1

DYNAMIC PROPERTIES
OF AN UNDISTURBED CLAY
FROM RESONANT COLUMN TESTS
By
DAN ZAVORAL
B.A.Sc, University of B r i t i s h Columbia, 1988
A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF

THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF APPLIED SCIENCE
in
THE FACULTY OF GRADUATE STUDIES
DEPARTMENT OF CIVIL ENGINEERING
We accept t h i s thesis as conforming
to the required standard
THE UNIVERSITY OF BRITISH COLUMBIA
» December, 1990
© Dan Zavoral, 1990

In presenting this thesis in partial fulfilment of the requirements for an advanced
degree at the University of British Columbia, I agree that the Library shall make it
freely available for reference and study. I further agree that permission for extensive
copying of this thesis for scholarly purposes may be granted by the head of my
department or by his or her representatives. It is understood that copying or
publication of this thesis for financial gain shall not be allowed without my written
permission.
Department of C int- £?tV&tAft£
The University of British Columbia

Vancouver, Canada
Date £>ecenêr AS^ I^Ô
DE-6 (2/88)
ABSTRACT
The dynamic properties of clay deposits under seismic or

wave loading conditions must be well understood to assure
dynamic s t a b i l i t y of structures founded on such soil. The
dynamic shear modulus and damping appear to be a complex
function of many variables, and a wide range of values have
been reported i n the l i t e r a t u r e . Consequently, considerable
uncertainty e x i s t s i n choosing the appropriate values of shear
modulus and damping f o r a p a r t i c u l a r problem.
This thesis presents a study of the influence of various

factors on the shear modulus and damping of a marine clay using
a resonant column/torsional shear device. In p a r t i c u l a r , the
influence of factors such as shear s t r a i n amplitude, e f f e c t i v e
confining stress, stress history, frequency ( s t r a i n rate), and
secondary time-dependent behaviour are examined. The pore
pressure response i s also studied.
The shear modulus was found to degrade f o r shear strains

above 0.005%. The strain dependency was found to be well
represented by a single normalized modulus reduction curve
regardless of the confining pressure or overconsolidation
ratio. Slower s t r a i n rates resulted i n smaller values of shear
modulus.
ii
Of the variables studied, the duration of sample
confinement was found to be the most imporant factor a f f e c t i n g
the material damping. Above 0.005% s t r a i n , the damping of the
marine clay increased with shearing s t r a i n amplitude. No
s i g n i f i c a n t e f f e c t of confining pressure and stress h i s t o r y on
damping was observed at any strain level. As well, the
material damping was found to be relatively independent of
loading frequency.
Both the shear wave v e l o c i t y and damping obtained i n t h i s

study were found to be consistent with the i n s i t u values
determined using the seismic cone penetration t e s t .
iii
TABLE OF CONTENTS
page
ABSTRACT i i
TABLE OF CONTENTS . iv
LIST OF FIGURES vii
LIST OF TABLES xii
ACKNOWLEDGEMENTS xiii
CHAPTER 1. INTRODUCTION " 1

1.1 PURPOSE 1
1.2 ORGANIZATION 2
CHAPTER 2. LITERATURE REVIEW: DYNAMIC PROPERTIES OF
COHESIVE SOILS 4
2.1 SHEAR MODULUS 4
2.1.1 E f f e c t of Shear Strain Amplitude 5
2.1.2 E f f e c t of Void Ratio 7
2.1.3 E f f e c t of Confining Stress 10
2.1.4 Consolidation Stress History 14
2.1.5 Frequency E f f e c t 16
2.1.6 Secondary Time E f f e c t s 16
2.2 DAMPING RATIO 21
2.2.1 E f f e c t of Shear Strain Amplitude 23
2.2.3 Frequency E f f e c t 26
2.2.4 Confining Pressure 29
2.3 NEEDS FOR RESEARCH 31
CHAPTER 3. DESCRIPTION OF RESONANT COLUMN/TORSIONAL
SHEAR EQUIPMENT 32
3.1 INTRODUCTION 32
3.2 RESONANT COLUMN APPARATUS 32
3.2.1 Confining Chamber 33
3.2.2 Drive Mechanism 35
3.2.3 Torsional Accelerometer 37
3.2.4 Height Change Measurement 38
3.2.5 Volume Change and Pore Pressure Monitoring
Equipment 38
3.2.6 A i r Pressure Transducer 39
3.3 RESONANT COLUMN ELECTRONIC COMPONENTS 39
3.3.1 Function Generator 39
3.3.2 Signal Amplifier 41
3.3.3 Charge Amplifier 41
3.3.4 Storage O s c i l l i s c o p e 42
3.3.5 Frequency Counter 42
3.3.6 D i g i t a l Multimeter 43
3.3.7 LVDT Readout Unit 43
3.3.8 Strain Indicator 43
iv
3.4 TORSIONAL SHEAR EQUIPMENT 44
3.4.1 Rotary Transducer 45
3.4.2 DC Power Supply 46
3.4.3 Power Amplifier 46
3.4.4 Storage O s c i l l i s c o p e 47
CHAPTER 4. THEORETICAL BACKGROUND 48
4.1 SHEAR STRESS AND STRAIN DISTRIBUTION 48
4.2 RESONANT COLUMN TEST 51
4.2.1 Linear Vibration Theory 51
4.2.2 Shear Modulus 53
4.2.3 Damping Ratio 54
4.2.3.1 Amplitude Decay Damping 54
4.2.3.2 Steady State Damping 55
4.3 TORSIONAL SHEAR TEST 56
4.3.1 Shear Modulus 57
4.3.2 Damping Ratio 59
CHAPTER 5. TESTING PROCEDURE 61
5.1 RESONANT COLUMN TEST 61
5.2 TORSIONAL SHEAR TEST 64
5.3 PORE PRESSURE MEASUREMENT CONSIDERATIONS 66
5.3.1 Pore Pressure Measurement 66
5.3.2 A i r Diffusion E f f e c t s 67
CHAPTER 6. SITE DESCRIPTION 75

6.1 GEOLOGY AND GEOTECHNICAL SITE DESCRIPTION 75
6.2 LABORATORY SOIL SAMPLES 78
CHAPTER 7. RESULTS OF RESONANT COLUMN TESTS 79
7.1 LOW AMPLITUDE SHEAR MODULUS 79
7.1.1 E f f e c t of Confining Pressure and Void
Ratio 79
7.1.2 Stress History 95
7.2 HIGH AMPLITUDE SHEAR MODULUS 108
7.2.1 Nonlinear Behaviour 109
7.2.2 E f f e c t of Shear Strain 109
7.2.3 E f f e c t of Confining Pressure and Stress
History 114
7.3 LOW AMPLITUDE DAMPING 119
History 121
7.4 HIGH AMPLITUDE DAMPING 129
7.4.1 E f f e c t of Strain and Nonlinearity 129
History 133
7.5 PORE PRESSURE RESPONSE 138
v
CHAPTER 8. COMPARISON OF TORSIONAL SHEAR AND RESONANT
COLUMN RESULTS 147
8.1 SHEAR MODULUS 147
8.3 PORE PRESSURE RESPONSE 156
CHAPTER 9. COMPARISON TO FIELD RESULTS 159
9.1 INTRODUCTION 160
9.2 SHEAR WAVE VELOCITY 160
CHAPTER 10. CONCLUSIONS 166
BIBLIOGRAPHY 172
APPENDIX A: Shear Modulus vs. Strain Curves 176
APPENDIX B : Damping vs. Strain Curves 211
vi
LIST OF FIGURES
Figure Page
2.1 Normalized Modulus Reduction Curves f o r
Sands and Clays 6
2.2 E f f e c t of P l a s t i c i t y Index (PI) on Normalized
Modulus Reduction Curves of Cohesive S o i l s
(from Zen et a l . , 1978) 8
2.3 Comparison of Three Experimental Formulas on

Modulus vs. Void Ratio Diagram
(from Ishihara, 1982) 11
2.4 Small Strain Shear Modulus Equations f o r Seven
Naturally Deposited Clays 13
2.5 E f f e c t of Overconsolidation Ratio on the Small
Strain Shear Modulus of Five Naturally Deposited
Clays (from Weiler, 1988) 15
2.6 E f f e c t of Varying Frequency Content on the Shear

Modulus vs. Strain Curves f o r Cohesive S o i l s
(after Aggour et a l . , 1987) 18
2.7 Typical Time-Dependent Increase i n Low Amplitude
Shear Modulus f o r Clay at Constant Confining
Pressure (after Anderson and Stokoe, 1978) 20
2.8 Relationship Between Rate of Modulus Increase and
P l a s t i c i t y Index (from Kokusho et a l . , 1982) ... 22
2.9 Range of Damping Ratios vs. Shear Strain f o r
Cohesive S o i l s 24
2.10 Typical Time Dependent Decrease i n Damping Ratio
at Constant E f f e c t i v e Stress f o r Clay
(from Marcuson and Wahls, 1978) 27
2.11 E f f e c t of Varying Frequency Content on the
Damping vs. Strain Curves for Clay (after Aggour
et a l . , 1987) 28
2.12 E f f e c t of Confining Stress, PI, and OCR on the
Small Strain Damping of Clay (from Kokusho et
a l . , 1982) 30
3.1 Resonant Column/Torsional Shear Apparatus 34
3.2 Plan View of Resonant Column Top Drive Plate ... 36
3.3 Resonant Column/Torsional Shear Wiring

Schematic 40
vxi
LIST OF FIGURES (CONT.)
Figure Page
4.1 Shear Strain i n Sample 49
4.2 Secant Modulus and Hysteretic Damping Ratio From
Torsional Shear Test (from Isenhower, 1979) .... 58
5.1 E f f e c t of A i r Diffusion on E f f e c t i v e Stress
In Sample 69
5.2 Comparison of Consolidation Rate with and without
Radial Drains 72
5.3 Increased A i r Diffusion E f f e c t at High Confining

Stress 74
6.1 Lower 232nd St. S i t e Location 76
6.2 P r o f i l e s of Atterberg Limits and Grain Size .... 77
7.1 Shear Modulus vs. Confining Pressure Relationship

for Block Samples 80
7.2 Shear Modulus vs. Confining Pressure Relationship
for Tube Samples 82
7.3 Variation of Modulus M u l t i p l i e r with P l a s t i c i t y
Index 84
7.4 Variation of Modulus Exponent with P l a s t i c i t y
Index 85
7.5 Variation of Shear Wave V e l o c i t y with Void Ratio
for Normally Consolidated Block Samples 87
7.6 Variation of Shear Wave Velocity with Void Ratio
for Normally Consolidated Tube Samples 88
7.7 Void Ratio Modified Shear Modulus vs. Confining
Pressure f o r Normally Consolidated Block
Samples 89
7.8 Comparison of Predicted and Measured Shear Moduli
for Normally Consolidated Block Samples 91
7.9 Void Ratio Modified Shear Modulus vs. Confining
Pressure f o r Normally Consolidated Tube
Samples 93
7.10 Comparison of Predicted and Measured Shear Moduli
for Normally Consolidated Tube Samples 94
viii
Figure Page
7.11 E f f e c t of OCR on Shear Modulus f o r Block

Samples 96
7.12 E f f e c t of OCR on Void Ratio Modified Shear
Modulus f o r Block Samples 98
7.13 Time-Dependent Increase i n Shear Modulus f o r
Normally Consolidated Block Sample 101

Tube Sample of Lower 232nd St. Clay 103
Overconsolidated Block Sample 104
7.16 E f f e c t of Aging on Normalized Modulus Reduction
Curves and on Low Amplitude Shear Modulus
Measured After High Amplitude Testing 107
7.17 Frequency Response Curves f o r Lower 232nd St.
Clay f o r Three Strain Levels 110
7.18 Normalized Frequency Response Curves Ill
7.19 Typical Shear Modulus vs Strain Curve f o r Lower
232nd St. Clay 112
7.20 Normalized Modulus Reduction Curves f o r Block
Samples 115
7.21 Normalized Modulus Reduction Curves f o r Tube
Samples from 11.75m 117
7.22 Normalized Modulus Reduction Curves f o r Tube
Samples from 13. 2m 118
Various Strain Levels 120
7.24 V a r i a t i o n of Low Amplitude Damping Ratio with

E f f e c t i v e Confining Stress f o r Block Samples ... 122
7.25 Variation of Low Amplitude Damping Ratio with
E f f e c t i v e Confining Stress f o r Tube Samples .... 123
7.26 Decrease i n Damping Ratio with Time f o r Block
Samples . s
125
ix
Figure Page
7.27 Decrease i n Damping Ratio with Time f o r
Tube Samples 127
7.28 Steady State and Amplitude Decay Damping vs.
Shear Strain 130
7.29 E f f e c t of Non-Linearity on determination of
Amplitude Decay Damping 132
7.30 Comparison of Accelerometer and RVDT Decay

Response 134
7.31 Damping vs. Shear Strain f o r Block Samples at
Various Confining Pressures and OCR 135
7.32 Damping vs. Shear Strain f o r Tube Samples at

Various Confining Pressures and OCR 137
7.33 E f f e c t of Duration of Confinement on High

Amplitude Damping 139
7.34 Pore Pressure Response f o r Normally
Consolidated Block Samples i n Resonant Column
Test 141
7.35 Pore Pressure Response f o r Normally
Consolidated Block Samples i n Resonant Column
Test 142
7.36 Pore Pressure Response f o r Overconsolidated
Block Sample i n Resonant Column Test 143
7.37 Normalized Pore Pressure Response f o r Normally
Consolidated Clay i n Resonant Column 144
7.38 Comparison of Pore Pressure, Damping, and Shear
Modulus vs. Strain Curves f o r Normally
Consolidated Clay 146
8.1 Shear Modulus vs. Strain Curves from Resonant
Column and Torsional Shear Tests f o r Sample
from 11.75m 148
8.2 Shear Modulus vs. Strain Curves from Resonant
from 13.2m 149
8.3 Frequency (Strain Rate) Dependency of Shear
Modulus vs. Strain Curves 151
x
Figure Page
8.4 Frequency (Strain Rate) Dependency of Shear
Modulus vs. Strain Curves 152
8.5 Damping Attenuation Curves Determined at

Various Frequencies f o r Clay from 11.75m 154
8.6 Damping Attenuation Curves Determined at
Various Frequencies f o r Clay from 13.2m 155
8.7 Comparison of Normalized Residual Pore Pressure
Response f o r Samples from Resonant Column and
Torsional Shear Tests 158
9.1 Comparison of Shear Wave V e l o c i t i e s Determined
from Resonant Column and SCPT Methods 161
9.2 Spectral Ratio Slope Method f o r Calculation
of In S i t u Damping 164
xi
LIST OF TABLES
Table Page
2.1 Values of k 17
7.1 Values of N G f o r Normally Consolidated Tube
Samples 106
7.2 Values of Nd f o r Samples at Various 02c' a n d
OCR 128
xii
ACKNOWLEDGEMENTS
I wish t o express gratitude to my supervisor, Professor

R.G. Campanella, f o r giving me the opportunity to work on t h i s
research under h i s guidance. His advice, and encouragement
throughout t h i s project are greatly appreciated.
I would l i k e t o thank John S u l l y f o r h i s valuable insight

and suggestions during the course of t h i s work. Thanks are also
due t o the technical staff i n the Department of Civil
Engineering who maintained and modified tha equipment used i n
t h i s research.
The comments and suggestions of Professor P. Byrne i n

reviewing t h i s t h e s i s were most u s e f u l l and welcome.
xiii
1. INTRODUCTION
1.1 PURPOSE
I t i s generally recognized that l o c a l s o i l c h a r a c t e r i s t i c s
a f f e c t the ground response when seismic waves propogate through
a soil profile. Advances i n earthquake ground response
analyses have greatly surpassed the knowledge of the basic
dynamic soil properties which are required i n order to
c o r r e c t l y predict the actual ground response.
In the 1985 Mexico C i t y earthquake, areas underlain by

soft Mexico City clay greatly amplified the ground motions
causing severe damage (Sun et a l . , 1988). The low shear
modulus and damping values associated with Mexico c i t y clay
were key factors i n producing the large ground motion
amplifications. More knowledge i s required of the dynamic
properties of undisturbed s o f t clays and the factors a f f e c t i n g
them. This thesis examines the dynamic c h a r a c t e r i s t i c s of a
soft clay near Vancouver, B.C. - an area susceptible to large
magnitude earthquakes.
This research addresses the complex e f f e c t of various
factors upon the shear modulus and damping of an undisturbed
clay, including: the effect of shear strain amplitude;
1
e f f e c t i v e confining stress; stress h i s t o r y ; frequency (strain
r a t e ) ; and secondary time-dependent behaviour.
To perform these studies a resonant column device was used

and l a t e r modified i n order to perform t o r s i o n a l shear tests on
the same sample. The clay used i n t h i s research was obtained
from the Lower 232nd Street University of B r i t i s h Columbia i n
s i t u research s i t e along the Trans-Canada Highway about 40 km
east of downtown Vancouver.
1.2 ORGANIZATION
A brief literature review emphasizing the e f f e c t s of

various factors on the shear modulus and damping f o r cohesive
s o i l s i s given i n Chapter 2.
The resonant column/torsional shear apparatus i s described

i n Chapter 3 and the method of analysis f o r both types of tests
i s given i n Chapter 4.
Chapter 5 describes the research s i t e , the geotechnical
aspects of the s o i l deposit, and the sampling methods used.
In Chapter 6 testing procedure considerations are discussed
while the t e s t r e s u l t s themselves are presented i n Chapters 7
and 8.
2
Lastly, Chapter 9 compares the shear modulus and damping
obtained in the laboratory with the in situ values as
determined from the seismic cone penetration t e s t .
3
2. LITERATURE REVIEW: DYNAMIC PROPERTIES OF COHESIVE SOILS
Several researchers have published comprehensive reports
on dynamic s o i l properties of cohesive s o i l s (Sun at a l . , 1988;
Kokusho et a l . , 1982; Hardin and Drnevich, 1972a and 1972b).
This chapter reviews t h e i r main findings and outlines the key
factors a f f e c t i n g dynamic shear modulus and damping of cohesive
soils.
2.1 SHEAR MODULUS
Hardin and Black (1968) concluded that shear modulus of
soils i s a function of:
i / e f f e c t i v e octahedral normal stress

i i / ambient stress and v i b r a t i o n h i s t o r y
i i i / degree of saturation
i v / deviatoric component of ambient stress
v/ grain c h a r a c t e r i s t i c s and mineralogy

v i / amplitude of v i b r a t i o n
v i i / frequency of v i b r a t i o n
v i i i / secondary time e f f e c t s (aging) and magnitude of load
increment
i x / s o i l structure, and
x/ temperature.
4
For sands, the modulus at small s t r a i n s (10~ % or
J
less) i s primarily a function of void ratio and effective

confining stress, with an additional small aging e f f e c t . For
clays, the findings are s i m i l a r except that secondary time
effects (aging) and clay mineralogy also appear to be
important. The s i g n i f i c a n t factors a f f e c t i n g the shear modulus
of cohesive s o i l s are reviewed i n the following sections.
2.1.1 E f f e c t of Shear Strain Amplitude
It i s clear that the shear modulus of soils reduces

rapidly with increasing strain amplitude as indicated . by
F i g . 2.1. The modulus versus strain amplitude curves are
commonly expressed by a normalized modulus reduction curve i n
which modulus values at any s t r a i n are normalized by the small
s t r a i n modulus, Gmax or G . Q The value of Gjjj ax i s approximately
constant f o r strains below the order of 10~ 3
percent where the
deformations are l i n e a r e l a s t i c and recoverable. For higher
strains from about 10~ -1%,
2
the soil behaviour becomes
nonlinear e l a s t o - p l a s t i c and r e s u l t s i n irrecoverable permanent
deformations. Indeed, the shear modulus at 0.5% strain i s
typically only one tenth of the i n i t i a l small s t r a i n shear
modulus.
Unlike the modulus reduction curves f o r sands which show a
relatively small variation from one sand to another, the
5
1.2
c
Fig. 2 . 1 Normalized Modulus Reduction Curves f o r

Sands and Clays
6
modulus reduction curves f o r cohesive s o i l s have a much larger
scatter (Fig. 2.1). Therefore, c e r t a i n factors a f f e c t i n g shear
modulus - s t r a i n relationships f o r clays must be considered to
obtain a more usable modulus reduction r e l a t i o n s h i p .
Kokusho et a l . (1982) and Sun et a l . (1988) reported that
p l a s t i c i t y index, an i n d i r e c t measure of clay mineralogy, has a
marked influence on the form of the normalized modulus

reduction curves (Fig. 2.2). Higher p l a s t i c i t y clays show a
slower rate of modulus reduction as well as a gradual s h i f t of
the curve to the r i g h t .
2.1.2 E f f e c t of Void Ratio
It i s not possible to independently vary both the void

r a t i o and confining pressure f o r normally consolidated samples.
Hence, i t i s d i f f i c u l t to separately determine the e f f e c t of
these two parameters on the small-strain shear modulus.
However, a l i n e a r v a r i a t i o n of shear wave v e l o c i t y with void
r a t i o enables the void r a t i o e f f e c t to be examined (Hardin and
Richart, 1963). This observation allowed Hardin and Black
(1968) to show that a void r a t i o factor, F(e) , can be used i n
an equation of the form
Gmax = K*F(e)(P ) - (a ')

a
1 n
3 c
n
(2.1)
7
Shear Strain - percent
F i g . 2.2 E f f e c t of P l a s t i c i t y Index (PI) on Normalized

Modulus Reduction Curves of Cohesive S o i l s
(from Zen et a l . , 1978)
8
i n which K i s a dimensionless constant, 02c' ^ s
the e f f e c t i v e
isotropic confining pressure in the same units as the
atmospheric pressure (P ) and G^x, a and i n which the void r a t i o
factor i s
F(e) = (2.973 - e ) / ( l + e) 2
(2.2)
Although widely used, Equation 2.2 is based on tests on

laboratory-prepared k a o l i n i t e and been found to be applicable
only f o r low p l a s t i c i t y clays or r e l a t i v e l y stiff clays with
void r a t i o smaller than 1.5 (Ishihara, 1982).
For clays with high p l a s t i c i t y index, Marcuson and Wahls

(1972) defined the void r a t i o factor as
F(e) = (4.4 - e ) / ( l + e) 2
(2.3)
Undisturbed alluvial clay samples are more compressible

than laboratory samples on which Eqns. 2.2 and 2.3 are based.
To evaluate such s o i l s , Kokusho et a l . (1982) performed tests
on undisturbed samples of s o f t alluvial clay from Teganuma,
Japan, and proposed the following formula:
F(e) = (7.32 - e ) / ( l + e) 2
(2.4)
Equation 2.4 is based on tests using laboratory-prepared
bentonite clay i n which time and overconsolidation e f f e c t s were
9
not eliminated. I t applies to clays with void r a t i o s between
1.5 and 2.5. The PI of t h i s clay varied from 40 to 100 and the
void r a t i o values ranged from 1.5 to 4.0.
Equations 2.2, 2.3, and 2.4 are a l l of the same form. The
d i f f e r e n t constants i n each of the equations are derived from a
regression analysis of the respective data sets which also
provides the value of K i n Eq. 2.1. These equations are
plotted i n F i g . 2.3 f o r three d i f f e r e n t confining pressures.
The above three formulae give approximately the same shear
modulus f o r void ratios of about 1.5 but otherwise large
differences exist. Considerable judgement is therefore
required i n choosing the appropriate void r a t i o r e l a t i o n s h i p .
2.1.3 E f f e c t of Confining Stress
Increased confining stress results in higher shear

modulus. For equations employing a void ratio factor as i n
Eq. 2.1, the modulus exponent, n, f o r a v a r i e t y of clays ranges
from 0.5 to 0.6 (Kokusho et a l . , 1982, Marcuson and Wahls,
1978, Hardin and Black, 1968) . To overcome the uncertainty i n
choosing the correct void r a t i o factor, shear modulus i s often
simply reported as
Gmax - K ( P > ' ( » 3 c ' )

a
1 n n
( -5)
2
10
10 20 10 40
Void ratio . e
F i g . 2.3 Comparison of Three Experimental Formulas on

Modulus vs. Void Ratio Diagram
(from Ishihara, 1982)
11
in which K and n vary according to the type of clay. An
equation of the form of Eq. 2.5 may be convenient when the void
r a t i o of the clay i s not accurately known. However, both the
modulus m u l t i p l i e r (K) and the modulus exponent (n) vary with
clay type.
Weiler (1988) gathered data found i n the l i t e r a t u r e . on

shear modulus-confining pressure relationships for normally
consolidated, naturally-deposited clays. The shear modulus-
confining pressure equations f o r the respective data sets are
presented i n table and graph form i n F i g . 2.4. Also included
in Fig. 2.4 is the shear modulus-confining pressure
r e l a t i o n s h i p f o r Teganuma Clay (Kokusho et a l . , 1982).
It appears that the modulus m u l t i p l i e r values shown i n

F i g . 2.1 decrease with increasing p l a s t i c i t y index, while the
modulus exponent values increase. This is intuitively
s a t i s f y i n g since a more p l a s t i c clay would be expected to be
s o f t e r (lower modulus) than a lean clay, and should therefore
have a lower modulus m u l t i p l i e r . Since a higher plasticity
clay generally undergoes a more rapid void r a t i o reduction with
increasing confining pressure (higher compressibility), the
shear modulus increase (modulus exponent) would be expected to
be greater. The modulus exponent values f o r a l l natural clays
reported i n the literature fall i n the range 0.8-1.2. The
range i n shear modulus values f o r the various clays shown i n
F i g . 2.4 i s almost an order of magnitude. However, the trend
12
Q_
100000-
o
CD
CO
3
T>
O
10000-
100 500
Confining Pressure (kPa)
# CLAY PI Gmax REFERENCE
1. Gulf o f Alaska 14-15 372.1P * a

1 5
( 7
< 3 C ')- 8 5
OCR- 5 9
Weiler (1988)
2 AGS CL Clay 16-22 436.3P ' (a3c')• OCR' 16 84 27 n

a
3 Boston Blue Clay 19-23 248.2P a

, 1 4
(<7 3 C ') - 0CR-
86 60 II
4 San Fran. Bay Mud 43 164.6P - a

0 5
( 7
< 3 C
/
) - "OCR- 51 n
5 AGS CH Clay 32-38 126.2P -' (a ') ' OCR* 18 1 18 69 n

a 3C
6 Leda Clay 37-44 97 . 4P "' 0 8

((73c') 1- 0 8
OCR* 69 ll
a
7 Teganuma Clay 40-103 53.6P "'

a
0 8
(<7 3 C ')
1
- OCR'
08 59
Kokusho e t a l
(1982)
Fig. 2.4 Small Strain Shear Modulus Equations f o r Seven

Naturally Deposited Clays
13
i n the curves i s predictable based on the p l a s t i c i t y index of
the clay as suggested by the table i n F i g . 2.4.
2.1.4 Consolidation Stress History
The e f f e c t of overconsolidation on the small s t r a i n shear

modulus of n a t u r a l l y deposited cohesive soils i s to increase
the modulus as compared to the normally consolidated condition.
The shear modulus increase due to stress h i s t o r y i s often
expressed by an overconsolidation r a t i o (OCR) term incorporated
into Eq. 2.5 such that
Gmax = K ( P ) - ( a ' ) O C R
a
1 n
3 C
n i n
(2.6)
The table i n F i g . 2.4 presents equations of t h i s form f o r s i x

natural clays. The influence of OCR on s m a l l - s t r a i n shear
modulus f o r f i v e of the clays examined by Weiler (1988) i s
i l l u s t r a t e d i n F i g . 2.5. The slope of t h i s p l o t (m) i s the OCR
exponent. Almost a l l of the data f a l l s within a range of m of
0.3 to 0.7, with an average of about 0.6. Although there i s
some v a r i a t i o n between the various clays, no trend of OCR
exponent with clay p l a s t i c i t y i s evident.
However, i f an equation employing the void r a t i o factor i s
used, then
14
10 /
/
/
/
•
^SlopaH>.39
• X/
*
5 •
/
/
/
X
/
/
/
CDo
i
/
'' /*
X Boston Slut Clay
- * z
JO* f /
s
J7\
*
*
/
/ 0 Buton Blu* Clay liat
• /&
/ / /
* S F Bay Hud I13J

P 2 •
CD / ' S
/ ' /
/ ' s + A6S CL Clay (91
/ * X
AAfiS CH City (SI
- G u l f of Alaska Clays (IS
10 100
OVERCONSOLIDATION RATIO
F i g . 2.5 E f f e c t of Overconsolidation Ratio on the Small

Strain Shear Modulus of Five Naturally Deposited
Clays (from Weiler, 1988)
15
Gmax = K*F(e) (P ) ^ ^ ' ^ O C R *
a (2.7)
and k varies with p l a s t i c i t y index (PI) as shown i n Table 2.1.
2.1.5 Frequency Effect
Aggour et a l . (1987) applied random vibrations with

varying cut-off frequencies to study the e f f e c t of loading
frequency on the shear modulus reduction curves f o r cohesive
soils. From these t e s t s , i t was found that random loading with
d i f f e r e n t frequency content a f f e c t s the shear modulus reduction
curve ( F i g . 2.6). The s o i l modili i n the case of random
loading were found to increase with higher frequency content
and were lower than the moduli obtained from sinusoidal
testing. The differences increased with increasing strain
amplitude. However, the d i f f e r e n t frequency cut-off values are
higher than the frequencies i n the 0.1 Hz t o 30 Hz range of
i n t e r e s t f o r most earthquakes (Sun et a l . , 1988).
2.1.6 Secondary Time E f f e c t s
Several investigators have shown that time-dependency i s a
s i g n i f i c a n t parameter a f f e c t i n g the dynamic properties of s o i l s
16
Table 2.1 Values of k
PI k
0 0
20 0.18
40 0.30
60 0.41
80 0.48
>100 0.50
17
70
60
to
W • Sinusoidal
3
iH 50 • 0-50 Hz
3
•O O 0-100 Hz
o <D 0-500 Hz
X,
u 40 • 0-1000 Hz
to
0) & 0-10000 Hz
30
20
10 -5 10 "
4
io -3
ior 2
10 -1
rms s t r a i n %
F i g . 2.6 E f f e c t o f V a r y i n g Frequency Content on t h e Shear

Modulus v s . S t r a i n Curves f o r C o h e s i v e S o i l s
( a f t e r Aggour e t a l . , 1987)
18
(Hardin and Black, 1968; Marcusson and Wahls, 1972; Anderson
and Woods, 1976; Anderson and Stokoe, 1978; Kokusho et a l . ,
1982). The r e s u l t s of these investigations confirm that small
s t r a i n dynamic shear moduli of both sands and clays increase
with time of confinement under sustained pressures.
Results of these studies suggest that time-dependent
change i n shear modulus can be separated into a period of
primary behaviour and a period of secondary behaviour. The
primary behaviour has been shown (Fig. 2.7) t o coincide with
the period of primary consolidation (Kokusho et a l . , 1982;
Anderson and Stokoe, 1978).
During the long-term time effect the values of shear
modulus generally increase with the logarithm of time. Figure
2.7 indicates that t h i s secondary increase i s much larger than
that predicted from the secondary decrease i n void ratio
estimated by Eg. 2.1. A time or t h i x o t r o p i c effect i s
therefore seen as being responsible for t h i s behaviour.
The rate of secondary modulus increase has been quantified
by the r a t i o AG/Giooo ^ n
which A G i s the increase i n shear
modulus per logarithmic cycle of time, and G ^ Q O O i- s

the modulus
measured 1000 minutes a f t e r the s t a r t of primary consolidation.
Generally, the rate of secondary modulus increase becomes
greater with decreasing mean grain size (Afifi and Richart,
1973). However, Kokusho et a l . (1982) suggest that a
19
F i g . 2.7 T y p i c a l Time-Dependent I n c r e a s e i n Low Amplitude
Shear Modulus f o r C l a y a t Constant C o n f i n i n g
P r e s s u r e ( a f t e r Anderson and Stokoe, 1978)
20
relationship involving plasticity index (PI) may be more
logical since PI i s a parameter which r e f l e c t s the chemical
activity of clay minerals (Fig. 2.8). The rate of secondary
modulus increase becomes larger as PI increases, with values as
large as 25% per log cycle of time. Clearly, time e f f e c t s must
be considered when conducting laboratory tests for shear
modulus or when extrapolating laboratory tests to field
conditions.
2.2 DAMPING RATIO
Damping e x i s t s i n a l l vibrating systems. I t serves to

remove energy from the o s c i l l a t i n g system. The energy loss can
be determined from the v i b r a t i n g system response. This loss i n
energy r e s u l t s i n the decay i n amplitude during free v i b r a t i o n .
For steady state forced v i b r a t i o n , the energy loss i s balanced
by the energy input of the d r i v i n g force.
Damping i s commonly distinguished as being either viscous

or hysteretic i n nature. The damping i s said to be viscous
when the dissipated energy per cycle increases proportionally
with frequency. This i s the assumption i n determination of
damping from the amplitude decay method i n the resonant column
test. The damping i s said to be hysteretic when the dissipated
energy per cycle i s independent of the frequency of loading.
This i s the assumption when the damping i s determined from the
21
A Koolinita LMorcuscn
3 Bentonile Jetal.11972)
Kaoline Afifi etal.

Anderson etc. (1973)
etol. Zen et al.
Hommoku
(1976) Clay + Sand (1978)
^^7<p>Cp Teganuma
20 40 60 80 100
PLASTICITY INDEX Ip
Fig. 2.8 Relationship Between Rate of Modulus Increase and
P l a s t i c i t y Index (from Kokusho et a l . , 1982)
22
stress s t r a i n response curves. Details on damping calculations
are given i n Chapter 4.
The damping of a p a r t i c u l a r s o i l has been found to depend
primarily on:
i / the s t r a i n amplitude of v i b r a t i o n
i i / the ambient state of e f f e c t i v e stress

i i i / v i b r a t i o n frequency
i v / void r a t i o , and
v/ confining time
as well as other, less s i g n i f i c a n t factors. These variables

are e s s e n t i a l l y the same as those a f f e c t i n g the shear modulus
(section 2.1) and are discussed i n the following sections.
2.2.1 E f f e c t of Shear Strain Amplitude
There is not an abundance of data reported in the

l i t e r a t u r e on the v a r i a t i o n of damping r a t i o with shear s t r a i n
amplitude. The limited data suggest that the damping r a t i o
curves f o r cohesive s o i l s have a wider range than those f o r
sands. Much of the early work during the 1960's was summarized
by Seed and Idris (1970). Figure 2.9 shows the range in
damping values summarized by Seed and I d r i s as well as the
23
30
oH 1—i—i 111111 1—i—IIIIIII i—i—i 11111
0.001 0.01 0.1

Shear Strain (*)
F i g . 2.9 Range of Damping Ratios vs. Shear Strain f o r
Cohesive S o i l s
24
range i n damping curves i d e n t i f i e d by Kokusho e t a l . (1982)
from studies performed i n the 1970's and early 1980's.
I t i s quite l i k e l y that the smaller range i n damping r a t i o

presented by Kokusho et a l . (1982) r e f l e c t s the improvement i n
dynamic testing equipment and procedures since the early
studies i n the 1960's. Nevertheless, i t can be seen that the
damping r a t i o f o r cohesive s o i l s increases from a minimum value
less than 5% at a shear strain of about 0.01% to a value
greater than 8% at a shear s t r a i n of 1.0%.
I t i s well known that a strong c o r r e l a t i o n e x i s t s between

damping ratio and shear modulus; i n p a r t i c u l a r , the damping
r a t i o decreases as shear modulus increases. Hardin and Drnevich
(1972b) suggested the following simple equation to relate
damping r a t i o (D) to shear modulus:
D = D ax(l-G/G
m max ) (2.8)
where Dmax i - s t n e
maximum damping at large s t r a i n .
The e f f e c t of duration of confinement of cohesive s o i l s i s

to decrease the damping logarithmically as a function of time.
This behaviour has been observed i n laboratory-prepared clays
25
(Marcusson and Wahls, 1978) as well as naturally deposited
clays (Kokusho e t a l . , 1982; H a r d i n and Drnevich, 1972a).
Due t o the r e l a t i o n s h i p between damping r a t i o and shear
modulus (Eg. 2.8), i t i s reasonable that the damping ratio
should decrease with duration of confinement since shear
modulus increases with confining time (see section 2.1.6).
U n l i k e the two time-dependent stages o f b e h a v i o u r (primary and
secondary) of shear modulus, a consistent linear logarithmic
decrease o f the damping r a t i o e x i s t s w i t h no k i n k apparent at
the completion o f primary c o n s o l i d a t i o n ( F i g . 2.10).
The r a t e of damping r a t i o decrease has been found t o be as
h i g h as 25% p e r l o g a r i t h m i c c y c l e o f time for bentonitic clay
(Marcusson and Wahls, 1978).
2.2.3 Frequency Effects
A series o f random v i b r a t i o n tests with varying c u t - o f f
f r e q u e n c i e s were performed by Aggour e t a l . (1987) t o study the
effect of loading frequency on dynamic properties of clay.
These t e s t s suggest t h a t h i g h e r f r e q u e n c i e s a r e a s s o c i a t e d with
lower v a l u e s o f damping ( F i g . 2.11), though the chosen cut-off
f r e q u e n c i e s were w e l l above the frequency range o f i n t e r e s t f o r
most earthquakes (0.1-30 Hz). As w e l l , the damping curve f o r
conventional sinusoidal loading does not appear to f i t this
26
0.06 i 1 — i — I I I I
T 1—r
LEGEND
0.0S CHAMBER
PRESSURE
PSI
0 10
0 20
< 7 40
cr A 80
O 0.03
z
CL _o_ -O—
2 o
Q 0.02 -7-
0.01
i i i i i i i i i
6 8 10 20 40 60 8 0 100
L O G T RATIO
F i g . 2.10 T y p i c a l Time Dependent Decrease i n Damping Ratio

a t Constant E f f e c t i v e S t r e s s f o r C l a y
(from Marcuson and Wahls, 1978)
27
15
• Sinusoidal
A 0-50 Hz
13 O0-100 Hz
0 0-500 Hz
• 0-1000 Hz
11 & 0-10000 Hz
M
9
C
•H
O.
1
10-5 10 -4 io-3 10 -2 10 -1
rms s t r a i n %
F i g . 2.11 E f f e c t of Varying Frequency Content on the

Damping vs. Strain Curves f o r Clay (after Aggour
et a l . , 1987)
28
trend since i t l i e s below the damping curve with the highest
frequency content.
Hardin and Drnevich (1972a), i n t h e i r t e s t s on Lick Creek
S i l t , suggested that frequencies above 0.1 Hz have only a minor
e f f e c t on damping. However, t h i s finding requires more study
since i t i s based on only a few data points.
2.2.4 Confining Pressure
The e f f e c t of confining pressure on damping i n cohesive

s o i l s has not been conclusively established i n the l i t e r a t u r e .
Kokusho et a l . (1982) presented data on small s t r a i n damping
for clay with PI ranging from 40 to 100 which, despite large
v a r i a t i o n s , appear to show only a s l i g h t l y decreasing damping
with increasing confining pressure (Fig. 2.12). Furthermore,
the differences i n the damping f o r OCR values up to 15 were
found to be i n s i g n i f i c a n t , on the average.
Hardin and Drnevich (1972b) proposed that damping

decreases with the square root of confining stress, but the
large v a r i a t i o n i n damping values at low s t r a i n makes t h e i r
finding questionable.
29
.06
.04-
O
cr N.C-| rO.C IP OCR

o » S-1-2 103 5
o .02 A S-2-2 85 5
• S-4-4 59 10
• • S-5-2 50 10
<
o o • S-6-3 40 15
... O 9 S-8-2 40 10
ICO 200 300 400
CONFINING STRESS <ri ( k N / m )
a
F i g . 2.12 E f f e c t of Confining Stress, PI, and OCR on the

Small Strain Damping of Clay (from Kokusho et
a l . , 1982)
30
2.3 NEEDS FOR RESEARCH
I t i s well known that the shear s t r a i n amplitude affects

both the shear modulus and damping of cohesive s o i l s . However,
since the dynamic properties appear to be a complex function of
many variables, there i s considerable uncertainty i n choosing
the correct values of shear modulus and damping for a
particular problem. There is a need to establish more
thoroughly the manner i n which shear modulus and damping vary
with factors such as confining pressure, stress history,
loading frequency, and aging. The p l a s t i c i t y index may o f f e r a
guide since i t i s a r e f l e c t i o n of the clay chemistry.
This research addresses some of the aforementioned
concerns.
31
3. RESONANT COLUMN/TORSIONAL SHEAR APPARATUS
3.1 INTRODUCTION
The resonant column apparatus used i n t h i s study i s known

as a Drnevich Long-Tor Resonant Column Apparatus. I t was
purchased from S o i l Dynamics Instruments Inc. i n 1981 f o r use
in the Graduate Soil Mechanics Laboratory at the Civil
Engineering Department of UBC. The term Long-Tor denotes the
capability of the apparatus to vibrate specimens i n either a
longitudinal or torsional mode of v i b r a t i o n . For this
research, however, only the t o r s i o n a l aspect of the equipment
was used since the focus of t h i s study was primarily on shear
modulus and shear damping c h a r a c t e r i s t i c s . Young's modulus and
longitudinal damping are not examined herein as they are
usually of secondary importance i n analyses of earthquake
response.
3.2 RESONANT COLUMN APPARATUS
Several modifications were made to the apparatus i n order

to allow f o r testing of saturated samples and to enable
torsional shear tests to be performed. The basic theory
behind t h i s equipment i s introduced i n Chapter 4.
32
The resonant column apparatus i s bolted securely to a
heavy s t e e l base which stands 80 cm high. This i s necessary to
ensure that the base of the sample i s relatively free of
ambient v i b r a t i o n (Drnevich Long-Tor Resonant Column Apparatus
Operating Manual, 1981). The apparatus i n p o s i t i o n f o r either
resonant column or torsional shear testing i s shown i n
Fig. 3.1. Except f o r the bottom aluminum pedestal or platen,
a l l components are made of s t a i n l e s s s t e e l .
3.2.1 Confining Chamber
The confining chamber encloses the s o i l sample and the

mechanical components of the resonant column device so that a
confining pressure can be applied to the s o i l sample. The
chamber, which i s designed f o r a maximum safe working pressure
of 700 kPa, e s s e n t i a l l y consists of a cast a c r y l i c tube which
i s secured between the bottom base plate and top chamber l i d by
four long s t a i n l e s s s t e e l b o l t s . The chamber l i d contains a
port connecting the chamber t o a regulated a i r pressure supply
and also houses the connections which j o i n the i n t e r i o r and
exterior electronic component cables. The sample pedestal,
which i s attached t o the r i g i d bottom chamber base plate, i s
designed to accommodate 3.5 cm diameter samples.
33
SUPPORT < SUSPENDING
STAN-D
SPRING
RVOT GUIDE
BRACKET
^•BASE PLATE DRAINAGE LINES
F i g . 3.1 Resonant Column/Torsional Shear Apparatus
34
3.2.2 Drive Mechanism
The 1.27 kg top drive platen consists of the top sample

cap, four permanent magnets, one horizontally and one
vertically oriented accelerometer, and a support rod which
allows a v e r t i c a l displacement transducer core and suspending
spring to be attached. This counterbalancing spring i s
suspended from a bracket at the top of a support stand, and i s
necessary to counteract the weight of the top drive platen
which would otherwise be acting as an a x i a l force on the
specimen. Isotropic confining pressures are thus maintained.
During resonant t e s t i n g , the top drive platen i s essentially
f l o a t i n g and able to rotate i n the horizontal plane so that
forced t o r s i o n a l e x c i t a t i o n can be applied at the top of the
specimen.
The drive mechanism consists of four drive c o i l s which are

encircled by the permanent magnets located at the four corners
of the top drive platen (Fig. 3.2). This configuration i s
superior to that found i n some other resonant column devices i n
that i t can accommodate fairly large torsional rotations
without mechanical interference between the magnets and c o i l s .
Such a configuration allows t e s t i n g at a much wider range of
strains (approximately .0005-2%) than i s possible f o r some
other device types. In addition, t h i s configuration can allow
for approximately a 1.5 cm decrease i n sample height during
consolidation or t e s t i n g .
35
::v//////A
TORSIONAL
ACCELEROMETER
777777A
DRIVE
MAGNET
COIL
F i g . 3.2 Plan View of Resonant Column Top Drive Plate
36
The o s c i l l a t i n g t o r s i o n a l r o t a t i o n force i s created by the
i n t e r a c t i o n of magnetic fields; specifically, the i n t e r a c t i o n
of magnetic f i e l d s generated by the s i n u s o i d a l l y varying drive
c o i l input current with those caused by the permanent magnets
of the drive plate.
Complete coupling between the specimen and the end platens

is ensured by using bronze porous discs with embedded
razorblade vanes which protrude about 1.5 mm. The discs
themselves are securely attached to the top and bottom platens
by means of machine screws.
3.2.3 Torsional Accelerometer
The t o r s i o n a l motion of the top platen i s monitored by a

horizontally oriented accelerometer (Columbia Research Labs,
Model 200-1-H) bolted to the top platen. The accelerometer i s
mounted at a distance of 3.16 cm from the axis of r o t a t i o n . An
acceleration s e n s i t i v i t y range of 0.01 g-lOk g i s s p e c i f i e d by
the manufacturer.
37
3.2.4 Height-Change Measurement
Changes i n specimen height during consolidation and
shearing are determined using a l i n e a r v a r i a b l e displacement
transducer (LVDT-Schaevitz 300-HR) which i s mounted between the
top drive plate and the suspending spring (Fig. 3.1). The LVDT
c o i l casing i s mounted on the support stand which provides a
fixed reference f o r the r e l a t i v e displacements of the LVDT
core.
3.2.5 Volume-Change and Pore Pressure Monitoring Equipment
Provision f o r drainage and pore pressure measurement i s

made by means of two drainage l i n e s which connect to the bottom
pedestal; no drainage i s provided through the top end of the
sample. The other ends of the drainage l i n e s are attached to a
volume change device which consists of several valves and a
volume-calibrated pipette through which a back pressure can be
applied t o the sample. This i s precisely the same system
commonly used for t r i a x i a l t e s t s .
Pore pressures are measured with a 150 p s i capacity pore

pressure transducer (Data Sensors Inc. PB519A) mounted on the
volume change device.
38
3.2.6 A i r Pressure Transducer
The confining pressures applied t o the sample are measured
with a 200 p s i capacity a i r pressure transducer (Statham
TP-A-1064-EX) mounted on the confining chamber lid.
3.3 RESONANT COLUMN ELECTRONIC COMPONENTS
The e l e c t r o n i c components used f o r resonant column t e s t i n g

are described i n this section, except f o r the pressure
transducers, accelerometers, and LVDT which are described i n
the preceding section as part of the resonant column apparatus.
A wiring schematic i s shown i n F i g . 3.3. The switch and
control boxes enable the operator to switch between t o r s i o n a l
and longitudinal modes of v i b r a t i o n as well as channelling the
accelerometer output and drive coil input currents t o the
oscilliscope.
3.3.1 Function Generator
The sinusoidal signal which i s input t o the drive c o i l s i s

generated by an EXACT Model 508 Function Generator. It is
capable of generating a wide range of frequencies (0.001 Hz-
20,000 Hz). The logarithmic frequency scanning feature proved
to be convenient as i t allowed f o r rapid location of the
39
RVDT Output RVDT Excitation
Tors. Power Input
Tors. Accel. Out.

LVDT
READOUT
RESONANT
COLUMN
TORSIONAL
APPARATUS CHARGE
AMPLIFIER
OSCILLI- OSCILLI-
SCOPE 1 SCOPE 2
(X-Y)
VERT. HOR.
Y-IN X-IN
SWITCH
BOX
DIGITAL
VOLTMETER
CONTROL Tors. Cur. Out.
BOX
TORSIONAL DC POWER
POWER SUPPLY
AMPLIFIER
DIGITAL
SIGNAL FREQUENCY
GENERATOR METER
F i g . 3.3 Resonant Column/Torsional Shear Wiring

Schematic
40
resonant frequency. For low amplitude t e s t s , the output l e v e l
of the function generator was sufficient to power the drive
coils. High amplitude and t o r s i o n a l shear t e s t i n g required an
amplifier to magnify the signal to the s p e c i f i e d amplitude.
3.3.2 Signal Amplifier
For high amplitude resonant column t e s t i n g , the sinusoidal

output s i g n a l from the function generator i s amplified by a
power amplifier (Phase Linear 300 Series Two) before being
input to the drive coils. The frequency response of the
amplifier is from 10Hz-190kHz necessitating the use of a
d i f f e r e n t amplifier f o r t o r s i o n a l shear tests which are run at
much lower frequencies.
3.3.3 Charge Amplifier
The t o r s i o n a l accelerometer output signal i s conditioned

by a charge amplifier (Columbia model 4102) p r i o r to being
measured and recorded. A transducer sensitivity control
permits balancing of the amplifier to the accelerometer
sensitivity.
41
3.3.4 Storage O s c i l l i s c o p e
A two channel d i g i t a l storage o s c i l l i s c o p e (Nicolet N 310)

is used to monitor the sinusoidal drive coil input and
accelerometer output. The X-Y feature of the o s c i l l i s c o p e
provides an on-screen display of one channel as the X-ordinate
and the other channel as the Y-ordinate. This y i e l d s an
e l l i p s e - known as a Lissajous figure - when both channels have
sinusoidal signals. The resonant condition of the v i b r a t i n g
column can be r e a d i l y i d e n t i f i e d because the on-screen figure
rotates to a v e r t i c a l position and stretches as the column
comes to resonance.
The storage capability of the o s c i l l i s c o p e was used

primarily to capture the decay i n accelerometer response during
free v i b r a t i o n from which damping was determined.
3.3.5 Frequency Counter
Resonance frequencies are read from the frequency counter

(FLUKE 1900A) when the column comes to resonance - as
determined from the Lissajous figure on the oscilliscope
screen. The frequency counter displays frequency values to
0.1 Hz accuracy.
42
3.3.6 D i g i t a l Multimeter
At resonance, a d i g i t a l multimeter (Fluke 8000A) i s used
to read the rms voltages associated with the drive c o i l input
current and the conditioned accelerometer output current. The
multimeter i s also useful i n checking that the maximum current
capacity of the drive coils (about 1 amp) i s not exceeded
during high amplitude t e s t i n g .
3.3.7 LVDT Readout Unit
An LVDT readout u n i t (Schaevitz LPM-210) converts the LVDT

output signal directly into a relative displacement
(millimeters) i f the LVDT i s zeroed p r i o r to t e s t i n g . Seperate
phase and gain adjustments on the readout unit enable the unit
to be c a l i b r a t e d .
3.3.8 S t r a i n Indicator
The pore pressure and a i r pressure transducer outputs are

read with a s t r a i n indicator box (Budd P-350). This device
requires manual balancing with a d i a l i n order to obtain a
reading. The output readings are given i n units of microstrain
and must be m u l t i p l i e d by the respective pore pressure or a i r
43
pressure transducer calibration factors f o r a conversion to
units of pressure.
The manual balancing procedure was simple for s e t t i n g c e l l

pressures and back pressures f o r consolidation. However,
during high amplitude t e s t i n g i n which pore pressure changes
can be rapid, the manual balancing and recording procedure
required considerable s k i l l since simultaneous adjustments of
the input current frequency were necessary to keep the column
i n resonance.
3.4 TORSIONAL SIMPLE SHEAR EQUIPMENT
Torsional shear testing employed essentially the same

equipment used f o r resonant t e s t s (section 3.2, 3.3) although
some apparatus modifications and additional electronic
equipment were required. In these tests, the t o r s i o n a l
accelerometer and charge amplifier were not required. However,
since resonant column t e s t i n g was always run i n conjunction
with the t o r s i o n a l shear tests, the full complement of
equipment was needed. A description of the required additional
equipment follows.
44
3.4.1 Rotary Transducer
A rotary differential displacement transducer (RVDT,
Pickering & Co., Inc., Model 23384) i s used f o r measuring
r o t a t i o n a l displacements during t o r s i o n a l shear t e s t s . Since
the angular acceleration during these lower-frequency tests i s
below the s e n s i t i v i t y of the t o r s i o n a l accelerometer, the
accelerometer could not be used to calculate torsional

displacements.
The RVDT has a range of motion of +10° and a maximum non-

linearity of 0.01%. By using a d i g i t a l low-pass f i l t e r to
eliminate high frequency noise from the recorded RVDT signal,
i t was found that angles as small as 0.018 degrees (typically
0.005% strain) could be accurately and repeatably measured.
C a l i b r a t i o n of the RVDT f o r such small rotations was done by
mounting the RVDT on the axis of rotation of a highly accurate
theodolite.
A s p e c i a l guide bracket and RVDT casing (Fig. 3.1) were

needed t o allow the RVDT t o move down with the top platen
during specimen consolidation. This was achieved by using a
fixed cross-beam with a carefully machined cylindrical bore
containing two v e r t i c a l guide s l o t s i n which the guide pins of
the RVDT casing could descend as the sample consolidated. The
guide s l o t s and pins enabled the RVDT to remain aligned with
the sample axis and at the same time constrained the RVDT
45
casing from rotating thereby providing a fixed reference f o r
the rotating RVDT core.
A RVDT was used instead of proximity sensors since the

RVDT i s able t o measure a larger range of motion (0 t o +10°)
and does not require such a careful setup procedure.
3.4.3 DC Power Supply
The regulated 6 v o l t excitation required

+ by the RVDT was
supplied by a DC tracking power supply (Leader model LPS-151).
3.4.5 Power Amplifier
In order t o supply the high voltage and d r i v i n g current

needed during torsional shear tests, a DC t o 20 KHz power
amplifier with a 50 v o l t , 1.0 amp peak output was employed (HP
Model 6824A) . The maximum 1.0 amp output i s p r e c i s e l y the
l i m i t i n g current of the t o r s i o n a l drive c o i l s .
The sinusoidal driving signal was produced by the Exact

Model 508 Function Generator which was then amplified and
adjusted by means of a variable gain control on the power
amplifier. The amplified output signal has v i r t u a l l y no phase
shift.
46
3.4.6 Storage O s c i l l i s c o p e
In addition to the storage o s c i l l i s c o p e used f o r resonant

column tests, a second digital oscilliscope was used f o r
t o r s i o n a l shear t e s t s . This second o s c i l l i s c o p e (Nicolet 4094)
is equipped with two disk drives which allow storage of
recorded signals on computer disks f o r subsequent conditioning
and analysis. This feature was used f o r recording the drive
coil input and RVDT output signals from which stress-strain
loops are determined. Also, the second storage o s c i l l i s c o p e
was used occassionally during resonant column tests to check
and compare strains calculated from the t o r s i o n a l accelerometer
response with those determined from the RVDT.
47
4. THEORETICAL BACKGROUND
The dynamic shear modulus and damping can be determined
either from resonant t e s t i n g or from slow t o r s i o n a l testing.
The specific methods and theory f o r each type of t e s t are
presented i n t h i s chapter.
4.1 SHEAR STRESS AND STRAIN DISTRIBUTION
The v a r i a t i o n of shear displacement i n the s o i l specimen

as a r e s u l t of forced t o r s i o n a l rotation of the top cap i s
shown i n F i g . 4.1. The t o r s i o n a l displacement v a r i e s from zero
at the fixed base to a maximum of r m a x at the top of the
sample. A l i n e a r v a r i a t i o n of displacement from the top t o the
bottom of the sample implies that there i s no vertical
variation i n shear strain. This has been experimentally
verified (Drnevich, 1972a). Radially, displacements vary from
zero along the longitudinal axis of the specimen t o a maximum
at the sample radius. Therefore, there e x i s t s a radial strain
gradient since the shear s t r a i n (7) at any point depends on the
distance, r , from the center of rotation according to:
7 = r*/l (4.1)
where 1 i s the sample length
48
49
9 i s the angular displacement
The fact that stress and s t r a i n vary with radius i n such a

specimen does not create a serious error. Hardin and Drnevich
(1972a) reported no s i g n i f i c a n t differences i n the behaviour of
hollow and s o l i d c y l i n d r i c a l samples when compared on the basis
of average s t r a i n (7 v) ^
a
n
the specimen. I t can be e a s i l y
shown that the average s t r a i n i s given by
7av = 2/3*(r m a x t9 m a x /l) (4.2)
In order to avoid the confusion a r i s i n g from inadequately

defined shear s t r a i n s found i n the l i t e r a t u r e , i t should be
noted that a l l resonant column and t o r s i o n a l shear strains
reported herein are based on the average shear s t r a i n defined
by Eq. 4.2. Shear strains are reported as s i n g l e amplitude
values. These are the average strains determined a t peak
rotational displacement with respect to the o r i g i n a l null
position. Although Eq. 4.2 i s s t r i c t l y applicable only i n the
e l a s t i c range where a l i n e a r r e l a t i o n s h i p e x i s t s between stress
and strain, no attempt has been made i n this research to
determine a nonlinear equivalent radius. Ni (1987) reported
that an equivalent radius of 3 / 4 * r m a x may be appropriate f o r
shear s t r a i n s up to 0.1 percent.
50
4.2 RESONANT COLUMN TEST
The determination of dynamic soil properties from the
resonant column t e s t i s based on l i n e a r v i b r a t i o n theory. The
fundamental aspects of the theory and the equations used t o
calculate shear modulus and damping are outlined i n the
following sections.
4.2.1 Linear Vibration Theory
The analysis of the resonant column test i n terms of

l i n e a r v i b r a t i o n theory was given by Hardin (1965), Drnevich
(1967), and Ni (1987). The a n a l y t i c a l solution assumes a
Kelvin-Voigt soil model having constant stiffness, G, and
constant viscous damping, c (amplitude independent). Also,
i t i s assumed that the shear s t r a i n does not vary with height
in the specimen. The equation of motion f o r torsional
v i b r a t i o n of the resonant column i s :
a e/at
2 2
= G/ P (a e/ax )
2 2
+ c/ P (a e/ax at)
3 2
(4.3)
or e = v * + i/*
tt s
2
xx xxt
where V s = shear wave v e l o c i t y = (G//>) 0,5
9 = angle of twist, p = mass density of s o i l
51
v = dynamic v i s c o s i t y = c/p
with subscripts x and t corresponding to derivatives with

respect to length and time, respectively.
Using the separation of variables procedure, and by
applying the appropriate boundary conditions, the frequency
equation of motion i s obtained:
I/I Q = r>l/V )tan (col/V )

s s (4.4)
where I Q represents the mass moment of i n e r t i a of the top cap

and drive plate, I i s the mass moment of i n e r t i a of the
assembled s o i l column, and u i s the undamped natural frequency.
The frequency equation (Eq. 4.4) can be routinely used to

calculate the shear wave v e l o c i t y of the e l a s t i c homogeneous
s o i l column. The equation applies to the case of steady-state
vibration and i s also valid f o r the case of decaying free
vibration. While Eq. 4.4 i s rigorously correct only f o r s o i l
exhibiting linear visco-elastic behaviour (strains less than
about 0.001%), much of the non-linear behaviour can be
described by performing the linear dynamic analysis
incrementally at various strains thereby approximately
establishing the non-linear s t r a i n amplitude dependent dynamic
s o i l properties.
52
4.2.2 Shear Modulus
The methods of dynamic modulus and damping determination

are covered by Drnevich et a l . (1978) and the d e t a i l s of the
apparatus calibration procedure are given in the resonant
column operating manual (Soil Dynamics Instruments Inc., 1981).
Knowing the resonant frequency of the s o i l column, sample

length, and mass moment of i n e r t i a of the sample, the shear
wave v e l o c i t y can be calculated from the frequency equation
(Eq. 4.4). A system c a l i b r a t i o n procedure gives the value of
the mass moment of i n e r t i a ( I ) of the top cap and drive plate.
Q
Solution of Eq. 4.4 was f a c i l i t a t e d through the use of a

Fortran data reduction computer program modified from Drnevich
et al. (1978). Nomographs for solving Eq. 4.4 are also
available and were used to check the computer generated values.
Shear modulus ( G m a x ) can be obtained from the shear wave

velocity (V )
s since they are related through the theory of
elasticity:
Gmax = PVS
2
(4-5)
53
where p i s the s o i l density. The shear modulus determined from
the resonant column t e s t i s the secant modulus since only the
points of peak response are measured.
4.2.3 Damping Ratio
As mentioned i n section 4.2.1, the theory underlying the
resonant column t e s t assumes a viscous dashpot to represent the
soil damping. Two methods can be employed to determine the
damping r a t i o : the amplitude-decay method and the steady state
method.
4.2.3.1 Amplitude Decay Damping
Energy loss i n a freely vibrating system results in a
decaying v i b r a t i o n amplitude. The response of a single degree
of freedom (linear) s o i l column i s described by the solution of
Eg. 4.3:
0(x,t) = C e _ D n t
s i n ( w t + <£) sin(«x/V )
d s (4.6)
where 8 = r o t a t i o n a l angle
C = a constant
D = damping r a t i o = cw/2G
4> = phase angle
= damped natural frequency = (1-D ) * 2 0 5
n = natural frequency
54
Considering the ratio of two successive peaks of vibration
gives:
'n/'n+l - e D w T
(4.7)
where T i s the period of v i b r a t i o n =
The logarithmic decrement, d, i s defined as the natural
logarithm of Eq. 4 . 7 :
d = ln(0n/0 n+1 ) = DOT
= 2wOD/w(j
= 2wD/(l-D ) * 2 0 5
(4.8)
Rearranging the terms i n Eq. 4 . 8 gives the damping r a t i o which
can be r e a d i l y determined from the free v i b r a t i o n response:
D = [d /(4w
2 2
+ d) ]
2
(4.9)
4.2.3.2 Steady State Damping
During steady state resonant vibrations the effect of
transient vibrations becomes i n s i g n i f i c a n t , and the damping can
be determined from the system resonant response. The dynamic
amplification of the viscously damped system at resonance i s
given by:
55
X/X Q = 1/(2D) (4.10)
where X i s the peak dynamic displacement, X Q i s the peak s t a t i c
displacement, and D i s the viscous damping r a t i o . The damping
can be calculated using Eq. 4.10 and the magnification factor
(X/X ) which
0 i s determined from the system response and the
apparatus constants.
4.3 TORSIONAL SHEAR TEST
The resonant column device can also be used as a t o r s i o n a l

shear t e s t device by applying a slow sinusoidal torque to the
sample cap and measuring the resulting rotation angle.
Apparatus modifications and additional equipment required to be
perform such tests are described i n section 3.4. The torque-
twist response can be r e a d i l y converted to a s t r e s s - s t r a i n loop
from which shear modulus and visco-elastic damping can be
determined. The methods and calculations involved in
determining these parameters from the t o r s i o n a l shear t e s t have
been given previously by Isenhower (1979) and Ni (1987), and
are b r i e f l y outlined below.
56
4.3.1 Shear Modulus
The shear modulus i s measured as the secant shear modulus
joining the end points of the stress-strain hysteresis loop
(Fig. 4.2). Mathematically, the secant shear modulus is
expressed by:
G = r/ 7 (4.10)
The shear stress, which i s determined from the applied

torque, i s based on the theory of e l a s t i c i t y of a c i r c u l a r bar
in torsion. Assuming a linearly varying shear stress with
radius, the average shear stress ( r g ) i s given by: aV
'avg = r T / J e q (4.11)
where T i s the applied torque, r q i s the equivalent radius e
(2/3r m a x ) , and J i s the polar moment of i n e r t i a given by:
J = jrr /4 4
(4.12)
The applied torque i s determined from the applied voltage to

the drive c o i l s , V^, and the l i n e a r torque c a l i b r a t i o n factor,
Kf Thus, the average shear stress r e l a t i o n i s :
r
avg ~ *eqKtVt/J ( - >
4 13
57
Shear
Stress
Fig. 4.2 Secant Modulus and Hysteretic Damping Ratio From

Torsional Shear Test (from Isenhower, 1979)
58
The shear s t r a i n ( 7 ) , on the other hand, i s calculated from
Eq. 4.2 with the rotation angle determined from the rvdt
c a l i b r a t i o n factor, K r (radians/volt), m u l t i p l i e d by the RVDT
output voltage ,Vr:
Teq - r e q K V /l
r r (4.14)
The limitation of non-linear soil behaviour i n the

resonant column discussed i n section 4.1 s i m i l a r l y applies t o
equations 4.13 and 4.14. That i s t o say, the equations are
s t r i c t l y applicable i n the low range of s t r a i n (0.001% or less)
where s o i l behaviour i s e l a s t i c . As such, both the stress and
the strain defined by equations 4.13 and 4.14 tend t o be
s l i g h t l y underestimated a t high s t r a i n s .
4.3.2 Damping Ratio
The h y s t e r e t i c damping r a t i o can be determined from the

stress-strain hysteresis loop for a cycle of loading
(Fig. 4.2). By r e l a t i n g the work done (hysteresis loop area),
W(j, t o the stored elastic energy (area of t r i a n g l e ) , W,
s a
simple expression f o r damping can be developed. For a single
degree of freedom system i t can be proven that
Wd = 2wDkx 2
(4.15)
59
where D i s the damping r a t i o , k i s the Kelvin-Voigt spring
constant, and x i s the peak displacement. As well, i t is
e a s i l y shown that the stored e l a s t i c energy i s
Ws = kx /2 2
(4.16)
Combining equations. 4.15 and 4.16 and solving for the damping
term r e s u l t s i n the damping expression
D = W /(4 W )
d W S (4.17)
Equation 4.17 enables the c a l c u l a t i o n of damping for any cycle

of loading from the measured s t r e s s - s t r a i n response. The
disadvantage of Eq. 4.17 i s that i t provides a method of
calculating damping which requires the system to be at
resonance. However, since the e f f e c t of frequency on damping
i s r e l a t i v e l y minor (Ni, 1987), the method can be extended to
other loading frequencies.
60
5. TESTING PROCEDURE
5.1 RESONANT COLUMN TEST
The procedure followed to measure dynamic s o i l properties

from the resonant column was similar to that outlined in
Drnevich et a l . (1978) and Novak and Kim (1981). Since dynamic
soil properties are significantly dependent on the testing
procedures and methods used, a summary of the testing
procedures used i n t h i s research i s given i n the following.
The setup of RCTS tests entailed a number of steps.

First, the apparatus pore pressure system was saturated, the
specimen prepared, and the entire system assembled. At t h i s
point, i n i t i a l readings and baselines were taken of the various
monitoring equipment.
The sample setup procedure i s similar to that used i n

conventional t r i a x i a l testing, though two notable differences
exist. For one, filter paper strips were used around the
e n t i r e surface of the sample to f a c i l i t a t e rapid consolidation.
As well, double silicone-greased membranes were used to l i m i t
a i r d i f f u s i o n (see section 5.3.2).
Prior to consolidation, the pore water pressure
coefficient, B, was checked using Skempton's equation
61
B=Au/A<73c'' where Au i s the pore pressure increase. The sample
was assumed to be saturated i f the value of B was greater than
0.95.
Consolidation was i n i t i a t e d by opening the drainage valves

connecting the sample to the volume change device. To monitor
the consolidation, readings were frequently taken of the water
level i n the volume change device as well as of the LVDT
output. During the consolidation period, low amplitude
v i b r a t i o n s were applied to the specimen to determine the change
in shear modulus and damping with time which can be very
s i g n i f i c a n t i n cohesive s o i l s .
For a l l t e s t s , consolidation was allowed to proceed at

l e a s t u n t i l the end of primary consolidation which was always
l e s s than 1000 minutes when r a d i a l filter paper drains were
used. Then, the drainage valves were closed and pore pressures
i n the sample were allowed to equalize before high amplitude
dynamic t e s t i n g was performed.
The high amplitude testing sequence commenced at a low

s t r a i n amplitude (about 0.001 percent) and progressed to a high
strain level (usually 0.5 to 1 percent). For some t e s t s , pore
pressures were monitored during and a f t e r each l e v e l of dynamic
t e s t i n g - commonly conducted at twice the s t r a i n amplitude of
the previous increment.
62
The procedure f o r conducting a high amplitude t e s t was to
f i r s t increase the drive c o i l input power to the pre-selected
level. Then, the frequency of the sinusoidal input current was
varied until resonance was established. Since the ratio
between power and resonance frequency is not linear, a
simultaneous adjustment f o r both the power and input frequency
is necessary to achieve resonance at a particular strain
amplitude (Novak and Kim, 1981). Resonance was i d e n t i f i e d by
adjusting the frequency to obtain a v e r t i c a l l y oriented e l l i p s e
on the x-y oscilliscope display of input current versus
accelerometer output. At t h i s time readings were taken of the
frequency, input signal voltage, output signal voltage, and
lvdt output.
Damping by the amplitude decay method was determined from

the decaying free vibrations measured by the torsional
accelerometer response a f t e r power was cut o f f to the drive
coils. This usually required the capture of at l e a s t twenty
cycles of accelerometer response on the d i g i t a l oscilliscope.
The entire procedure described above was performed f o r each
s t r a i n l e v e l of a high amplitude t e s t .
Lastly, to minimize the e f f e c t of the number of cycles on
the dynamic soil behiour, the duration of resonance was
generally maintained at about 60 seconds.
63
5.2 TORSIONAL SHEAR TEST
Since t o r s i o n a l shear tests could be performed on the same

samples as resonant column t e s t s , the sample setup procedure
was i d e n t i c a l . The t e s t i n g procedure, however, was e n t i r e l y
d i f f e r e n t since the theory underlying the t o r s i o n a l shear t e s t
i s psuedo-static whereas the resonant column t e s t i s based on
l i n e a r v i b r a t i o n theory.
Contrary t o the resonant column t e s t , the stresses and

strains f o r the t o r s i o n a l shear test were measured in a
r e l a t i v e l y d i r e c t manner. That i s , the stress and s t r a i n were
determined from a l i n e a r c a l i b r a t i o n f o r torque and r o t a t i o n a l
displacement, respectively.
P r i o r to commencing a torsional shear t e s t sequence the

shear modulus at small s t r a i n (about 0.001%) was measured i n
the resonant column mode. This was done so that the shear
modulus from the resonant column test sequence which
immediately followed the t o r s i o n a l shear testing could be
corrected f o r pre-straining.
First, the testing frequency was selected and the

amplitude of the drive c o i l input current was adjusted to give
the estimated torque required. From a t r i a l run i t was found
that the maximum frequency at which the t o r s i o n a l shear t e s t
could be run without significant dynamic effects was about
64
1 Hz; therefore, the torque calibration was v a l i d only f o r
psuedo-static conditions. A f t e r the frequency and amplitude
were selected, the t o r s i o n a l shear t e s t was conducted with the
sinusoidal drive c o i l input and RVDT output d i g i t a l l y recorded
for the v i b r a t i o n cycles of i n t e r e s t . These two data records
were l a t e r modified to give the s t r e s s - s t r a i n response by means
of a digital data processing program (VU-POINT, Maxwell
Laboratories, Inc.) on a personal computer. The progam also
allowed f o r integration of the s t r e s s - s t r a i n loops thereby
enabling quick c a l c u l a t i o n of the hysteretic damping r a t i o .
The pore pressure response was monitored both during and

after the t o r s i o n a l shear testing, and significant pore
pressure changes were recorded. Equalization of pore pressures
was allowed before the next increment of t e s t i n g was performed.
Commonly, t e s t i n g was done at three or four shear s t r a i n l e v e l s
before the maximum capacity of the drive c o i l s was reached.
After the strain-dependent response was determined from

the t o r s i o n a l shear test, the strain-dependent response was
also determined using the resonant column t e s t which allowed
t e s t i n g to a somewhat higher maximum s t r a i n .
65
5.3 PORE PRESSURE MEASUREMENT CONSIDERATIONS
5.3.1 Pore Pressure Measurement
It i s well known that sufficient time i s required to

measure pore pressures i n laboratory tests on f i n e grained
soils because of the time l a g involved i n r e g i s t e r i n g pore
pressure changes - even when small volume compliance
transducers are used (Bishop and Henkel, 1962, Sangrey et a l . ,
1978) . This i s an important consideration i n the case of
resonant column t e s t i n g since the specimen i s subjected to very
fast rates of c y c l i c s t r a i n . However, no pore pressures are
developed until a sufficiently high strain amplitude is
surpassed.
The r a d i a l d i s t r i b u t i o n of stress within the sample i s not

uniform (see section 4.1), r e s u l t i n g i n a nonuniform pore
pressure d i s t r i b u t i o n within the specimen. Thus, the maximum
value of pore pressure occurring within the sample w i l l not be
measured, even i f enough time i s allowed f o r pore pressure
equalization. The problem of high pore pressure gradients w i l l
only be important i n high shear s t r a i n amplitude tests since
shear-induced pore pressures are presumably not s i g n i f i c a n t at
very low s t r a i n s .
66
During high s t r a i n amplitude staged t e s t s , pore pressures
were allowed to come to equilibrium before the next l e v e l of
strain was applied. This was done to minimize the
nonhomogenous state of stress i n the sample r e s u l t i n g from pore
pressure gradients generated during high amplitude t e s t i n g .
5.3.2 A i r Diffusion E f f e c t s
Several researchers have reported the problem of a i r

d i f f u s i o n through the membrane and i t s e f f e c t on pore pressure
measurements (Novak and Kim,1981, Sangrey et a l . , 1978,
Marcuson and Wahls, 1978, Pollard et a l . , 1977). Due to the
presence of an air-water interface within the t e s t i n g chamber,
air i s forced into solution with the water and progressively
d i f f u s e s through the rubber membrane. A f t e r passing through
the membrane the a i r can come out of solution i n the form of
a i r bubbles r e s u l t i n g i n erroneous volume change readings and
inaccurate excess pore pressure measurements. Pollard et a l .
(1977) examined the pore pressure change within a triaxial
sample and concluded that the process could be described by
Ps - P c [ l - e - < ~
d T T i )
] (5.4)
where
67
specimen pore pressure
c e l l pressure, and
time.
The value d i s an empirical diffusion parameter which i s

approximately constant but varies somewhat with pressure,
degree of c e l l f l u i d saturation by a i r , and several secondary
factors (Sangrey et a l . , 1978). The variable T^ i s an i n i t i a l
reference time depending primarily on the degree of a i r
saturation i n the c e l l and pore f l u i d s . Pore pressures are not
affected by a i r diffusion until time T^ when the p a r t i a l
pressure gradient of a i r becomes sufficient to establish
s i g n i f i c a n t d i f f u s i o n through the membrane. However, the pore
pressure errors become prominent as continued a i r d i f f u s i o n
causes the pore pressures to increase.
Figure 5.1 shows the r e s u l t of an a i r d i f f u s i o n t e s t . The

t e s t was performed i n a standard t r i a x i a l c e l l using p r e c i s e l y
the same sample dimensions and setup d e t a i l s as used i n a l l
resonant column and t o r s i o n a l shear tests i n this research.
Deaired water was used as the confining fluid and s i l i c o n e -
greased double prophylactic memranes were used around the
sample. The zero hour time i n F i g . 5.1 corresponds to the time
a f t e r i s o t r o p i c consolidation and pore pressure equalization of
the sample. At that time an a i r water interface was introduced
into the c e l l by draining a l l the c e l l water leaving water only
i n the small inner plexiglass containment c e l l d i r e c t l y around
68
4'
Air Diffusion Test: Sample AD3
constant cr =400 kPa

T =60 hours
hours -? 0
o
back pressure, a y
-Q- during consolidation = 100 kPa
•TP"
Uexcess=(0-c-CT )*(l-e- b
d C r
- T o )
)
O
Q_
i-
o
ro
b
\
\
o \
o
\
100 11111 u 111111111111111111 n 111111111111111111111111 I I 111 fi i

50 100 150 200 250 300
Time (hours)
F i g . 5.1 E f f e c t of A i r D i f f u s i o n on E f f e c t i v e Stress
In Sample
69
the specimen. The jacket of water around the specimen was
required to avoid d i r e c t contact of pressurized a i r with the
sample memrane.
Under a constant applied c e l l pressure, no e f f e c t of a i r

d i f f u s i o n on the measured pore pressure i s evident f o r about
the f i r s t 60 hours. Thereafter, continued a i r d i f f u s i o n and
pore pressure increase causes the e f f e c t i v e confining pressure
to decrease to 50 percent of the i n i t i a l value a f t e r only 170
hours (Fig. 5.1).
To circumvent these d i f f i c u l t i e s , some researchers have

successfully used mercury as a containment f l u i d because a i r
does not r e a d i l y dissolve i n mercury, even under high pressures
(Novak and Kim, 1981, Marcuson and Wahls, 1978). The apparatus
modifications and special safety precautions required to use a
mercury jacket precluded i t s use i n the UBC S o i l Mechanics
Laboratory.
A t e s t i n g program constraining a l l t e s t s involving pore

pressure measurements to 60 hours duration was therefore
undertaken. This was considered to be a safe time l i m i t since
most t e s t s were performed at lower c e l l pressures than the 400
kPa c e l l pressure used f o r the a i r d i f f u s i o n t e s t of F i g . 5.1.
Lower c e l l pressures delayed onset of a i r d i f f u s i o n effects
( i . e . longer T^) i n tests performed by Pollard et a l . (1977).
70
However, i n order to be able to complete staged tests
within the 60 hour time l i m i t , i t was necessary to use filter
paper drains around the specimen to shorten the drainage path
thereby decreasing the time required f o r consolidation. Based
on the theory of consolidation, the use of r a d i a l drains o f f e r s
a greater than 70-fold decrease in the time required for
completion of primary consolidation (Bishop and Henkel, 1962).
The actual consolidation time decrease using r a d i a l drains was
somewhat l e s s - a 28-fold decrease at a consolidation stress of
70 kPa (Fig. 5.2). At higher consolidation stresses, the
effectiveness of the f i l t e r drains was reduced, presumably as a
r e s u l t of the decreasing permeability of the f i l t e r paper with
increasing confining stress. Nevertheless, the reduction in
consolidation times was sufficient to enable any test to be
completed within the 60 hour l i m i t .
The degree of saturation of the specimen was checked

immediately following every stage of t e s t i n g by determining the
Skempton B-parameter. A B-value of 0.95 or greater was
considered s u f f i c i e n t l y high to indicate that saturation of the
specimen was maintained. Pore pressures generated during
t e s t i n g were considered accurate only i f such a high B-value
was determined following the t e s t stage.
For t e s t s requiring consolidation stresses greater than
about 150 kPa, the decrease i n sample volume as a r e s u l t of
consolidation caused the water l e v e l i n the confining jacket
71
F i g . 5.2 Comparison of Consolidation Rate with and without
Radial Drains
72
surrounding the specimen to drop low enough to d i r e c t l y expose
the uppermost portion sample memrane to the pressurized a i r i n
the confining chamber. This l e d to an e a r l i e r onset of a i r
d i f f u s i o n e f f e c t s f o r these higher-pressure tests as indicated
by the more rapid decrease i n B-value (Fig. 5.3). For tests
performed at effective consolidation stresses less than 150
kPa, F i g . 5.2 shows that B-values of about 0.95 were maintained
for a time period greater than 60 hours. Thus, information on
generated pore pressures had to be obtained shortly after
consolidation (several hours) f o r high pressure t e s t s .
In future testing, i t i s recommended that a method be

developed to eliminate t h i s problem. A telescoping containment
c e l l with a method to p e r i o d i c a l l y flush the containment cell
with deaired water i s suggested.
73
1.00
0.90 H
| 0.80
>
m
§0.70
•+->
CL
E
a>
^ 0.60
0.50 ***** o-c' <150 kPa

3
A A A A A C73C >1 50 kPa
0.40 T—i—i—|—i—i—i—|—i—i—i—|—i—i—i—j—i—i—r
20 40 60 80 100
Time (hours)
F i g . 5.3 Increased A i r Diffusion E f f e c t at High Confining

Stress
74
6. SITE DESCRIPTION
6.1 GEOLOGY AND GEOTECHNICAL SITE DESCRIPTION
The Lower 232nd St. research s i t e i s located at the 232nd

St. interchange of the Trans Canada Highway, about 40 km east
of downtown Vancouver (Fig. 6.1). The Quaternary sequence at
the s i t e i s part of the Fort Langley Formation consisting of
marine s i l t to clay with occasional sand lenses, deposited
during the glacial regression (Armstrong, 1978). These fine
grained soils have become s l i g h t l y s e n s i t i v e as a result of
leaching subsequent to deposition. The silts and clays are
s l i g h t l y organic and are underlain by dense sands and gravels.
Due to dessication, the soil is overconsolidated at the
surface, becoming normally consolidated below a depth of about
5 m.
Numerous Atterberg limit tests and hydrometer tests

indicate that the s o i l below 3 m i s reasonably uniform, except
for occasional sandy layers (Fig. 6.2). Above 2.5 m depth, the
surficial soil i s of higher p l a s t i c i t y (PI=40) while the soil
below 3 m i s of lower p l a s t i c i t y (PI=20).
75
F i g . 6.1 Lower 232nd S t . S i t e L o c a t i o n
PL* ; WN* ; Lbs Clay55 ; siltss ; s a n d *
0 20 40 60 80 100 0 20 40 60 80 100
0 | i i i i • i t i »
• o
• o
• • o
• • o
• o
a
• • o
Lr 232 St
a • o
• • o
Lr 232 St
J I 1 I L 1
15 ' ' ' I I I L
ooooo Liquid limit

• • • • • Water content
• a o a a Plastic limit
F i g . 6.2 P r o f i l e s o f A t t e r b e r g L i m i t s and G r a i n Size
77
6.2 LABORATORY SOIL SAMPLES
Two types of undisturbed s o i l samples were obtained; block

samples and 2.6" inner diameter tube samples. The block
samples were carefully trimmed at a depth of 2 m from an
excavated p i t whereas the wireline tube samples continuously
sampled the s o i l to a depth of 16 m. Both types of samples
were sealed and stored i n a humidity room u n t i l needed.
The block samples were obtained i n order to have a large

number of i d e n t i c a l samples from the same horizon which could
be used t o study the e f f e c t s of various parameters on the
dynamic s o i l properties. On the other hand, only a few tube
samples were tested since they were used mainly to compare the
dynamic properties determined i n the laboratory with those
derived from i n s i t u techniques. The difference i n p l a s t i c i t y
index (PI) between the two types of samples also enabled the
comparison of dynamic properties on the basis of PI.
78
7. RESONANT COLUMN RESULTS
7.1 LOW AMPLITUDE SHEAR MODULUS
It i s natural to examine separately the e f f e c t s on low
amplitude and high amplitude shear modulus of various
parameters such as confining pressure and duration of

confinement.
7.1.1 E f f e c t of Confining Pressure and Void Ratio
As mentioned i n section 2.1.2, the e f f e c t s of confining

pressure and void ratio on shear modulus of normally
consolidated cohesive soils are difficult to examine
independently since these two factors are i n t i m i t e l y linked.
When considered together, however, t h e i r i n d i v i d u a l e f f e c t s can
be distinguished.
From fourteen undisturbed block samples taken from a depth

of 2.0 m, 19 normally consolidated determinations of low
amplitude shear modulus (G^x) were made at e f f e c t i v e confining
pressures ranging from 70 to 500 kPa. These low amplitude
dynamic shear modulus values are plotted against the respective
isotropic effective confining stresses i n F i g . 7.1. Despite
some data scatter, the shear modulus increases almost i n d i r e c t
79
Block Samples 2.0 m
G0=122.2*(2.97-e) /0 +e) 2
*Pa- (oV)'
5 S
Hardin a n d Drnevich
100000 -
9-
S-
o 7-
CL
8-
on
5-
3
"O
o
3-
a
J C
2-
^Go=143.2*Pa-0l*(c-3c')
10000 9 -
7 —•
i —•
i—r
8 T -r
X
9
100
Osc (kPa)
F i g . 7.1 Shear Modulus v s . C o n f i n i n g P r e s s u r e R e l a t i o n s h i p

f o r B l o c k Samples
80
proportion t o the increase i n e f f e c t i v e confining stress. The
relationship between G^x a n d
confining pressure can be
approximated as
Gâx = 143.2Pa- 0 1
(<x 3 c ')- 9 9
(7.1)
where i s a^c' the e f f e c t i v e confining pressure and P a i s the

atmospheric pressure i n the same units as G^ x and a • N o
attempt t o separate the influence of void ratio i s made i n

Eq. 7.1.
The standard error i n estimating shear modulus using

Eq. 7.1 i s 3900 kPa. Due to the change i n the p r o f i l e of s o i l
characteristics (Fig. 6.2) at the 2.0m depth from which the
block samples were obtained, some of the scatter i n shear
modulus values may be the r e s u l t of sample differences.
Indeed, v e r t i c a l v a r i a t i o n i n f i n a l water content within some
block samples were found to be as high as 15% suggesting that
specimens were non-uniform.
Low amplitude shear modulus ( G m a x ) was also determined f o r

several tube samples obtained from 8-14m and tested at
e f f e c t i v e confining pressures ranging from 70 t o 400 kPa. Once
again, the shear modulus values c l e a r l y increase with confining
pressure f o r the normally consolidated tests i n F i g . 7.2. The
least squares best f i t line describing the relationship i s
given by
81
G=122.2*(2.97-e)/(1 +e) *Pa-(oV)'
0
2 5 S
Hardin and Drnevich ^
100000-
o
a.
CO
3
TJ
o
o
JC
CS)
G =292.1*Pa *(a3 ')
0
,10
C
Lower 232 St. Clay

nd
Normally Consolidated
10000- Tube Samples: 8—14m
T—i—r
100 500
Effective Confining Pressure (kPa)
Fig. 7.2 Shear Modulus vs. Confining Pressure Relationship

for Tube Samples
82
Gnax = 292.1*P - a
1 0
(a 3 C ')- 9 0
(7.2)
w i t h a sample standard e r r o r o f 2000 kPa. Void r a t i o effects
are not s e p a r a t e d from Eg. 7.2. The soil index p r o p e r t i e s (PI
and g r a i n s i z e ) i n the 8-14m depth range from which these tube
samples were o b t a i n e d i s r e l a t i v e l y uniform; the s t a n d a r d error
i s about h a l f t h a t o f the b l o c k samples.
The dimensionless modulus m u l t i p l i e r values, K, for the
b l o c k samples and tube samples - 143.2 and 463.6, respectively
fit the trend of increasing modulus multiplier with
decreasing PI identified in F i g . 2.4 for seven undisturbed
clays of varying p l a s t i c i t y . These are shown g r a p h i c a l l y in
F i g . 7.3 which p r o v i d e s a f i r s t estimate of K on the b a s i s of
PI.
The respective modulus exponent values for the block
samples and tube samples are 0.99 and 0.90. These v a l u e s are
shown p l o t t e d against PI i n F i g . 7.4 along with the modulus
exponents f o r the c l a y s presented i n F i g . 2.4. The modulus
exponents o f a l l the c l a y s l i e i n a f a i r l y narrow range between
0.84 and 1.18 with the less plastic clays generally having
s m a l l e r modulus exponents.
The trend of i n c r e a s i n g modulus exponent and decreasing
modulus m u l t i p l i e r with increasing p l a s t i c i t y index might be
83
Plasticity Index
ro J> o> oo o ro
o o o o o o o
to o 111111111111n111111111111111111111111111II1111II11111111IIi
H <
3 ft)
& n
(D H-
x o»
rt
O f-4
o
Hi f BLOCK SAMPLES
s
O
0-
H
(fl
s H
c
H
rt
f-f TUBE SAMPLES
•o
M
H-
fl>
rt
tr
ft>
(fl
rt
H-
o
o
o
120
100 H co
LU
<
CO
80 H
o
o
_J
CQ
60 4 CO
LU
<
CO
40H LU i
CQ
ID i
20 H I
11111111111111111111111M11111111111111111111M11
0.70 0.80 0.90 1.00 1.10 1.20
Modulus Exponent, n
Fig. 7.4 V a r i a t i o n o f Modulus Exponent w i t h Plasticity-

Index
85
intuitively predicted; since higher plasticity clays are
generally softer (lower modulus) than less p l a s t i c clays, a
lower modulus m u l t i p l i e r would be expected. Higher p l a s t i c i t y
clays would also be expected to have a higher modulus exponent
than l e s s p l a s t i c clays since they generally undergo a more
rapid reduction in void ratio with increasing confining
pressure.
In order to examine the e f f e c t of void ratio on shear

modulus of both the tube and the block samples, a void ratio
factor i n the form of Eq. 2.1 was employed. The form of the
equation is derived from the assumption of a linear
relationship of shear wave v e l o c i t y with void ratio which
appears to be v a l i d f o r the block samples (Fig. 7.5) as well as
for the tube samples (Fig 7.6). Much of the scatter in the
data i s the r e s u l t of sample differences since only small
variations from a linear relationship were observed for
individual samples tested at more than one normally
consolidated state.
For the normally consolidated block samples tested, the

combined e f f e c t of void r a t i o (e) and confining pressure (02c')
on low amplitude shear modulus (G^x) i s shown i n F i g . 7.7 with

a best f i t l i n e given by
Gmax = 2 9 . 3 * F ( e ) P a - 4 0 ( a 3 C ' ) - 6 0 (7.3)
86
200.0 ~T
Lower 232 St. Cloy
nd
o o Normally Consolidated
180.0 - Block Samples: 2.0m
-^160.0 -
^140.0 -
*o
_o
£ 120.0 -
Q>
>
i 100.0 -
o
£ 80.0 -
60.0 -
40.0 —" 1 1 1 " 1 1 1 1 1 1 1 1—

0.60 0.80 1.00 1.20 1.40 1.60 1.80 2.00
Void Ratio, e
F i g . 7.5 Variation of Shear Wave Velocity with Void Ratio

for Normally Consolidated Block Samples
87
240.0
Lower 232 St. Clay
nd
220.0 - Tube Samples: 8—14m
CO 200.0 -
E
180.0 -
"
_o
o
5 160.0 -
>
S 140.0 -
i_
D
lo 120.0
100.0 H
80.0 T T
0.60 0.70 0.80 0.90 1.00
Void Ratio, e
F i g . 7 . 6 Variation of Shear Wave V e l o c i t y with Void Ratio

for Normally Consolidated Tube Samples
88
10000
Lower 232 St Cloy
nd
Block Samples: 2.0m
oo
40/ ,\.602
G =29.3*F(e)*Pa- (03c')
0
w
F(e)=(5.20-e) /(l+e)
2
1000 T 1 1 1 1 T r
100 600
Effective Confining Stress (kPa)
Fig. 7.7 Void Ratio Modified Shear Modulus vs. C o n f i n i n g

Pressure for Normally Consolidated Block
Samples
89
where P a = atmospheric pressure i n the same units as
a ' and Gjjax
3 C
and F(e) = (5.2-e) /(l+e)2
Equation 7.3 has the atmospheric pressure term, Pa, included to

make the equation v a l i d f o r any system of units and to make the
modulus m u l t i p l i e r (29.3) dimensionless. The constant of 5.2
in the numerator of the void ratio function, F(e), was
determined by regession analysis which involved varying the
constant until the linear log G max /F(e) versus log ajc'
r e l a t i o n s h i p with the smallest standard error was identified.
This constant appears to be dependent on the clay type as
values ranging from 2.97 to 7.32 have been i d e n t i f i e d i n the
literature (Kokusho et a l . , 1982). I f the commonly used form
of F(e) with a constant of 2.97, as suggested by Hardin and
Drnevich (1972b) i s used, a poor f i t to the block sample data
is observed and the shear modulus exponent f o r the best f i t
line i s only 0.27. Data presented by Zen et a l . (1978)
indicates that the Hardin and Drnevich expression with F(e)
constant of 2.97 does not yield a good agreement to the
experimental r e s u l t s f o r clays with PI greater than about 25.
The best f i t line described by Eq. 7.3 has a standard

error i n shear modulus of 3000 kPa which i s not a very large
improvement on Eq. 7.1 with a standard error of 3900 kPa.
Shown i n F i g . 7.8 are the measured values of shear modulus
which compare reasonably well to the predicted values based on
Eq. 7.3.
90
Fig. 7 . 8 Comparison of Predicted and Measured Shear Moduli
for Normally Consolidated Block Samples
91
For the normally consolidated tube samples tested, the
combined e f f e c t of void r a t i o and confining pressure i s shown
i n F i g . 7.9 with best f i t l i n e given by
Gjjax = 1 3 7 . 2 * F ( e ) P - a
3 8 4
(* ')-
3 c
6 1 6
(7.4)
where P a = atmospheric pressure i n the same units as G m a x
and 02c'
and F(e) = (2.97-e) /(1+e) 2
The void r a t i o factor with a constant of 2.97 i s that given by

Hardin and Drnevich (1972b) to be applicable f o r a l l s o i l s .
However, data presented by Zen et a l . (1978) suggests that i t
i s only v a l i d f o r clays with PI less than 25. The predicted
shear moduli based on Eg. 7.4 and the measured shear moduli
show good agreement (Fig. 7.10), with a standard error i n shear
modulus of 2700 kPa.
Seperating out the void r a t i o influence by means of F(e)

i n equations 7.3 and 7.4 makes the shear modulus exponent 0.6
f o r both the block samples and tube samples. This i s the same
value of modulus exponent given by Kokusho et a l . (1982) f o r
Teganuma Clay, though the shear modulus m u l t i p l i e r and void
r a t i o factor values are d i f f e r e n t since they appear to depend
on the c h a r a c t e r i s t i c s of the p a r t i c u l a r s o i l .
92
40000
Lower 232 St. Clay
nd
Tube Samples: 8-14m
G=137.2*F(e)*Pa (o-c)°'
0
,38
3
, 62
F(e) = (2.973-e) /(1+e) 2
10000 T I 1 1 I 1
100 600
Effective Confining Pressure (kPa)
F i g . 7.9 Void Ratio Modified Shear Modulus vs. Confining

Pressure f o r Normally Consolidated Tube
Samples
93
120000n
100000 ~
80000i
60000 z
40000-
20000 z
1111111111111111 n 111111111111111 n 111111111111111111111111

20000 40000 60000 80000 100000 120000
^measured (kPd)
Fig. 7.10 Comparison of Predicted and Measured Shear Moduli

for Normally Consolidated Tube Samples
94
As can be seen i n figures 7.1 and 7.2, the well-known
equation f o r shear modulus (Eq. 2.1, 2.2) given i n Hardin and
Drnevich (1972b) s i g n i f i c a n t l y overpredicts the shear modulus
values f o r both the block samples and tube samples of Lower
232nd St. Clay. This finding confirms data presented by Zen et
al. (1978) which suggests that the Hardin and Drnevich equation
should not be used with higher p l a s t i c i t y clays (PI>25).
7.1.2 Stress History
To study the e f f e c t of stress history on the low

amplitude shear modulus f o r Lower 232nd St. Clay, block samples
were tested at various states of overconsolidation. In general,
the shear modulus increases with increasing overconsolidation
r a t i o (OCR).
The normalized small strain shear modulus values f o r

overconsolidated block samples are shown i n F i g . 7.11 f o r
various OCR. The normalized shear modulus i s the shear modulus
measured i n the overconsolidated (OC) condition divided by the
shear modulus f o r normally consolidated (NC) samples given by
Eq. 7.1. Despite some data scatter, the shear modulus can be
expressed as
Gmax = (Gmax)ncOCR' 55
(7.5)
95
Block Samples 2.0 m o
Over-consolidated
OCR
F i g . 7.11 E f f e c t o f OCR on Shear Modulus f o r B l o c k

Samples
96
where G ^ x i s the measured shear modulus and ( G m a x ) n c i s the
shear modulus f o r a normally consolidated sample measured at
the same e f f e c t i v e confining s t r e s s . Substituting Eq. 7.1 f o r
normally consolidated shear modulus i n Eq. 7.5 gives the shear
modulus expression valid f o r both normally consolidated and
overconsolidated block samples
Gmax = 1 4 3 . 2 * P - ( a ' ) - O C R -
a
01
3C
99 55
(7.6)
The OCR exponent of .55 i n Eq. 7.6 i s i n the range of 0.5-0.7
identified f o r a l l but one of the seven undisturbed cohesive
s o i l s l i s t e d i n F i g . 2.4.
Alternatively, i f Eq. 7.4 i s used t o represent the

normally consolidated s o i l with F(e) to account for void r a t i o
changes, then the expression for shear modulus i s
Gmax = (Gmax)ncOCR* 26
(7.7)
As can be seen i n F i g . 7.12, a r e l a t i v e l y large scatter e x i s t s

which suggests that the same void r a t i o factor, F(e), does not
apply t o both NC and OC clay, though some of the data scatter
may be the r e s u l t of errors i n void r a t i o measurement. The
reason f o r the reduced OCR exponent i n Eq. 7.7 as compared to
Eq. 7.5 i s that the e f f e c t of overconsolidation on G m a x is
p a r t i a l l y accounted for by the void r a t i o function, F(e).
97
Block Samples 2.0 m
Overconsolidated
F i g . 7.12 E f f e c t of OCR on Void Ratio Modified Shear

Modulus f o r Block Samples
98
The OCR exponent, however, agrees with the 0.18-0.30 range
specified by Hardin and Drnevich (1972b) f o r clays with PI
between 20 and 40 (Table 2.1). The void r a t i o modified shear
modulus expression f o r both NC and OC block samples i s
Gmax = 2 9 . 3 * F ( e ) P - ( a ' ) - O C R -
a
4 0
3 C
6 0 2 6
(7.8)
where F(e) = (5.2-e) /(l+e) 2
Tube samples were not tested at overconsolidated states as

the focus of the tests was on normally consolidated behaviour
since the i n s i t u s o i l i s normally consolidated at the depths
tube samples were obtained. Furthermore, only a l i m i t e d number
of samples were available f o r t e s t i n g .
To study the time dependent increase i n low amplitude

shear modulus discussed i n section 2.1.6, the shear modulus f o r
some samples was monitored at various times during the
consolidation period. This generally involved taking
measurements well beyond the time required f o r completion of

primary consolidation i n order to establish the time dependent
increase at constant e f f e c t i v e stress.
99
The time dependent increase in shear modulus for a
normally consolidated block sample of Lower 232nd St. Clay i s
shown f o r three e f f e c t i v e confining pressures i n F i g . 7.13.
A f t e r about the f i r s t ten minutes, the modulus versus log T
relationship may be approximated by two straight lines
corresponding to the primary and secondary consolidations.
Good agreement exists between the kink i n the curve and the
completion of primary consolidation determined from the
consolidation curve (not shown) f o r each confining pressure
which i s represented by the arrow on each curve. Modulus
values determined prior to the end of primary consolidation
correspond to an unknown e f f e c t i v e stress since excess pore
water pressures s t i l l e x i s t within the specimen.
The rate of secondary increase i n shear modulus per log

cycle of time, AG, increases with confining pressure
(Fig. 7.13). The normalized rate of shear modulus increase,
NQ, as defined by Anderson and Stokoe (1978) i s given by
N G = AG/G 100 o ( -9)

7
where vA = secondary increase i n G per log cycle

of time
G
1000 = G
measured at 1000 minutes of confinement
The values of N Q f o r the tests shown i n F i g . 7.13 range
from about 18-22%. Although the duration of each test in
F i g . 7.13 was less than 2000 minutes, two tests (B14C2 and
100
arrows represent end of
primary consolidation
AG due to void
ratio change
• • • • • B9C4
A A A A A B9C3 146
0 0 v B9C2
AA
74
' "I T—r I I I I| T i—i—r
i i i 11—
10 100 1000
Confining Time (min)
F i g . 7.13 Time-Dependent Increase i n Shear Modulus f o r

Normally Consolidated Block Sample
101
B8C2) were continued to 6000 minutes with no deviation from the
linear semi-log relationship. Anderson and Stokoe (1978)
report a s i m i l a r modulus-time response f o r periods as long as
20 weeks.
For comparison, the time dependent shear modulus increase

for a tube sample obtained from a depth of 11.75m i s shown i n
F i g . 7.14 for three effective confining pressures. The
behaviour i s very s i m i l a r to that shown by the block samples
(Fig. 7.13). Once again, the kink i n the curve corresponds
c l o s e l y to the end of primary consolidation represented by the
arrow on each curve and the value of AG increases with
confining pressure. The normalized rate of shear modulus
increase, NQ, given by Eq. 7.9 ranges from 15-21% - slightly
less than the 18-22% range determined f o r the block samples.
The secondary increase i n shear modulus i s not unique to

p o s i t i v e confining pressure increments such as those shown i n
Figs. 7.13 and 7.14. Figure 7.15 shows the shear modulus
versus time relationship f o r two unloading confining pressure
increments to overconsolidation r a t i o s of 2.9 and 9.3. After
the initial modlus reduction as a result of swelling to
equilibrium with a lower confining pressure, a secondary
increase i n shear modulus i s again observed. The values of NQ
are about 12% and 18% f o r unloading to OCR of 2.9 and 9.3,
respectively.
102
120.0
Lower 2 3 2 nd
St. Clay 11.75m
TEST
100.0 -
3DDDDTS1C2 69 kPa .206
\AAAATS1C3 198 .212
">0000TS1C4 397 .149 j
cl 80.0 -
AG due to void
ratio change
CO
60.0 -
O
D
CD
$ 40.0 H
20.0 -
arrows represent end of

0.0 primary consolidation
T T T TT- i i i 1111 1 T 1 I I I II |
10 100 f 1000
F i g . 7.14 Time-Dependent Increase i n Shear Modulus f o r

Tube Sample of Lower 232nd St. Clay
103
35.0 zr
Block Samples 2.0m
30.0 z
£25.0
in
-§20.0
in
15.0 z
10.0 z
TEST OCR CT ' C
No
AAAAA B9C6 9.3 32 kPa .119
2.9 101 .177
00000 B9C5
5.0 T T T—i—I I i i 11 T i — r
i i i 111—
10 100 , 1000
Fig. 7.15 Time-Dependent Increase in Shear Modulus f o r

Overconsolidated Block Sample
104
Values of NQ determined f o r a number of t e s t s are l i s t e d
i n Table 7 . 1 . Values f o r normally consolidated t e s t s range
from 13.8 to 23.4% with an average of 18.4%. Time, therefore,
must be considered when interpreting laboratory t e s t s since the
shear modulus i s time dependent. Unless otherwise stated, the
shear moduli reported herein are those values measured at
approximately 1 0 0 0 minutes of,confining time.
The rate of secondary modulus increase, NQ, i s such that

it can not be accounted f o r by changes i n void r a t i o during
secondary consolidation. When these void ratio changes are
considered, the increase i n shear modulus estimated from Eqns.
7.3 and 7.4 i s much less than that observed. This predicted
increase i s shown by the dashed l i n e s i n Figs. 7 . 1 3 and 7 . 1 4 .
Some time-dependent physio-chemical bonding may be the cause of
the large secondary shear modulus increase.
Insight into the aging e f f e c t can be obtained by examining

F i g . 7 . 1 6 . The normalized modulus reduction curves f o r a block
sample are shown f o r two d i f f e r e n t confining times ( 3 and 9 3
hours) by the open symbols. Correspondingly, the s o l i d symbols
represent the low amplitude shear modulus measured immediately
after each increment of high amplitude cycling. The solid
symbols therefore reflect the amount of degredation or
disturbance of the soil structure bonds. No significant
decrease i n the low amplitude shear modulus occurs f o r strains
less than about 0.005%. Thereafter, the degredation increases
105
Table 7.1 Values of NQ f o r Normally
Consolidated Samples
TEST <73c' N G
(kPa)
B8C2 77.4 .234
B9C2 74.3 .184
B9C3 146.4 .222
B9C4 223.5 .187
B12C3 292.5 .138
B14C2 70.0 .164
B18C2 146.2 .197
B19C2 109.3 .145
TS1C2 69.0 .206
TS1C3 198.0 .212
TS1C4 397.0 .149
106
1.20
Block Sample 2.0m
o- '=70 kPa
3c
1.00 __OD . |
n°
•
•
9
0.80 -
X
o 9
E
CD
O 0.60 -
ooooo confined 93 hours

D D D D D confined 3 hours
0.40 - end of primary consol. = 25 min.
CP
solid symbols represent low—amplitude (1.5x1 a" *) 3
modulus immediately after high—amplitude testing
0.20 1 — I I I I I I I I — I — I I I I '111
0.001 0.01 0.1

Strain Amplitude (%)
F i g . 7.16 E f f e c t of Aging on Normalized Modulus Reduction

Curves and on Low Amplitude Shear Modulus
Measured After High Amplitude Testing
107
with shear strain amplitude. However, the amount of
degredation i s considerably larger f o r the longer confining
time suggesting that proportionately more physio-chemical
bonding has taken place.
It should be mentioned that the low amplitude shear

modulus always recovered to i t s original value given a
sufficient period of time after high amplitude testing,
although the length of time required f o r recovery increased f o r
longer confining times. The low amplitude modulus regain
behaviour was also observed by Isenhower (1979) i n h i s tests on
San Francisco Bay Mud and i s consistent with a time dependent
physio-chemical bonding mechanism.
7.2 HIGH AMPLITUDE SHEAR MODULUS
The high amplitude shear modulus i s the shear modulus

corresponding to shear strains greater than about 0.005%, where
the shear modulus becomes strain dependent. The factors
affecting high amplitude shear modulus are discussed i n the
following sections.
108
7.2.1 Nonlinear Behaviour
In resonant column t e s t i n g the method used to establish
the resonant frequency of the s o i l sample involves sweeping a
range of frequencies i n order to i d e n t i f y the peak v i b r a t i o n
response. A typical low-strain frequency response curve is
shown by the lowermost curve in F i g . 7.17. The curve is
symmetrical about the resonant frequency implying that the s o i l
behaviour i s l i n e a r e l a s t i c at small s t r a i n s .
With increasing strain the frequency response curves
become increasingly asymmetrical and s h i f t to the l e f t . The
strain dependency i s shown i n F i g . 7.18 where the frequency
response curves at three s t r a i n s have been normalized by the
peak amplitude. Since the shear modulus i s calculated from the
resonant frequency (section 4.2.1), the decrease in the
resonant frequency with s t r a i n amplitude implies that the s o i l
s t i f f n e s s i s non-linear. The s t r a i n softening c h a r a c t e r i s t i c s
of s o i l s are best i l l u s t r a t e d on a semi-logarithmic p l o t , known
as a modulus reduction curve.
7.2.2 E f f e c t of Shear Strain
A t y p i c a l shear modulus reduction curve f o r a block sample
of Lower 232nd St. Clay at 2.0m depth i s shown i n F i g . 7.19.
The curve was obtained from a high amplitude t e s t sequence i n
109
0.10
Shear Strain
nnnnn | w
0 input current (12 mV) 2.26x10" * 3
AAA&A medium input current (96 mv) 1.55x10"

* * * * * high input current (830mV) 7.70x10"
0.08 -
c 0.06 -
"a
m
| 0.04
m
0.02 -
0.00 —r - Ijl ^ F f ^ ^ - f l ^
10 15 20 25 30 35 40
Frequency (Hz)
F i g . 7.17 Frequency Response Curves f o r Lower 232nd St.

Clay f o r Three Strain Levels
110
Shear Strain
nnnnn | w0 input current (12 mV) 2.26x10"*"*
/WÂA medium input current (96 mv) 1.55x10*18
* * * * * high input current (830mV) 7.70x10"l8
Frequency (Hz)
F i g . 7.18 N o r m a l i z e d Frequency Response Curves
111
F i g . 7.19 T y p i c a l Shear Modulus v s S t r a i n Curve f o r Lower
232nd S t . C l a y
112
which the shear modulus i s determined first at a low s t r a i n
(about 0.001%) and then at progressively larger s t r a i n s . The
shear modulus i s approximately constant f o r s t r a i n s less than
0.005% suggesting l i n e a r e l a s t i c behaviour i n that range. For
shear s t r a i n s greater than 0.01%, the clay s t i f f n e s s exhibits a
large s t r a i n softening behaviour; indeed, the shear modulus at
0.4% strain i s less than h a l f of the constant small strain
shear modulus, usually termed Gmax o r G
o- should be noted
that beyond the l i n e a r e l a s t i c threshold of 0.005%, both the

frequency and s t r a i n rate vary along the curve. These e f f e c t s
are further discussed i n section 8.1.
The shear modulus reduction curves f o r a l l successful high

amplitude tests are given i n Appendix A. These include the
t e s t s performed on the c l a y - s i l t from block samples at 2.0m
depth as well as tests on 2.6" diameter tube samples from
various depths t o 14m.
When comparing modulus reduction curves, i t i s often

desireable t o present the results i n the form of a normalized
modulus reduction curve i n which the shear modulus at a given
strain i s normalized by Gjnax* Such a curve i s shown i n
F i g . 7.16. The percent reduction i n shear modulus at any
strain i s readily apparent since the low amplitude shear
modulus has a r a t i o of one. Also shown i n F i g . 7.16 i s the
small strain shear modulus (measured at 0.0015% strain)
measured immediately after each increment of high amplitude
113
testing. The Gmax values remain e s s e n t i a l l y the same when
measured a f t e r high amplitude tests up t o 0.07% s t r a i n and only
a small (<5%) reduction i n Gmax occurs a f t e r the highest s t r a i n
tested (about 0.4%).
This finding suggests that very little structural

breakdown occurs i n the clay f o r the range of s t r a i n s tested i n
the resonant column. An important implication to testing
procedure i s that more than one high amplitude test can be
carried out on the same sample without significant
deterioration of the clay structure. This serves t o reduce the
number of samples required as well as t o enable the comparison
of high amplitude non-linear behaviour f o r various stress and
time conditions on the same sample, therby eliminating some
problems with sample differences.
7.2.3 E f f e c t of Confining Pressure and Stress History
For the block samples obtained from a depth of 2.0 m with

PI ranging from 33 t o 61, there i s e s s e n t i a l l y no influence of
effective confining stress (50 to 500 kPa) and
overconsolidation r a t i o (1.0 to 9.3) on the normalized modulus
reduction curve. This can be seen i n F i g . 7.20 where the
normalized modulus values f o r a wide range of stress states
fall on a single curve. Thus, knowing the shape of the
normalized modulus reduction curve allows the high amplitude
114
1.2
Lower 232 St. Clay
nd
Block Sample: 2.0m
1.0 -
0.8 - t.
Jo-6
CD 0
TEST cr c'
3
OCR
0.4
* * * * * B17C1 484 kPa 1.0 A
O O D O D B17C2 103 4.7
A A A A A B17C3 53 9.2
ooooo B19C2 109 1.1
0.2 00000 B19C3 268 1.0
<X*X><>-B19c5 50 5.4
0.0 i—r i i 11111— I I 11 III— T — r I I 11 III T 1 IIIIIII

0.0001 0.001 0.01 0.1 1
Strain Amplitude (ss)
F i g . 7.20 N o r m a l i z e d Modulus R e d u c t i o n Curves f o r B l o c k

Samples
115
shear modulus to be predicted f o r a wide range of effective
confining stress and OCR values i f the low amplitude shear
modulus f o r the particuluar stress state i s known.
The normalized modulus reduction curves f o r samples from

depths of 11.75 and 13.2 m are shown f o r several e f f e c t i v e
confining pressures in Figs. 7.21 and 7.22, respectively.
These samples have Pi's between 20 and 25. A small e f f e c t of
confining pressure i s seen f o r both depths i n the 0.01-0.1%
range of s t r a i n amplitudes where the s t r a i n amplitude values
are s t i l l f a i r l y accurate. Above 0.1% s t r a i n , differences i n
the number of cycles s h i f t the G/Gmax values so as to obscure
the e f f e c t of confining pressure. Nevertheless, i t can be seen
that higher confining pressures i n the 0.01%-0.1% range s h i f t
the modulus reduction curve to the right. Also shown i n
F i g . 7.21 is the modulus reduction curve for an
overconsolidated sample (OCR=3.9) which suggests l i t t l e effect
of stress history on the normalized modulus values.
I t appears that the differences i n p l a s t i c i t y between the

block samples and the tube samples affect the normalized
modulus curves. These r e s u l t s agree with those summarized by
Sun et a l . (1988) i n which the influence of confining pressure
on the normalized modulus reduction curve was shown to
gradually diminish with increasing p l a s t i c i t y index.
116
1.2
Lower 232 St. Clay nd
Tube Sample: 11.75m
1.0 - A DO • A
A
*3
A
*
A
* •
0.8 -
•
X * A
£0.6
o -
0°
0.4 -
TEST cr 3C OCR
* * * * * TS1C2 •69 kPa 1.0
• DDDD TS1C3 198 1.0
0.2 - A A A A A JS1C4 397
103 1.0
00 0 0 0 TS1C5
3.9
0.0 T 1 llllll| 1 1 IIIIIII 1 1 llllll| T—I II I I I
0.0001 0.001 0.01 0.1

Fig. 7.21 N o r m a l i z e d Modulus R e d u c t i o n Curves f o r Tube

Samples from 11.75m
117
1.2
Lower 2 3 2 St. Claynd
Tube Sample: 13.2m

1.0 - A a A*° • £
* A
•
A
0.8 -
*
•
X A
o
cl " 0 6
* D
0.4 - *
TEST 0-30' OCR A
* * * * * TS2C2 76 kPa 1.0 •

0.2 - D O D O O TS2C3 137 1.0
A A A A A TS2C4 268 1.0
0.0 I I I 11 III 1—I I I 111II 1—I I I 111II 1—I I I 11II
0.0001 0.001 0.01 0.1 1

F i g . 7.22 Normalized Modulus Reduction Curves f o r Tube

Samples from 13.2m
118
Based on F i g . 7.16 i t may be concluded that the duration

of confinement has virtually no effect, within the range
investigated, on the normalized modulus reduction curve. The
reason f o r t h i s can be seen from F i g . 7.23 where the normalized
rate of shear modulus increase, Ng, defined .in Eg. 7.9 i s
essentially independent of s t r a i n amplitude. Thus, both the
small strain and large strain amplitude moduli increase at
about the same rate leaving t h e i r r a t i o (G/Gjjax) unchanged.
In summary, f o r the Lower 232nd St. Clay studied there i s

relatively little influence of confining stress, stress
history, and secondary time effects on the normalized shear
modulus reduction curves, e s p e c i a l l y f o r the higher p l a s t i c i t y
block samples obtained from a depth of 2.0m. For the tube
samples, a small s h i f t of the curves to the r i g h t occurs with
increasing confining stress.
7.3 LOW-AMPLITUDE DAMPING RATIO
As with the shear modulus presented i n sections 7.1 and
7.2, damping r a t i o i s presented seperately i n terms of low
amplitude and high amplitude r e s u l t s .
119
18.0
Strain NQ
4.0 - QQQQQ 1.5x1O'J* .158
arjpati 2.1x10" * .171
2
- A A ^ A A 1.0x10"]* .165
0&.00i> 3.6x10~ g .161
1
2.0 -rr\ 1 i—i i i i 111 1 1—i i i i 111 1 1—i i i i 11
10 100 1000 10000

F i g . 7.23 Time-Dependent I n c r e a s e i n Shear Modulus f o r

Various S t r a i n Levels
120
The low amplitude damping r a t i o f o r three block samples of
Lower 232nd St. Clay are shown i n F i g . 7.24 f o r a wide range of
effective confining stress and OCR values. Very little
influence of confining stress or stress history i s seen on the
low amplitude damping r a t i o , which i s determined at a s t r a i n of
0.001% by the amplitude decay method. Regardless of the state

of stress, the damping r a t i o f o r a l l tests shown i s between 1.0
and 1.7%.
On the other hand, the damping r a t i o values determined

from the tube samples, which are of lower p l a s t i c i t y , decrease
with confining pressure (Fig. 7.25). Based on the limited
data, i t appears that the damping r a t i o decreases i n direct
proportion to the increase in confining stress. However,
damping ratio differences between individual samples are
greater than the e f f e c t of confining pressure over the range
tested. Considering the range i n confining pressure and OCR
values, the 0.9% to 2.4% range i n damping f o r the tube samples
i s r e l a t i v e l y narrow. The damping r a t i o s presented herein are
those values determined at approximately 1000 minutes of
confinement.
121
5.0
Lower 232 nd
St. Clay Block Sample 2.0m
solid symbols represent

overconsolidated states TEST 0"3C OCR
.£ 4.0 f B9C2 74 kPa 1.6
a
i_ B9C3 146 1.0
-int> V B9C4 224 1.0
B9C6 101 2.9
LB9C7 32 9.3
fB17C1 484 1.0
o 3.0 \ B17C2 103 4.7
6 LB17C3 53 9.2
B19C2 109 1.1
a B19C3 268 1.0
B19C4 134 2.0
B19C5 50 5.4
200 300 500

Confining Stress (kPa)
F i g . 7.24 V a r i a t i o n of Low Amplitude Damping Ratio with

E f f e c t i v e Confining Stress f o r Block Samples
122
5.0
Lower 232 nd
St. Clay Tube Samples
solid symbols represent

overconsolidated states OCR
TEST oW
£ 4.0
8.2 m O T1C2 71 kPa 1.0
D T2C3 250 1.0
2.6 m
•+->
CO TS1C2 69 1.0
TS1C3 198 1.0
11.75m TS1C4 397 1.0
8
D
3.0 .TS1C5 103 3.9

d TS2C2 76 1.0
TS2C3 137 1.0
D 13.2 m TS2C4 268 1.0
2.0 - *"Q
c
'a.
£
0.0 —I 1 1 1 1 r
100 200 300 400 500
Confining Stress (kPa)
Fig. 7.25 V a r i a t i o n of Low Amplitude Damping Ratio with

E f f e c t i v e Confining Stress f o r Tube Samples
123
To study the time dependent decrease i n damping ratio

modulus at constant effective confining stress, damping
measurements f o r a number of t e s t s were taken well beyond the
time required f o r completion of primary consolidation.
The decrease i n damping r a t i o with the logarithm of time

i s shown i n F i g . 7.26 f o r a block sample of Lower 232nd St.
Clay from a depth of 2.0 m. For a l l f i v e stress states shown,
the damping ratio determined by the steady state method
decreases l i n e a r l y with the logarithm of time. However, f o r
the two unloading increments shown, the damping r a t i o remains
constant until the completion of swelling and decreases
thereafter. In a l l cases, swelling or consolidation was
complete before 250 minutes.
The rate of secondary decrease i n damping r a t i o , N , can d
be quantified by
N d - AD/D 1 0 0 0 (7.10)
where AD = secondary decrease i n damping r a t i o per

log cycle of time
D
1000 =
damping r a t i o determined at 1000 minutes
of confinement
The value of N d i s shown i n F i g . 7.26 f o r each stress state.
Values range from 39-60% which suggests that the damping has an
124
5.0
Lower 2 3 2 nd
St. Clay Block Sample 2.0 m
OCR N,
4.0 H 03C
74 1.6 .60
3 146 1.0 .39 300-i
cn 4 224 1.0 .52
c ooooo 6 101 2.9 .42
*|-3.0 A A A A A 7 32 9.3 .43 Q_
o
Q 7
a Load Sequence
^ 2.0
>s
"O
O
<D
•*->
m
1.0 H
0.0 ''M i—i—i i i 111 1 1—i—i i i 111—
10 100 1000
Time (min)
F i g . 7.26 Decrease i n Damping R a t i o w i t h Time f o r B l o c k

Samples
125
even greater relative time dependency than the 14-23%
determined f o r shear modulus.
For tube samples, the time dependent damping response i s

much the same as that identified f o r the block samples.
Referring to F i g . 7.27, the normalized rate of damping
decrease, N , d ranges from about 35-43% per log cycle of time.
The N d values determined from a number of tests are shown

i n Table 7.2 along with the respective OCR and e f f e c t i v e stress
values. The range i n N d values i s 15-60% with an average of
38%. This corresponds to a decrease i n damping ratio of
approximately 0.5% per log cycle of time ( i n minutes) which
compares quite well to the 0.4-0.7% decrement of damping r a t i o
i d e n t i f i e d by Kokusho et a l . (1982) i n t h e i r tests on Teganuma
Clay.
Applying an increment of confining pressure erases much of

the aging e f f e c t of the previous confining pressure. This i s
evident from Figs. 7.26 and 7.27 where the damping r a t i o always
increases a f t e r a change i n confining pressure and decreases
with time thereafter.
126
10.0
Lower 232 St. Cloy
nd
Tube Sample: 11.75 m

400-1
8.0 H a
CL
Load Sequence
6.0 H
# °3C OCR N d
ooooo 2 75 kPa 1.0 .415

• ••DP 3 200 1.0 .355
4 397 1.0 .349
4.0 AAA4A
103 3.9 .430
5
2.0 H
0.0 i i 111— -i—i—i i i 111 1— T — r i i i 111—

10 100 1000
Time (min)
F i g . 7.27 Decrease i n Damping R a t i o w i t h Time f o r

Tube Samples
127
Table 7.2 Values of f o r Samles at
Various a^c' a n d 0 C R
TEST 3C'
ff
OCR Nd
(kPa)
B8C2 77.4 1.0 .350
B9C2 74.3 1.6 .600
B9C3 146.4 1.0 .388
B9C4 223.5 1.0 .515
B9C6 101.0 2.9 .421
B9C7 32.0 9.3 .429
B12C3 292.5 1.0 .152
B14C2 70.0 1.7 .207
B19C2 109.3 1.0 .515
TS1C2 75.0 1.0 .415
TS1C3 198.0 1.0 .355
TS1C4 397.0 1.0 .349
TS1C5 103.0 3.9 .430
128
7.4 HIGH-AMPLITUDE DAMPING RATIO
7.4.1 E f f e c t of Strain and Nonlinearity
A t y p i c a l high amplitude damping curve f o r lower 232nd St.

Clay i s given i n F i g . 7.28 with values given by both the steady
state and amplitude decay methods. The strain amplitude
reported f o r tests employing the amplitude decay method i s that
immediately p r i o r to power cutoff ( i e . the maximum value). At
shear s t r a i n s less than 0.01%, both methods give e s s e n t i a l l y
the same constant low amplitude damping value. But, with
increasing strain the steady state method gives a smaller
damping attenuation than the amplitude decay method. This high
amplitude discrepancy occurred i n v i r t u a l l y a l l high amplitude
tests conducted i n t h i s study as well as i n t e s t s previously
performed on sand. Since the steady state method assumes a
symmetrical frequency response curve which i s not a good
assumption f o r high amplitude t e s t s , the amplitude decay method
i s preferable.
As discussed i n section 7.2.1, the frequency response of a

s o i l sample becomes nonlinear as the s t r a i n s exceed a threshold
value of approximately 0.005%. S i m i l a r l y , the c a l c u l a t i o n of
damping r a t i o w i l l be affected by the nonlinear response.
129
16.0
Lower 232 St. Clay
nd
Tube Sample: 13.2m
12.0 H
•DODO Steady State

8.0 H * * * * * Amplitude Decay
TEST 0-30' OCR

•
TS2C4 268 kPa 1.0
4.0 H
0.0 —1—1 1 1 11111 1—i 1 i 11111 1—1 1 1 1111| -i—1 1 11111
0.0001 0.001 0.01 0.1

Shear Strain (%)
F i g . 7.28 Steady S t a t e and A m p l i t u d e Decay Damping vs

Shear S t r a i n
130
In the amplitude decay of free vibrations method discussed
i n section 4.3.2.1, the damping r a t i o i s calculated from the
logarithmic decrement of free vibrations monitored a f t e r the
d r i v i n g current i s turned o f f . The damping r a t i o i s calculated
from Eg. 4.9 i n which the logarithmic decrement, d, i s assumed
to be constant i n order to y i e l d a unique value of damping
ratio. However, for s t r a i n s i n excess of the threshold value
the logarithmic decrement w i l l not be unique but w i l l vary with
the number of free v i b r a t i o n cycles selected. This e f f e c t can
be seen i n F i g . 7.29 where the normalized amplitude i s plotted
on a logarithmic axis against the number of free v i b r a t i o n
cycles. The normalized amplitude i s the peak accelerometer
output f o r a p a r t i c u l a r cycle divided by the peak accelerometer
output prior to drive current cutoff. The logarithmic
decrement becomes increasingly nonlinear with number of cycles
as the shear s t r a i n increases. In order to minimize the e f f e c t
of such nonlinearity on the calculated damping ratio, the
logarithmic decrement for a l l tests was established at a
normalized amplitude of 0.2 to 0.3 where the deviation from the
initial linear logarithmic decrement i s negligible for a l l
s t r a i n s , within the range tested (Fig. 7.29).
Since the decay of free vibrations i s determined from the
accelerometer response, it is not the true displacement
response required by Eq. 4.8 i n the c a l c u l a t i o n of damping.
However, t h i s introduces only a small error as the displacement
and acceleration are e s s e n t i a l l y related by a constant f o r any
131
TEST TS2C4 o- c'=268 kPa
3
0 3 5 8 10 13 15 18 20 23
Cycle No.
F i g . 7.29 E f f e c t o f N o n - L i n e a r i t y on d e t e r m i n a t i o n o f
A m p l i t u d e Decay Damping
132
particular test. As can be seen i n F i g . 7.30 the shapes of the
accelerometer decay and rotational displacement (RVDT) decay
curves are v i r t u a l l y i d e n t i c a l . In t e s t s where the r o t a t i o n a l
transducer was used, very good agreement between the curves was
observed f o r both high s t r a i n and low s t r a i n t e s t s confirming
that the use of the accelerometer decay curve to determine the
material damping r a t i o i s a v a l i d procedure.
The influence of confining pressure and stress history on

the high amplitude material damping can be evaluated by
comparing the damping attenuation curves at d i f f e r e n t stress
states. Damping attenuation curves f o r a l l high amplitude
t e s t s are includes i n Appendix B.
For block samples stage tested, there i s no noticeable

inluence of e f f e c t i v e confining pressure and stress history on
the damping attenuation curves, within the range tested. This
is clearly evident i n F i g . 7.31 where the damping data from
different samples and testing conditions fall on the same
curve. Thus, a single damping attenuation curve can be
extended to apply to a wide range of confining stress and OCR
values. Since the corresponding normalized modulus reduction
values formed a unique curve (Fig. 7.20), the use Eq. 2.8 to
relate high amplitude shear modulus and damping appears
133
8.3C— 1 1 1 1
1 1 1 1 1
1 1 1 1 r
1.6s 1.5s 2.Bs

TIME
Fig. 7.30 Comparison of Accelerometer and RVDT Decay

Response
134
16.0
Lower 232 St. Clay nd
Block Sample: 2.0m
12.0 -
c - TEST 0"3C OCR
£ * He * * *B17C1 484 kPa 1.0
o • • • • • B17C2 103 4.7
Q A A A A A B17C3 53 9.2
>% - OOOOO B19C2 109 1.1
D 8.0 - 00000 B19C3 268 1.0
Q
O
<D ***** B19C5 50 5.4
q>
xi
E
< 4.0 - AO
*
tf
0.0 I I I I I I I I — i—i i IIIIIII I—i—i 11111

IIIIIII 1
0.0001 0.001 0.1
0.01
Shear Strain (*)
F i g . 7.31 Damping v s . Shear S t r a i n f o r B l o c k Samples a t

V a r i o u s C o n f i n i n g P r e s s u r e s and OCR
135
reasonable, though Eg. 2.8 i n c o r r e c t l y suggests a low amplitude
damping of zero when G=G max .
However, stage tested tube samples obtained from depths of

11.75 m and 13.2 m show some dependence of damping r a t i o on the
e f f e c t i v e confining stress (Fig. 7.32). Generally, the higher
the confining pressure the lower the damping attenuation curve;
t h i s e f f e c t i s masked by differences i n the number of loading
cycles f o r s t r a i n s greater than about 0.1%.
Also shown i n F i g . 7.32 i s the damping attenuation curve

identified f o r the tests on block samples i n F i g . 7.31. The
comparison indicates that the higher p l a s t i c i t y block samples
have a lower damping r a t i o f o r the e n t i r e range of strains
tested. The shape of the modulus attenuation curves f o r both
types of samples, however, i s essentially the same. These
findings are very s i m i l a r to those established by Kokusho et
al. (1982) i n t h e i r tests on Teganuma Clay.
7.4.3 Secondary Time E f f e c t
Although a comprehensive series of tests to e s t a b l i s h the
e f f e c t of confinement time on the high amplitude damping r a t i o
were not performed, i t nevertheless appears that aging has an
effect on the high amplitude damping. Based on the limited
data i t appears that the duration of confinement influences not
136
16.0
Lower 232 St. Clay
nd
Tube Samples: 11.75- 13.2m
TEST 0"3c' OCR

12.0 - ***** TS1C2 69 kPa 1.0
CD ooono TS1C3 198 1.0
c AAAAA TS1C4 397 1.0
'CL 00000 TS1C5 103 3.9
E ooooo TS2C2 76 1.0
D ***** TS2C3 137 1.0
Q
>% x xxxx TS2C4 268 1.0
O 8.0 -
o
Q
Q>
•o
"5.
E 4.0 -
<
average curve for

block samples
0.0 —j 1—i i 111111 1—i i 111 i i | 1—i i 111 H I T—I I I I I 11
0.0001 0.001 0.01 0.1

Shear Strain (ss)
Fig. 7.32 Damping vs. Shear Strain f o r Tube Samples at

Various Confining Pressures and OCR
137
only the low amplitude damping r a t i o (section 7.3.2) but also
the high amplitude values (Fig. 7.33); longer confining times
decrease the damping r a t i o over the entire s t r a i n range tested.
Therefore i t may be possible to use the "unaged" damping
attenuation curve along with the "aged" low amplitude damping
value i n order to establish the aged high amplitude damping
ratio.
7.5 PORE PRESSURE RESPONSE
In saturated s o i l subjected to a c y c l i c stress-strain

history, excess pore pressures may develop thereby reducing the
shear resistance of the s o i l . I t i s therefore of i n t e r e s t to
examine the relationship between excess pore pressures and
factors such as confining pressure and shear s t r a i n amplitude.
During resonant column testing, dynamic pore pressures are

r a r e l y reported because of the fast rate of t o r s i o n a l loading
which results in a non-uniform pore pressure distribution
within the specimen. However, i f pore pressures are allowed to
equalize between v i b r a t i o n intervals, an i n d i c a t i o n of the
magnitude of the pore pressure response may be obtained from
the residual excess pore pressure.
For several tests the excess pore pressure was monitored
both during and a f t e r the usual one minute v i b r a t i o n i n t e r v a l ,
138
10.0
Lower 232 St. Clay
nd
Block Sample 2.0m
8.0 -
Steady State Damping
6.0 -
D
CH
cn
c
"Q.
E 4.0 - •
o
Q
2.0 - * *
• •
* * * * * confined 3 hours
• noon confined 93 hours
0.0 T 1 I I I I| 1 1 1 I I I I T 1—I—IIII
0.01 0.1
Fig. 7.33 Effect of Duration o f Confinement on H i g h

Damping
/ A m p l i t u d e
139
until a steady value was obtained. One minute of vibration
corresponds to 600 to 1800 cycles, depending on the resonant
frequency which in turn depends primarily on the strain
amplitude and confining pressure. The pore pressure response
for a block sample of Lower 232nd St. Clay is shown in
F i g . 7.34 f o r four l e v e l s of shear s t r a i n . For each strain
l e v e l , some time beyond the end of the v i b r a t i o n i n t e r v a l i s
required f o r the pore pressures to reach a steady value. This
was generally achieved within f i v e minutes a f t e r c y c l i n g was
terminated. The residual value of excess pore pressure
increases with increasing s t r a i n amplitude. Precisely the same
behaviour i s seen for a tube sample from a depth of 8.2m
(Fig. 7.35). Overconsolidated samples, however, displayed a
dilatent behaviour showing increasingly negative excess pore
pressures with s t r a i n amplitude (Fig. 7.36).
Six specimens normally consolidated at e f f e c t i v e confining

pressures ranging from 72 kPa to 486 kPa were subjected to
shear strains as high as 1.0%. Figure 7.37 shows the
r e l a t i o n s h i p between the shear s t r a i n and the residual excess
pore pressure, u, normalized by the effective confining
pressure. Again, i t should be noted that the residual pore
pressures are those r e s u l t i n g from about 600 to 1800 vibration
cycles (one minute of v i b r a t i o n ) , depending on the resonant
frequency of the sample which i n turn depends primarily on the
confining pressure and shear strain amplitude. A l l six
specimens show e s s e n t i a l l y the same behaviour with very little
140
50.0
40.0 -
D
D_
CO
30.0
CO
o 20.0 -
Q_
CO
CO
<D
O
X
LU
10.0 -
i i I i i i i i i i i i I i i i i i i i i i | i i II
2.0 4.0 6.0 8.(
Time (min)
Fig. 7.34 Pore P r e s s u r e Response f o r Normally

C o n s o l i d a t e d Block Samples i n Resonant Column
Test
141
20.0
0.0 2.0 4.0 6.0 8.0

Time (min)
F i g . 7.35 Pore Pressure Response f o r Normally

C o n s o l i d a t e d Block Samples i n Resonant Column
Test
142
10.0
TEST B 1 7 C 3 : oV=50 kPa 0CR=9.7
7=8.4x10 %
7=3.3x10
7=1.2x10°%
-10.0 -
15.0 I i i i i I i i ii ii ii ii i i i i i i i i i i i i i i | i i i i i i i i i
0.0 2.0 4.0 6.0 8.0
Time (min)
F i g . 7.36 Pore P r e s s u r e Response f o r O v e r c o n s o l i d a t e d

B l o c k Sample i n Resonant Column T e s t
143
0.001 0.01 0.1 1
Shear Strain (%)
Fig. 7.37 Normalized Pore Pressure Response for Normally-

Consolidated Clay in Resonant Column
144
pore pressure generation below 0.01% strain. Above 0.1%
strain, the normalized residual pore pressure rises rapidly
with shear strain suggesting the onset of large plastic
strains. Included i n F i g . 7.37 i s the pore pressure response
for a tube sample from 8.2m represented by the s o l i d circles
which shows essentially the same behaviour as the block
samples. The resonant v i b r a t i o n frequencies f o r shear strains
greater than 0.1% were between 8 Hz and 19 Hz, depending on the
confining pressure. Due to the rapid r i s e i n pore pressure,
f a i l u r e of the samples would have occurred f o r s t r a i n s much i n
excess of 1.0%.
When compared to the damping attenuation curves shown i n

F i g . 7.38, the normalized pore pressures respond in a very
s i m i l a r fashion with s t r a i n . Also, i t appears that the onset
of pore pressure development and, hence, p l a s t i c strains at
about 0.01% corresponds to the s t a r t of s t r a i n softening i n
shear modulus. Thus, for saturated normally consolidated
samples the normalized residual excess pore pressure, the
damping r a t i o , and the normalized shear modulus are similarly
related to shear s t r a i n .
145
7.38 Comparison of Pore Pressure, Damping, and Shear
Modulus vs. Strain Curves f o r Normally
Consolidated Clay
146
8. COMPARISON OF TORSIONAL SHEAR & RESONANT COLUMN RESULTS
Towards the end of the research program, the resonant

column apparatus was modified to enable t e s t s to be performed
at any chosen frequency while maintaining the c a p a b i l i t y of
performing resonant tests on the same sample. The apparatus
modifications are discussed i n section 3.4 and the theory
underlying the t o r s i o n a l shear t e s t i s given i n section 4.3.
These changes enable the comparison of dynamic properties
obtained at various testing frequencies by means of the
combined resonant column/torsional shear method. S p e c i f i c a l l y ,
the shear modulus, damping r a t i o , and pore pressure response
are considered i n the following sections.
8.1 SHEAR MODULUS
Shear modulus values from both resonant column and

t o r s i o n a l shear tests for two tube samples from depths of 11.75
and 13.2m are shown i n Figs. 8.1 and 8.2, respectively. The
resonant column tests, shown by open symbols, and the 0.1 Hz
t o r s i o n a l shear t e s t s , represented by s o l i d symbols, show good
agreement at the low confining pressures. As the e f f e c t i v e
confining pressure increases, the difference between shear
modulus values obtained by the two methods also increases. In
addition, the discrepency between resonant column and t o r s i o n a l
shear values generally decreases with increasing strain
147
open symbols : resonant "column test
120.0 closed symbols : torsional shear test, 0.1 Hz
TEST
ooooo TS1C2 69 kPa
100.0
£ 80.0
CO
3
•o 60.0
o
20.0 -
Lower 232"° St. Clay 11.75m

0.0 i 1111— T 1—I I I I I I I —I—I—I I I 1111 i — i — i i i 111
0.001 0.01 0.1

Fig. 8.1 Shear Modulus vs. Strain Curves from Resonant

from 11.75m
148
open symbols : resonant column test
80.0 closed symbols : torsional shear test, 0.1 Hz
TEST
kPa
60.0 -
D
Q_
CO
3
4? 40.0
o
0)
20.0 -
Lower 232 nd
St. Clay 13.2m
0.0 -n— T 1 I I I I I| T 1 I I I I I| T 1—I I I I I
0.001 0.01 0.1

F i g . 8.2 Shear Modulus v s . S t r a i n Curves from Resonant

Column and T o r s i o n a l Shear T e s t s f o r Sample
from 13.2m
149
amplitude. The d i s p a r i t y i n shear modulus values appears to be
l a r g e l y due to differences i n the o s c i l l a t i o n frequencies which
represent differences in strain rate between t o r s i o n a l shear
and resonant column t e s t s .
Figures 8.3 and 8.4 show the shear modulus values from
t o r s i o n a l shear tests, conducted at frequencies of 0.01, 0.1,
and 1.0 Hz as well as the shear modulus values obtained from
the resonant column. For the t o r s i o n a l shear tests i t is
evident that the shear modulus increases with increasing
frequency, and hence, with increasing s t r a i n rate. The higher
shear modulus values obtained i n the resonant column t e s t might
be expected since the resonant column tests are run at higher
frequencies.
In order to extend the torsional shear results to

frequencies greater than about 1.0 Hz, a torque c e l l would
have to be built into the apparatus to measure the applied
dynamic shear stress. For frequencies of about 1 Hz or less, a
s t a t i c c a l i b r a t i o n was found to be s u f f i c i e n t . On average i t
was found that the rotational transducer (RVDT) gave shear
strains which were about 15% lower than those given by the
accelerometer calibration in the resonant column test.
However, the differences in observed shear modulus values
between resonant column and torsional shear tests cannot be
accounted f o r by such a small difference i n shear s t r a i n s . In
fact the shear s t r a i n difference i s such that the discrepency
150
50.0
Lower 232 St. Clay
nd
13.2m
TEST TS2C3
03c'=137 kPa
40.0 -
D
CL
30.0 -
to
X»
O
20.0
o
CO
JZ
I/)
10.0 - * * * * * Resonant Test (10-23 Hz)

ooooo 1 Hz Torsional
• D D D D 0.1 Hz Torsional
A A A A A 0.01 Hz Torsional
0.0 i i i 1111 1—i—i i i 1111 -i—i—i i 111
0.001 0.01 0.1

F i g . 8.3 Frequency ( S t r a i n Rate) Dependency o f Shear

Modulus v s . S t r a i n Curves
151
80.0
Lower 2 3 2 St. Clay
nd
13.2m
TEST TS2C4
03C-268 kPa
70.0 -
.60.0 -
D
CL
to 50.0 -
•o
o
2
i_
40.0 H
D
CO
co
30.0 -
* * * * * Resonant Test (20-30 Hz)

20.0 ooooo 1 Hz Torsional
• •••• 0.1 Hz Torsional
- A A A A A 0.01 Hz Torsional
10.0 •n— T—r i i i 111 1 1—i i i i i 11 T—I I I I I I
0.001 0.01 0.1

F i g . 8.4 Frequency (Strain Rate) Dependency of Shear

Modulus vs. Strain Curves
152
between resonant column and torsional shear values would be
s l i g h t l y greater i f shear s t r a i n corrections were to be made.
Since i t i s not known which method y i e l d s a more accurate value
of s t r a i n , no corrections were made.
Based s o l e l y on the r e s u l t s of t o r s i o n a l shear tests, a

100-fold increase i n frequency from 0.01 to 1.0 Hz r e s u l t s i n
an increase i n shear modulus of only 10% at a s t r a i n of 0.01%.
The e f f e c t of s t r a i n rate therefore appears to be f a i r l y minor,
i n the range investigated.
It should be mentioned that no frequency effect was

observed i n a t o r s i o n a l shear/resonant column t e s t conducted on
a dry sand sample. Dynamic and s t a t i c t o r s i o n a l shear tests
performed by Bolton and Wilson (1989) confirm that the dynamic
response of dry sand i s independent of frequency.
8.2 DAMPING RATIO
The damping attenuation curves f o r tube samples from 11.75

and 13.2m consolidated to the i n s i t u octahedral stress are
shown in Figures 8.5 and 8.6, respectively. Regardless of
whether the t o r s i o n a l shear or resonant column method i s used,
the damping i s approximately the same f o r any p a r t i c u l a r shear
strain. For the resonant column tests, the amplitude decay
method was used to determine the damping whereas the damping
153
16.0
nd
Lower 232 St. Clay 11.75m
Test TS1C2 Osc-69 kPa
14.0
* * * * * Amplitude Decay (resonant)

12.0
ooooo Torsional, 1.0 Hz
aoann Torsional, 0.1 Hz
A A A A A Torsional, 0.01 Hz
10.0
.£ 8.0
E
o
Q
6.0
4.0 -
2.0 -
0.0 —i—i i 111 H I 1—i i 111111 1—i i 1111n T—I I I II 11

0.0001 0.001 0.01 0.1
Shear Strain (%)
F i g . 8.5 Damping Attenuation Curves Determined at

Various Frequencies f o r Clay from 11.75m
154
16.0
Lower 232 nd
St. Clay 13.2m
Test TS2C2 cf =76 kPa
3c
,
14.0
* * * * * Amplitude Decay (Resonant)
ooooo Torsional Shear, 1.0 Hz
12.0 ••nnn Torsional Shear, 0.1 Hz *A
A A A A A Torsional Shear, 0.01 Hz
10.0 / a
•§ 8.0
*/
E
Q
6.0
4.0
2.0
0.0 1 1—1 1 1 1 III 1 1 IIIIIII 1 IT I llll| T 1 1 1 1 1 ll|
0.0001 0.001 0.01 0.1

Shear Strain (55)
Fig. 8.6 Damping Attenuation Curves Determined at

Various Frequencies f o r Clay from 13.2m
155
ratio from the t o r s i o n a l shear t e s t s was determined from the
measured s t r e s s - s t r a i n hysteresis loops.
Based on the limited data, the damping r a t i o appears to be

unaffected by frequency, and hence strain rate, since the
t o r s i o n a l shear tests conducted at 0.01, 0.1, and 1.0 Hz and
the resonant column tests give e s s e n t i a l l y the same r e s u l t .
Since the dissipated energy per cycle seems to be independent
of the frequency of loading, the damping ratio can be
considered to be h y s t e r e t i c . Viscous damping, on the other
hand, would require the dissipated energy to increase
proportionally with frequency.
8.3 PORE PRESSURE RESPONSE
For the t o r s i o n a l shear tests conducted, f i n a l excess pore

pressures were monitored only f o r tests at a frequency of
1.0 Hz. In order to compare r e s u l t s of t o r s i o n a l shear and
resonant column tests at a comparable number of loading cycles,
at least 300 cycles of loading were performed i n the t o r s i o n a l
shear t e s t before the pore pressures were allowed to equalize.
This large number of cycles would require too long a v i b r a t i o n
interval at 0.01 Hz or 0.1 Hz to be completed i n a reasonable
period of time. Thus, only the 1.0 Hz values are considered
herein.
156
The normalized residual excess pore pressures versus
s t r a i n f o r both block samples and tube samples i n the resonant
column was given i n F i g . 7.37. This data i s shown i n F i g . 8.7
along with the data from t o r s i o n a l shear tests conducted on
tube samples at a frequency of 1.0 Hz. Clearly, the normalized
residual pore pressures are s i g n i f i c a n t l y higher f o r the
t o r s i o n a l shear tests than f o r the resonant column t e s t s which,
depending on confining pressure, have v i b r a t i o n frequencies of
8 to 22 Hz i n the 0.1 to 1% shear s t r a i n range. Though the
t e s t s are such that they are stress controlled as opposed to
s t r a i n controlled, the threshold s t r a i n below which no excess
pore pressures are generated appears to be about 0.01% - the
same as f o r resonant column t e s t s .
The r e s u l t s are consistent with those reported f o r c y c l i c

simple shear tests (Ansal and Erken, 1989) as well as those f o r
cyclic triaxial tests (Matsui et a l . , 1980), both of which
report a dependence of excess pore pressures on loading
frequency i n addition to a threshold stress r a t i o below which
no excess pore pressures are generated.
157
0.40
Normalized Pore Pressure Response
for Normally Consolidated
/
nd
0.30 -
/
o oW TEST
tn
0000 129 B16C1^
b B16C2
0.20 - A A A A 293
B18C1 Resonant
* * * * 146 B17C1 Column
486 B19C2 8-22 Hz
•••a 109 T1C2 _
tboooo 72 TS1C2 1 Torsional
• • • • • 69 TS2C2 J Shear
mm 76
0.10
0.00 •i rn q 1 D| | | I I 11 i — i — i i 1111
0.001 0.01 0.1

Shear Strain (ss)
F i g . 8.7 Comparison of Normalized Residual Pore Pressure

Response f o r Samples from Resonant Column and
Torsional Shear Tests
158
9. COMPARISON TO FIELD RESULTS
9.1 INTRODUCTION
I t i s often desireable to determine the dynamic properties

of soil in situ since laboratory tests may not always be
representative of f i e l d conditions. One of the best i n s i t u
techniques i s the seismic cone penetration t e s t (SCPT) which
has been used extensively at UBC since 1980 (Robertson et a l . ,
1986). At f i r s t , research was focused on determining the i n
situ small strain shear wave v e l o c i t y from which the shear
modulus can be e a s i l y calculated. These methods are now well
established (Campanella and Stewart, 1990). More recently,
emphasis has been placed towards determination of the i n s i t u
damping c h a r a c t e r i s t i c s (Stewart and Campanella, 1990).
The SCPT method involves h o r i z o n t a l l y s t r i k i n g a weighted

beam at the ground surface thereby producing a polarized shear
wave which t r a v e l s through the ground and i s sensed at depth by
a h o r i z o n t a l l y oriented accelerometer located near the t i p of
the cone. This procedure i s repeated as the cone i s advanced
i n one meter depth increments. Comparison of the successive
accelerometer records i s the basis on which the shear wave
v e l o c i t y and damping c h a r a c t e r i s t i c s are determined.
159
9.2 SHEAR WAVE VELOCITY
The shear wave v e l o c i t y p r o f i l e s determined from two SCPT
soundings are shown i n F i g . 9.1. The shear wave v e l o c i t i e s
from both tests increase almost linearly with depth from a
value of about 85 m/s at a depth of 3m to a value of
approximately 140 m/s at a depth of 20m. The shear strains
associated with the i n s i t u shear wave v e l o c i t i e s are i n the
range 0.001% to 0.0001% where the soil behaviour i s linear
elastic.
In order to compare laboratory and i n s i t u properties, the

stresses i n laboratory tests should c o r r e c t l y model those i n
the f i e l d . However, the i s o t r o p i c confining pressures applied
to laboratory specimens are rarely representative of the
anisotropic in situ stresses. Resonant column tests on
anisotropically consolidated clay have shown that the
deviatoric component of stress has no e f f e c t on the shear wave
velocity f o r stress ratios less than 2.5 (Hardin and Black,
1968). Only the octahedral normal stress was found to control
the shear wave v e l o c i t y .
In situ and laboratory tests performed by Sully (1990)
indicate that: the soil at the Lower 232nd St. site is
essentially normally consolidated below a depth of 6m; the
lateral coefficient at rest pressure (K )

Q f o r the normally
consolidated clay i s 0.55; and the s o i l can be considered as
160
Shear Wave Velocity (m/s)
0.0 50.0 100.0 150.0 200.0
O.o -I i i i i I i i i i I i i i i I i i i i
F i g . 9.1 Comparison of Shear Wave V e l o c i t i e s Determined

from Resonant Column and SCPT Methods
161
being i n a cross-anisotropic state of stress. These findings
coupled with the aforementioned findings by Hardin and Black
suggest that i t i s reasonable to compare the shear wave
v e l o c i t i e s from the i s o t r o p i c a l l y consolidated resonant column
t e s t s with the normally consolidated SCPT i n s i t u v e l o c i t i e s at
the same e f f e c t i v e octahedral stress. The i n s i t u octahedral
stress f o r the normally consolidated clay was calculated based
on a cross-anisotropic state of stress with a K Q value of 0.55.
The v a r i a t i o n of shear wave v e l o c i t y with depth f o r the

resonant column tests on tube samples i s shown by the thick
dotted line i n F i g . 9.1. This relationship was derived by
substituting Eq. 7.2 into Eq. 4.5 and solving f o r the shear
wave v e l o c i t y . On average, the laboratory derived v e l o c i t i e s
are about 10 m/s lower than the i n s i t u v e l o c i t i e s and the rate
of v e l o c i t y increase with depth i s the same. Lower laboratory
values might be expected since they were generally determined
a f t e r 1000 minutes of sample confinement, with no aging e f f e c t s
considered; aging, as exists in situ, always increases
s t i f f n e s s response or modulus. Nevertheless, the agreement
between laboratory and i n s i t u r e s u l t s i s quite good. Lastly,

i t should be noted that while the shear s t r a i n s corresponding
to the resonant column v e l o c i t i e s are s l i g h t l y greater than
those i n s i t u , they are s t i l l i n the l i n e a r e l a s t i c range and
should therefore give the same r e s u l t .
162
9.3 DAMPING RATIO
Several methods are available t o calculate the i n s i t u
damping r a t i o from the SCPT. Experience at UBC (Stewart and
Campanella, 1990) suggests that only the spectral slope method
developed by Redpath et a l . (1982) gives consistently
meaningful r e s u l t s .
The spectral slope method i s based on the equation
d[ln(A2/Al)]/(df*dR) = -z (9.1)
where A2/A1 = r a t i o of signal amplitudes i n

frequency domain from successive depths
dR = depth derivative
df = frequency derivative
z — a/f, where a i s an attenuation
coefficient
which has i t s development given i n Stewart and Campanella

(1990). The solution of Eq. 9.1 requires a signal processing
computer program. The c o e f f i c i e n t z i s determined by f i r s t
c a l c u l a t i n g the fast Fourier transform (FFT) of one signal at a
given depth, then f o r each deeper signal computations are made
of the FFT, the r a t i o of the FFT's, and the negative of the
natural logarithm (ln) of the r a t i o . Next, the slope of the
l n ( r a t i o ) versus frequency p l o t (spectral r a t i o slope) f o r each
depth i s plotted against depth. Such a p l o t i s given i n
F i g . 9.2 f o r the Lower 232nd St. s i t e between 5 m and 20 m.
163
Spectral Ratio Siope
0.000
n n m m m m
0.002
m
0.004 0.006 0.008
m m m m n i i m n n n i n m n M m l
Slope of best fit is L290SC1

3.6180x10 s/rn
Fig. 9.2 Spectral Ratio Slope Method f o r Calculation

of In S i t u Damping
164
The slope of t h i s depth p l o t gives the z c o e f f i c i e n t f o r the
particular soil layer. Finally, the fraction of critical
damping (D) can be computed from
D - (zV )/(2*)
s (9.2)
where V s i s the shear wave v e l o c i t y .
Using eqn. 9.2, the spectral slope method y i e l d s a damping

r a t i o of 0.6% f o r the s o i l between 5 and 20 m. Analysis of two
other SCPT soundings gave damping ratios of 1.2% and 1.9%,
respectively. These values compare well with the 0.9% to 2.4%
range of unaged damping r a t i o values determined from resonant
column tests on normally consolidated samples from 8-14 m
(section 7.3.1). As discussed i n section 7.3.2, the aging
e f f e c t on damping i n resonant column tests appears to be very
s i g n i f i c a n t with the damping r a t i o decreasing about 0.5% per
log cycle of time (minutes) at constant e f f e c t i v e stress. If
the aging effect were considered the damping ratio from
resonant column tests would be lower. However, since the
length of time to which the laboratory aging e f f e c t can be
extended i s not known, a time extrapolation of resonant column
r e s u l t s to f i e l d conditions i s d i f f i c u l t to make at t h i s time.
165
10. CONCLUSIONS
The following i s a summary of the dynamic c h a r a c t e r i s t i c s
and response of Lower 232nd St. clay, based on the r e s u l t s of
resonant column and t o r s i o n a l shear t e s t s performed f o r t h i s
study.
LOW AMPLITUDE SHEAR MODULU8 (Gmax)
Under i s o t r o p i c confinement, a l i n e a r log-log relationship

e x i s t s between dynamic shear modulus (Gmax) a n d
effective
isotropic confining pressure (03c'' )• F o r t n e
higher
plasticity block samples (PI=40) the r e l a t i o n s h i p can be
expressed as
Gâx = 29.3*F(e)P a (a c')- OCR-

3
60 26
(10.1)
where F(e) = (5.2-e) /(1+e) 2
i n which P a i s the atmospheric pressure i n the same units as

G
max a n c
* 3c''
a a n a
- e
*- the void r a t i o .
s
For normally consolidated tube samples having Plasticity
index of 20-25, the relationship i s given by
Gmax = 1 3 7 . 2 * F ( e ) P a - 3 8 ( a 3 C ' ) * 6 2
(10.2)
166
where F(e) = (2.97-e) /(l+e)
2
The commonly used Hardin and Drnevich equation f o r shear

modulus s i g n i f i c a n t l y overpredicts the value of Gmax f o r
both clay types. Although i t i s widely used, i t appears

that the Hardin and Drnevich void r a t i o factor with constant
of 2.97 i s v a l i d only f o r clays with PI l e s s than 25.
Overconsolidated samples generally have higher shear modulus

than normally consolidated specimens. The e f f e c t of stress
history on Gjaax ^ s
expressed by the OCR term in
equation 11.1 with exponent 0.26. This agrees with the
range i n OCR exponents suggested by Hardin and Drnevich f o r
clays with PI between 20 and 40.
Increasing duration of confinement at constant effective

stress was found to result in a significant increase i n
shear modulus. Typically, the rate of shear modulus
increase normalized by the 1000 minute value was found to
range from 14-23% per log cycle of time (minutes).
From t o r s i o n a l shear tests, i t was found that increasing the
t o r s i o n a l frequency from 0.01 Hz to 1.0 Hz resulted i n a 10%
increase i n shear modulus. Shear modulus values from the
resonant column t e s t were always higher than t o r s i o n a l shear
tests since they were always run at higher frequencies
(strain rate).
167
The resonant column shear wave v e l o c i t i e s , and hence shear
moduli, were found t o be s l i g h t l y lower than the i n s i t u
values determined from the seismic cone penetration t e s t .
The difference can be explained due t o the increased
s t i f f n e s s of the i n s i t u s o i l as a r e s u l t of aging (5,000 to
10,000 years).
HIGH AMPLITUDE SHEAR MODULUS
The shear modulus decreases with increasing shear s t r a i n f o r

strains greater than about .005%. Regardless of the
confining pressure or overconsolidation ratio, normalized
modulus reduction relationship (G/Gmax v s
* shear strain) can
be represented by a single curve f o r the higher p l a s t i c i t y
(PI=40) clay. A s l i g h t s h i f t of the G/Gmax curve towards
higher values occurs with increasing effective confining
pressure f o r the lower p l a s t i c i t y (PI=20) clay. Thus, the
influence of confining pressure diminishes with increasing
plasticity.
Based on limited data i t appears that duration of
confinement causes both the small s t r a i n and large strain
amplitude moduli t o increase at about the same rate thereby
leaving the normalized modulus reduction curve unaffected.
168
LOW AMPLITUDE DAMPING RATIO
Very little influence of confining pressure or stress
history was observed on the small s t r a i n (.001%) damping
ratio f o r the higher p l a s t i c i t y clay (PI=40). Damping
r a t i o s obtained from resonant column t e s t s ranged from 1.0-
1.7%. For the lower p l a s t i c i t y samples (PI=20) the damping
ratio was found to decrease slightly with effective
confining pressure. Damping ratios ranging from 0.9-2.4%
were obtained.
The damping r a t i o was found to s i g n i f i c a n t l y decrease with

increasing time at constant e f f e c t i v e stress. On average,
the damping r a t i o decreased about 0.5% per l o g cycle of time
(minutes).
The damping r a t i o appears to be unaffected by frequency, and

hence, strain rate, since the torsional shear tests
conducted at various frequencies and the resonant column
tests gave e s s e n t i a l l y the same r e s u l t .
The range of damping r a t i o s determined from the seismic cone

penetration test using the spectral slope method yielded
damping values from 0.6-1.9%. This range compares well with
the 1.2-2.4% i d e n t i f i e d from unaged resonant column t e s t s .
169
HIGH AMPLITUDE DAMPING
There i s no e f f e c t of confining pressure or stress history

on the high amplitude damping f o r the clay with PI of 40.
For the lower plasticity (PI=20) clay, there i s a s l i g h t
decrease i n damping with confining pressure. The higher PI
clay was found to have s l i g h t l y lower damping f o r the entire
s t r a i n range tested. This implies that s i t e s with higher
plasticity soils would tend to have higher ground motion
amplifications.
Based on r e s u l t s from resonant column and t o r s i o n a l shear

tests, there appears to be no significant effect of
frequency (strain rate) on the high amplitude damping r a t i o .
Limited data indicates that the increased duration of

confinement at constant effective stress decreases the
damping r a t i o over the entire s t r a i n range tested.
PORE PRESSURE RESPONSE
In the resonant column t e s t , no s i g n i f i c a n t excess pore

pressure i s generated u n t i l shear strains of 0.05-0.1% are
surpassed. Pore pressures increase rapidly thereafter.
170
Results of t o r s i o n a l shear t e s t s indicate that the magnitude
of the residual excess pore pressure increases s i g n i f i c a n t l y
with slower frequency.
171
BIBLIOGRAPHY
A f i f i , S.S. and Richart, F.E., J r . (1973), "Stress-History

E f f e c t s on Shear Modulus of S o i l s " , S o i l s and Foundations,
Japan, V o l . 13, No. 1, March, pp. 77-95.
Aggour, M.S., Tawfiq, K.S. and Amini, F. (1987), "Effects of
Frequency Content on Dynamic Properties f o r Cohesive
S o i l s " , Proceedings, Third International Conference on
Earthquake Engineering and S o i l Dynamics, Princeton, June,
Vol. 42, pp. 31-39.
Anderson, D.G. and Stokoe, K.H., I I , (1978), "Shear Modulus:

A Time Dependent S o i l Property", Dynamic Geotechnical
Testing, ASTM STP 654, American Society f o r Testing and
Materials, pp. 66-90.
Anderson, D.G. and Woods, R.D. (1976), "Time-Dependent Increase
in Shear Modulus of Clay", Journal of the Geotechnical
Engineering Division, ASCE, Vol. 102, GT5, May,
pp. 525-537.
Ansal, A.M. and Erken, A. (1989), "Undrained Behavior of Clay

Under C y c l i c Shear Stresses", Journal of Geotechnical
Engineering, ASCE, V o l . 115, No. 7, July, pp. 968-983.
Armstrong, J.E. (1978), "Post Vachon Wisconsin Glaciation,
Fraser River Lowland, B r i t i s h Columbia", Goelogical Survey
of Canada, B u l l e t i n 332.
Bolton, M.D. and Wilson, M.R. (1989), "An Experimental and
Theoretical Comparison Between S t a t i c and Dynamic
Torsional S o i l Tests", Geotechnique, V o l . 39, No. 4,
pp. 585-599.
Bishop, A.W. and Henkel, D.J. (1962). The Measurement of S o i l
Properties i n the T r i a x i a l Test. 2nd Ed., William Clowes
and Sons, London.
Campanella, R.G. and Stewart, W.P. (1990), "Seismic Cone
Analysis Using D i g i t a l Signal Proccessing f o r Dynamic S i t e
Characterization", Proceedings, 43rd Canadian Geotechnical
Conference, Quebec, Canada.
Drnevich, V.P. (1967), "Effect of Strain on the Dynamic

Properties of Sand", Ph.D. Thesis, University of Michigan,
at Ann Arbor.
Drnevich Long-Tor Resonant Column Apparatus Operating Manual
(1981), S o i l Dynamics Instruments, Inc., Lexington,
Kentucky.
172
Drnevich, V.P., Hardin, B.O. and Shippy, D.J. (1978), "Modulus
and Damping of S o i l s by the Resonant Column Method",
Dynamic Geotechnical Testing, ASTM STP 654, American
Society f o r Testing and Materials, pp. 91-125.
Hardin, B.O. (1965), "Nature of Damping i n Sands", Journal of

the S o i l Mechanics and Foundations D i v i s i o n , ASCE,
Vol. 91, No. SMI, Jan., pp. 63-97.
Hardin, B.O. and Black, A.M. (1968), "Vibration Modulus of

Normally Consolidated Clay", Journal of the S o i l Mechanics
and Foundations Division, ASCE, V o l . 94, No. SM2, March,
pp.353-369.
Hardin, B.O. and Drnevich, V.P. (1972a), "Shear Modulus and
Damping i n S o i l s : Measurement and Parameter E f f e c t " ,
Journal of the S o i l Mechanics and Foundations Division,
ASCE, V o l . 98, SM6, June, pp. 603-624.
Hardin, B.O. and Drnevich, V.P. (1972b), "Shear Modulus and

Damping i n S o i l s : Design Equations and Curves", Journal of
the S o i l Mechanics and Foundations Division, ASCE,
Vol. 98, No. SM7, July, pp. 667-692.
Hardin, B.O. and Richart, F.E. J r . (1963), " E l a s t i c Wave
V e l o c i t i e s i n Granular S o i l s " , Journal of the S o i l
Mechanics and Foundations Division, ASCE, V o l . 89,
No. SMI, pp. 33-65.
Isenhower, W.M. (1979), "Torsional Simple Shear/Resonant Column
Properties of San Fransisco Bay Mud", MASc Thesis. C i v i l
Engineering Department, University of Texas at Austin,
Austin, Texas, 307 p.
Ishihara, K. (1982), "Dynamic Response Analysis", Proceedings,

Int. Symposium on Numerical Models i n Geomechanics,
Zurich, September, pp. 237-259.
Kokusho, T., Yoshida, Y. and Esashi, Y. (1982), "Dynamic

Properties of Clay f o r Wide Strain Range", S o i l s and
Foundations, V o l . 22, No. 4, December, pp. 1-18.
Marcuson, W.F., I I I , and Wahls, H.E. (1972), "Time E f f e c t on

Dynamic Shear Modulus of Clays", Journal of the S o i l
Mechanics and Foundations Division, ASCE, V o l . 98,
No. SM12, pp. 1359-1373.
Marcuson, W.F., I I I , and Wahls, H.E. (1978), "Effect of Time

on Damping Ratio of Clays", Dynamic Geotechnical Testing,
ASTM STP 654, American Society f o r Testing and Materials,
pp. 126-147.
173
Matsui, T., Ohara, H. and Ito, T. (1980), " C y c l i c Stress-
S t r a i n History and Shear Characteristics of Clay", Journal
of Geotechnical Engineering, ASCE, Vol.106,
No. GT10, October, pp. 1101-1120.
Ni, Sheng-Huoo (1987), "Dynamic Properties of Sand under True

T r i a x i a l Stress States from Resonant/Column Torsional
Shear Tests. Ph.D. Thesis. C i v i l Engineering Department,
University of Texas at Austin, Austin, Texas, 421 p.
Novak, M. and Kim, T.C. (1981), "Resonant Column Technique for

Dynamic Testing of Cohesive S o i l s " , Canadian Geotechnical
Journal, Vol. 18, pp. 448-455.
Pollard, W.S., Sangrey, D.A. and Poulos, S.J. (1977), " A i r

Diffussion Through Membranes i n T r i a x i a l Tests",
Journal of the Geotechnical Engineering D i v i s i o n , ASCE,
Vol. 103, No. GT10, pp.1169-1173.
Redpath, B.B., Edwards, R.B., Hale, R.J. and Kintzer, F.C.

(1982), " Development of F i e l d Techniques to Measure
Damping Values for Near-Surface Rocks and S o i l s . Prepared
for NSF Grant No. PFR-7900192.
Robertson, P.K., Campanella, R.G., Gillespie,D. and Rice, A.,

(1986), "Seismic CPT to Measure In-Situ Shear Wave
V e l o c i t y . Journal of Geotechnical Engineering Division,
ASCE, Vol. 112, No. 8, pp. 791-804.
Sangrey, D.A., Pollard, W.S. and Egan, J.A. (1978), "Errors
Associated with Rate of Undrained C y c l i c Testing of Clay
S o i l s " , Dynamic Geotechnical Testing, ASTM STP 654,
American Society for Testing and Materials, pp. 280-294.
Seed, H.B. and I d r i s , I.M. (1970), " S o i l Moduli and Damping
Factors f o r Dynamic Response Analysis", Report No.
UCB/EERC-70/10, University of C a l i f o r n i a , Berkeley,
December.
Stewart, W.P. and Campanella, R.G. (1991), "Insitu Measurement
of Damping i n S o i l s " , Proc. 2nd International Conference
on Geotechnical Earthquake Engineering and S o i l Dynamics,
St. Louis, March.
Sully, J.P. (1990), "Evaluation of In S i t u Lateral Stress From
F u l l Displacement Tests", Ph.D. Thesis, C i v i l Engineering
Department, University of B r i t i s h Columbia.
Sun, J . I . , Golesoskhi, R. and Seed, B.H. (1988), "Dynamic

Moduli and Damping Factors f o r Cohesive S o i l s " , Report No.
UCB/EERC-88/15, University of C a l i f o r n i a , Berkeley,
August.
174
Weiler, W.A, J r . (1988), "Small-Strain Shear Modulus of Clay",
Earthquake Engineering and S o i l Dynamics II - Recent
Advances i n Ground-Motion Evaluation, ASCE Geotechnical
Special Publication, No. 20, pp. 331-345.
Zen, K., Umehara, Y. and Hamada, K. (1978), "boratory Tests

and In-Situ Seismic Survey on Vibratory Shear Modulus of
Clayey S o i l s with Various P l a s t i c i t i e s " , F i f t h Japan
Earthquake Engineering Symposium, pp. 721-728.
175
APPENDIX A:
Shear Modulus vs. Strain Curves
176
25.0
nd
Tube Sampie: 8.2m
20.0 H
o
CL
co
15.0 H
O 10.0 H
0)
co
5.0 H
TEST OCR
* * ** T1C2 71 kPa 1.0
0.0 — i — i i IIIIII 1—i I I I I I I I 1—i • 111 I I i 1—i i 11 i n

0.0001 0.001 0.01 0.1 1
Strain Amplitude (*)
8Z.I
Shear Modulus (MPa)

rO cn
O o o o o o o
o 0 0
o 0 0 0 0
O
O 111 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 111111 1 1 1 1 1 1 1 1 1 1 1 1 1 1
O
*
*
*
*
ro —i
—j.
m
o CO
o —i
O" ro c o
to""" m
o oq_
S"^
CD a>
—I CO
Q ~U * Q ro
MM* D 3 C J
> O *
3 o b O " CO
C oo
Q.
a>
6Z.T
Shear Modulus (MPa)

08T
Shear Modulus (MPa)

181
Shear Modulus (MPa)

ro o 00
o o o o
P b o
o o o
o I I I ! « I
i i H i i i t t I i i i i i i i i i I i i t i i t i i i I i t i i i i i i
CO _)
n
o oo
—i
o
O" C O
o
(/)""* CD CD
co
s
q
—\
CO
"0 Q ro
a 3 ^
-o ^
S8
cF
Is '
a
0
cOL
" CO
70
a> O
D
^ H
_
o -
381
Shear Modulus (MPa)

70.0
nd
Tube Sample: 11.75m

60.0 4
.50.0 -3
o
0-
co 40.0
n
o
"5
30.0 ^
o
4)
JC
20.0 9
10.0 3 TEST oV OCR

• • * • TS1C3 198 kPa 1.0
0.0 — i — I I IIIIII 1 — i n i t n i i — I T I nm 1—i i i run

0.0001 0.001 0.01 0.1
fr8T
Shear Modulus (MPa)

S8T
Shear Modulus (MPa)

0)
o o o o o o
o
o • • • • • • o
o
o o o o o o o
I I I I 1^1 I I I I I I I I I 1 I I I I I I I I I 1 I I I I 1 t I I I I I I I I I I I I I I I I I I I I I I I I I I I
—!
CO _|
• ^ n
o co
o
o
Hr
c o
o CD CD
3
AM • CO
Q ro
>p a q OJ
CD* *
3
C*J o
o- CD O " CO
70
c
Q.
CD
nd
Block Sample: 2.0m

16.0 H
* *
CL 12.0
3
TJ
J 8.0
4.0
TEST oV OCR
-f***» B2C1 100 kPa 1.2
0.0 H 1—» i 111111 1—r-rrnm 1—i i 11 ni| 1—r i 11 in
0.0001 0.001 0.01 0.1
10.0
nd
Block Sample: 2.0m
8.0 H
6.0 • *
4.0 H
2.0
TEST oV OCR
B4C1 57 kPa 1.8
0.0 1 1 I I I 1111 1 1 I I I 1111 1 1 I I I Mil 1 1 111111
0.0001 0.001 0.01 0.1

881
Shear Modulus (MPa)

30.0
Lower 2 3 2 St. Claynd
Block Sample: 2.0m
25.0 H
o
D_
2
20.0 H
0)
X)
o
15.0
o
4)
XL
10.0 H
TEST CT ' c OCR
3**** B5C2 103 kPa 1.6
5.0 — i — i i n i ii| 1—i I I I I I I I 1—i i i n i i | T I I I I 111
0.0001 0.001 0.01 , s 0.1

nd
Block Sample: 2.0m

16.0 H
cu 12.0
CO
3
O
8.0 H
4.0 H
TEST CT '
c OCR
4 * • * * B5C3 55 kPa 3.0
0.0 — i — i i 111111 1—» i i IIii| 1—i i i IiJ11 T — i i 11 m l
0.0001 0.001 0.01 0.1

T6I
Shear Modulus (MPa)

30.0
nd
Block Sample: 2.0m
25.0 H
D
CL
20.0 H
o
2
fc 15.0 H
o
4)
x:
CA)
10.0
TEST OCR
B7C1 144 kPa 1.0
5.0 — i — i l 11 n i ) 1—r I I I ni] 1—i i i1 m i T 1 1 I I 111

0.0001 0.001 0.01 0.1
£61
Shear Modulus (MPa)

v6T
Shear Modulus (MPa)

36.0
nd
Block Sample: 2.0m
32.0 H
28.0 H
O
2
o 24.0 H
XL
CO
20.0
TEST OCR
!]•»••«, B9C3 146 kPa 1.0
16.0 -| 1—i i i 111H 1—i I i 1111| 1—ill 1—i I Mill
0.0001 0.001 0.01 0.1

16.0
nd
Block Sample: 2.0m
12.0 H
D
CL
CO
3
•S 8.0
o
"5
a
0)
x:
4.0 H
CO
TEST oV OCR
* * * * * B9C7 32 kPa 9.3
0.0 1 1 I I Hll| 1 1 1 I 1 1 I I Mll| 1 1 I I I III

0.0001 0.001 0.01 , 0.1 1
40.0
nd
Block Sample: 2.0m
35.0 -
* •
o
2
30.0 -
CO
_D
3
X P
o
2E
25.0 H
JC
to
20.0 -
TEST oV OCR
q , , , * * B13C1 200 kPa 1.0
15.0 -\ 1 i i 11mi 1—i i i MIII 1—» n mil 1—ri 11MI
0.0001 0.001 0.01 , 0.1 1

20.0
nd
Block Sample: 2.0m
16.0 • * *
o *
*
0)
3
% 12.0
o
Z2
o
x:
to
• *
8.0 H
TEST oV OCR
•»••* B13C2 50 kPa 4.0
4.0
0.0001 0.001 0.01 0.1
14.0
Lower 232 St. Clay
nd
Block Sample: 2.0m
12.0 - 3
CL 10.0
-o 8.0
o
-c 6.0
to
4.0 H
TEST oV OCR
• D D o n n B14C2 70 kPa 1.7
2.0 1 1—II M i l l 1—I—I I 1 1111

i—i—i » 1111
0.001 0.01 0.1

24.0
Lower 232 St. Clay
nd
Block Sample: 2.0m
20.0 H
o
Q_
16.0 H
m
3
"O
o
fe 12.0
0)
JC
CO
8.0 H
TEST OCR
D D D D O B15C2 140 kPa 1.0
4.0 1 1—II i n n 1—i—i i i 1111 -

r ri 111
1
T
0.001 0.01 0.1

14.0
Lower 232 St. Clay
nd
Block Sample: 2.0m

12.0 3
10.0 3
a
m 8.0
"5
TJ
O
2 6.0 d
o
JC
to
4.0 -3
2.0 3 JEST oV OCR

qnnano B15C3 31 kPa 4.5
0.0 T 1—1 I M I 11 1 1—I I I I I 11

T — i — i i 111
0.001 0.01 0.1

zoz
Shear Modulus (MPa)

ro ro
0 0
P t° P P o ro
h o o o o o h h
i i l i m i t t u l i m u m h u t t i n i l n n i n u l i i t n m i
50.0
Lower 232 St. Clay
nd
n
Block Sample: 2.0m
40.0 H
D
D_
30.0 -\
(0
X)
o
20.0 H
o
JO
to
10.0
TEST oV OCR
• D D D D D B16C3 293 kPa 1.0
0.0 11111 1—i—i i 1111| T — i — i i 11111 1—i—I I I I I 11

0.001 0.01 0.1 1
70.0
Lower 232 St. Clay
nd
Block Sample: 2.0m

60.0 4
^50.0 -3
o
CL
IS
m 40.0
3
O
2
30.0 -3
D
4)
to
20.0 -3
10.0 -3 TEST OCR

D D D U P B17C1 484 kPa 1.0
0.0 1 I » 11| 1 1—I I M l l | 1 1—I "I I I l l | T — i — I I I I II
0.001 0.01 0.1

30.0
Lower 232 St. Clay
nd
Block Sample: 2.0m
25.0 H
c? 20.0
-u 15.0
o
I-
10 0
5.0 H
TEST oV OCR
• nDoan B17C2 103 kPa 4.7
0.0 1 1 1—I I I I 111 1 1—I I I I I 11 T 1—r i I I I I
0.001 0.01 0.1

902
Shear Modulus (MPa)

30.0
Lower 232 St. Clay
nd
Block Sample: 2.0m
25.0 d
c?20.0
to
•o 1 5 . 0
o
| 10.0
5.0 d
TEST OCR
Il o • • n • B18C2 146 kPa 1.0
0.0 1 — i — i i 11111 1 — i — i i 11111 1 — i — I I I I II
0.001 0.01 0.1

21.0
Lower 232"° St. Clay
Block Sample: 2.0m
20.0 •
•
19.0
D
D_
S18.0
to a
•o 17.0
o
g16.0
J C
in
15.0 •
TEST oV OCR
14.0
DUDDD B19C2 109 kPa 1.1
13.0 -j 1—i—i i i 1111 1—i—i i i 1111 1—i—i i 1111

0.001 0.01 0.1 1
40.0
Lower 232 St. Clay
nd
Block Sample: 2.0m
35.0 d
c?30.0
CO
•o 25.0
o
2
o
0)
JC 20.0 A
co
15.0 - 3
TEST oV OCR
• •QDDD B19C3 268 kPa 1.0
10.0 1 1—I I I I 111 1 1—I I I I 111 i — i — i t i i n
0.001 0.01 0.1

oxz
Shear Modulus (MPa)

APPENDIX B:
Damping vs. Strain Curves
211
16.0
nd
Tube Sample: 8.2m
12.0 H
steady State
•i 8.0 D D Q D D
***** Amplitude Decay

E
o
Q
TEST <T'
C OCR
T1C2 71 kPa 1.0 a
4.0 H
•
0.0 —i—i Minn 1—i i 111111 1—i M i n n T 1 1 1 1 III

0.0001 0.001 0.01 0.1
Shear Strain (»)
16.0
nd
Tube Sample: 8.2m
12.0
or. aanan Steady State

* * * * * Amplitude Decay
.§ 8.0
Cu
E
o
o TEST OCR
T2C3 250 kPa 1.0
4.0 H
•
0.0 —i—i I I 11111 1—i i i 11 ii| 1—i i i 11111 i — i i 111 H I
0.0001 0.001 0.01 0.1

Shear Strain (»)
16.0
nd
Tube Sample: 13.2m
12.0 H
n a n a steady State
a
.£
o. 8.0 * * * * * Amplitude Decay
£ a
o
Q
TEST oV OCR
TS2C2 76 kPa 1.0
4.0 H
S 5
0.0 —i—i i i m u 1—i i 111 ii| 1—i i 111111 T 1 I I I III
0.0001 0.001 0.01 0.1

Shear Strain (*)
16.0
Lower 2 3 2 St. Clay nd
Tube Sample: 13.2m
12.0
a n n u o steady State
£ 8.0 * * * * * Amplitude Decay
Qu
E
o
Q TEST cr 'c OCR
a
TS2C3 137 kPa 1.0
4.0 H
0.0 — i — i i 111111 1—i i 11 nn 1—i i 111111 T 1 I I I III
0.0001 0.001 0.01 v 0.1

Shear Strain (*)
912
Damping (sc)
16.0
Lower 232nd St. Clay 11.75m
Test TS1C2 oV= 69 kPa
12.0 H
cn
c ***** Steady State
to 8.0 H • • D • a Amplitude Decay
H
E
D
Q
4.0 H
D
*
0.0 T — r i 111111— 11111

~i—r~r i i—r • 111111 T II I II III
0.0001 0.001 0.01 0.1

Shear Strain (%)
16,0
Lower 232nd St. Clay 11.75m
Test TS1C3 cr '=198 kPa
c
12.0
cn
bteady State
***** Steady
£ 8.0 AAAAA Ampliuude Decay
Q.
E
o
Q
4.0 H A
*
0.0 T — r ~ r i 11111— ! I 1111 i IIII

T—r T—r
0.0001 0.001 T—r~r 0.01 0.1
Shear Strain
16.0
Lower 232 nd
St Clay 11.75m
Test TS1C4 oV=397 kPa
12.0 H
* * * * * Steady State
£ 8.0 A A & A A Amplitude Decay
*
4.0 A
0.0 T 1 llllll| "I—I I I 1111 T 1 IIIIIII T 1 IIIIIII

0.0001 0.001 0.01 0.1
Shear Strain (%)
16.0
nd
Tube Sample: 11.75m
12.0 H
D D D D P Steady State
.£ 8.0 * * * * * Amplitude Decay
•
Q.
E
o
Q
TEST oV OCR
TS1C5 103 kPa 3.9
4.0 H
•
fi
0.0
— i — i i 111111 1—nn tii| 1—i i 111 ii| 1—i i 111111
0.0001 0.001 0.01 0.1 1

Shear Strain (%)
16.0
no
Block Sample: 2.0m
12.0 H
steady State
cE 8.0 H
D D n o a
E
o
o a
TEST OCR
B2C1 100 kPa 1.2 * a
4.0 H
D
a
* ft •o "
a
0.0 1 1 I I I I ll| 1 1 I I Mll| 1 1 I I Mll| 1 1 I I I III
0.0001 0.001 Shear 0.01

Strain .(*). 0.1
16.0
nd
Block Sample: 2.0m
12.0 H
cn D D D o a steady State
E
o
a
TEST OCR
B4C1 57 kPa 1.8
4.0 H
D ° s „ *
0.0 — i — i i ii n n 1 — ii in m 1 — i IIIIIII i—IIIIIII

0.0001 0.001 0.01 0.1
Shear Strain (*)
zzz
Damping (%)
J> CO f\5 CO
• • • •
o o o o
t I t I t I I I I t I I I I I I I I 1 1 I I t t'l I I 1 I I I
*zz
Damping (*)
4k oo 10 a>
• • • •
0 o o o
1 i t i i t t i i i i i i i i i i i i i t i i i i i t i i i
szz
Damping (*)
16.0
Lower 232 St. Clay
na
Block Sample: 2.0m
12.0 H
onnao Steady State

£o. 8.0 ***** Amplitude Decay
E
o
Q
TEST oV OCR
B6C1 116 kPa 1.0
4.0
S «
0.0 1 1 I »Mll| 1 1 1 I I Ml) 1 1 I I IUI| T II I I 111

0.0001 0.001 0.01 0.1
Shear Strain (*)
16.0
Lower 232™ St. Clay
Block Sample: 2.0m
12.0 H
v5>
cn • noan Steady State
£ 8.0
E
o
Q TEST oV OCR
B7C1 144 kPa 1.0
4.0 H
•
0.0 —i—i i i ini| 1—i i 11 iii| 1—i i 11 nu 1—i i 11 in

1
0.0001 0.001 0.01 0.1
Shear Strain (*)
16.0
Lower 232 St. Clay
nd
Block Sample: 2.0m
12.0 H
ci 8.0
anoDo Steady State
E
o
a
TEST OCR
B8C2 144 kPa 1.0
4.0
a
•
a
0.0 "1 1—i i 11 ui| 1—i i 111 ii| 1—i i II mi 1—i i 11 M I

0.0001 0.001 0.01 0.1 1
Shear Strain (*)
16.0
Lower 232 nd
St. Clay Block Sample: 2.0 m
Test B9C2 o-'=74 kPa 0CR=1.6
12.0 H
to 1 8.0 H
AAAAA Steady State
& A A & A Amplitude Decay
to
vo E
o
o
4.0 H
A A A
A
0.0 -j 1—i i 111 m 1—II 111 m 1—i IIIIIII T 1 I 1 11 III
0.0001 0.001 0.01 , x 0.1

Shear Strain ( » )
16.0
Lower 232 nd
St. Clay Block Sample: 2.0 m
Test B9C3 a'=146 kPa OCR=1.0
12.0 H
AAAAA Steady State

£ 8.0 ***** Amplitude Decay
Q.
E
o
o
4.0 H
A
*
0.0 1 1 I I I 1111 1 1 I I 11 l l | 1 1 I 1I H l | T—I I M III

0.0001 0.001 0.01 0.1
Shear Strain ( » )
16.0
nd
Block Sample: 2.0m
12.0 H
cn Steady State
c D D O D D
8.0 H * * * * * Amplitude Decay *

E
o D
Q
TEST OCR *
D
B9C7 32 kPa 9.3

4.0 H *
a
*•
•
*
* •
•
0.0 —i—i i i 11111 1—i i i i m i 1—i I I I I I I I T—I I I I III
0.0001 0.001 0.01 0.1

Shear Strain (ss)
16.0
Lower 232 St. Clay
n0
Block Sample: 2.0m
12.0 H
Steady State
£ 8.0
A A A A A
Q.
E
o
o
TEST OCR
B13C1 200 kPa 1.0
4.0 H
*
A
•
A
0.0 —i—i i i 11111 1—i i i 1111| T — r r •i•I T 1 I I I III

0.0001 0.001 0.01 , v
0.1
Shear Strain (*)
16.0
Lower 232 St. Clay
na
Block Sample: 2.0m
12.0 H
A A A A A Steady State
Q.
E
o
Q TEST OCR
A B13C2 50 kPa 4.0
4.0 H
0.0 1 1 I I I 1111 1 1 I 1 1 1111 1 1 I I I 1111 T — i i i nn

0.0001 0.001 0.01 0.1
Shear Strain (*)
16.0
Lower 232 St. Claynd
Test B14C2 cr '=70 kPa c OCR=1.7

3 hours conFi*n«rnent
12.0
AAAAA Steady State

8.0 H Amplitude Decay
4.0
1
A
X *
0.0 —i—r i i 11 IT*] 1—r~i i 11 rq I—r~T~r T n r T 1 I l"TTTT|
0.0001 0.001 0.01 0.1

Shear Strain (%)
16.0
Lower 232 St. Clay
nd
Test B14C2 a '=70 c kPa OCR=1.7

13 hours cohFYntmerxt
12.0
8.0 H * * * * * Amplitude Decay
4.0
A
*
0.0 —i—i i IIIII) 1—i i 111 m 1—i i 111111 T—I 1 I II m

0.0001 0.001 0.01 0.1
Shear Strain (%)
16.0

Test B15C2 oV=141 kPa OCR=1
12.0 H
AAAAA Steady State

•i 8.0 * * * * * Amplitude Decay
*
A
4.0
* ft
A A
0.0 T—~r TT I I 111 T — I I I I irrj 1 -r-r-rrrrrf-

0.0001 0.001 0.01 0.1
Shear Strain (*)
16.0
Lower 232 nd
St. Clay
Test B15C3 oV=31 kPa 0CR=4.5
12.0 H
8.0 * * * * * Amplitude Decay
*
A
4.0
0.0 II i—r r i irnj i—r T I T rrrr ~i—r i ri m i

0.0001 0.001 0.01 0.1
Shear Strain
16.0

Test B16C2. or '=130 kPa
c 0CR=1
12.0 H
AWv\ Steady State

8.0 ***** Amplitude Decay
*
A
4.0
*
A A
A
0.0 -J 1—i i i • • • 11 1—r i n IIlj 1—i i IIMI] ~I—i M i n i
0.0001 0.001 0.01 0.1

Shear Strain (SB)
15.0 Lower 232 St. Clay
nd
Test B16C3 a'=293 kPa OCR=1
10.0
cn
C A A A A A Steady State
E
O
Q
5.0
A A
*
*
0.0 1 1 I I I II11 1 1 IIIMI| 1 1 IIIIIII i—i IIIIIII

0.0001 0.001 0.01 0.1
Shear Strain (%)
16.0
Test B17C1 o- '=484 kPa
c 0CR=1
12.0 H
cn AAAAA Steady State

•g 8.0
E
D
O
4.0 4
*
A
A
A A
*
0.0 1 1 I I I I 11| 1 1 I I 1 I 11| 1 1—I I I IIII T 1 I llllll
0.0001 0.001 0.01 0.1

Shear Strain (ss)
16.0
Lower 232 St. Clay
nd
Test B17C2 a'-=103 kPa 0CR=4.7
12.0 H
Steady State
A A A A A
CL ***** Amplitude Decay

E
o
Q
*
4.0 H
A
*
A
*
A A
0.0 i—r i—r~T 11111 ] i—i i i II II

i i 1 1 1 1 1 — ~i—i—r
0.0001 0.001 0.1
0.01
Shear Strain
16.0

Test B17C3 oV=53 kPa 0CR=9.2
12.0
AAAAA Steady State

8.0 * * * * * Amplitude Decay
*
A
4.0 H *
A
A
*
0.0 1 1 l~T T ITT| 1 1 ! ITTTTf 1 1 I I I I 111 T — r n r m i
0.0001 0.001 0.01 0.1
Shear Strain (%)
16.0
Lower 232 St. Clay
nd
Test B18C2 <T-=146 kPa

c OCR=1
12.0 H
AAAAA Steady State

£ 8.0 H * * * * * Amplitude Decay
4.0 H
0.0 — i — i — i i I 11
11 1 — i IIIIIII 1—i IIIIIII i—I IMMII
0.0001 0.001 0.01 0.1

Shear Strain (%)
16.0
Lower 232 nd
St. Cloy
Test B19C2 a'=109 kPa 0CR=1.1
12.0
o n j AAAAA Steady State

o.U - | ***** Amplitude Decay
*
4.0 A
4
* A
A
0.0 -j 1—i IIIIIII t—i IIIIIII 1—i IIIIIII T—I I I I 1 ll|
0.0001 0.001 0.01 0.1

Shear Strain (%)
16.0
Test B19C3 o- '=268 kPa
c OCR=1
12.0 H
AAAAA Steady State

•i 8.0 ***** Amplitude Decay
4.0 H *
A
0.0 1 1 I I I I I 11 1 ! I I I I I 11 1 1 I rTTTT| T—I I I I I 111

0.0001 0.001 0.01 0.1
Shear Strain (%)
16.0
Lower 232 no
St. Clay
Test B19C5 oV= 50 kPa 0CR=5.4
12.0 H
AAAAA Steady State

.£ 8.0 H ***** Amplitude Decay
4.0 H
0.0 ~i—i i i I 11 i f i — r m T i r f i i i 111111

T 1—I I I I III
0.0001 0.001 0.01 0.1
Shear Strain (%)

1

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

1

Uploaded by

Copyright:

Available Formats

DYNAMIC PROPERTIES

FROM RESONANT COLUMN TESTS

B.A.Sc, University of B r i t i s h Columbia, 1988

A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF

We accept t h i s thesis as conforming

to the required standard

THE UNIVERSITY OF BRITISH COLUMBIA

© Dan Zavoral, 1990

Department of C int- £?tV&tAft£

The University of British Columbia

Date £>ecen^er AS^ I^^O

The dynamic properties of clay deposits under seismic or

This thesis presents a study of the influence of various

The shear modulus was found to degrade f o r shear strains

Both the shear wave v e l o c i t y and damping obtained i n t h i s

CHAPTER 1. INTRODUCTION " 1

CHAPTER 6. SITE DESCRIPTION 75

APPENDIX A: Shear Modulus vs. Strain Curves 176

APPENDIX B : Damping vs. Strain Curves 211

2.3 Comparison of Three Experimental Formulas on

2.6 E f f e c t of Varying Frequency Content on the Shear

3.1 Resonant Column/Torsional Shear Apparatus 34

3.2 Plan View of Resonant Column Top Drive Plate ... 36

3.3 Resonant Column/Torsional Shear Wiring

5.3 Increased A i r Diffusion E f f e c t at High Confining

7.1 Shear Modulus vs. Confining Pressure Relationship

7.11 E f f e c t of OCR on Shear Modulus f o r Block

7.14 Time-Dependent Increase i n Shear Modulus f o r

7.24 V a r i a t i o n of Low Amplitude Damping Ratio with

7.30 Comparison of Accelerometer and RVDT Decay

7.32 Damping vs. Shear Strain f o r Tube Samples at

7.33 E f f e c t of Duration of Confinement on High

8.5 Damping Attenuation Curves Determined at

7.2 Values of Nd f o r Samples at Various 02c' a n d

I wish t o express gratitude to my supervisor, Professor

I would l i k e t o thank John S u l l y f o r h i s valuable insight

The comments and suggestions of Professor P. Byrne i n

I t i s generally recognized that l o c a l s o i l c h a r a c t e r i s t i c s

a f f e c t the ground response when seismic waves propogate through

a soil profile. Advances i n earthquake ground response

analyses have greatly surpassed the knowledge of the basic

dynamic soil properties which are required i n order to

c o r r e c t l y predict the actual ground response.

In the 1985 Mexico C i t y earthquake, areas underlain by

This research addresses the complex e f f e c t of various

factors upon the shear modulus and damping of an undisturbed

clay, including: the effect of shear strain amplitude;

To perform these studies a resonant column device was used

A brief literature review emphasizing the e f f e c t s of

The resonant column/torsional shear apparatus i s described

Chapter 5 describes the research s i t e , the geotechnical

aspects of the s o i l deposit, and the sampling methods used.

In Chapter 6 testing procedure considerations are discussed

while the t e s t r e s u l t s themselves are presented i n Chapters 7

Several researchers have published comprehensive reports

on dynamic s o i l properties of cohesive s o i l s (Sun at a l . , 1988;

Kokusho et a l . , 1982; Hardin and Drnevich, 1972a and 1972b).

This chapter reviews t h e i r main findings and outlines the key

factors a f f e c t i n g dynamic shear modulus and damping of cohesive

2.1 SHEAR MODULUS

Hardin and Black (1968) concluded that shear modulus of

soils i s a function of:

i / e f f e c t i v e octahedral normal stress