?" - "I" 1.0/75.-
.75. .8T35- -:;- -:- -;- 3'31. - -:- 1.06/87.6
72" 'E'ZE'T 'T '5' 3'3?" - "" 1.06/79.-
77 '2'???" 'T ‘3' 3'3? - "" 1.05/79.-
73' '3'37'17 'ZZ' 372' 32" 76/3.3 Z'IZ -

Notes: Silicon grease applied to pile before sample prep.
Short duration loads described in Table 4.3.

 

 

 

79 3.38 -3 su n/a n/a - - 1.06/25.8
80 3.39 -2 su 57.1 0.90 - - 1.06/54.0
81 3.40 -3 su 57.8 0.88 76/62.l 0.79 -

Notes: Silicon grease applied to pile before sample prep.
Short duration loads described in Table 4.3.

-- .... m -..". “a:

 

82 3.41 -3 su ’V57 ‘V0.89 76/63.3 1.59 -
Notes: Silicon grease applied to pile before sample prep.
Short duration loads described in Table 4.3.

 

 

83 n/a -3 su 57.3 0.89 152.8/- 0 -
Notes: Silicon grease applied to pile before sample prep.
Pile load - 0.30 kN caused sudden failure.

 

84 4.26 -3 su 60.2 0.91 - - O.35/51.6

Table 3.1 (cont.)

85

86

87

88

89

90

91

92

93

94

95

96

97

98

99

100

101

la
b
N

&
U
H

 

Notes:

 

3.43

Notes:

 

Notes:

 

 

Notes:

 

A.7

Notes:

A.7

Notes:

8.44

n/a

Notes:

 

A.1

Notes:

 

A.5

Notes:

 

A.1

Notes:

A.5

Notes:

 

A.1

Notes:

 

 

 

 

271

 

 

 

 

 

 

 

 

 

 

-3 su 56.5 0.90 - - 1.06/51.6
-3 su 57.8 0.95? - - 1.06/51.2
-3 su 56.4 0.89 76/63.3 1.59 -
Short duration loads described in Table 4.3.

-3 su 59.6 0.90 76.62.l 0.79 -
Short duration loads described in Table 4.3.

-3 su 56.7 0.85 76/63.7 1.59 -
Short duration loads described in Table 4.3.

-3 su 55.9 0.90 - - 1.59/60.3
-3 su 58.3 0.89 - - 1.06/52.4
-3 su 57.8 0.88 76/63.7 1.59 -
Short duration loads described in Table 4.3.

-3 su 57.9 0.90 76/63.7 1.59 -
Short duration loads described in Table 4.3.

Lever arm adjusted at 8.72 and 14.37 mm.

-3 su 56.7 0.85 76/63.7 1.59 -
Short duration loads described in Table 4.3.

Lever arm adjusted at 11.58, 25.12, 34.44 mm.

-3 su 56.8 0.90 12.7/60.5 1.59 -

-3 su 55.3 0.90 12.7/62.2 1.59 -
Sample damaged during load system set-up.

-3 su 57.4 0.87 12.7/61.7 3.18 -
Silicon grease applied behind pile lug.

-3 su 59.1 0.88 15.2/65.5 1.59 -
Silicon grease applied behind pile lug.

-3 su 57.1 0.88 15.2/65.5 3.18 -
Silicon grease applied behind pile lug.

-3 su 56.8 0.87 l7.8/6l.9 1.59 -
Silicon grease applied behind pile lug.

-3 su 57.3 0.87 10.2/60.7 3.18 -
Silicon grease applied behind pile lug.

Table 3.1 (cont.)

102

I >
f u:

 

 

 

 

?’
kn

 

&
...;

 

?’
'n-I

 

8.45

 

.> o
H

 

e»i
H

 

A.1

 

?’
UI

 

b
H

 

 

k
H

 

 

 

 

......

 

 

 

?’
h)

 

272

 

-:;- .22. 522; 028;. l7.8/63.2 I259 22 .....
: Silicon grease applied behind pile lug.

'22' 2'22' 172/65.7 2'22 """"
'22" '22' 22'2 2'22' 25.4/53.3 -
'32" '22' 22'2 2'22' 25.4/53.7 2'22 -
"2" '22' 3'2 2'22' 25.4/50.7 2'22 -
'72- '22' 2222 2'22' 19.0/51.2 2'22 -
.:;.- -::- 5:22 028;. 15.2/60.5 32:8 - -

2' 22'2 19.0/51"; 2'22 -
".2" '22' 22'2 2'22' 25.4/52.5 2'22
".2" '22' 22.2 222' 19.0/52.5 2'22 -
T .2:- 272' 32;;- 25.4/63.7 T's-9. - ’
-:;- -::- 522; 02;;— 25.4/59.3 125; -
"2" '22' 22'2 222' 172/593 2'22
"2" '22' 2'22' 25.4/62.1 2'22 """"
"2" '22' 22.2 2.22' 25.4/53.2 2'22 -
'22" '22' 22'2 2'22' 24.1/51.5 2.22 -
.73.. -::- 5:25 02;;— 24.l/60.3 02;; -
"2" '22' 222 222' - "" 1.59/51'
'22" '2' 22'2 24.1/552 2'22 -
"2" '22' 22'2 2'222 24.1/61.5 2'22 -
"2" 22'2 2'22' 76/28.8 2'22 -
'12" '22' 22'2 222 75/30..) 2'22 -

273

 

 

23016-
NH 600'0 + SBZ'I--

 

NH LLZ‘I = 9901 IPIIIUI

I l

 

 

30"

O O
N H

(mm) S‘Juemaoetdsrq

312d

30

25

20

15

10

t (hr)

Time,

Pile Test #20.

Figure 3.1.

0v

mm

om

mm

25 s

.NN‘ “may «Ham .~.m mosmflm

.uafia

om ma o“ m

 

274

 

‘ 1 I 1‘

NX 6V'F“

q 1 1

1

q

1 1‘

J1

J

1

1 1 1

fl 1 411114 ‘1 1 1 14* fl 1 1‘ 1 1 —

Mn IS'I a 9901 terurur

o
.—o

.8

5‘1uamaoetdsra 911d

5%

 

 

(mm)

275

 

 

 
 
  

‘ZHOI B

an 220 o 2 Z8“?—

._.zH 01 a N1 220'0 T BI'Z

Mn 8€°I = 9901 terzrur

 

45

l
35

AllAle J
20 25 30
t (hr)

Time,

1 L; A J J 1 I
15

10

l A l A l

 

 

40r-

30'

' 2_L—~
O O
N 0.0

(mm) y‘auamaoetdsrq atld

Pile Test #23.

3.3.

276

 

31 29°1
23 01‘a 33 950'0 2 75'1
- 23 01 a 33 220°o + 9£°1
.. 23 01 a 33 910'0 2 92'1

23 01 a -
31 L90'o + 91°1—
2301 a 33 990 0 2 91'1-
2301 a 33 290:0 I 91';—
23019 33 £50°0 2 91'1-—

2301 a 33 220°0 T 91°1-

2301 a 31 c10'0 2 91'1-

NX 91'1"

Nx 88'0 = 9201 1212131

 

 

 

l 2_L I,
O O O 0

(mm) g‘nuamaaetdstq 311d

25

20

15

t (hr)

10

Time,

Pile Test #24.:

Figure 3.4.

277

 

 

 

33 [9’1 = 9901 IBIJIUI

 

 

 

L L I
O O O
m N I"

(mm) ‘3‘1uamaoetds10' BIId

 

4D

35

30

25

20

15

10

t (hr)

Time,

Pile Test #25.

Figure 3.5.

278

 

 

O
‘2

 

D W2 O In-
N

m
n

O
M

2301 a 3>1 8100 2 55'1—

In
N

(mm)

._ 23 01 a 3:1 120'0 2' 99'1

NX SS'I 8 P901 IeliluI

5‘3uamaaetdsta GIId

 

20

15

t (hr)

Time,

10

Pile Test #26.

Figure 3.6.

279

 

 

 

 

 

In
.b~
1
1
1
3
1
3.:
jun
1
1
3
1
1
.4
_juu
.1v
23 01 a 33 8100 + LE'I- 3
1
.4
.1
33 [2'1 = peo'I 12111111 3
l L 1 . o
O O O O
s? M N .-0

(mm) 3‘1uamaoetdstq and

t (hr)

Time,

Pile Test #27.

Figure 3.7.

280

 

 

 
   
  
  
  
  

 

 

—.23. 01 a 33 [90°C 2 02'2 . .‘
..23 01 a 33 090°o 2' 102 3
:19
23 01 a ._ 1
33 9500 + 68'1— _-
is
.(D
_ :8
23 01 a 33 £500 + 91%.. j
23 01 0 33 ”0'0 2 99‘1_ -f
3::
2,
5o
‘0')
1
1
ED
.N
'O
33 11'1 = 0901 121:1u1
1 1 L o
O O O D
M N H

(mm) 3‘1uamaoetdstq 3115

t (hr)

Time,

Figure 3.8.

Pile Test #28.

281

 

 

H 013 _
XL90'O + 12'3—

    

23 019 33 190'0 2 26'1—-

.Jiii.iiil.

I

10

J

 

 

 

O‘ML

N3 £I°I = P901 IPIIIUI
I I I
o «G <3 fl:
‘1' M N u-o
(mm) S‘iuamaoeIdSIQ 311d

30

20-

t (hr)

Time,

Pile Test #30.

Figure 3.9.

282

 

 

 

40f-

 

1
5
m

NH 77 3

N1 [6' -—

NX OS'Z— ‘

 
     
 
 

23 01 a 33 2900 7 90'2—

23 019 33 1100 2 20'2—

23 01 a 33 9100 '+' £02—

NX 20'5- ‘

 

33 ZZ'I = 9991 12131UI ‘

3 #15
N 1-4

(mm) 3 ‘3uamaoetdsm and

 

20 25

15
)

t (hr

10
Time,

Pile Test #31.

Figure 3.10.

283

 

 

-\\_“

 

 

33 sz'z— i
33 1s'z— ]
2301 9 33 950°0 1 68'1-
2301 9 33 620'0 1 68’1- '1
2301 9 33 £10'0 + 68'1- 1
33 69‘1— j
23 01 9 33 250'0 T IL'I — J
23 01 9 33 LZO'O '+' 11°1- 2
23 01 933 £10'0211°1—— '
.4
J
1
mzv 12221 pausnfpv j
.4
33 IL‘I = 3901 IBIIIUI ‘
l L L L
«o ‘3 ‘3 o
<r ‘9 N F.

(3131) 9 ‘mamaoetdsrq 312d

20 25 30 35 40
Time,

15

10

t (hr)

Pile Test #32.

Figure 3.11.

284

 

 

15'—

   

0.2- on c- 3013 'dmai 38193 ‘33 19'1--

OT'Z ‘ ._
2H OI 3 N3 670'0 + 99'1

   
     
    
 

23 01 9 33 320'0 I 19°1

“ ZH OI 6 NH ZIO'O + 99'1

23 01 9 33 330°0 2 99'1 -
23 01 9 33 210'0 2 99°1--

N)! '79'1—

23 01 9 33 810'0 2 GZ'I-

 

33 62°1_.
23 01 9 33 290°0 + 62 1-

  
 

23 019 33 910'0 I 6Z'I—-

33 62°1 a 3801 12121UI
l I

O m
v-d

(mm) 3 ‘3uamaoetdsgq 312d

 

lllll'UJLlllLJJlJlj
80

llLAAL|AMJlLAIL11Lll

All

.1

ILLLIIILI

lll;jlllilllLlLl
40

Llljlll‘lll

ll]l_l11!11

  

 

60

t 2hr)

Time,

20

Pile Test #34.

Figure 3.12.

285

 

 

N3! 01'0-

N3! 29'0-

23 01 9 33 c10°0 2,33'0-

23 01 9 33 100'0 I 99°0-

23 01 9 33 100 o 2 vv°o__

33 91°1 = 0901 12121UI ‘33 33'

 

 

L J l
13 0* ac
M N H

(umu ‘3‘1uamaoetds10 3119

t (hr)

Time,

Pile Test #38.

Figure 3.13.

286

 

 
       
     
 

 
    
 

 

 

 

3
fl
1
__ 33 '7'7'0 .. j
_ 23 01 9 33 2200 + u;0 ,
— 23 01 9 33 1100 + u°0 _,
'-NX CL'O J
23 01 9 _ 2
33 9100 + 950- 1
#2301 9 33 600'0 2 950— '2
\ d
33 950— 2
J
1
23 01 9 33 £10 0 2 93' .
1
23 01 9 33 100'0 2 vv°o-— 2
4
cull
\ .
\\. _
33 98'1 = 3201 1212121 \ j
! 1 1 , 33 vrox
o» O O O
M N c-i

(mm) lg‘nuamaoetdstq 312d

t (hr)

Time,

Pile Test #40.

Figure 3.15.

287

 

 

2301 9.33 210'0 7 93'0-.

..23 01 9 33 £10'0 T 9v°o

    
 

23 01 9 33 100'0 I 93°0-

\

33 93'0 = 3201 1212121
1

 

 

20

O
.—0

(mm) g‘zuamaoetdsm 311d

t (hr)

Time,

Pile Test #41.

Figure 3.16.

288

 

 

 

 

 

.1
‘WQ
.1
2
1
«~dcfi
.1
1
j
1
~1cu
I
23 01 9 33 £100 32 ”'0 - 2
\ 1
\ 1
_11—9
23 01 9 33 1000 + 1170. j
\ 1
\ -4
\1
33 13'0 . 9201 12121UI ‘
1 1 L c,
O O C CD
on o1 —4

(mm) 3‘3uamaoetdsxq 312d

t (hr)

Time,

Pile Test #42.

Figure 8.17.

289

 

 

1-NX €£°O

    

2301 9 _ " N2 0
33 600'0 + £1'0-

 

NH EL'O-

33 0 uaqi ‘33 06'I = 3901 12111UI
[- P l I

 

 

O o C O
4' m N c—a

(mm) 3‘3uamaoetdstq 3115

20

15

10

t (hr)

Time,

Pile Test #43.

Figure 3.18.

 

290

 

/

NX SL'O—

33 9S“0-—

   

23 01 9 33 110'0 2 99‘0-

23 01 9 33 600‘0 7 91'0~

N3! ES'I = P901 I9¥3IUI 'N‘l 9S'O\

 

 

L 1 1 1 ‘
o» ‘O ‘3 o C:
a: 9‘ °‘ ”

(mm) s‘auamaoetdsrq SIId

t (hr)

Time,

Pile Test #44.

Figure 3.19.

291

 

 

33 95'0 = 9201 1212121
L

 

 

20

(mm)

C
.—4

3‘3uamaoetdsrq 9115

[5 20 25

t (hr)

10

Time,

Pile Test #45.

Figure 3.20.

292

 

N1 09'0-—

ZH 01 0- _
NX LZO'O + 95'0—
ZH OI 0
N1 810'0 +

23 01 9
33 910'0 '

2H OI @ NH 600'

 

23 019 33 17000 2 99'0—

 

 

t (hr)

Time,

Pile Test #46.

Figure 3.21.

 

 

33 990 K‘
33 290 “‘ —\\
33 15°1 = 3201 1212121 J
O O O C C)
<7 m N 1—1

(mm) 5‘2uamaoetds1q 311d

293

 

 
  
  
  
 
   
 
  

NH 09'0—-

2H OI 0 _
33 [ZO'O + 99'0-
ZH OI 9 _
810'0 + 9S'0 -
ZH 01 9

33 £10°0 2 91'0-

23 019 33 600’0 2 95‘0-

23 01 9 33 17000 '2' 99'0—

33 99°0-L .

 

 

 

 

.\
33 12'1 . 3201 1212121
I l i
3 3 5?. 3- 3:

(mm) 3‘2uamaoetds1a 3113

t (hr)

Time,

Pile Test #48.

Figure 3.22.

294

 

 

    

 

 

 

N
2
41n
NH 99°0-
NX EL'O ..
33 0- 2‘“
NH 9S'0.
33 ES‘I = 9201 1212121 2]
I 1 1 ‘3
:2 E} 53 <3

(mm) 3‘3uamaoetdstq 312d

t (hr)

Time,

Pile Test #49.

Figure 3.23.

295

 

 

NH 95°C 4

 

 

N3 16°C 8 P901 1923?;T\\\\
L l_ L i
<3 <3 <3 <3 <3
‘1’ M N v-‘
(mm) 3‘3uamaaetdstq 3115

C.)

t (hr)

'Time,

Pile Test #52.

Figure 3.24.

296

 

 

. Na 9SOO\
' = 80 9131“
331891 1) '11 L I 33 39'01

 

<3 1D
2 N 1'"

(mm) 3‘1uamaoetdstq 312d

t (hr)

Time,

Pile Test #54.

Figure 3.25.

 

 

297

 

 

Alli-Ill-

All

.1

 

23 01 9 293 9:0'0 2 22's.
23 01 9 293 600'0 + 92'0—

adfl 9'7'0"
PdN 92°C—

23 01 9 293 100'0 2 92'0--

   

A lIAIIAAA-llfifilIll-llfllljl-IAAIA‘AIII‘llllll

40

 

 

adN 9E'0—-
PdN 62'0—-
c3
n1
BJN ZZ'O—
PdN LVI'O = P901 IBIIIUI
1 L. -ca
~o ~¢ of :1
<3 <3 ’ <3

(mm) 3‘1uamaoetdsra 3119

t (hr)

Time,

Pile Test #61.

Figure 3.26.

298

 

 

     
  

 

 

— 8
1
3
;
1:)
‘1co
can u'o- :
can 25°o- 1
1
2H 01 9 Ban £20°0 1 29'0— 3
1 ’3
a .C
99»
1 o“
u E
can ZS'O— i E
can 9c°o-— ‘1
23 01 9 em LIO'O I 9£°o- 4]
18
can 9£°o_. 1
am zero- -§
2H 01 9 8cm 110‘0 1' zero- 3
A
van z91°o = 9901 taIIIUI 5
L i l *o
m o In 0
.3 _3 o

(mm) ‘g‘uuamaaetdstq BITd

Pile Test #62.

Figure 3.27.

299

.m0& gums mafia .m~.m muswwm

om CV AMSV U 03H“.

ON a

Fl, F

 

 

‘qq“““‘dd*+‘l‘.‘J4‘fl‘4fi‘q1‘44-‘4**q‘“‘1—‘i“dlq‘1q“d‘d‘d““j

\u‘ll‘
\\

P901 IPIIIUI

edfl 29°C
1

 

 

3‘3uamaaetdstq attd

(mm)

300

 

 

 

 

1.21—

~53
p4
(’\
~22
4%
9d“ LI'I~. d
mo
('6
Pan BS'O . P901 I9131UI
I 19 l 1 c3
8 :2 R "N‘ o
.s o‘ o‘ <5
(mm) y‘auamaaetdsrq 311d

t (hr)

Time,

Pile Test #67.

Figure 8.29.

301

 

 

 

 

 

I
1
q
q
.1
4
J
Han vc'I— "j
2H 01 9 can L90'0 I vz'I—- 1
23 01 9 can “0'0 7 98‘1- 1
.4
I
1
I
R '1
g ..
9dN vc°I = 9901 IPIIIUI j
l J, L, 1
on O on O I!“
N C N V‘ N O
.4 .3 <3 ‘o 0

(mm)

3‘1uamaoetdsrq BIId

3O 45

t (hr)

15

Time,

Pile Test #69.

Figure 8.30.

302

.o~* “may mafia .fim.m muzwfim

Anne 8 .mafia

 

 

 

om 0v ON a
ddfiddqduud4dedde4deA-J«Adjdd444444411djqdfiq441..JldJJfilfi .C

I
\|\\ u
T?

3 I... .

. I mN o
x 9
‘_ T.
I m.
. 8

I

8 p

l o

m“ a 0m 0
a mu
S
6
N
d

e Irm~.o

usoo.~

mm.~

 

 

 

S‘Juamaaetdsrq BIId

(mm)

303

on

u u
av A av

.HN‘ “may «Hum .mm.m museum

.mafie
om ca

 

 

1 _ ‘ d 4 I u q d

1 1 q 1 d 3 1 .1 I‘ l d N - d‘ 4 G 4 q 1 141 I -

dldfld~14¢d_fidid—Jfi1flfl-4141~J41

9801 IPIIIUI

9d“ SL'O =

 

 

 

3‘1uamaaetdsrq SIId

(mm)

304

cm

.mne 3999 «Hum .mm.m museum

Augv u .osfia
om oe om ow as

o

 

 

‘ - I d i

0
\\

..\
.\
..\
\

I41di“‘d‘dlI‘d1d‘dI‘ll“.‘1fi“l~“ddl‘“u+luiqdqd“
‘

9901 IPIIIUI

9JH €L°0 =

 

 

y‘auamaaetdstq GIIJ

(mm)

305

 

 

 

 

(mm) y‘auamaoetdsta GIId

J
..c:
¢CO
d
.4
1.8
.jc:
‘ V
_fic:
‘0‘)
‘\ 1
\ is
\ 2
\ q
4
1
can 6L‘0_ .
2H 019 can 020'0 + 61'0- lo
1
1
can ez'o . p901 IEIJIUI -
L ' . l O
N . F‘ O

t (hr)

Time,

Pile Test #74.

Figure 3.34.

 

306

 

saw 99 0‘;
(awn ananflH 60 o 9 can no 0 + 99 0A

can 99 o-
29 60'0 9 9cm 110'0 7 99'0 X

BdN 89°0"

can 99'0 = 9801 IHIIIUI
L 1 ' i

29 60'0 9 9cm LIO'O 399'0 «X

LJLILLIII

ILLILJI

A L. LA I A 1 LA I

j

.4
q
1

1
4

 

 

M N , F‘
(mm) 3‘1uamaaetdstq 311d

C3

60

45

t (hr)

30
Time,

15

Pile Test #75.

Figure 8.35.

307

 

 

“\ (aAen azenbs) _
A. 29 9o°o 9 can 610'0 + sz'o
\x

-

 

 

'\
\
\.
\
\.
\
\ .
\ u
!\
\
\\
.\\
\
\
Paw S£.0‘—'\
\
\
K
\
\
\
‘|
K
\i
x
9d“ €£°0 3 9901 IBI3IUI \
l 1 L S
‘0. Q N
C O <5

(mm) 3‘1uamaaetdstq GIId

t (hr)

Time,

Pile Test #76.

Figure B.36.

308

 

 

 

 

'1
l
l
1
.1
, 1
\\‘ J
“K van 20's _j
\ 4
\\ .1
\. '1
\\ .
.\\ +
\ \ 1
\ Ban 9S'IT
9cm {6'0 . 9901 Iarqu \\ f
L L I \g
m N o-A

(mm) 3‘3uamaaetdsta 3115

40

35
t (hr)

30
Time,

Pile Test #78.

Figure 8.37.

309

 

 

 

 

\\\‘
._ ~ - o
9 N
\.‘\\ J
\
QJN L8'I"\
(anen azenbs) _ \
2H LI'O 9 Hem 590'0 + can L8'I-\\
.\
Ban [8.1 -\\\,
(amen annbsfln [1'0 9 2cm ”0'0 I z9°I -\
\‘\ d
\
PdN £8'I-\~
' <0
1 o
\ 1
i
‘.
i _
Han II'I = 9901 IPIIIUI X'
L 1 L 9 o
U“ o ‘n. o
d ,4 ‘o

(um)

5‘1uamaaetdsiq BIId

t (hr)

Time,

Pile Test #79.

Figure 8.38.

310

cm

.om‘.umma ofiflm

Augv u .mafla

on on on UN

 

 

-dl

.¢m.m musmfim

Dd

.jjiuld . . .IJ. {11444d41dq414 11.14qu 1411jfiu4111.44 . «1444

 

Q

J

\.

9d“ 8E'I

999 [6‘0 = 9901 IBIIIUI

 

 

3‘1uamaoetdsrq attd

(mm)

311

 

 

 

 

.1
W
-t:
N
\\
\
\ _ m
\ H
\\
\
\
\ 4
\\
\
\. 'l
“K‘
\\ d S
.\\ 4
\
4
"‘ m
\
Ban 96'0 = 9801 1913131
I L D
Q N 0

(mm) 3‘3uamaoetdstq alld

t (hr)

Time,

Pile Test #81.

Figure 8.40.

312

 

29 01 9 N1 09'0 I 95‘!

29 01 9 _
99 09°o + 9S”I \\\

29 019 99 099 o I 9S'I
99 99'! ‘~

29 01 9 99 09°o I ze°o

29 01 9 99 090 o I 26'0

 

NH Z6'0 = P901 IPIIIUI
I I I

90

69

39

. 120

 

 

as \D ‘1‘

10

(mm) 9 “nuamaaetdsxq 9114

t (hr)

Time,

Pile Test #82.

Figure 8.41.

* 313

 

 

 

 

(mm)

3‘1uamaaetdstq 311d

L")
‘Q
\\\\
.\\\
\ ‘ "‘
\
\
\ .
. %N
\ . _,
\
\__
\
can ZO'Z = 9801 ternrur
l L L I
N N .3 ..2 o'

t (hr)

Time,

Pile Test .85.

Figure 8.42.

314

 

 

NX 9S'I

44
I:

(mm)

\

P901 IEIIIUI

I

N

g‘uuamaaetdsyq BIId

 

t (hr)

Time,

Pile Test #88.

Figure 8.43.

REFERENCE LIST

REFERENCES

Alkire, B.D. (1972). "Mechanical Properties of Sand-Ice
Materials", unpublished Ph.D. Dissertation, Michigan State
University, East Lansing, Michigan.

Alwahab, M. R. M. (1983). "Bond and Slip of Steel Bars in
Frozen Sand," unpublished Ph.D. Dissertation, Michigan
State University, East Lansing, Michigan.

Andersland, O.B. (1987). Chapter 8 in GROUND ENGINEER’S
REFERENCE BOOK, edited by F.G. Bell, Butterworth and Co.
(Publishers) Ltd., London.

Andersland, O.B.; and Akili, W. (1967). "Stress Effects on
Creep Rates of Frozen Clay Soil," Geotechnique, London,
Vol. 17, no. 1, pp. 27-39.

Andersland, 0.8.; and AlNouri, I. (1970). "Time-Dependent
Strength Behavior of Frozen Soil," J. Soil Mech. Found.
Div. Am. Soc. Civil Eng., Vol. 96, No. 8M4, pp. 1249-1265.

Andersland, 0.3.; and Alwahhab, M.R.M. (1983). "Lug
behavior for model steel piles in frozen sand,"
PERMAFROST, Proceedings 4th International Conference,
National Academy Press, Washington, D.C., p. 16-21.

Andersland, 0.8.; and Douglas, A.G. (1970). "Soil
Deformation Rates and Activation Energies," Geotechnique,
London, Vol. 20, No. 1, pp. 1-16.

Andersland, O.B.; Sayles, F.H.; and Ladanyi, B. (1978).
"Mechanical Properties of Frozen Ground," Chapter 5 in
GEOTECHNICAL ENGINEERING FOR COLD REGIONS, ed. by 0.8.
Andersland and D.M. Anderson., McGraw-Hill Book Co., New
York.

Anderson, D. M. ; Pusch, R. ; and Penner, E. (1978).

"Physical and Thermal Properties of Frozen Ground, " Chapter
2 in GEOTECHNICAL ENGINEERING FOR COLD REGIONS (ed. by O. B.
Andersland and D.M. Anderson), McGraw-Hill Book Co., New
York.

315

316

Anderson, D.M.: and Tice, A.R. (1972). "Predicting Unfrozen
Water Contents in Frozen Soils from Surface Area
Measurements," Highway Research Record, No. 393, pp. 12-18.

Assur, A. (1963). "Discussion on Creep of Frozen Soils,"
PERMAFROST, Proc. lst Int. Conference, Purdue University,
Lafayette, Indiana, pp. 339.

ASTM (1988). "Piles Under Static Axial Compressive Load,"
ASTM-01143.

Baker, T.H.W. (1978a). "Effect of End Conditions on the
Uniaxial Compressive Strength of Frozen Sand", PERMAFROST,
Proc. 3rd Int. Conference, Edmonton, Alberta, Vol. 1, pp.
608-614.

Baker, T.H.W. (1978b). "Strain Rate Effects on the
Compressive Strength of Frozen Sand", Engineering Geology,
13, pp. 223-231.

Baker, T.H.W. (1979). "Strain Rate Effect on the
Compressive Strength of Frozen Sand," GROUND FREEZING,
First International Symposium, Bochum. (Edited by H. L.
Jessberger), Elsevier Sci. Publ. Co., pp. 223—231.

Bishop, R.F.; Hill, R.; and Mott, N.G. (1945). "The Theory
of Indentation and Hardness Tests," Proc. Phys. Soc., Lond.
57(3) pp. 147-159.

Bragg, R.A. (1981). "Properties for Sand-Ice Structural
Systems," unpublished Ph.D. Dissertation, Michigan State
University, East Lasning, Michigan.

Bragg, R.A.; and Andersland, 0.8. (1980). "Strain Rate,
Temperature, and Sample Size Effects on Compression and
Tensile Properties of Frozen Sand," GROUND FREEZING, Second
International Symposium, Trondheim, Norway, pp. 34-47.

Brown, J.; Rickard, W.; and Vietor, D. (1969). "The Effect
of Disturbance on Permafrost Terrain," U.S. Army Cold Reg.
Res. Eng. Lab., Spec. Rep. 138, Hanover, N.H.

Chaichanavong, T. (1976). "Dynamic Properties of Ice and
Frozen Clay Under Cyclic Triaxial Loading Conditions,"
unpublished Ph.D. Dissertation, Michigan State University,
East Lansing, Michigan.

Chamberlain, E.; Groves, C.; and Perham, R. (1972). "The

Mechanical Behavaior of Frozen Earth Materials Under High
Pressure Triaxial Test Condition," Geotechnique, London,

Vol. 22, No. 3, pp. 469-483.

317

Clough R.; and Penzien J. (1975). DYNAMICS OF STRUCTURES,
McGraw-Hill Book Co., New York.

Dillon, H.B.; and Andersland, O.B. (1966). "Predicting
Unfrozen Water Contents in Frozen Soils," Canadian
Geotechnical Journal, Vol. 3, No. 2, pp. 53-60.

Dufour, 8.; Sego, D.C.; and Morgenstern, N.R. (1983).
"Vibratory Model Pile Driving in Frozen Sand," PERMAFROST,
Proceedings Fourth International Conference, National
Academy Press, Washington. D.C., pp. 255- 260.

Eckardt, H. (1982). "Creep Tests With Frozen Soils Under
Uniaxial Tension and Uniaxial Compression," Roger J.E.
Brown Memorial Volume. Proceeding of the 4th Canadian
Permafrost Conference Calgary, National Research Council of
Canada, pp. 365-373.

Frost, H.; and Ashby, M.F. (1982). "Deformation Mechanism
Maps" THE PLASTICITY AND CREEP OF METALS AND CERAMICS,
Pergamon Press.

Glasstone, 8.; Laidler, K.; and Eyring, H. (1941). THE
THEORY OF RATE PROCESS, MCGraw-Hill Book Co., New York.

Gold, L.W. (1970). "Process of Failure in Ice," Canadian
Geotechnical Journal, Vol. 7, No. 4, pp. 405-413.

Gold, L.W. (1970). "Process of Failure in Ice," Canadian
Geotechnical Journal, Vol. 7, No. 4, pp. 405-413.

Goughnour, R.R.; and Andersland, O.B. (1968). "Mechanical
Properties of a Sand-Ice System," J. Soil Mech. Found. Div.
Am. Soc. Civil Eng., Vol. 94, No. 8M4, pp. 923-950.

Griffith, A.A. (1924). "The Theory of Rupture," lst Int.
Congress for Appl. Mech., Delft, pp. 55-63.

Haynes, F.D.; Karalius, J.A.; and Kalafut, J. (1975).
"Strain Rate Effect on the Strength of Frozen Silt," U.S.
Army Cold Reg. Res. Eng. Lab., Res. Rep. 350, p. 27.

Heydinger, A. G. (1987). "Piles in Permafrost," Journal of
Cold Regions Engineering, American Society of Civil
Engineers, Vol. 1, No. 2, pp. 59-75.

Hill, R. (1950). "THE MATHEMATICAL THEORY OF PLASTICITY,"
Clarendon Press, Oxford, England.

Hoekstra, P. (1969). "The Physics and Chemistry of Frozen
Soils," Spec. Rep. 103, High. Res. 8., pp. 78-90.

318

Hult, J.A.H. (1966). "CREEP IN ENGINEERING STRUCTURES,"
Blaisdell Publ. Co., Waltham, Mass.

Jellinek, H.H.G. (1957 a). "Tensile Strength Properties of
Ice Adhering to Stainless Steel," USA SIPRE, Corps of
Engineers, Wilmette, Ill., Res. Rep. 23.

Jellinek, H.H.G. (1957 b). "Adhesive Properties of Ice,
Part 1," USA SIPRE, Corps of Engineers, Wilmette, Ill.,
Res. Rep. 39.

Jellinek, H.H.G. (1960 a). "Bonding of Flat Ice Surfaces-
Some Preliminary Results," USA SIPRE, Corps of Engineers,
Wilmette, Ill., Res. Rep. 61.

Jellinek, H.H.G. (1960 b). "Adhesive Properties of Ice,
Part 2," USA SIPRE, Corps of Engineers, Wilmette, Ill.,
Res. Rep. 62.

Jellinek, H.H.G. (1960 c). "Some Frictional Properties of
Thin Water Films," USA SIPRE, Corps of Engineers, Wilmette,
Ill., Spec. Rep. 37.

Jellinek, H.H.G. (1967). "Liquid-Like (Transition) Layer on
Ice," Journal Of Colloidal and Interface Science, Vol. 25,
pp. 192-205.

Johnson, K.E. (1970). "The Correlation of Indentation
Experiments" J.Mech. Phys. Solids. 18, pp. 118-126.

Johnston, G.H. (1963). "Pile Construction in Permafrost,"
PERMAFROST, Proceedings lst International Conference,
National Academy Press, Washington, D.C., pp. 477-481.

Johnston, G.H.; and Ladanyi, B. (1972). "Field Tests of
Grouted Rod Anchors in Permafrost." Can. Geotechnical J.,
Vol. 9, pp. 176-194.

Kaplar, C.W. (1969). "Laboratory Determination of Dynamic
Moduli of Frozen Soils and of Ice," U.S. Army, Cold Reg.
Res. Eng. Lab., Hanover, N.H., Res. Rep. 163.

Kaplar, C.W. (1971). "Some Strength Properties of Frozen
Soil and Effect of Loading Rate," U.S. Army, Cold Reg. Res.
Eng. Lab., Spec. Rep. 159, Hanover, N.H.

Kovacs, A.; and Michitti, F. (1970). "Pile Driving by Means
of Longitudinal and Torsional Vibrations," U.S. Army, Cold
Reg. Res. Eng. Lab., Spec. Rep. 141, Hanover, N.H.

319

Lachenbruch, A.H. (1988). "Permafrost Temperature and the
Changing Climate" PERMAFROST, Proceedings Fifth
International Conference on Permafrost, Tapir Publishers,
Trondheim, Norway, Vol. 2, pp. 1165-1170.

Ladanyi, B. (1966). "Failure Mechanism of Rock Under a
Plate Load," Proc. First Cong. Int. Soc. Rock Mech.,
Lisbon, Port. 1, pp. 415-420.

Ladanyi, B. (1972). "An Engineering Theory of Creep of
Frozen Soils," Canadian Geotechnical Journal, Vol. 9, No.
1, pp 63-80.

Ladanyi, B. (1976). "Use of the Static Penetration Test in
Frozen Soils," Canadian Geotechnical Journal, Vol. 13, No.
2, pp. 95-110.

Ladanyi, B. (1981). "Mechanical Behavaior of Frozen Soils,"
MECHANICS OF STRUCTURED MEDIA (ed. By A.P.S. Selvadurai)
Elsevier Scientific Publ. Co., New York, pp. 205-245.

Ladanyi, B. (1982). "Determination of Geotechnical
Parameters of Frozen Soils by Means of the Cone Penetration
Test", Proc. Symp. ESOPT II, Amsterdam, Vol. 2, pp. 671-
678.

Ladanyi, B. (1985). "Use of the Cone Penetration Test for
the Design of Piles in Permafrost", ASME Journal of Energy
Resources Technology, 107, pp. 183-187.

Ladanyi, 8.; and Guichaoua, A. (1985). "Bearing Capacity
and Settlement of Shaped Piles in Permafrost", Proc. XI
Intl. Conf. Soil Mech Found. Eng., San Francisco, pp. 1421-
1427.

Ladanyi, B.; and Guichaoua, A. (1986). "Time Dependent Bond
Regain in Driven Piles in Permafrost," Proceedings of the
International Conference on Deep Foundation; Beijing, Vol.
1, pp. 2164-2169.

Ladanyi, 8.: and Johnston, G.H. (1973). "Evaluation of In-
Site Creep Properties of Frozen Soils with the
Pressuremeter," PERMAFROST, Proc. 2nd Intl. Conf., N.
American Contribution, Yakutsk, U.S.S.R., pp. 310-318.

Ladanyi, B.; and Johnston, G.H. (1974). "Behavior of
Circular Footings and Plate Anchors Embedded in
Permafrost," Canadian Geotech. Journal, 11, pp. 531-553.

Li, J.C. (1979). "Dynamic Properties of Frozen Granular
Soils," unpublished Ph.D. Dissertation, Michigan State
University, East Lansing, Michigan.

320

Linell, K.; and Lobacz, E.F. (1980). "Design and
Construction of Foundations in Areas of Deep Seasonal Frost
and Permafrost," U.S. Army Cold Reg. Res. Eng. Lab.,
Hanover, New Hampshire, Special Report 80-34.

Luscsher, U.; Black, W.T.; and McPhail, J.F. (1983).
"Results of Load Test on Temperature Controlled Piles in
Permafrost," PERMAFROST, Proceedings of the Fourth
International Conference, National Academy of Science
Press, pp. 756-761.

Manikian, V. (1983). "Pile Driving and Load Tests in
Permafrost for the Kuparuk Pipelone System," PERMAFROST,
Proceedings of the Fourth International Conference,
National Academy of Science Press, pp. 804-810.

Marin, J.; Pao, Y.H.; and Cuff, G. (1950). "Creep
Properties of Lucite and Plexiglas for Tension,
Compression, Bending and Torsion," Trans. Am. Soc. Mech.
Eng., 73, pp 705-718.

Mellor, M. (1979). "Mechanical Properties of
Polycrystalline Ice," In: Per Tryde (Ed), Physics and
Mechanics of Ice, IUTAM-Symp. Copenhagen, Springer-Verlag,
Berlin, pp. 217-245.

Mitchell, J.K.; Campanella, R.G.; and Singh, A. (1968).
"Soil Crep as a Rate Process," J. Soil Mech. Found. Div.
Am. Soc. Civil Eng., Vol. 94, No.SM1, pp. 231-253.

Nadai, A. (1963). THEORY OF FLOW AND FRACTURE OF SOLIDS,
Vol. 2, McGraw-Hill Book Co., New York.

Neukirchner, R.J. (1988). "Standard Method for Pile Load
Tests in Permafrost" PERMAFROST, Proceedings Fifth
International Conference, Tapir Publishers, Trondheim,
Norway, Vol. 2, pp. 1147-1151.

Nixon, J.F. (1987). "Pile Load Tests on Saline Permafrost
at Clyde River, N.W.T." Canadian Geotechnical Journal, Vol.
25, No. l.

Nixon, J.F.; and Lem, G. (1984). "Creep and Strength
Testing of Frozen Saline Fine-Grained Soils," Canadian
Geotechnical Journal, Vol. 21, pp. 518-529.

Nixon, J.F.; and McRoberts, E.C. (1976). "A Design Approach
for Pile Foundations in Permafrost," Canadian Geotechnical
Journal, Vol. 13, pp. 40-57.

321

Odquist, F.K.G. (1966). "Mathematical Theory of Creep and
Creep Rupture," OXFORD MATHEMATICAL MONOGRAPH, Oxford
University Press, London.

Odquist, F.K.G.; and Hult, J. (1962). CREEP IN METALLIC
MATERIALS, Springer, Berlin.

Orth, W. (1985). "Deformation Behavior of Frozen Sand and
Its Physical Interpretation," GROUND FREEZING, Proceedings
Fourth International Symp., Sapporo, Japan.

Parameswaran, V.R. (1979). "Creep of Model Piles in Frozen
Soil," Canadian Geotechnical Journal, Vol. 16, No. 1, pp.
69-77.

Parameswaran, V.R. (1982). "Displacement of Piles Under
Dyunamic Loads in Frozen Soils," The Roger J.E. Brown
Memorial Volume Proceedings, Fourth Canadian Permafrost
Conference, Calgary, National Research Council of Canada,
pp. 555-559.

Parameswaran, V.R. (1984). "Effect of Dynamic Loads on
Piles in Frozen Soils," Third International Cold Regions
Engineering Specialty Conference, Edmonton, Alberta,
Canada. Vol. 1, pp. 41-52.

Parameswaran, V.R. (1985). "Cyclic Creep of Frozen Soils"
GROUND FREEZING, Proceedings of the Fourth International
Symposium, Sapporo, Japan, Vol. 2, pp. 201-206.

Richart, F.E., Jr.; Hall, J.R., Jr.; and Woods, R.D.
(1970). VIBRATIONS OF SOILS AND FOUNDATIONS, Prentice-
Hall, Inc., Englewood, N.J.

Ruedrich, R.A.; and Perkins, T.K. (1974). "A Study of
Factors Influencing the Mechanical Properties of Deep
Permafrost," Journal of Petroleum Technology, pp. 1167-
1177.

Sayles, F.H. (1966). "Low Temperature Soil Mechanics,"
Technical Note 143, U.S. Army Cold Reg. Res. Eng. Lab.,
Hanover, New Hampshire.

Sayles, F.H, "Creep of Frozen Sand," U.S. Army Cold Reg.
Res. Eng. Lab., Hanover, New Hampshire, Technical Report
190.

Sayles, F.H. (1973). "Triaxial and Creep Tests on Frozen
Ottawa Sand," PERMAFROST, Proc. 2nd International
Conference, North American Contribution, Yakutsk, U.S.S.R.,
pp. 384-391.

322

Sayles, F.H, (1974). "Triaxial Constant Strain Rate Tests
and Triaxial Creep Tests on Frozen Ottawa Sand," U.S. Army
Cold Reg. Res. Eng. Lab., Hanover, New Hampshire, Tech.
Report 253.

Sayles, F.H.: and Epanchin, N.V. (1966). "Rate of Strain
Compression Tests on Frozen Ottawa Sand and Ice" U.S. Army
Cold Reg. Res. Eng. Lab., Hanover, New Hampshire, Technical
Note 140.

Sayles, F.H.; et al (1987). "Classification and Laboratory
Testing of Artificially Frozen Ground," Journal of Cold
Regions Am Soc. Civil Eng., Vol. 1, No. 1, pp. 22-48.

Schmid, W.E. (1969). "Driving Resistance and Bearing
Capacity of Vibro-Driven Model Piles," Performance of Deep
Foundations, American Society for Testing and Materials,
Special Technical Publication 444, p. 362-375.

Scott, F.S. (1969). "The Freezing Process and Mechanics of
Frozen Ground," Cold Regions Science and Engineering
Monograph ll-Dl, U.S. Army Cold Reg. Res. Eng. Lab.,
Hanover, New Hampshire.

Soo, S. (1983). "Experimental and Numerical Analysis of
Frozen Sand Structures," unpublished Ph.D. Dissertation,
Michigan State University, East Lansing, Michigan.

Soo, 8.; Wen, R. K.; and Andersland, O. B.(1986). "Flexural
Behavior of Frozen Soil," Canadian Geotechnical Journal,
Vol. 23, pp. 355-361.

Stelzer, D. L.; and Andersland, O.B. (1988). "Dynamic Load
Effect on Settlement of Model Piles In Frozen Sand,"
PERMAFROST, Proceedings Fifth International Conference,
Tapir Publishers, Trondheim, Norway, Vol. 2, pp. 1165-1170.

Stelzer, D.L.; and Andersland, O. B. (1988). "Creep
Parameters for Pile Settlement Equations," OFFSHORE AND
ARTIC OPERATIONS SYMPOSIUM - 1989, ed. by A. Ertas, D. Hui,
and R.G. Urquhart, Am. Soc. Mech. Eng., New York,

pp. 133-138.

Stelzer, D.L.; and Andersland, O. B. (1989). "Pile
Roughness and Load Capacity in Frozen Soils", COLD REGIONS
ENGINEERING, ed. by R.L. Michalowski, Proceedings of the
Fifth International Conference, Am. Soc. Civil Eng., New
York, pp. 226-235.

Stevens, H.W. (1975). "The Response of Frozen Soils to
Vibratory Loads," U.S. Army Cold Reg. Res. Eng. Lab.,
Hanover, New Hampshire, Technical Report No. 265.

323

Thomas, H. P.; and Luscher, V.(l980). "Improvement of
Bearing Capacity of Pipe Piles by Corrugations," Collection
of Papers from a U.S. Soviet Joint Seminar, Leningrad,
USSR, pp. 229-234.

Ting, J.M.; Martin, R.T.; and Ladd, C.C. (1983).
"Mechanisms of Strength for Frozen Sand," Journal of
Geotechnical Engineering, Am. Soc. Civil Eng., Vol. 109,
No. 10.

Trimble, J.R.; and Mitchell, R.J. (1982). " A Comparison of
Static and Repeated Loading Tests on Natural Frozen Soils,"
Proceedings of the 4th Canadian Permafrost Conference, pp.
433-439.

Tsytovich, N.A. (1960). "Bases and Foundations on Frozen
Soil," NAS NRS, Publ. 804, Highway Research Board, Spec.
Rep. 59.

Tsytovich, N.A. (1975). MECHANICS Of FROZEN GROUND, McGraw-
Hill Book Co., New York.

Vinson, T.S. (1978). "Response of Frozen Ground to Dynamic
Loading," Chapter 8 in GEOTECHNICAL ENGINEERING FOR COLD
REGIONS, McGraw-Hill Book Co., New York.

Vinson, T.S.; Czajkowski, R.; and Li, J. (1977). "Dynamic
Properties of Frozen Cohesionless Soils Under Cyclic
Triaxial Loading Conditions," Michigan State University,
Div. Eng. Res. Rep. MSU-CE-77-l.

Vyalov, 8.8. (1959). "Rheological Properties and Bearing
Capacity of Frozen Soils," U.S. Army Cold Reg. Res. Eng.
Lab., Hanover, New Hampshire, Translation 74.

Vyalov, 8.8. (1962). "The Strength and Creep of Frozen
Soils and Calculations for Ice-Soil Retaining Structures,"
U.S. Army Cold Reg. Res. Eng. Lab., Hanover, New Hampshire,
Translation 76.

Vyalov, 8.8. (1963). "Rheology of Frozen Soils,"
PERMAFROST, Proc. lst Int. Conf., NAS-National Research
Council Publ. 1287, pp. 332-337.

Vyalov, 5.8. (1965). "The Strength and Creep of Frozen
Soils and Calculations for Ice-Soil Retaining Structures,"
U.S. Army Cold Reg. Res. Eng. Lab., Hanover, New Hampshire,
Translation 76.

324

Vyalov, 8.8.; Gorodetskii, V.F; Ermakov, V.F.; Zatsarnaya,
A.G.: and Pekarskaya, N.K. (1969). "Methods of Determining
Creep, Long Term Strength And Compressibility
Characteristics of Frozen Soils," Technical Translation
1364, National Research Council of Canada, Ottawa.

Weaver, J.S. (1979). "Pile Foundations in Permafrost,"
unpublished Ph.D. Dissertation, University of Alberta,
Edmonton.

Weaver, J.S., and Morgenstern, N.R. (1981 a). " Simple
Shear Creep Tests on Frozen Soils," Canadian Geotechnical
Journal, 18:217-229.

Weaver, J.S., and Morgenstern, N.R. (1981 b). " Pile Design
in Permafrost," Canadian Geotechnical J., 18:357-370.