STATISTICAL AND NUMERICAL ANALYSES OF

PRESSUREMETER TESTS IN GLACIAL TILLS

by

Kanagaratnam Balachandran,

B.Eng. (Hons), University of Moratuwa, 1997

M.Eng, University of Moratuwa, 2009

A thesis

presented to Ryerson University

in partial fulfillment of the

requirements for the degree of

Master of Applied Science

in the program of

Civil Engineering

Toronto, Ontario, Canada, 2016

© (Kanagaratnam Balachandran) 2016

AUTHOR’S DECLARATION

I hereby declare that I am the sole author of this thesis. This is a true copy of the thesis, including
any required final revisions, as accepted by my examiners.

I authorize the Ryerson University to lend this thesis to other institutions or individuals for the
purpose of scholarly research.

I further authorize the Ryerson University to reproduce this thesis by photocopying or by other
means, in total or in part, at the request of other institutions or individuals for the purpose of
scholarly research.

ii
STATISTICAL AND NUMERICAL ANALYSES OF
PRESSUREMETER TESTS IN GLACIAL TILLS

Kanagaratnam Balachandran
Master of Applied Science, 2016
Department of Civil Engineering
Ryerson University, Toronto, Canada

ABSTRACT

This study is performed on pressuremeter tests (PMT) in glacial tills based on comprehensive
geotechnical investigation programs for a light rail transit project in the City of Toronto. The
main objectives are to establish a correlation between SPT-N values and PMT parameters, and
the Menard “α” factors for glacial tills. Currently, there are no such relationships available. So
first, the pairs of PMT data and SPT-N values are collected at the same depth and test area. With
these paired data, two linear correlation equations are established. Then, the numerical
simulation is performed for PMTs in glacial tills by using finite element software, Plaxis 2D. The
Mohr-Coulomb material model is used to model the different types of soil. The Menard “α”
factor is suggested based on the best match between numerical prediction and field PMT. Ranges
of SPT-N, EPMT and PL are also suggested for glacial tills.
.

iii
ACKNOWLEDGEMENTS

This thesis arose in part out of years of research. In that time, I worked with several people
whose contributions to the research and the making of the thesis deserve special mention. It is a
pleasure to convey my gratitude to them in this humble acknowledgement.

In the first place I would like to record my sincere thanks and gratitude to Dr.Jinyuan Liu for his
supervision, advice and guidance from the very early stage of this research as well as giving me
extraordinary experiences throughout the work. Above all and the most needed, he provided me
encouragement and support in various ways. His truly engineering intuition has made him as a
constant oasis of ideas and passion in geotechnical engineering, which exceptionally inspired and
enriched my growth as a student, a researcher and geotechnical engineer want to be.

In the second place I would like to record my sincere thanks and gratitude to Dr. Laifa Cao for
his supervision. He provided a continuous source of ideas relating to the technical approach,
numerical modelling, and the practical applications of the research.

Many thanks go to the Dean, Faculty of Engineering, Head of the Department of Civil
Engineering and Dr A.El-Rabbany Professor and Graduate Program Director, Civil Engineering
Program, for making all the necessary arrangements to carry out the project.

I would like to thanks Rachel Harpley, Graduate Program Administrator, Department of Civil
Engineering, for her continuous communication and support during this study.

I must also thank the Department of Civil Engineering and Natural Sciences and Engineering
Research Council of Canada for financial assistance and Metrolinx and SPL Consultant Limited
for granting permission to study the ECLRT project data.

iv
It is a pleasure to pay tribute to the academic staff during the course studying stage for helping
me in various ways for the successful completion of this thesis.

I would like to give sincere thanks to my wife and my children for their endless love,
encouragements and patience throughout my educational career.

v
TABLE OF CONTENTS

AUTHOR’S DECLARATION .................................................................................................... ii

ABSTRACT .................................................................................................................................. iii
ACKNOWLEDGEMENTS ........................................................................................................ iv
TABLE OF CONTENTS ………………………………………………………………............vi
LIST OF TABLES………………………………………………………………………………ix
LIST OF FIGURES………………………………………………………………………….......x
CHAPTER 1: INTRODUCTION ................................................................................................ 1
1.1 RESEARCH BACKGROUND ........................................................................................................................... 1
1.2 ENGINEERING BACKGROUND ..................................................................................................................... 1
1.3 NEED FOR RESEARCH ....................................................................................................................................4
1.4 OBJECTIVES OF THE STUDY ........................................................................................................................ 5
1.5 RESEARCH METHODOLOGY ........................................................................................................................ 5
1.6 THESIS OUTLINE .............................................................................................................................................6
CHAPTER 2: LITERATURE REVIEW .................................................................................. 8
2.1 INTRODUCTION ...............................................................................................................................................8
2.2 PREVIOUS RESEARCH ON STATISTICAL CORRELATION BETWEEN SPT AND PMT ....................... 8
2.2.1 STANDARD PENETRATION TEST (SPT) .............................................................................................. 8
2.2.1.1 Equipment and Test Procedure ............................................................................................. 9
2.2.1.2 SPT-N correction ................................................................................................................ 10
2.2.1.3 Interpretation of SPT ........................................................................................................... 13
2.2.2 PRESSUREMETER TEST (PMT) ............................................................................................................ 13
2.2.2.1 Equipment and Test Procedure ........................................................................................... 14
2.2.2.2 Calibration........................................................................................................................... 16
2.2.2.3 Test results interpretation .................................................................................................... 17
2.2.2.4 Application of these two parameters ................................................................................... 20
2.2.2.5 Pressuremeter test correlation in soils ................................................................................. 21
2.2.3 CORRELATION BETWEEN SPT-N WITH PMT PARAMETERS ........................................................ 26
2.3 PREVIOUS RESEARCH ON NUMERICAL SIMULATIONS OF THE PMT .............................................. 28
2.3.1 GEOMETRY AND BOUNDARY CONDITION OF THE MODEL…………………………………….28

2.3.2 PROB LENGTH (L) TO DIAMETER (D) RATIO (L/D RATIO)……………………………………... 31

2.3.3 TYPE OF MODEL……………………………………………………………………………………… 31

vi
 2.3.4 DEFORMED MESH AND ANALYSIS ………………………………………………………………....33

2.3.5 ANALYSIS RESULTS AND GRAPHES……………………………………………………………..….36

2.3.6 COMPARISON OF PM AND YOUNG MODULUS …………………………………………………....39

2.3.7 CONVENTIONAL LIMIT PRESSURE………………………………………………………………….39

2.3.8 ANALYTICAL METHOD ……………………………………………………………………………….41

2.4 SUMMARY....................................................................................................................................................... 42
CHAPTER 3: STATISTICAL CORRELATION BETWEEN SPT-N VALUE WITH PMT
PARAMETERS FOR GLACIAL TILLs ................................................................................. 44
3.1 INTRODUCTION ............................................................................................................................................. 44
3.2 DATA SELECTION, SPT-N VALUE CORRECTION AND FILTERING .................................................... 44
3.3 GENERAL RANGE OF SPT-N, EPMT AND PL FOR GLACIAL TILLS ........................................................ 46
3.3.1 RANGE OF SPT-N VALUE ................................................................................................................... 46
3.3.2 RANGE OF EPMT VALUE ...................................................................................................................... 49
3.3.3 RANGE OF PL VALUE ............................................................................................................................ 53
3.4 CORRELATION BETWEEN SPT-N AND EPMT VALUE ............................................................................ 56
3.5 CORRELATION BETWEEN SPT-N AND PL VALUE ............................................................................... 59
3.6 COMPARISON OF CORRELATION BETWEEN SPT-N VALUE AND BOTH EPMT AND PL ................ 65
3.6.1 COMPARISON OF CORRELATION BETWEEN SPT-N VALUES AND EPMT FOR SAND .............. 65
3.6.2 COMPARISON OF CORRELATION BETWEEN SPT-N VALUES AND PL FOR SAND ................... 67
3.6.3 COMPARISON OF CORRELATION BETWEEN SPT-N VALUES WITH BOTH ..................................
EPMT AND PL FOR GLACIAL TILL ......................................................................................................... 68
3.6.4 COMPARISON OF CORRELATION BETWEEN SPT-N VALUES AND EPMT /PL ............................ 71
RATIO FOR SAND ................................................................................................................................... 71
3.6.5 COMPARISON OF CORRELATION BETWEEN SPT-N VALUES AND EPMT /PL ................................
RATIO FOR GLACIAL TILL ................................................................................................................... 73
3.7 SUMMARY ...................................................................................................................................................... 75
CHAPTER4: FINITE ELEMENT SIMULATION OF PMT ................................................ 77
4.1 INTRODUCTION ............................................................................................................................................. 77
4.2 FINITE ELEMENT METHOD ......................................................................................................................... 77
4.2.1 CONSTITUTIVE BEHAVIOUR .............................................................................................................. 77
4.2.2 STEPS INVOLVED IN THE FEM ........................................................................................................... 81
4.3 MOHR-COULOMB MATERIAL MODEL ..................................................................................................... 83
4.4 2D PMT MODELLING AND VERIFICATION .............................................................................................. 85
4.4.1 2D PMT MODELLING ............................................................................................................................. 85
4.4.2 VERIFICATION OF THE MODEL .......................................................................................................... 86
4.5 SENSITIVITY STUDY .................................................................................................................................... 92

vii
 4.5.1 MESH COARSENESS .............................................................................................................................. 92
4.5.2 HORIZONTAL BOUNDARY CONDITION ........................................................................................... 93
4.5.3 VERTICAL BOUNDARY CONDITION ................................................................................................. 94
4.6 CASE STUDY AT MOUNT DENNIS STATION ........................................................................................... 96
4.6.1 FINITE ELEMENT ANALYSES.............................................................................................................. 97
4.6.2 2D FINITE ELEMENT RESULTS AND ANALYSIS ........................................................................... 102
4.6.3 COMPARISON OF PRESSURE VS RADIAL STRAIN CURVES FROM PMT AND PLAXIS ......... 110
4.6.4 VALIDATE THE RESULTS .................................................................................................................. 116
4.7 SUMMARY .................................................................................................................................................... 117
CHAPTER 5: CONCLUSIONS AND RECOMMENDATIONS FOR FUTURE
RESEARCH .............................................................................................................................. 119
5.1 MAIN CONCLUSIONS ................................................................................................................................. 119
5.2 RECOMMENDATIONS FOR FUTURE RESEARCH ................................................................................. 120
REFERENCES .......................................................................................................................... 122
APPENDICES ........................................................................................................................... 127

viii
LIST OF TABLES

Table 1.1 Layout of thesis ............................................................................................................................. 6

Table 2.1 Approximate correction factor for SPT-N values (after Skempton, 1986) ................................. 12
Table 2.2 Compactness condition of sands from SPT ................................................................................ 13
Table 2.3 Approximate common values of the PMT parameters for clay and sand .................................. 21
Table 2.4 Typical Menard PMT values ...................................................................................................... 22
Table 2.5 Typical Menard α factor (Briaud, 1992) .................................................................................... 24
Table 2.6 Typical Young’s modulus for different types of soils (Bowles, 1996) ....................................... 25
Table 2.7 Review of the Literature survey regards FEM simulation of PMT ............................................ 42
Table3.1 Summary of SPT-N value for different types of soils ................................................................. 47
Table3.2 Summary of EPMT value for different types of soils ..................................................................... 50
Table 3.3 Summary of PL value for different types of soils ....................................................................... 53
Table3.4 Summary of correlation between SPT-N value with both EPMT and PL value for different types of
soils ............................................................................................................................................................ 64
Table 3.5 Summary of range of SPT-N, EPMT and PL for corrected and filtered data ................................. 76
Table 3.6 Summary of correlation equation for EPMT and PL for 𝑁160 .................................................... 76
Table 3.7 Summary of EPMT /PL ratio for filtered data ................................................................................ 76
Table 4.1 Parameters used in the M-C model for dense Hostun sand ........................................................ 86
Table 4.2 Comparison of horizontal displacement related to mesh coarseness .......................................... 93
Table 4.3 The values of volumetric strain related to horizontal boundary condition ................................. 94
Table 4.4 The value of volumetric strain related to vertical boundary condition ....................................... 95
Table 4.5 Summary of soil parameters used in the FEM analysis ............................................................. 98
Table 4.6 Linear correlation equations for glacial till at MD Station ....................................................... 112
Table 4.7 Linear with intercept correlation equations for glacial till at MD Station ................................ 115
Table 4.8 Summary of predicted E and calculated and analytical EPMT at MD Station ............................ 116
Table 4.9 Summary of the EPMT, E and Menard “α” factor for MD Station.............................................. 118

ix
LIST OF FIGURES

Figure 1.1 Typical glacial till (http://www.free-stockillustration.com ......................................................... 2

Figure 1.2 Crosstown route map (http://www.thecrosstown.ca/the-project)............................................... .3
Figure 2.1 Standard split barrel sampler used in SPT (ASTM, 2014) .......................................................... 9
Figure 2.2 The Menard pressuremeter equipment (ASTM, 2000).............................................................. 14
Figure 2.3 Sequences of the PMT (ASTM, 2000) ...................................................................................... 16
Figure 2.4 Calibration curves obtained before the PMT (Bowles, 1997) ................................................... 17
Figure 2.5 Typical pressuremeter test curve (ASTM, 2000) ...................................................................... 18
Figure 2.6 Determination of limit pressure from inverse of volume vs pressure curve .............................. 20
Figure 2.7 Correlation between SPT N and net limit pressure (PL) (Briaud, 1992).................................... 26
Figure 2.8 Correlations between SPT N and Ep (Briaud, 1992) ................................................................. 27
Figure 2.9 Correlation between SPT N and EPMT for clays (Ohya et al., 1982) .......................................... 27
Figure 2.10 Correlation between SPT N and EPMT for sands (Ohya et al., 1982) ....................................... 28
Figure 2.11 The 2D axisymmetric model and associated mesh diagram (Malecot et al.(2012)) ................ 30
Figure 2.12 The 2D axisymmetric model and associated mesh diagram (Levasseur et al.(2012))............ .30
Figure 2.13 Geometry model for PMT (Plaxis 2D (2012)) ........................................................................ 31
Figure 2.14 FE discretization of the calibration chamber (Schanz et al. (2000)) ....................................... 33
Figure 2.15 Points from which the displacements are read (Rita (2008))................................................... 34
Figure 2.16 The 2D axisymmetric model and associated mesh diagram (Malecot et al.(2009)) ................ 34
Figure 2.17 The 2D axisymmetric model and associated mesh diagram (Levasseur et al.(2009))............. 35
Figure 2.18 Deformed geometry of the PMT (Plaxis (2012))..................................................................... 35
Figure 2.19 Geometry of axisymmetric model (Fawaz (2014)) ................................................................. 36
Figure 2.20 Comparison between experimental and numerical results of the PMT (Schanz et al.(20000) 37
Figure 2.21 Model and test curves at 500 kPa of effective vertical stress (Rita (2008)) ............................ 37
Figure 2.22 Comparison of numerical results and PMT data (Plaxis (2012)) ............................................ 38
Figure 2.23 Experimental and numerical pressure-volume curves at different depth (Fawaz (2014)) ....... 38
Figure 3.1Range of SPT-N value for cohesive glacial tills ........................................................................ 48
Figure 3.2 Range of SPT-N value for cohesionless glacial tills ................................................................ 49
Figure 3.3 Range of EPMT value for cohesive glacial tills .......................................................................... 51
Figure 3.4 Range of EPMT value for cohesionless glacial tills .................................................................... 52

x
Figure3.5. Range of PL value for cohesive glacial tills .............................................................................. 54
Figure 3.6 Range of PL value for cohesionless glacial tills ........................................................................ 55
Figure 3.7 Correlation between SPT-N vs EPMT for cohesive glacial tills ................................................. 57
Figure 3.8 Correlation between SPT-N vs EPMT for cohesionless glacial tills ........................................... 59
Figure 3.9 Correlation between SPT-N vs PL for cohesive glacial tills ..................................................... 61
Figure 3.10 Correlation between SPT-N vs PL for Cohesionless glacial tills ............................................ 63
Figure 3.11 Comparison of correlation between SPT-(N1)60 and EPMT for sand ........................................ 67
Figure 3.12 Comparison of correlation between SPT-(N1)60 and PL for sand ............................................. 69
Figure 3.13(a) Comparison of correlation between SPT-(N1)60 vs EPMT for linear with intercept
relationship for glacial till ........................................................................................................................... 70
Figure 3.13 (b) Comparison of correlation between SPT-(N1)60 vs PL for linear with intercept relationship
for glacial till ............................................................................................................................................... 71
Figure 3.14 (a) Correlation between EPMT vs PL for sand ............................................................................ 72
Figure 3.14 (b) Correlation between EPMT/PL vs SPT-(N1)60 for sand ......................................................... 73
Figure 3.15 (a) Correlation between EPMT vs PL for glacial till ................................................................... 74
Figure 3.15 (b) Correlation between SPT-(N1)60 vs EPMT/PL for glacial till ................................................ 75
Figure 4.1 Stresses on a typical element .................................................................................................... 78
Figure 4.2 Example of axisymmetric co-ordinate axis .............................................................................. 81
Figure 4.3 Stresses in the axisymmetric element ........................................................................................ 81
Figure 4.4 Mohr’s circles of effective stress (Ports and Zdravkovic (2001)) ............................................. 84
Figure 4.5 Mohr - Coulomb yield surfaces in principal stress space (Ports and Zdravkovic (2001))......... 85
Figure 4.6 2D axisymmetric model and associated mesh (Levasseur et al. (2009)) ................................... 87
Figure 4.7 Geometry of the PMT model .................................................................................................... 87
Figure 4.8 Deformed mesh diagrams .......................................................................................................... 88
Figure 4.9 Zoomed view of deformed mesh diagram ................................................................................. 89
Figure 4.10 Horizontal displacement diagram ……………………………………………………………90
Figure 4.11 Zoomed view of horizontal displacement diagram…………………………………………..91

Figure 4.12 Pressure (p) vs volumetric strain (𝛥𝑣/𝑉) curves for verification of the model ...................... 92
Figure 4.13 Pressure (p) vs volumetric strain (𝛥𝑣/𝑉) curves for horizontal boundary conditions ............ 94
Figure 4.14 Pressure (p) vs volumetric strain (𝛥𝑣𝑉) curves for vertical boundary conditions .................. 94
Figure 4.15 Soil profile at MD Station according to borehole MD101-PMT and test @6.0m ................ 100
Figure 4.16 Typical FE mesh for numerical simulation at MD Station ................................................... 101

xi
Figure 4.17 Water table diagram for soil profile at MD Station @ test depth 6m after generate the in-situ
initial stress condition. .............................................................................................................................. 101
Figure 4.18 Typical deformed meshes @ 6.0 m depth at MD Station ...................................................... 103
Figure 4.19 Zoomed views of the deformed meshes diagram@ 6.0 m depth at MD Station ................... 104
Figure 4.20 Typical horizontal displacements arrow diagram @ 6.0m depth at MD Station................... 105
Figure 4.21 Zoomed views of the horizontal displacements arrow diagram @ 6.0m depth at MD Station
.................................................................................................................................................................. 106
Figure 4.22 Typical horizontal displacement shaded diagram @ 6 m depth at MD Station……………107
Figure 4.23 Zoomed view of horizontal displacement shaded diagram @ 6 m depth at MD Station…..108
Figure 4.24 Typical horizontal displacement cross section @6.0m depth at MD Station ........................ 108
Figure 4.25 Zoomed view of horizontal displacement cross section (along the probe) @6.0m depth at MD
Station ...................................................................................................................................................... 109
Figure 4.26 Pressure vs radial strain graph at 6m depth for MD Station .................................................. 111
Figure 4.27 Linear relationship for EPMT vs E for cohesionless glacial till at MD Station ....................... 114
Figure 4.28 Linear relationship for EPMT vs E for cohesive glacial till at MD Station ............................. 115

xii
CHAPTER 1: INTRODUCTION

1.1 RESEARCH BACKGROUND

Statistical correlations between in-situ soil testing results and soil parameters are increasingly
used during various stages of geotechnical engineering work. Statistical correlation equations are
used during the early stage of engineering design work since they are more practical ways to
proceed than extensive in-situ testing programs. In geotechnical design work statistical
correlations are widely used to predict unknown parameters from simple known parameters, and
save the time and cost.

Many geotechnical design parameters of the soil can be derived from Standard Penetration Test
(SPT) and SPT is widely used around the world. On the other hands, Pressuremeter Test (PMT)
is becoming increasing more popular for site investigation and geotechnical design especially in
estimating soil properties for foundation design.

The numerical simulation of the PMT in use of the Finite Element Method (FEM) becomes more
and more popular in geotechnical engineering. This method assumes to model the soil behavior
by constitutive equation. A lot of soil constitutive models deal with a large variety of
geotechnical problems. Nevertheless, these constitutive models have most of the time a large
number of parameters whose values are unknown.

1.2 ENGINEERING BACKGROUND

This study is performed in the Eglinton Crosstown Light Rail Transit (ECLRT) project in
Toronto. The site is situated along Eglinton Avenue from the existing Kennedy Subway Station
in the east to the Mount Dennis Station in the west, in Toronto, Ontario, Canada.

The Toronto area acquired at least three glacial and two interglacial periods from the published
geological data (Karrow (1967) and Sharpe (1980)). The geological history of the Toronto area

1
has included several advances and retreats of glaciers of Illinoian and Wisconsinan age (Karrow
and White (1998)). The glacial tills in this area were generally deposited during the early to late
Wisconsinan periods, represented by the Sunnybrook, Seminary, Meadowcliffe, Newmarket and
Halton tills (Sharpe et al. (1999)). The glacial till deposits in Toronto can be divided into low
plasticity cohesive glacial tills (silty clay to clayey silt glacial till) and cohesionless glacial tills
(sandy silt to silty sand glacial till) (Manzari et al. (2014)). This kind of soil is derived due to the
wearing away and entrainment of material as a result of the moving ice of a glacier. As shown in
Figure 1.1, this type of soil can be described as high variability materials in both horizontal and
vertical axis, and it normally contains complex non-linear stress-strain characteristics (Baker et
al. (1998)).

Figure 1.1 Typical glacial till (Source- Mark Clark, http://www.free-stockillustration.com)

In addition to that, the tills consist of a heterogeneous mixture of gravel, sand, silt and clay size
particles in varying proportions. Cobbles and boulders are common in these deposits (Robert et
al. (2011)). The recorded maximum boulder size founded in Toronto so far has been about 3m in

2
the maximum dimension. Boulder volume ratios (BVR) (total boulder volume per volume of
excavated earth material) of 0.12% and 0.17% for interglacial deposits and glacial tills
respectively have been recommended for TTC Subway projects such as the Sheppard Subway
(Boone and Shirlaw (1996)) and the Toronto – York Spadina Subway extension (Boone and
Westland (2008)).

The proposed ECLRT is approximately 33 km in length and located approximately 7 km north of

Lake Ontario. There are 25 proposed stations along the alignment as shown in Figure 1.2.

Figure 1.2 Crosstown route maps (http://www.thecrosstown.ca/the-project)

A series of laboratory and in-situ tests were conducted in advance at the above stations. The in-
situ tests included SPTs, field vane shear tests, pre-bored TEXAM PMTs and seismic tests. The
laboratory tests included density and moisture content measurements, grain size and hydrometer
analysis, consistency (Atterberg) limit tests, consolidation tests, consolidated undrained and
drained triaxial compression tests.

3
Based on these tests, the soil is classified as a glacial till which further classified as low plasticity
cohesive glacial till and cohesionless glacial till according to the current version of TTC
geotechnical standards (2014). In this area, the low plasticity cohesive glacial till mostly consists
of the following soil types such as (i) silty clay till (ii) clayey silt till. The cohesionless glacial till
mostly consists of following soil types such as (iii) sandy silt till (iv) silty sand till. The glacial
tills are interbedded with silty clay, clayey silt, sandy silt, sand and silt and silty sand. However,
the behavior of glacial tills in southern Ontario is not fully understood.

1.3 NEED FOR RESEARCH

The SPT is a well-established method of investigating soil properties. The differences in testing
practices can be at least partly compensated by changing the measured N to (N1)60. There are
many possible applications to correct the field measured SPT-N. There is no any general
agreements on these applications of corrections of field measured SPT-N. In contrast to heavy
criticisms about the SPT-N correction, there is strong needed to recommend a suitable correction
method for more suitable for local conditions.

Estimation of the PMT parameters such as PM modulus (EPMT) and pressure limit (PL) from SPT
–N value has been studied by a few researchers in the past. Attempted correlations have usually
been weak because of the differences in the methods and uncertainties involved in the tests. Even
though, they are widely used in practice to get an idea about the level of the geotechnical
parameters used in the design. The most of the correlation work done in the past was for sand
and clay. There is almost no correlative work on glacial till especially Toronto glacial till. Hence
research is needed to avoid these short comes and recommended a suitable relationship more
suitable for local condition especially glacial tills in the city of Toronto.

In the case of numerical simulation of PMT, there is bulk of information available but none of
these simulations is performed for real soil profile. The length to diameter ratio of the probe is
influenced in the plain strain condition of the probe. A very little information is available on
back calculating the PM modulus from pressure-volume curve which is obtained from
simulation. However, this modulus differs from the elastic Young’s modulus (E) which is the

4
principal soil parameter. The deduction of the Young’s modulus from PM modulus is still under
𝐸𝑃𝑀𝑇
research. Menard developed ratio of for peat, clay, silt, sand, sand and gravel. Currently
𝐸

there is no such a relationship available for glacial tills in the city of Toronto. Hence there is a
strong need for an in-depth research to develop and recommend suitable relationship for glacial
tills in the city of Toronto.

1.4 OBJECTIVES OF THE STUDY

The main objectives of this thesis are listed below.

(i) To correct the field measured SPT-N and develop a ratio of corrected SPT-N ((N1)60)
(𝑁1 )60
to field measured SPT-N (NF) which is ratio of ( ) for glacial tills.
𝑁𝐹

(ii) To establish the ranges of SPT-N, EPMT and PL for glacial tills.
(iii) To develop the statistical correlation relationships between SPT – N values with PMT
parameters such as PM modulus (EPMT) and pressure limit (PL) for glacial tills.
(iv) To develop the statistical correlation equations between PM modulus (EPMT) with
Young’s modulus (E) for glacial tills by using Finite Element Method (FEM).
𝐸𝑃𝑀𝑇
(v) To develop the Menard’s “α’ factors which is the ratio of for glacial tills.
𝐸

1.5 RESEARCH METHODOLOGY

Statistical analysis is carried out to investigate the relationship between SPT-N values with PMT
parameters such as PM modulus (EPMT) and pressure limit (PL) for glacial tills in the city of
Toronto based on soil investigation for the ECLRT project, Canada. The first step is to collect
the pairs of PMT data and SPT-N value at the same depth in the same test area. After collect
these data, the field measured SPT-N values are corrected and filtered. Then an attempt is made
to develop a correlation between corrected and filtered SPT-N values with PMT parameters. As
emphasized by Phoon and Kulhawy (1999), local correlations that are developed within a
specific geologic setting generally are preferable to generalized global correlations because they
are significantly more accurate.

5
Numerical analysis of PMT is performed using Plaxis 2D software. The Mohr–Coulomb (M–C)
material model is used in this simulation. The appropriate parameters required are grasped from
ECLRT geoengineering factual data reports for different types of soils. Those identified and
extracted values are used in M-C material model which is in the constitutive model. The
pressure-radial strain curve obtained from this simulation is used to compute the PM modulus
(EPMT). This is determined from the quasi-linear part of the pressure vs radial strain curve.
However, this modulus differs from the elastic Young’s modulus (E) which is the principal soil
parameter. The PM modulus has been related to the elastic Young’s modulus for the glacial tills.
𝐸𝑃𝑀𝑇
Then Menard’s rheological factors “α” which is the ratio of are derived for glacial tills.
𝐸

These findings will help geotechnical community in evaluating and interpreting geotechnical
parameters for their clients.

1.6 THESIS OUTLINE

This thesis consists of five chapters including references and annex. The thesis outline is shown
in Table1.1.

Table1.1 Layout of thesis

Chapter Title Content
Chapter 1 Research background Introduction, engineering background, need for
research, objectives and methodology.
Chapter 2 Literature review Literature review on SPT, PMT, correlation between
SPT and PMT, and simulation of PMT.
Chapter 3 Statistical correlation Correction for field measured SPT-N and SPT-N
between SPT-N and correction ratio.
PMT for glacial tills Develop the ranges of SPT-N, EPMT and PL for glacial
tills.
Develop correlation equations between SPT-N values
with PMT parameters such as EPMT and PL for glacial
tills.

6
 Compare these values and correlation equations with
literature.
Chapter 4 Finite Element FEM simulation is performed for PMTs by using
Method simulation of Plaxis 2D. The M-C material model is adopted in this
PMTs in glacial tills simulation.
Develop correlation between PM modulus (EPMT) and
Young’s modulus (E) for glacial tills.
Develop Menard “α” factors for glacial tills.
Chapter 5 Conclusions and Content of conclusions and recommendations for
Recommendations future research.
References

Appendices Content of borehole reports, PMTs results, SPT-N

corrections calculation sheets.

7
CHAPTER 2: LITERATURE REVIEW

2.1 INTRODUCTION

This chapter presents the results of the literature review conducted during the study. Information
available from specific research studies on statistical correlation between SPT with PMT are
very few, as only a few researchers have studied for sand and clay, even rare for Toronto glacial
till. Such information, as it was considered very valuable, is presented in the first part of this
chapter. Further literature survey was conducted on numerical simulations of the PMT. As a bulk
of information on modelling has been emanated through actual practice and through available
theory in the past researchers, the information gathered from those sources are presented in the
second part of this chapter. This literature survey provided the background information for the
formulation and execution of the research study.

2.2 PREVIOUS RESEARCH ON STATISTICAL CORRELATION

BETWEEN SPT AND PMT

2.2.1 STANDARD PENETRATION TEST (SPT)

Standard penetration test (SPT) was first introduced in early 1900’s by driving an open end pipe
into soil during wash boring process and it has become the most extensively used in-situ test in
site investigation practice.

Originally, the test was used to determine the relative density of granular soils. The idea of the
SPT at the beginning was the comparison of blows required to penetrate the tested soil. If the
number of blows for a tested location was larger than another tested location, it was concluded
that the denser soil is the one with the largest blow count. Although SPT had been performed
only for granular soils in the past, it is executed in almost all kinds of soil today including weak
rock.

8
2.2.1.1 Equipment and Test Procedure

The SPT is a well-established method for soil investigation. As many forms of the test are in use
worldwide, standardization is essential to facilitate the comparison of results from different
investigations, even at the same site (Thorburm (1986)). In 1958, the test method was
standardized by ASTM D1586. This means that the test standardized using a 50 mm outside
diameter split spoon sampler with dimensions shown in Figure 2.1, is driven into soil with a 64
kg weight having a free fall of 760 mm auto hammer. The blows required to drive the split–
barrel sampler a distance of 305 mm, after an initial penetration of 152 mm, is referred as the
SPT–N value. Procedure is repeated after the drilling to the depth of the next test.
(Conventionally test is performed at every 1.0 to 1.5 m intervals). In this study, SPT is performed
in accordance with the ASTM D 1586 method.

Figure 2.1 Standard split barrel sampler used in SPT (ASTM (2014))

9
2.2.1.2 SPT-N correction

The main reasons for extensive use of SPT in the site exploration can be related to many factors
such as availability of equipment, simplicity of the operation, applicability in the type of soils
and process of sampling. Due to all of these practical aspects, the results of SPT–N value can be
dramatically affected by drilling operation, the type of equipment, capability of the operator,
presence of size of particles (such as gravels, cobbles and boulders) and ground water conditions.

The SPT is usually stopped on the following situation where 50 or more blows are required for
150mm penetration, 100 blows are obtained to drive the required 300 mm and 10 successive
blows produce no advance. If any of the above situations is encountered during the test, SPT-N
value for the relevant depth is recorded as “refusal” and indicated with the letter “R” in the
borehole logs.

In the above refusal situation the recorded field SPT-N value has to be corrected using Equation
2.1 according to Cao et al. (2015) with measured penetration depth.

305 𝑁
𝑁𝐹 = [2.1]
𝛥𝑠

Where 𝑁𝐹 - Corrected SPT-N value

N - Field recorded SPT-N value
Δs - Measured penetration depth in mm

Because of the variability in equipment and operating conditions, the direct use of SPT results
for geotechnical design is not recommended. As a result, many corrections shall be done on the
field SPT-N values. These corrections can be summarized in an equation formatted as given
below according to the Canadian Foundation Engineering Manual (2006).

(𝑁1 )60 = CECNCRCBCS𝑁𝐹 [2.2]

(𝑁1 )60 = CN 𝑁60 [2.3]

10
𝑁60 = CE 𝑁𝐹 [2.4]

𝑃 0.5
CN = (𝜎1 ) [2.5]
𝐸𝑅𝑅
CE = [2.6]
60

Where CE - Hammer energy correction factor

𝐸𝑅𝑅 – Rod energy ratio
CN - Overburden pressure correction factor
P - Atmospheric pressure
σ' - Effective overburden pressure
CR - Rod length correction factor
CB - Borehole diameter correction factor
CS - Sampler correction factor
NF - Corrected SPT-N value for penetration depth
N60 - SPT-N value corrected to 60% of theoretical free fall hammer energy
(𝑁1 )60 - SPT-N value correctd for both vertical effective stress and input energy

In the literature, most researchers express their concerns regards energy correction which was
elaborated as follows. The energy delivered to the rods during a SPT expressed as a ratio of the
theoretical free fall potential energy, can vary from 30% to 90% (Kovacs and Salomone (1982)
and Robertson et al. (1983)). Schmertmann and Palacios (1979) have shown that the SPT blow
count is inversely proportional to the delivered energy. Kovacs et al. (1984), Seed et al. (1984)
and Robertson et al. (1983) have recommended that the SPT-N value has to be corrected to an
energy level of 60% (CFEM (2006)). The SPT N-values corresponding to 60% efficiency are
termed N60. The practice in the United States/Canada the SPT N-value measured to an average
energy ratio of 60% (ERR=60%) according to ASTM D1586-11 (2014). In this study energy ratio
of 60% (ERR=60%) is adopted.

Other correction factors such as CR, CS and CB are mentioned in Table 2.1 according to CFEM
(2006) (after Skempton (1986)) is adopted in this study.

11
Table2.1 Approximate correction factors for SPT-N values (after Skempton (1986))
Correction factor Item Correction factor value
CR Rod length (below anvil)
>10 m 1.0
6 – 10 m 0.95
4–6m 0.85
3–4m 0.70
CS Standard sampler 1.0
US sampler without liner 1.2
CB Borehole diameter
65 – 115 mm 1.0
150 mm 1.05
200 mm 1.15

Bowles (1997) suggested that there are three possible approaches about correction of SPT-N
value.

(i) Do nothing on the field recorded N value

(ii) Adjust only for overburden pressure
(iii) Apply all of the mention corrections

Since there is no any general agreement on the application of corrections to the field record SPT-
N value, many of the correlation with SPT-N value only suggests energy correction, in some
cases overburden correction was recommended. However, overburden correction for fine grained
soils is still considered as controversial issue and not preferred in practice (Sivrikaya and Togrol,
(2007)).

In contrast to serious criticisms about SPT for being destructive and sensitive to many factors, it
is still most commonly used in-situ test in the geotechnical engineering practices.

12
2.2.1.3 Interpretation of SPT

A term describing the compactness condition of a cohesionless soil is often interpreted from the
results of a SPT. Compactness and penetration values are often related to Table 2.2, which was
proposed by Terzaghi and Peck (1967). Notice that the term “compactness condition” replaces
the earlier term “relative density” used in the past according to CFEM (2006).

Table 2.2 Compactness condition of sands from SPT

Compactness condition SPT-N Index (blows per 0.3 m)
Very loose 0–4
Loose 4 – 10
Compact 10 – 30
Dense 30 – 50
Very dense Over 50

2.2.2 PRESSUREMETER TEST (PMT)

The PMT was invented by German Kogler (1930) in order to measure a soil deformation
modulus. Due to the technology of that time, the unit was not operational. Furthermore, the
inventor has failed to correctly interpret the results and the unit was abandoned.

In 1954 a young French engineer, Louis Menard, took up the idea in the refining the inflatable
cylinder Kogler, he added two guard cells to the central measuring cell, avoiding the expansion
of the drilling and thus making interpretable test. The unit became operational quickly because of
advances in technology.

Now a day the PMT is becoming more popular in Ontario for site investigation and geotechnical
design especially in estimating soil properties for foundation design. Louis Menard developed
the pre-bored PMT device and considered it to be one of the most precise testing methods
available for almost any type of soil (Menard (1965)).

13
2.2.2.1 Equipment and Test Procedure

Equipment

The pressuremeter consists of three main parts which are a probe, a control unit and tubing for
inflation as shown in Figure 2.2.

Figure 2.2The Menard pressuremeter equipment (ASTM (2000))

(i) Probe: A typical Menard type PM probe includes three separate cells namely as top
cell, loading cell and bottom cell. Top and bottom cells are usually called “guard”
cells which are filled with gas before the test in order to isolate the loading cell from
end condition effects. Load cell is a flexible membrane (usually made from rubber)
that is filled with water after the guard cells are inflated. The two guard cells are used
to reduce end effects on the middle cell which will produce predominantly radial
strains in the soil. Lateral displacements are measured only in the middle cell.

14
 (ii) Control unit: A metal case that houses the main cylinder, four quick connectors and
the control valve. It consist a manual actuator to operate the piston and digital
pressure gauge. It is used for both controlling the pressure given to the probe and
monitoring the volume changes with respect to pressure increase by the dial gauges.
(iii) Tubing: A high pressure single conduit fitted with a shut off quick connector to keep
the probe and tubing saturated.

Currently, many types of PMs have been developed besides original Menard type PM such as
self -boring and push in PMs which can be used for different in-situ soil conditions.

Borehole preparation

It is extremely important to minimize disturbance of the borehole wall during, the drilling
process. Drilling methods should be selected to prevent collapse of the borehole wall, minimize
erosion of the soil and prevent softening of the soil (Finn et al. (1984)). Good test results begin
with a high quality borehole having minimal disturbance to its side walls, typically requiring
mud wash rotary techniques which was recommended (Briaud (2013)). Maintain the drilling
mud level at or near the top of the borehole minimizes the horizontal stress release from drilling.
During drilling, the operator should carefully monitor the rotation rate, advance rate and mud
flow to obtain a high quality borehole.

Test procedure

PMT is performed either by application of pressure in equal increments (stress controlled) or

equal volume increments (strain controlled). In stress control test, apply the pressure on the
control unit in about equal increments, until the expansion of the probe during one load
increment exceeds about 1⁄4 of V0. Generally 25, 50, 100 or 200 kPa pressures are selected for
testing soils. Too small steps will result in an excessively long test, too large steps may yield
result with inadequate accuracy. The pressure steps should be determined in such a way that
about 7 to 10 load increments are obtained. In a strain control test, increase the volume of the
probe on volume increments of 0.05 to 0.1 times the volume V0 until the limit of the equipment is

15
reached. For both procedures, take readings after 30s and 1 min after the pressure or volume
increment have been applied. Volume readings are recorded to an accuracy of 0.2% of V0 and
pressure readings to an accuracy of 5% of the limit pressure. The sequences of the PMT are
shown in Figure 2.3.

Figure 2.3 Sequences of the PMT (ASTM (2000))

2.2.2.2 Calibration

Before the test, two main calibration namely as volume and pressure calibration are required in
order to correct the raw data obtained during the test. Those calibrations are explained below.

Volume calibration: Volume calibration is performed for detection of the leaks in the system
and making necessary adjustments about system compressibility. PM probe is usually placed in a
steel tube before the volume calibration and the pressure is increased in steps. For a given

16
pressure, the volume lost is determined since the probe is confined by the tubes. A typical
volume calibration curve is given in Figure 2.4.

Pressure calibration: Pressure calibration is performed to determine the self - resistance of the
rubber membrane to expansion. Before the pressure calibration, probe is taken out from the steel
tube and calibration is performed in atmospheric pressure condition. A typical pressure
calibration curve is given in Figure 2.4.

Figure 2.4 Calibration curves obtained before the PMT (Bowles (1997))

Calibration in the PMT is essential for obtaining accurate results from the test and if the
calibrations are not carried out properly, and then the data obtained from the test can be
considered as useless.

2.2.2.3 Test results interpretation

Plot the pressure–volume curve by entering corrected volume and corrected pressure on a
coordinate system. Connect the points by a smooth curve. This curve is the corrected PMT curve
which is shown in Figure 2.5 and used in to determine the PMT modulus.

17
Figure 2.5 Typical pressuremeter test curve (ASTM (2000))

The PMT modulus is determined as follows.

𝛥𝑃
EPMT=2(1+υ) (V0+Vm) 𝛥𝑉 [2.7]

Where EPMT - Pressuremeter modulus

υ - Poisson’s ratio
V0 - Volume of the measuring portion of the uninflated probe at 0 volumes reading at
ground surface
Vm - Corrected volume reading in the center portion of the ΔV volume increase
ΔP - Corrected pressure increase in the center part of the straight line portion of the
pressure – volume curve (see Fig 2.5)
ΔV - Corrected volume increase in the center part of the straight line portion of the
pressure – volume curve, corresponding to ΔP pressure increase (see Fig 2.5)

18
The PMT modulus can also be calculated as represented by the slope of the pressure vs radial
strain curve along its linear portion is shown in Equation 2.8.

2 2
𝛥𝑅 𝛥𝑅
[1+(𝑅 ) ] + [1+(𝑅 ) ]
0 2 0 1
EPMT=(1 + 𝜐)(𝑝2 − 𝑝1 ) 2 2 [2.8]
𝛥𝑅 𝛥𝑅
[1+(𝑅 ) ] − [1+(𝑅 ) ]
0 2 0 1

𝛥𝑅
Where p and are the pressure and the corresponding radial strain recorded at the beginning
𝑅0

(subscript 1) and at the end (subscript 2) of the linear portion of the PMT pressure vs radial strain
curve respectively. The Poisson’s ratio is given by υ. For soils under drained conditions (ie, zero
excess pore pressure) a Poisson’s ratio of 0.33 is typically used, in which case the PMT modulus
is designated as the Menard’s modulus EPMT (Baguelin et al. (1978)).

Conventional limit pressure is determined as follows, the limit pressure (PL) is defined as the
pressure where the probe volume reaches twice the original soil cavity volume, defined as the
volume V0 +Vi (see Fig 2.5) where Vi is the corrected volume reading at the pressure where the
probe made contact with the borehole. The volume reading at twice the original soil cavity
volume is (V0 + 2Vi). The limit pressure is usually not obtained by direct measurements during
the test due to limitation in the probe expansion or excessively high pressure.

If the test was conducted to read sufficient plastic deformation, the limit pressure can be
determined by a 1/V to P plot, as shown in Figure 2.6. Points from the plastic range of the test
generally fall in an approximate straight line. The extension of this line to twice the original
probe volume will give the limit pressure (PL) on the plot.

19
Figure 2.6 Determination of limit pressure from inverse of volume vs pressure curve (ASTM
(2000))

2.2.2.4 Application of these two parameters

These two parameters are directly used in the Menard method (1965) to design the foundation. It
is assumed that the ultimate bearing capacity is related to the pressure limit (PL) and the
settlement is related to PM modulus (EPMT). The relationships have been shown below. The
pressuremeter bearing capacity factor, k, is defined by following equation.

𝑞 −𝜎
k=𝑃 𝑢 −𝜎𝑣 [2.9]
𝐿 ℎ

Where 𝑞𝑢 – Ultimate bearing capacity

𝜎𝑣 - Total vertical stress at the formation level
𝜎ℎ - Total horizontal stress at the pressuremeter test level

The Menard method for settlement is based on a modulus of elasticity which is expressed in
terms of the PM modulus, EPMT. The settlement is given by following equation.

20
 𝑞− 𝜎𝑣
S=(9𝐸 ) x a shape factor [2.10]
𝑃𝑀𝑇

Where q is the total bearing capacity

In this study the PMT was performed accordance with Procedure B, volume-controlled loading,
as outlined in the ASTM D 4719-00, Pre-bored PMT was completed using a TEXAM unit.

2.2.2.5 Pressuremeter test correlation in soils

Pressuremeter test results are used for identification of the soils. Briaud (1992) developed an
approximate common value of the pressuremeter parameters for clay and sand are given in Table
2.3 below.

Table 2.3 Approximate common values of the PMT parameters for clay and sand
Sand
Soil type Loose Compact Dense Very dense
PL(kPa) 0-500 500-1500 1500-2500 >2500
EP(kPa) 0-3500 3500-12000 12000-22500 >22500
Clay
Soil type Soft Medium Stiff Very stiff Hard
PL(kPa) 0-200 200-400 400-800 800-1600 >1600
EP(kPa) 0-2500 2500-5000 5000-12000 12000-25000 >25000

Yield pressure (Py)

The yield pressure indicates the end of the linear pseudo – elastic deformation and the onset of
plasticity. This yield pressure is useful in indicating beyond which pressure significant creep
deformation may occur.

𝐸𝑃𝑀𝑇 𝑃𝐿
Two useful ratios, such as 𝑃𝐿
and 𝑃𝑦
can be used as a general guideline for soil identification,
as follows (Briaud (1992)).

21
 𝐸𝑃𝑀𝑇
For sand 7 < < 12
𝑃𝐿

𝐸𝑃𝑀𝑇
For clay 12 < 𝑃𝐿

Typical Menard PMT values are presented in the CFEM (2006) is shown in Table 2.4.

Table 2.4 Typical Menard PMT values

Types of soil Limit pressure (kPa) 𝐸𝑃𝑀𝑇
𝑃𝐿
Soft clay 50 – 300 10
Firm clay 300 – 800 10
Stiff clay 600 – 2500 15
Loose silty sand 100 – 500 5
Silt 200 – 1500 8
Sand and gravel 1200 – 5000 7
Till 1000 – 5000 8
Old fil 400 – 1000 12
Recent fill 50 – 300 12

For most soil types the ratio between the limit and the yield pressures may be expressed as
𝑃𝐿
(Briaud (1992)) 1.3 < < 2.0
𝑃𝑦

Deformation modulus of soils (Es)

The slope of the PMT curve in the elastic range is defined as PMT modulus or Menard modulus
(EP or EM) of the soil. PMT modulus is commonly used in geotechnical practice for foundation
design because in many cases, the soil or rock shows elastic behavior before the failure
conditions. This deformation modulus is one of the most important parameters in any
geotechnical engineering projects. Its determination is not a fully solved theoretically. (Serrano
and Romana(2002)).

22
Expansion of a cylindrical cavity in an infinite elastic medium can be defined from cavity
expansion theory (Lame (1852) cited in Baguelin et al. (1978)) as

𝛥𝑃
G=V(𝛥𝑉) [2.11]

Where G – Shear modulus

V - Volume of the cavity
P – Pressure in the cavity

Shear modulus can be substituted with Young’s modulus by using the equation obtained from
theory of elasticity as follows.

𝐸𝑆
G= [2.12]
2(1+𝜐)

The critical parameter in the equation above is the Poisson’s ratio (υ) which varies with the type
of soil. For practical purposes a value of 0.33 is commonly selected for the Poisson’s ratio.
However, it is not appropriate to use for the undrained behavior of cohesive soils because
volume of the soil does not change during the loading. Thus, saturated clay would have a
Poisson’s ratio of o.45. Since Menard accepted the υ as 0.33 in his original study, PMT modulus
is calculated as follows.

EM=2(1+0.33) G [2.13]

EM=2.66G [2.14]

Even though PMT modulus describes elastic behavior of a soil, it shall be used cautiously for
design purposes because of the reasons listed below (Briaud (1992)).

(i) Strains on the soil are generally in large ranges which may not be realistic for the real
loading conditions.

23
 (ii) Tensile stresses are likely to occur in the circumferential direction during the test. In
spite the PM is a compression test, since the soil is known to be weak under tension;
measured modulus is reduced due to tensile stresses.
(iii) Disturbances on the walls of borehole significantly reduce the modulus.
(iv) Aspect ratio (L/D) of the probe has been found to be a factor that can be affecting the
modulus.
(v) Loading of the soil is relatively fast and in short time duration whereas the real
superstructure loads act slowly during a larger time span.
(vi) PMT modulus is a horizontal modulus. For vertical loading on the soil vertical
modulus should be considered which differs from horizontal modulus especially in
anisotropic soils.

As above reasons, the PMT modulus can be considered as a relatively low value compared with
Young’s modulus. Menard (1975) proposed that the PMT modulus should be divided by a
correction factor α in order to relate with Young’s modulus (Briaud (1992)). Typical α value is
proposed by Menard for different types of soil and rock are given in Table 2.5.

Table 2.5 Typical Menard α factors (Briaud (1992))

24
The pressuremeter modulus has been related empirically to the elastic modulus of the soil as
EM/E =α, (Menard (1995)), in which α is termed by Menard as the rheological coefficient and
has a value between 0 and 1.

Elastic Young’s modulus of soil

Soil Young’s modulus (E), commonly referred to as soil elastic modulus, is an elastic soil
parameter and a measure of soil stiffness. It is define as the ratio of the stress along an axis over
the strain along that axis in the range of elastic soil behavior. The elastic modulus is often used
for estimation of soil settlement and elastic deformation analysis. Soil elastic modulus can be
estimated from laboratory or in-situ tests or based on correlation with other soil properties. In
laboratory, it can be determined from triaxial test or indirectly from oedometer test. On field, it
can be estimated from SPT, CPT and PMT.

Typical Young’s modulus was recommended for different types of soils by Bowles (1996) were
shown in Table 2.6.

Table 2.6 Typical Young’s modulus for different types of soils (Bowles (1996))
Soil type Young’s modulus (Mpa)
Clay Very stiff 2 – 15
Soft 5 – 25
Medium 15 – 50
Hard 50 – 100
Sandy 25 – 250
Sand Silty 5 – 20
Loose 10 – 25
Dense 50 – 81
Silt 2 – 20
Glacial till Lose 10 – 150
Dense 150 – 720
Very dense 500 – 1440

25
2.2.3 CORRELATION BETWEEN SPT-N WITH PMT PARAMETERS

Estimation from SPT-N value of the two pressuremeter parameters such as pressuremeter
modulus (EPMT) and pressure limit (PL) has been studied by a few researchers in the past (Briaud
(1992) and Ohya et al. (1982)). Attempted correlations have usually been weak because of the
differences in the methods and uncertainties involved in the tests. Even though, they are widely
used in practice to get an idea about the level of the geotechnical parameters used in the design.
One linear relationship with zero intercept was proposed by Briaud (1992) for the EPMT and PL
from SPT-N value for sands (Figure 2.7 and Figure 2.8), while one non-linear relationship was
proposed by Ohya et al. (1982) on the basis of data obtained from alluvial and dilluvial deposits
in Japan (Figure 2.9 and Figure 2.10). Both researches indicated the wide scatter of data.

Figure 2.7 Correlation between SPT N and limit pressure (PL) (Briaud (1992))

26
Figure 2.8 Correlations between SPT N and Ep (Briaud (1992))

Figure 2.9 Correlation between SPT N and EPMT for clays (Ohya et al. (1982))

27
Figure 2.10 Correlation between SPT N and EPMT for sands (Ohya et al. (1982))

Further nonlinear relationships between SPT and both EPMT and PL for sand and clay were
proposed by Bozbey (2010) for data measured during an extensive geotechnical investigation
conducted in Istanbul, Turkey. In glacial tills, there is a study conducted by Yagiz (2008), with a
linear relationship with an intercept between the corrected SPT-N values (Ncor) with both EPMT
and PL in Gumusler country, 10 km north of the city of Denizli, Turkey.

2.3 PREVIOUS RESEARCH ON NUMERICAL SIMULATIONS OF THE

PMT

2.3.1 GEOMETRY AND BOUNDARY CONDITION OF THE MODEL

The PMT was simulated by an axisymmetric model in the Schanz et al. (2000), Michel et al.
(2000), Jacques (2007), Rita (2008), Levasseur et al. (2009), Malecot et al. (2009), Plaxis 2D
(2012), Monnet (2012), Sedran et al. (2013), Fawaz et al. (2014). The geometry sizes, diameter
of 1.2 m and height of 1.5 m were used by Schanz et al. (2000), Malecot et al. (2009) and Plaxis

28
2D (2012). The Levasseur et al. (2009) used a diameter of 6 m and height of 5 m in their
geometry model. Jacques (2007) and Monnet (2012) used a geometry model with diameters of 5
and 10 m respectively and 2 m of height above the midpoint of the probe. But both of them did
not mention the dimensions below the probe. Fawaz et al. (2014) used 7 m of diameter and 7 m
of height below the probe but did not mention the dimensions above the probe. The Michel et al.
(2000), Rita (2008) and Sedran et al. (2013) did not define the geometry sizes of their model.

In the boundary conditions, bottom of the model was vertical fixity and both vertical faces of the
model were horizontal fixity as used by Schanz et al. (2000) and Plaxis 2D (2012). Bottom of
the model was total fixity (both horizontal and vertical) and both vertical faces of the model were
horizontal fixity as used by Levasseur et al. (2009), Malecot et al. (2009) and Fawaz et al.
(2014). Bottom of the model was vertical fixity and right side of the model was free in both
directions (horizontal and vertical) and the left side of the model was horizontal fixity as used by
Jacques (2007) and Monnet (2012). The Michel et al. (2000), Rita (2008) and Sedran et al.
(2013) once again did not define the boundary conditions of their model.

Each authors used different geometry dimensions and boundary conditions. The geometry
models with dimensions and boundary conditions that were available from the literature are
shown in Figure 2.11 to Figure 2.13.

To simulate the test at a deeper elevation, a vertical stress was applied at the top of the mesh
(load B). The pressuremeter applies a radial stress (load A) at the lower part of the borehole with
an imposed stress at each loading steps as stated in the Jacques (2007). But in the Plaxis 2D
(2012) and Schanz et al. (2000) load A was applied as a vertical surcharge stress and load B was
applied in a radial stress on the probe. In the Levasseur et al. (2009) load B was used for both
radial stress and overburden vertical stress.

29
Figure 2.11 The 2D axisymmetric model and associated mesh diagram (Malecot et al. (2009))

Figure2.12 The 2D axisymmetric model and associated mesh diagram (Levasseur et al. (2009))

30
Figure 2.13 Geometry model for PMT (Plaxis 2D (2012))

2.3.2 PROB LENGTH (L) TO DIAMETER (D) RATIO (L/D RATIO)

The PMT was modeled as an axisymmetric problem. In this case, where L/D ratio is higher than
6, the test results were similar to those obtain considering plain strain conditions (Houlsby and
Carter (1993)). The L/D ratio used by Waschkowski (1976), Schanz et al. (2000), Jacques
(2007), Rita (2008), Levasseur et al. (2009), Plaxis 2D (2012), Monnet (2012) and Sedran et al.
(2013) were 6.6, 2.9, 7.5, 6.7, 8, 3.6, 6 and 6.5 respectively. The Michel et al. (2000), Malecot
et al. (2009), and Fawaz et al. (2014) did not mention the L/D ratio in their study.

2.3.3 TYPE OF MODEL

A lot of soil constitutive models can be found in the literature that permits to deal with a large
variety of geotechnical problems. Nevertheless, these constitutive models have, most of the time,
a large number of parameters whose values are not obvious to identify. Classically, the
parameters’ values are estimated from laboratory tests on small samples or from in- situ tests.

31
Unfortunately, parameters estimated from laboratory test were difficult because of the weak
representative of the soil sample size and the perturbations imposed to the samples during its
extraction. Likewise, in–situ tests do not allow the direct identification of the constitutive
parameters of the soil layers.

In the literature, a constitutive model was introduced which was formulated in the frame work of
classical theory of plasticity. Instead of using Hooke’s single stiffness model with linear
elasticity in combination with an ideal plasticity according to Mohr-Coulomb (M-C). A new
constitutive formulation using a double stiffness model for elasticity in combination with
isotropic strain hardening was used. Summarizing the existing double stiffness models the most
dominant type of model is the Cam Clay model (Hashiguchi (1985) and Hashiguchi (1993)). To
describe the non-linear stress strain behavior of soil, beside the Cam Clay model the pseudo
elastic (hypo-elastic) type of model has been developed. There an Hookean relationship was
assumed between increments of stress and strain and non-linearity was achieved by means of
varying Young’s modulus. The Duncan Chang (1970) model known as hyperbolic model
captures soil behavior in a very tractable manner on the basis of only two stiffness parameters
and it was very much appreciated among consulting geotechnical engineers. The major
inconsistency of this type of model which was the reason why it was not accepted by scientists is
that, in contrast to the elasto - plastic type of model, a purely hypo elastic model cannot
consistently distinguish between loading and unloading. In addition the model was not suitable
for collapse load computation in the fully plastic range. These restrictions will be overcome by
formulating a model in an elasto - plastic frame work in the constitutive models. The hardening
soil model supersedes the Duncan Chang model. Firstly by using the theory of plasticity rather
than theory of elasticity and secondly by including soil dilatancy and thirdly by introducing a
yield cap.

In case of simulation of PMT, the M-C model was used as a material model by Schanz et al.
(2000), Levasseur et al. (2009), Malecot et al. (2009), Monnet (2012) and Fawaz et al. (2014).
The hardening soil model (HSM) was used by Michel et al. (2000), Rita (2008), Plaxis 2D
(2012) and Sedran et al. (2013). Jacques (2007) used Tresca model as a material model in the
simulation study.

32
2.3.4 DEFORMED MESH AND ANALYSIS

In the PMT analysis, 15 nodes triangular elements were used in the mesh by Jacques (2007) and
Rita (2008). The deformed geometry of the PMT after the simulation from Schanz et al. (2000),
Rita (2008), Levasseur et al. (2009), Malecot et al. (2009), Plaxis 2D (2012) and Fawaz et al.
(2014) are shown in Figure 2.14 to Figure 2.19 respectively.

Figure 2.14 FE discretization of the calibration chamber (Schanz et al. (2000))

33
Figure 2.15 Points from which the displacements are read (Rita (2008))

Figure 2.16 The 2D axisymmetric model and associated mesh diagram (Malecot et al. (2009))

34
Figure 2.17 The 2D axisymmetric model and associated mesh diagram (Levasseur et al. (2009))

Figure 2.18 Deformed geometry of the PMT (Plaxis (2012))

35
Figure 2.19 Geometry of axisymmetric model (Fawaz (2014))

In this FE analysis of PMT, a vertical interface along the shaft of the PMT borehole and a
horizontal interface just above the PM were introduced to allow for a discontinuity in horizontal
displacement in the Plaxis 2D (2012). The vertical interface along the borehole face was
introduced in the Levasseur et al. (2009) and Malecot et al. (2009).

2.3.5 ANALYSIS RESULTS AND GRAPHS

The numerical results and pressuremeter test data were plotted and compared, which were shown
in Figure 2.20 to Figure 2.23 from Schanz et al. (2000), Rita (2008), Plaxis 2D (2012) and Fawaz
(2014) respectively. Graphs were drawn for pressure vs volumetric strain. The volumetric strain
was calculated from the original radius R0 and lateral expansion Ux of the PM. The volume
change cannot directly be measured from Plaxis and was calculated from Equation 2.15.

𝛥𝑉 (𝑅0+ 𝑈𝑥 )2 − (𝑅0 )2
= [2.15]
𝑉0 (𝑅0 )2

36
Figure 2.20 Comparison between experimental and numerical results of the PMT (Schanz et al.
(2000))

Figure 2.21 Model and test curves at 500kPa of effective vertical stress (Rita (2008))

37
Figure 2.22 Comparison of numerical results and PMT data (Plaxis (2012))

Figure 2.23 Experimental and numerical pressure-volume curves at different depth (Fawaz
(2014))

38
2.3.6 COMPARISON OF PM AND YOUNG MODULUS

The pressuremeter modulus (EM) and its comparison to the Young modulus (E) of soil were
performed by Sedran et al. (2013) and Fawaz et al. (2014). Pressuremeter testing provides stress-
strain data for pseudo-elastic and elastic–plastic ranges of soil deformation. The analysis of stress
and strain changes in a soil mass due to PMT loading is based in the theory of cavity expansion
as it pertains to an infinitely long cylinder expanding into an infinite soil range. Assuming
uniform isotropic and linear elastic soil behavior, the elastic property of the soil is represented by
the pressuremeter modulus (EM) (Briaud (1992)), and is calculated with the following
expression.

2 2
𝛥𝑅 𝛥𝑅
[1+(𝑅 ) ] + [1+(𝑅 ) ]
0 2 0 1
EM = (1 + 𝜐)(𝑝2 − 𝑝1 ) 2 2 [2.16]
𝛥𝑅 𝛥𝑅
[1+(𝑅 ) ] − [1+(𝑅 ) ]
0 2 0 1

𝛥𝑅
Where p and are the pressure and the corresponding radial strain recorded at the beginning
𝑅0

(subscript 1) and at the end (subscript 2) of the linear portion of the PMT pressure vs radial strain
curve respectively. The Poisson’s ratio is given by υ. For soils under drained conditions (ie, zero
excess pore pressure) a Poisson’s ratio of 0.33 is typically used, in which case the pressuremeter
modulus is designated as the Menard’s modulus EM (Baguelin et al. (1978)).

2.3.7 CONVENTIONAL LIMIT PRESSURE

The numerical validation of an elastoplastic formulation of the conventional limit pressure

measured with the pressuremeter test in cohesive soil studied by Jacques (2007). An elastoplastic
presssuremeter theory was used to determine the conventional limit pressure. Then conventional
limit pressure was computed by using Plaxis to check the validity of the theoretical results. In the
Plaxis, the Tresca failure model was used to determine the conventional limit pressure. The
behavior of cohesive soil around the pressuremeter was studied by Jacques (2007). The influence
of the permeability in a linear elastoplastic soil and of the geometry of the probe has been studied

39
(Nahra and Frank (1986)) by the numerical expansion of a cylindrical cavity. The well-known
theory of undrained behavior (Baguelin et al. (1972)) has been used (Prapaharan et al. (1989)) to
investigate the pressuremeter expansion as a function of the undrained shear strength which
varies with the strain rate. Numerical results with a constitutive model (Cambou and Bahar
(1993)) show that the test can be assumed to be an undrained one with permeability lower than
10-10 m/s.

The stress strain behavior of the cohesive soil is assumed to follow either a linear elasticity
relation (Gibson and Anderson (1961) and Silvestri (2003)) a hyperbolic elastoplastic relation
(Silvestri (2004)) or a power law (Bolton and Whittle (1999)).

The theoretical analysis was chosen for its ability to describe the pressuremeter test from
beginning to end with only a few parameters, when numerical analysis with sophisticated models
needs many mechanical parameters, which cannot be precisely fitted. As a matter of fact, on a
pressuremeter curve which is computed with an eight parameters model, only one or two
parameters can be fitted, while six or seven other parameters must be assumed (Cambou and
Bahar (1993)).

During the PMT, three different areas of soil were considered from the borehole wall to the
infinite radius. Plasticity appears in the first zone between the radial stress (𝜎𝑟 ) and the

circumferential stress (𝜎𝛳 ). This first plastic area extends between the radius ra (borehole wall)
and rb (external radius of the first plastic area). The second plastic area between radii r b and rc
(external radius of both plastic areas). An elastic area extends beyond radius rc.

In the horizontal and vertical planes the equilibrium of an element of soil is given by

𝑑𝜎𝑟
𝜎𝑟 - 𝜎𝛳 + r =0 [2.17]
𝑑𝑟

The conventional limit pressure (PlM) was obtained by using the following expression.

40
 2𝐺(√2−1)+𝑐𝑢
PlM = γz + cu ln [ (1−𝐾 ] [2.18]
0 )𝛾𝑧+𝑐𝑢

This relation is quite different from the Menard experimental correlation proposed by the
European Regional Technical Committee (Amar et al. (1991)).

PlM = 5.5𝑐𝑢 + 𝐾0 γz if PlM - 𝐾0 γz < 300Mpa [2.19]

PlM = 10(𝑐𝑢 - 25) + 𝐾0 γz if PlM - 𝐾0 γz > 300Mpa [2.20]

The in–situ pressuremeter tests were carried out for over consolidated plastic clay in Paris. The
theoretical pressuremeter curve and experimental one are drawn in one graph and compared.
Then they used finite element program Plaxis (Brinkgreve and Vermeer (1998)) with the Tresca
model to compute the value of the conventional limit pressure, which was compared to the result
of theory. The model used was elastoplastic with a constant shearing modulus and five
parameters (Young’s modulus, Poisson’s ratio, undrained shear strength, no friction and
dilatancy angle). The method used for the validation was a variation of only one parameter when
the other ones stay constant. Finally influence of the vertical stress, coefficient of pressure at rest
(K0), shearing modulus and shear strength in the conventional limit pressure was discussed.

2.3.8 ANALYTICAL METHOD

Analytical method was used to evaluate soil parameters which were used in the M-C model by
Fawaz et al. (2014). Different rheological laws have been developed to describe the behavior of
soils around the pressuremeter. The study of Combarieu (1995) based on Pasturel’s formula has
evolved a theoretical relation between the limit pressure (PL) and soil parameters E, υ, c and φ. In
case of cohesive and granular soils (c and φ different to 0) that relation is given below.

𝑠𝑖𝑛𝜑
𝐸 1+𝑠𝑖𝑛𝜑
PL + c cotφ = (P0+ccotφ) (1+sinφ) ( ) [2.15]
2(1+𝜐)(𝑃0 +𝑐𝑐𝑜𝑡𝜑)(𝑠𝑖𝑛𝜑)

41
This formula was used to calculate the cohesion and the friction angle by using the value of the
pressure at rest (P0) and limit pressure (PL) determined from the in-situ test and elastic modulus
obtained in the numerical analysis.

2.4 SUMMARY

This chapter discussed about the literature review of the statistical correlation between SPT and
PMT and numerical simulation of the PMT. In the case of statistical correlation, most of the
correlation work done in the past was for sand and clay and only one for tills. There is almost no
correlation work on Toronto glacial tills. So far there is no clear explanation on correlation
between these two parameters for glacial till especially Toronto glacial tills. The literature has a
significant lack of information concerning any glacial till such as cohesive or cohesionless
glacial tills.

In the case of numerical simulation of PMT, there is bulk of information available but no one has
simulated the model for real soil profile. Every author mentions the geometry of the model with
some dimensions but they don’t clearly mention the soil layers above and below the test depth.
In addition to soil profile, authors did not properly define the geometry size, width of the model
from axis of symmetry and depth below the mid-point of the probe. Further literature survey
show that commonly used models are HSM and MCM. Each author’s used different boundary
conditions, geometry size, probe L/D ratio and type of model which are shown in Table 2.7.

Table 2.7 Reviews of the literature survey regards FEM simulation of PMT
Author’s name Type of Probe Model geometry size Boundary condition
model L/D
ratio
Plaxis 2D HSM* 2.7 Diameter 1.2m Bottom – VF
manual (2012) Height 0.75m Vertical both side- HF
Fawaz et MCM Not Width 7m Bottom – Both fixity
al.(2014) mention Height 7m below the Vertical both side- HF
probe

42
Jacques (2007) Tresca 7.5 Diameter 5m Bottom - VF
Model Height 2m above the Right side vertical – both direction
mid of probe free
Left side vertical - HF
Levasseur et MCM Not Diameter 1.20m Bottom – Both fixity
al. (2009) mention Height 1.5m Vertical both side- HF
Schanz et al. MCM 2.9 Diameter 1.2m Bottom – VF
(2000) Height 1.5m Vertical both side- HF
Malecot et al. MCM 8 Diameter 6m Bottom – Both fixity
(2009) Height 5m Right side vertical- HF
Left side vertical- HF below probe
Rita (2008) HSM* 6.7 Not define Not define
Sedran et al. HSM* 6.5 Not define Not define
(2013)
Monnet (2012) MCM 6 Diameter 10 m Bottom –VF
Height 2 m above the Right side vertical – Free
mid of probe Left side vertical – HF above
probe
Michel et HSM* Not Not define Not define
al.(2000) define
Waschkowski 6.6
(1976)
* HSM – Hardening soil model
 MCM – Mohr Coulomb model
 VF – Vertical fixity of the model
 HF – Horizontal fixity of the model

43
CHAPTER 3: STATISTICAL CORRELATION BETWEEN SPT-N
VALUES WITH PMT PARAMETERS FOR GLACIAL TILLS

3.1 INTRODUCTION

The statistical analysis is carried out in this chapter to investigate the relationship between SPT-
N value with both EPMT and PL. The first step is to collect the pairs of PMT test data and SPT-N
value at the same depths in the same boreholes. Secondly selected data are corrected and filtered
according to the methodology discussed in the section 3.2. Thirdly the general ranges of SPT-N,
EPMT and PL values are discussed in the section 3.3. Then correlation between SPT-N value with
both EPMT and PL are discussed in the section 3.4 and 3.5 respectively. In addition the
comparisons are done between the studied correlation equations with literature equations and
values in the section 3.6. And finally the ranges of corrected and filtered SPT-N, EPMT and PL
values and correlation equations are summarized in the section 3.7.

3.2 DATA SELECTION, SPT-N VALUE CORRECTION AND FILTERING

SPTs conducted near the PMTs at similar depths are selected to develop the relationship
between SPT-N values and both EPMT and PL in this chapter for the following stations such as
Allen, Avenue, Bathurst, Bayview, Bermondsey, Blackcreek, Birchmount, Caledonia, Don
Mills, Kennedy, Lesile, Mount Dennis, Victoria, Warden, West Portal and Wynford. The
typical borehole report is attached in Appendix 3.1. The pairs of readings (SPT-N and EPMT) for
clayey silt and silty sand till are not available from these stations in this study.

SPT-N correction

The first correction for SPT-N values are performed according to Cao et al. (2015) for field
measured SPT-N for penetration depth. This means that some situations, the field SPT hammer
are refused while driving, when it reached boulders or cobbles. In this situation hammering is
stopped and number of blows is counted before full penetration of 305 mm. Then the field SPT-

44
N values are corrected according to equation 2.1 in Chapter 2. For example, when the sample
tube is driven 175 mm into the ground and the number of blows is 40, then the SPT N-value is
305 𝑥40
70 ( = ). Typical correction calculation sheet is attached in Appendix 3.2.
175

The second corrections are performed according to the CFEM (2006). Because of the variability
in equipment and operating conditions, direct use of SPT-N values for geotechnical design is not
recommended. As a result, many corrections shall be done on the field SPT-N values. Those
corrections are described in Chapter 2, section 2.2.1.2 based on CFEM (2006). In case of
cohesive glacial tills, overburden correction is not accommodated in this study. In these
situations, the SPT-N became SPT-(𝑁)60 . In the case of cohesionless glacial tills, overburden
corrections is accommodated. In these situations, the SPT-N became SPT-(𝑁1 )60.Typical
correction calculation sheets for each station is attached in Appendix 3.2. Based on these
corrections, the conclusion is made that after the correction, the SPT-N value became half of the
field measured SPT-N value, specifically in the deeper elevation.

Filtering data

The processing of these data is one of the most challenging works in this project. Reliability of
an analysis result is mostly defined by the accuracy of selected data rather than the method used
for the analysis. Therefore, the selection of the most representative parameters for a site is the
key to a successful analysis. With that in mind, in order to evaluate the correlation between SPT-
N values and both EPMT and PL more accurately, the compiled data are filtered by using the
following methodology.

(1) The SPT’s often reached refusal, i.e. blow count (N) values are greater than 100 for 300 mm
or less increment when the SPT sampler hits a cobble or boulder within the glacial till. As a
result the SPT-N values are assigned values of 100 or more than 100. The SPT-N values
greater than 100 are disregarded.
(2) The data situated far from the trend line is discarded by visual inspection compare to other
data.

45
(3) In such cases the same SPT-N value is associated with different values of both EPMT and PL
and this pair of readings are omitted.
Apparently more theoretical study is needed to develop a sound rationale to filter the data.

3.3 GENERAL RANGES OF SPT-N, EPMT AND PL FOR GLACIAL TILLS

The ranges of SPT-N, EPMT and PL values are determined for both groups in which all of the data
are collected from in-situ tests and in which the data are corrected and filtered.

3.3.1 RANGES OF SPT-N VALUES

The ranges of SPT-N values of cohesive and cohesionless glacial tills are shown in Figure 3.1
and Figure 3.2 respectively. Further ranges, means and standard deviations of SPT-N values for
different types of soil and all soil, for all data as well as corrected and filtered data are shown in
Table 3.1. The data that were discarded during the filtering process are shown in Figure 3.1, with
small circles in the all data analysis. The percentages (%) marked in Figure 3.2 represents most
of the range values that belong to the thick portion of the range diagrams.

46
Table 3.1 Summary of SPT-N values for different types of soil
Soil type Ranges of SPT-N values for all data (Corrected & Filtered data)
No. of data Range Mean Standard
deviation
Cohesive glacial tills
Silty clay 38 (22) 8-91 (6-62) 38 (26) 23 (16)
Silty clay till 25 (14) 5-98 (3-31) 32 (15) 24 (9)
Clayey silt till 21 (16) 6-152 (4-67) 46 (29) 32 (19)
All soil 84 (52) 5-152 (3-67) 38 (24) 26 (16)
Cohesionless glacial tills
Sand 23 (18) 21-150 (13-97) 61 (50) 29 (23)
Silt 16 (14) 8-123 (4-98) 66 (46) 33 (30)
Sandy silt 22 (20) 6-86 (4-91) 53 (45) 23 (23)
Silty sand 23 (18) 38-127 (25-76) 63 (49) 21 (18)
Sandy silt till 8 (7) 34-93 (16-80) 58 (60) 20 (25)
All soil 92 (77) 6-150 (4-98) 60 (49) 26 (23)

47
 160
140
120
SPT-N Range

100
80
60
40
20
0

Clayey silt till

Silty clay till
Silty clay

All soil
All data

70

60
SPT - (N) Range

50
60

40

30 55% 63%
58%
20
50%
10

0
Clayey silt till
Silty clay till
Silty clay

All soil

Corrected and filtered data

Figure 3.1 Ranges of SPT-N values for cohesive glacial tills

48
 160
140
120
SPT-N Range

100
80
60
40
20
0

Sandy silt till

Silty sand
Sandy silt

All soil
Sand

Silt

All data

100
SPT- (N ) Range

80

70%
60
1 60

61% 50% 56%

56%
40 60%

20

0
Sandy silt till
Silty sand
Sandy silt

All soil
Sand

Silt

Corrected and filtered data

Figure 3.2 Ranges of SPT-N values for cohesionless glacial tills

49
3.3.2 RANGES OF EPMT VALUES

The ranges of EPMT values of cohesive and cohesionless glacial tills are shown in Figure 3.3 and
Figure 3.4 respectively. Further ranges, means and standard deviations of EPMT values for
different types of soil and all soil, for all data as well as filtered data are shown in Table 3.2. The
data that were discarded during the filtering process are shown in Figure 3.3, with small circles
in the all data analysis. The percentages (%) marked in Figure 3.4 represents most of the range
values that belong to the thick portion of the range diagrams.

Table 3.2 Summary of EPMT values for different types of soil

Soil type Ranges of EPMT values (MPa) for all data (Filtered data)
No. of data Range Mean Standard
deviation
Cohesive glacial tills
Silty clay 38 (23) 11- 224 (11-150) 76 (65) 47 (39)
Silty clay till 25 (13) 4-223 (4-36) 49 (18) 51 (10)
Clayey silt till 21 (16) 16-288 (16-131) 82 (58) 76 (41)
All soil 84 (52) 4 -288 (4-150) 69 (51) 58 (40)
Cohesionless glacial tills
Sand 22 (14) 26-197 (26-149) 104 (91) 48 (46)
Silt 16 (14) 19-140 (19-140) 84 (82) 32 (33)
Sandy silt 22 (15) 2-163 (28-78) 71 (53) 47 (17)
Silty sand 23 (13) 10-231 (39-96) 105 (69) 58 (18)
Sandy silt till 8 (6) 18-273 (18-134) 112 (76) 79 (39)
All soil 91 (62) 2-273 (18-149) 93 (74) 53 (34)

50
 300

250
Range (MPa)

200

150

100
PMT
E

50

Clayey silt till

Silty clay till
Silty clay

All soil
All data

160
140
Range (MPa)

120
100
80 63%
52%
60
PMT

56%
40
E

20 54%
0
Clayey silt till
Silty clay till
Silty clay

All soil

Filtered data
Figure 3.3 Ranges of EPMT values for cohesive glacial tills

51
 E Range (MPa) E Range (MPa)
PMT PMT

All data

Filtered data
0
50
100
150
0
50
100
150
200
250
300

Sand

64%
Sand

Silt

57%
Silt

Sandy silt

60%

52
Sandy silt

Figure 3.4 Ranges of EPMT values for cohesionless glacial tills

Silty sand

62%
Silty sand

Sandy silt till

67%
Sandy silt till

All soil
55%
All soil
3.3.3 RANGES OF PL VALUES

The ranges of PL values of cohesive and cohesionless glacial tills are shown in Figure 3.5 and
Figure 3.6 respectively. Further ranges, means and standard deviations of PL values for different
types of soil and all soil, for all data as well as filtered data are shown in Table 3.3. The data that
were discarded during the filtering process are shown in Figure3.3, with small circles in the all
data analysis. The percentages (%) marked in Figure 3.4 represents most of the range values that
belong to the thick portion of the range diagrams.

Table 3.3 Summary of PL values for different types of soil

Soil type Ranges of PL values (MPa) for all data (Filtered data)
No. of data Range Mean Standard
deviation
Cohesive glacial tills
Silty clay 38 (22) 0.64-9.02 (1.25-5.56) 3.56 (3.62) 1.89 (1.23)
Silty clay till 25 (17) 0.41-7.78 (0.41-5.63) 2.72 (2.35) 2.13 (1.85)
Clayey silt till 20 (15) 1.0 - 14.15 (1.00-6.00) 4.0 (2.79) 3.55 (1.72)
All soil 83 (54) 0.41-14.15 (0.41-6.00) 3.42 (3.00) 2.47 (1.65)
Cohesionless glacial tills
Sand 23 (17) 0.21-13.32 (2.42-13.32) 7.23 (7.97) 3.55 (3.46)
Silt 16 (12) 0.97-14.57 (3.17-9.08) 7.10 (7.05) 3.04 (1.81)
Sandy silt 22 (18) 0.29-15.79 (1.33-9.03) 5.40 (4.52) 3.65 (2.13)
Silty sand 23 (18) 1.42-14.30 (1.42-13.55) 7.28 (6.37) 3.80 (3.67)
Sandy silt till 8 (4) 1.7-22.49 (6.00-8.04) 9.64 (7.02) 6.83 (0.87)
All soil 92 (69) 0.21- 22.49 (1.33-13.55) 7.00 (6.44) 4.00 (3.09)

53
 P Range (MPa) P Range (MPa)
L L

All data

0
1
2
3
4
5
6

Filtered data
0
5
10
15

Silty clay Silty clay

55%
Silty clay till Silty clay till

53%

54
Figure 3.5. Ranges of PL values for cohesive glacial tills
Clayey silt till Clayey silt till

67%
All soil All soil
50%
 P Range (MPa)
P Range (MPa) L
L

All data

Filtered data
0
5
10
15
20
25

0
2
4
6
8
10
12
14
Sand Sand

53%
Silt

50%
Silt

Sandy silt

55
Sandy silt

61%

Figure 3.6 Ranges of PL values for cohesionless glacial tills

Silty sand

61%
Silty sand

Sandy silt till

75%
Sandy silt till

All soil All soil

54%
3.4 CORRELATION BETWEEN SPT-N AND EPMT VALUES

(a) Low plasticity cohesive glacial tills

The correlation between SPT-N values and EPMT have been plotted for low plasticity cohesive
glacial tills in both original data as well as corrected and filtered data formats, as shown in Figure
3.7. The correlation functions are determined for both cases in which all the data were included
and in which the data were corrected and filtered. The correlation functions and coefficients are
given in Table 3.4. The corrected and filtered data analysis provides a much improved
correlation coefficient compared to all original data analysis.

56
 300
Silty clay
250 Silty clay till
Clayey silt till
200
(MPa)

All soil

150
PMT
E

100

50

0
0 50 100 150 200 250
SPT-N values
All data

160 y = M1*M0
Silty clay Value Error
140 m1 0.87194 0.15857
Silty clay till
120 Clayey silt till Chisq 70930 NA
(MPa)

2 -0.92967 NA
100 All soil R

80
PMT

60
E

40
20
0
0 20 40 60 80 100
SPT- (N) values
60

Corrected and filtered data

Figure 3.7 Correlations between SPT-N vs EPMT for cohesive glacial tills y = M1
Valu
m1 1.700
57
Chisq 867
2 0.5508
R
(b) Cohesionless glacial tills

The correlation between SPT-N values and EPMT have been plotted for cohesionless glacial tills
in both original data as well as corrected and filtered data formats, as shown in Figure 3.8. The
correlation functions and coefficients are given in Table 3.4. The corrected and filtered soil data
analysis shows that there is a better correlation relationship between SPT-N and EPMT. After
corrected and filtered, the sandy silt till does not have enough pairs of data.

58
 300
Sand
250 Silt
Sandy silt
Silty sand
(MPa)

200
Sandy silt till
150 All soil
PMT
E

100

50

0
0 20 40 60 80 100 120 140 160
SPT-N value
All data
y = M1*M0
C
200 Value Error
y = M1*M0
Sand y = M1*M0 m1 1.1819 0.11878
Value
Silt Error Value ErrorChisq 16391 NA
150
m1 1.0656
Sandy0.13472
silt m1 1.1276 0.14891 2 0.45722 NA
R
Chisq 23774 NA Chisq 17087 NA
(MPa)

Silty sand
2 0.14643 2
R All soil NA R -0.19513 NA
100
PMT
E

50

0
0 20 40 60 80 100
SPT- (N ) value
1 60

Corrected and filtered data

Figure 3.8 Correlations between SPT-N vsyE= for cohesionless glacial tills
M1*M0
PMT y = M1
Value Error Valu
m1 1.7845
59 0.13858 m1 1.580
Chisq 10236 NA Chisq 5981
2 0.62141 NA 2 0.4452
R R
3.5 CORRELATION BETWEEN SPT-N AND PL VALUES

(a) Low plasticity cohesive glacial tills

The correlation between SPT-N values and PL have been plotted for low plasticity cohesive
glacial tills in both original data as well as corrected and filtered data formats, as shown in Figure
3.9. The correlation functions are determined for both cases in which all the data were included
and in which the data were corrected and filtered. The correlation functions and correlation
coefficients are given in Table 3.4. The corrected and filtered data analysis provides a much
improved correlation coefficient compared to all original data analysis.

60
 10

8
P (MPa)

6
L

4 Silty clay
Silty clay till
2 Clayey silt till
All soil
0
0 20 40 60 80 100 120 140 160
SPT-N value

All data
y = M1*M0
10 Value Error
Silty clay m1 0.08145 0.0060293
Silty clay till Chisq 72.015 NA
8
2 0.26841 NA
Clayey silt till R
P (MPa)

6 All soil
L

0
0 20 40 60 80 100
SPT- (N) value
60

Corrected and filtered data y = M1*M0

Figure 3.9 Correlations between SPT-N vs PL for cohesive glacial tills Value Error
m1 0.10393 0.0078714
Chisq 13.52 NA
61
2 0.67741 NA
R
(b) Cohesionless glacial tills

The correlation between SPT-N values and PL have been plotted for cohesionless glacial tills in
both original data as well as corrected and filtered data formats, as shown in Figure 3.10. The
correlation functions and coefficients are given in Table 3.4. The all soil data analysis shows that
there is a weak correlation relationship between SPT-N and PL where the correlation coefficient
(R2) is 0.04 to 0.46. After corrected and filtered, sandy silt till does not have enough pairs of
data. The corrected and filtered data analysis provides a much improved correlation coefficient
(0.67 to 0.85) compared to all the original data analysis.

62
 25
Sand
Silt
20
Sandy silt
Silty sand
P (MPa)

15 Sandy silt till

All soil
L

10

0
0 20 40 60 80 100 120 140 160
SPT-N value
All data y = M1*M0 y = M1*M
Value Error Value
15
m1 0.076063 0.011053 m1 0.095178
Sand Chisq 160.01 NA Chisq 133.54
Silt 2
R -0.15682 NA R
2 -0.043919
10 Sandy silt
P (MPa)

Silty sand
All soil
L

0
0 20 40 60 80 100
SPT- (N ) value
1 60

Corrected and filtered data

Figure 3.10 Correlations between SPT-N vs PL for cohesionless glacial tills

63
Table 3.4 Summary of correlations between SPT-N values with both EPMT and PL values for
different types of soil

Soil type Correlation equations (EPMT) (MPa) Correlation equations (PL) (MPa)
(R2) (R2)
All data Corrected & All data Corrected &
Filtered data Filtered data
Cohesive glacial tills
Silty clay 0.87N (0.23) 1.70(𝑁)60 (0.55) 0.082N (0.27) 0.104(𝑁)60 (0.68)
Silty clay till 1.49N (0.72) 1.37(𝑁)60 (0.91) 0.075N (0.80) 0.092(𝑁)60 (0.89)
Clayey silt till 1.76N (0.48) 1.61(𝑁)60 (0.83) 0.054N (0.19) 0.104(𝑁)60 (0.90)
All soil 1.65N (0.41) 1.54(𝑁)60 (0.83) 0.070N (0.23) 0.103(𝑁)60 (0.87)
Cohesionless glacial tills
Sand 1.07N (0.15) 1.79(𝑁1 )60 (0.79) 0.076N (0.16) 0.126(𝑁1 )60 (0.71)
Silt 1.13N (0.20) 1.58(𝑁1 )60 (0.45) 0.095N (0.04) 0.141(𝑁1 )60 (0.85)
Sandy silt 1.18N (0.46) 1.68(𝑁1 )60 (0.77) 0.090N (0.46) 0.124(𝑁1 )60 (0.67)
Silty sand 1.27N (0.07) 1.47(𝑁1 )60 (1.0) 0.093N (0.08) 0.158(𝑁1 )60 (0.80)
Sandy silt till 1.67N (0.37) No correlation 0.146N (0.32) No correlation
All soil 1.10N (0.26) 1.54(𝑁1 )60 (0.58) 0.080N (0.26) 0.139(𝑁1 )60 (0.71)

64
3.6 COMPARISONS OF CORRELATIONS BETWEEN SPT-(𝑵𝟏 )𝟔𝟎
VALUES AND BOTH EPMT AND PL VALUES

There is a limited information available about the correlation between SPT-N values and both
EPMT and PL for sand and clay, and is particularly sparse for glacial tills. This section presents a
study on the correlation between SPT-(𝑁1 )60values and both EPMT and PL for glacial tills in the
city of Toronto. In addition, the comparison is performed between this study and the literature
for sand and glacial tills. For the sand there are two types of literature models are available. The
developed regression line by using corrected and filtered data is compared with available
literature models. Further, another comparison of the data is performed for glacial till within the
studied data. In this comparison, a linear correlation with intercept is used due to the available
linear literature model.

3.6.1 COMPARISONS OF CORRELATION BETWEEN SPT-(𝑵𝟏 )𝟔𝟎 VALUES AND EPMT

FOR SAND

The approximate correlation between SPT-N and EPMT proposed by Ohya et al. (1982) and
Bozbey (2010) are plotted on Figure 3.11(a) with the studied data. In this comparison, a
nonlinear power best fit line is plotted for the studied corrected and filtered data due to the
available nonlinear literature model. For the preliminary estimation of the EPMT for the sand, the
EPMT can be estimated from the SPT-(𝑁1 )60values using the following relationship.

EPMT (MPa) = 4.992 (𝑁1 )600.74 R2 = 0.81 [3.1]

The predicted EPMT values are calculated by using “Eq. 3.1” and the measured and predicted
EPMT values are also presented in Figure 3.11(a).

Another comparison of the data is performed with the research conducted by Briaud (1992), as
shown in Figure 3.11(b). In this comparison a linear correlation with a zero intercept has been
used due to the available linear literature model. For the preliminary estimation of the EPMT for
the sand, the EPMT can be estimated from the SPT-(𝑁1 )60 values using the following relationship.

65
EPMT (MPa) = 1.79 (𝑁1 )60 R2 =0.79 [3.2]

The predicted EPMT values are calculated by using “Eq. 3.2” and the measured and predicted
values are also presented in Figure 3.11(b).

Further, a comparison of the EPMT range is performed with the research by Briaud (1992). The
EPMT value for dense sand is greater than 22.5 MPa from Briaud (1992). In this study the EPMT
range is 26 – 149 MPa. It is higher than Briaud (1992)’s value. This is because the studied sand
in this project is dense to very dense with cobbles and boulders. In addition to that comparison,
the range of SPT-(𝑁1 )60value is performed with the CFEM (2006). The SPT-N value of dense
sand is greater than 50 in the CFEM (2006). The mean value of SPT-(𝑁1 )60 is 50 in this study. It
shows that the studied SPT-(𝑁1 )60 value of dense sand is a good agreement with CFEM (2006).

160
Measured -Epmt(MPa)
140 Bozbey -Epmt (MPa)
120 Ohya - Epmt(MPa)
Predicted - Epmt(MPa)
(MPa)

100
80
PMT

60
E

40
20
0
0 20 40 60 80 100
SPT-(N )60 value
1

(a) Non-linear relationship

66
 200
Measured -Epmt(MPa)
Briaud - Epmt(MPa)
150
Predicted - Epmt (MPa)
(MPa)

100
PMT
E

50

0
0 20 40 60 80 100
SPT-(N ) value
1 60

(b) Linear relationship

Figure 3.11 Comparison of correlation between SPT-(𝑁1 )60 and EPMT for sand

3.6.2 COMPARISON OF CORRELATION BETWEEN SPT-(𝑵𝟏 )𝟔𝟎 VALUES AND PL

FOR SAND

An approximate correlation between SPT-N and PL proposed by Bozbey (2010) is plotted on

Figure 3.12(a) with this investigation. In this comparison, a nonlinear power best fit line is
plotted for the studied corrected and filtered data due to the nonlinearity nature of the literature
model. For the preliminary estimation of the PL for the sand, the PL can be estimated from the
SPT-(𝑁1 )60 values using the following relationship.

PL (MPa) = 0.223 (𝑁1 )600.86 R2 = 0.73 [3.3]

The predicted PL values are calculated by using “Eq. 3.3” and the measured and predicted values
are also presented in Figure 3.12(a).

67
Another comparison of the corrected and filtered data is performed with the results from Briaud
(1992) and is plotted on Figure 3.12(b). In this comparison, a linear correlation with a zero
intercept has been used due to the linearity nature of the literature model. For the preliminary
estimation of the PL for the sand, PL can be estimated from the SPT-(𝑁1 )60 value using the
following relationship.

PL (MPa) = 0.126 ((𝑁1 )60 R2 =0.71 [3.4]

The predicted PL values are calculated by using “Eq. 3.4” and the measured and predicted values
are also presented in Figure 3.12(b).

Further, a comparison of range of PL is performed with Briaud (1992). The PL value for dense
sand is greater than 2.5 MPa from Briaud (1992). In this study the PL range is from 2.42 MPa –
13.32 MPa. It shows that the lower limit of the range is closer to Briaud’s (1992) value, but the
upper limit of the range is much higher than that reported by Briaud (1992).

12
Bozbey - PL (MPa)
10 Measured - PL (MPa)
Predicted - PL (MPa)
8
P (MPa)

6
L

0
0 20 40 60 80 100
SPT-(N ) value
1 60

(a) Non-linear power best fit

68
 50
Briaud - PL (MPa)
40 Predicted - PL (MPa)
Measured - PL (MPa)
P (MPa)

30
L

20

10

0
0 20 40 60 80 100
SPT-(N ) value
1 60

(b) Linear relationship

Figure 3.12 Comparison of correlation between SPT-(𝑁1 )60 and PL for sand

3.6.3 COMPARISONS OF CORRELATION BETWEEN SPT-(𝑵𝟏 )𝟔𝟎 VALUES WITH

BOTH EPMT AND PL FOR GLACIAL TILLS

The comparison of the corrected and filtered data is performed for glacial till with Yagiz (2008)
as shown in Figure 3.13. In this comparison, a linear correlation with an intercept has been used
due to the linearity nature of the literature model. For the preliminary estimation of the EPMT and
PL for the glacial till, EPMT and PL can be estimated from the SPT-(𝑁1 )60value using the
following relationships.

EPMT (MPa) = 1.492 (𝑁1 )60 + 1.635 R2 = 0.83 [3.5]

PL (MPa) = 0.094 (𝑁1 )60 + 0.211 R2 = 0.84 [3.6]

69
The predicted EPMT and PL values using “Eq. 3.5” and “Eq. 3.6”, and the measured EPMT and PL
values, are plotted in Figure 3.13(a) and (b) respectively. The comparison shows that measured
EPMT and PL are higher than the literature value. The reason for this is that the Toronto area
glacial tills deposit consists of cobbles and boulders (Ng et al. (2011)). The literature model
equation shows Ncor.. The Ncor means, either the SPT-N is corrected for whole factors which are
mentioned in the CFEM (2006) or only some factors. There are still uncertainties in this regard.

160
Measured - Epmt (MPa)
140
Yagiz - Epmt (MPa)
120
Predicted - Epmt (MPa)
(MPa)

100
80
PMT
E

60
40
20
0
0 20 40 60 80 100
SPT-(N ) value
1 60

Figure 3.13(a) Comparison of correlation between SPT-(𝑁1 )60vs EPMT for linear with intercept
relationship for glacial tills

70
 12
Measured - PL (MPa)
10 Yagiz - PL(MPa)
Predicted - PL (MPa)
8
P (MPa)

6
L

0
0 20 40 60 80 100
SPT-(N ) value
1 60

Figure 3.13 (b) Comparison of correlation between SPT-(𝑁1 )60 vs PL for linear with intercept
relationship for glacial tills

3.6.4 COMPARISONS OF CORRELATION BETWEEN SPT-(𝑵𝟏 )𝟔𝟎 VALUES AND EPMT

/PL RATIO FOR SAND

The range of EPMT /PL ratio for sand is between 9 and 20, with an average of 14 and standard
deviation of 4. The EPMT /PL ratio is compared to the studied by Bozbey (2010). The range in
Bozbey’s study is from 7 to 12. It shows that lower trend value differs with 2 and upper trend
value differs with 8. The studied mean value (14) is closer to the literature upper limit (12). The
main reason of the difference is the compactness of the sand. The compactness of sand may vary
from very loose, loose, medium, dense to very dense due to the inconsistencies. The CFEM
(2006) state that the ration of EPMT /PL for loose silty sand is 5 and sand and gravel is 7. The EPMT
/PL ratio in this study is higher than the literature value due to the presence of gravel and cobbles
in the sand. This was mentioned in the grain size analysis report which was conducted with the
selected sample from the proposed ECLRT project.

71
Figure 3.14 (a) represents the variation of the pressuremeter modulus with pressure limit. It is
clear that the EPMT varies linearly with the pressure limit with a relationship of EPMT = 14PL (R2
=0.83). Further Figure 3.14 (b) represents the variation of the EPMT/PL with SPT-(𝑁1 )60 values. It
2
shows that the linear relationship with an intercept gives better correlation coefficient (R =
0.62) compare to other relationships. The correlation equation between EPMT/PL ratio with SPT-
(𝑁1 )60 is given by EPMT/PL = 4.964 + 0.192(𝑁1 )60 .

200
Epmt(MPa)

150
(MPa)

100
PMT
E

50

0
0 2 4 6 8 10 12 14
P (MPa)
L

Figure 3.14 (a) Correlation between EPMT vs PL for sand

72
 22
Epmt/PL
20

18
L
/P

16
PMT
E

14

12

10

8
20 30 40 50 60 70 80
SPT-(N ) value
1 60

Figure 3.14 (b) Correlation between EPMT/PL vs SPT-(𝑁1 )60 for sand

3.6.5 COMPARISON OF CORRELATION BETWEEN SPT-(N1)60 VALUES AND EPMT

/PL RATIO FOR GLACIAL TILL

In this study, the EPMT /PL ratio is performed for glacial till in all stations which are mentioned in
section 3.2. The range of EPMT /PL ratio is between 5 and 26, with an average of 14 and standard
deviation of 4. The comparison of EPMT /PL ratio is performed with CFEM (2006) which shows
differences with the studied range. It is stated in the literature, the many commonly used
correlations in the geotechnical practice to estimate the geotechnical parameters from the in-situ
tests, contains a certain amount of inaccuracy. The reasons for this result can easily be related to
quality of the in-situ and laboratory tests. Since the database of this study is mainly comprised of
contracted construction projects, the quality of the site explorations and testing of the soils are
questionable parameters for this type of research. In addition, there is also a more important
reason that affects the obtained results which is the heterogeneous nature of the soil.

73
Figure 3.15 (a) represents the variation of the PMT modulus with pressure limit for glacial till. It
is clear that the EPMT varies linearly with the pressure limit with a relationship of EPMT = 14 PL
(R2 = 0.77). Further, Figure 3.15 (b) represents the variation of EPMT/PL with SPT-(𝑁1 )60values.
It shows that the linear relationship with an intercept gives better correlation coefficient (R2 =
0.77). The correlation equation between EPMT/PL ratio with SPT-(𝑁1 )60 is given by EPMT/PL =
9.304 + 0.140(𝑁1 )60.

200
Epmt (MPa)

150
(MPa)

100
PMT
E

50

0
0 2 4 6 8 10 12 14
P (MPa)
L

Figure 3.15 (a) Correlation between EPMT vs PL for glacial till

74
 20
Epmt/PL
18

16
L
/P
PMT

14
E

12

10

8
0 10 20 30 40 50 60 70
SPT-(N ) value
1 60

Figure 3.15 (b) Correlation between SPT-(𝑁1 )60 vs EPMT/PL for glacial till

3.7 SUMMARY

In this chapter, the study is performed based on an intensive site investigation program
conducted for the Eglinton Crosstown LRT Project, in the city of Toronto. Data is collected from
in-situ tests such as SPT and PMT and analyzed statistically. In this chapter, an attempt is made
to develop new relationships between corrected and filtered SPT-N values with both EPMT and PL
for various types of glacial tills. The ranges of SPT-N, EPMT and PL values are suggested in Table
3.5. The developed new correlation equations are summarized in Table 3.6. The EPMT /PL ratios
also are summarized in Table 3.7. In addition to that the comparison is made with available
literatures. There are good agreements with literature values in some extents. The accuracy of the
evaluated correlations can be increased by more carefully performed and well controlled in-situ
testing, borehole sampling and laboratory testing. In this way, some of the uncertainties can be
reduced and the reliability of the correlations would be enhanced.

75
Table 3.5 Summary of ranges of SPT-N, EPMT and PL for corrected and filtered data
Cohesive glacial tills
Soil type Silty clay Silty clay till Clayey silt till All soil
SPT- (𝑁)60 6 -62 3-31 4-67 3-67
EPMT (MPa) 11- 150 4-36 16-131 4-150
PL (MPa) 1.25-5.56 0.41-5.63 1.00 - 6.00 0.41-6.00
Cohesionless glacial tills
Soil type Sand Silt Sandy silt Silty sand Sandy silt till All soil
SPT- (𝑁1 )60 13-97 4-98 4-91 25-76 16-80 4-98
EPMT (MPa) 26-149 19-140 28-78 39-96 18-134 18-149
PL (MPa) 2.42-13.32 3.17-9.08 1.33-9.03 1.42-13.55 6.0-8.04 1.33-13.55

Table 3.6 Summary of correlation equations for EPMT and PL

Cohesive glacial tills
Soil type Silty clay Silty clay till Clayey silt till All soil
EPMT (MPa) 1.70N (𝑁)60 1.37 (𝑁)60 1.61 (𝑁)60 1.54 (𝑁)60
PL (MPa) 0.104 (𝑁)60 0.092 (𝑁)60 0.104 (𝑁)60 0.103 (𝑁)60
Cohesionless glacial tills
Soil type Sand Silt Sandy silt Silty sand All soil
EPMT (MPa) 1.79 (𝑁1 )60 1.58 (𝑁1 )60 1.68 (𝑁1 )60 1.47 (𝑁1 )60 1.54 (𝑁1 )60
PL (MPa) O.126 (𝑁1 )60 0.141(𝑁1 )60 0.124 (𝑁1 )60 0.158 (𝑁1 )60 0.139 (𝑁1 )60

Table 3.7 Summary of EPMT /PL ratio for corrected and filtered data
Soil type EPMT /PL Ratio
Sand 9-20
Glacial till 5-26

76
CHAPTER 4: FINITE ELEMENT SIMULATIONS OF PMTS

4.1 INTRODUCTION

The main aim of this chapter is to back calculate the PMT modulus (EPMT) by using the finite
element method (FEM) software, Plaxis 2D. For this purpose Mohr–Coulomb (M-C) model is
used. The modelling methods and procedures of using Plaxis 2D are discussed in Section 4.2.
The description of the M–C model is discussed in Section 4.3. The Section 4.4 described the 2D
- PMT modeling and verification. In this section the PMT model is verified by existing literature
model (Levasseur et al. (2009)). There is a good agreement between both models. In Section 4.5,
the sensitivity study is performed to investigate the influence of mesh coarseness and boundary
conditions. In Section 4.6, the case study is performed based on extensive PMTs are conducted
in the Mount Dennis (MD) Station in the ECLRT project in Toronto. In this section, the PMT
modulus (EPMT) is calculated for different types of glacial tills according to the borehole # MD
101 with assumed values of Young’s modulus. Then the PMT modulus (EPMT) is correlated with
the Young’s modulus (E) for various types of glacial tills. The Menard’s rheological factor (α) is
made for each type of glacial tills. Finally the results are summarized in Section 4.7.

4.2 FINITE ELEMENT METHOD

FEM plays a major role in the geotechnical engineering practice because it allows modeling
complicated nonlinear soil behavior through constitutive models. It will handle complex
problems where analytical solutions are nearly impossible. Nowadays, the FEM has become very
popular in the geotechnical engineering design. Even though, before the FEM can be used in
design, the accuracy of any proposed solution technique must be proved (Owen and Hinton
(1980)). In the FEM, the study object is divided into a number of finite elements, and the
interaction between each one of these elements is analyzed for various geometrics with different
boundary conditions and interfaces. It can predict the stresses, deformations and pore pressures
of a specified soil profile.

77
4.2.1 CONSTITUTIVE BEHAVIOUR

The constitutive behavior is the stress-strain behavior. The 3D constitutive equations (Eqs. 4.1 –
4.3) can be used to calculate the stresses in a soil mass with neglecting the inertia effects and all
body forces except the self-weight ɣ in x (vertical) direction (Timoshenko and Goodier (1951)).
Equilibrium equations (Eqs. 4.1 – 4.3) are in terms of total stresses that must satisfy the
boundary conditions. Fig. 4.1 shows the stresses on a typical element. Compressive stresses are
considered as positive.

 X  YX  ZX
    0 [4.1]
X Y Z
 XY  Y  ZY
  0 [4.2]
X Y Z
 XZ  YZ  Z
  0 [4.3]
X Y Z

Figure 4.1 Stresses on a typical element

In the equilibrium and compatibility conditions, the constitutive behavior of a soil can be
expressed mathematically as Eqs.4.4 or Eqs.4.5.

  D [4.4]

78
 X   D11 D12 D13 D14 D15 D16   X 
    
 Y   D21 D22 D23 D24 D25 D26   Y 
 Z   D31 D32 D33 D34 D35 D36   Z 
    [4.5]
 XY   D41 D42 D43 D44 D45 D46   XY 
 XZ   D51 D52 D53 D54 D55 D56   XZ 
    
 ZY   D61 D62 D63 D64 D65 D66   ZY 

For a linear elastic material, the [D] matrix is written as Eqs.4.6.

[4.6]
Where
E- Modulus of elasticity
υ- Poisson’s ratio

However, the soil behavior is usually non-linear. Therefore, increments of stress and strain can
be more realistically related using the constitutive equation as indicated in Eq. 4.4 (Potts and
Zdravkovic (2001)). It is also realistic for the [D] matrix to depend on the current and past
history. The constitutive behavior can be stated by means of total or effective stresses. If it is
needed to specify the constitutive behavior in terms of effective stress, the principle of effective
stress can be used to obtain total stresses required for use with equilibrium equations (Eqs.4.7 –
4.10)
   '  f [4.7]

 '  D' [4.8]

79
  
 f  D f  [4.9]

Therefore,
 
  D'  D f  [4.10]

Where
 - Total stress
’ – Effective stress
f - Pore fluid pressure
D′- Constitutive relationship relating the change in effective stress to the change in strain
Df - Constitutive relationship relating the change in pore fluid pressure to the change in strain

Plain strain condition

The constitutive relationship Eq 4.5 can be reduced to Eq 4.11 as follows for plain strain
conditions.

[4.11]

Axisymmetric condition

In the case of axisymmetric problems, it is usual to carry out analyses using cylindrical
coordinates as shown in Figure 4.2, r (radial direction), z (vertical direction) and 
(circumference direction). Due to the symmetry, there is no displacement in the  direction and
the displacement in the r and z directions are independent of  and therefore the strains reduce to
Eq 4.12 (Timoshenko and Goodier (1951)).

80
Fig 4.2 Example of axisymmetric co-ordinate axis

[4.12]
Where u and v are the displacements in the r and z direction respectively.
Stresses in the axisymmetric element are shown in Figure 4.3 and the [D] matrix is similar to
plain strain situation as shown in the Eq 4.11.

Fig 4.3 Stresses in the axisymmetric element

4.2.2 STEPS INVOLVED IN THE FEM

The following steps are involved in the FEM (Potts and Zdravkovic (2001)).

Step 1: Element discretization

In this process, the geometry of the problem is modelled by an assembly of small regions termed
as finite elements, which have nodes defined on the element boundaries, or within the element.

81
Step 2: Primary variable approximation

Primary variables such as displacements, stresses, etc., must be selected. The rules with regard to
how these variables have to vary over a finite element are established. Nodal values are used to
express the variations. Displacements are usually adopted as a primary variable in geotechnical
engineering.

Step 3: Element equations

The elemental equations (Eq. 4.13) are derived using an appropriate variational principle (e.g.,
the minimum potential energy).

K E d E   RE  [4.13]

Where
[KE ] - Element stiffness matrix
{ΔdE}- Vector of incremental element nodal displacements
{ΔRE}-Vector of incremental element nodal forces

Step 4: Global equations

The element equations are combined to form global equations (Eq. 4.14).

KG dG   RG  [4.14]

Where
[KG]- Global stiffness matrix
{ΔdG}-Vector of all incremental global nodal displacements
{ΔRG}- Vector of all incremental global nodal forces

82
Step 5: Boundary conditions

The global equations are modified by formulating boundary conditions. Loadings such as line
and point loads, pressures, body forces, etc. affect {ΔRG}, and the displacements affect {ΔdG}.

Step 6: Solve the global equations

The displacements {ΔdG} at all the nodes can be obtained by solving the global equations. These
nodal displacements are used to evaluate stresses and strains.

4.3 MOHR-COULOMB MATERIAL MODEL

In this research, the Mohr-Coulomb (M-C) elastoplastic constitutive model is used as a material
model for soils. The basic parameters used in this model with their standard units are listed
below.

E : Young’s modulus [𝑘𝑁/𝑚2 ]

υ : Poisson’s ratio [−]
c : Cohesion [𝑘𝑁/𝑚2 ]
 : Frictionangle [°]
 : Dilatancy angle [°]

As shown in Fig. 4.4, the Mohr circles of stress at failure are obtained by plotting the results of
the laboratory tests in term of effective stresses. The tangent line to the failure circles from
several tests, performed with different initial effective stresses, is called the Coulomb failure
criterion (Eqs. 4.15 – 4.17).

 f  c' 'nf tan ' [4.15]

Where
τf -Shear stress on the failure plane

83
 ϭ'nf -Normal effective stress on the failure plane
c' -Cohesion
 ' -Angle of shearing resistance

Figure 4.4 Mohr’s circles of effective stress (Ports and Zdravkovic (2001))

The M-C failure criterion is defined as:

1 ' 3 '  2c' cos  '1 ' 3 'sin  ' [4.16]

1 '   v ' , and  3 '   h '

Therefore, the yield function is given below

F ({1 '},{k})  1 ' 3 '2c' cos  '(1 ' 3 ' ) sin  ' [4.17]

This equation can be more conveniently written in terms of stress invariants p‫׳‬, J, and Ɵ (Eqs.
4.18 – 4.19).

 c'  
F ({ '1 },{k})  J     p' g ( )  0 [4.18]
 tan '  

84
 sin  '
g    [4.19]
 sin  sin  ' 
cos    
 3 

Figure 4.5 Mohr-Coulomb yield surfaces in principal stress space (Ports and Zdravkovic (2001))

As shown in Figure 4.5, an irregular hexagonal cone is plotted by the yield function in principal
effective stress space.

4.4 2D PMT MODELLING AND VERIFICATION

4.4.1 2D PMT MODELLING

For the 2D PMT modeling, the FEM software, PLAXIS 2D, is used. PLAXIS name was derived
from PLasticity AXISymmetry, a computer program developed to solve the cone penetrometer
problem by Pieter Vermeer and De borst. According to (Burd (1999)), the initiation of this
program was held at Delft University of Technology Netherland by Pieter Vermeer in 1974.
Earlier version of PLAXIS was in DOS interface. In 1998, the first PLAXIS 2D for Windows
was released. The new versions and modifications were carried out for the analysis of soil
behavior for geotechnical engineers. In this study Plaxis 2D version 2012 is used.

85
4.4.2 VERIFICATION OF THE MODEL

The 2D FEM model is validated by using a published case study on PMT, which was performed
by Levasseur et al. (2009) for Hostun sand. The geometry of the model is shown in Figure 4.6.
The PMT test depth is 3 m. In this analysis, an axisymmetric FEM is used which is the same as
the way used by Levasseur et al. (2009). In this model, the soil is represented by an M-C model
whose parameters are shown in Table 4.1. To validate the model same soil parameters are used
with same soil model as M-C by using Plaxis 2D in this verification study. The geometry of the
model has shown in Figure 4.7. The deformed mesh diagram and the zoomed view of the
diagram are shown in Figure 4.8 and Figure 4.9 respectively. The horizontal displacement
diagram and the zoomed view of the diagram are also shown in Figure 4.10 to Figure 4.11
𝛥𝑣
individually. The pressure (P) vs volumetric strain ( 𝑉 ) curve is plotted with the curve from the

literature. There is a good agreement between the two curves, as shown in Figure 4.12.

Table 4.1 Parameters used in the M-C model for dense Hostun sand
Parameter Levasseur et al. (2009) value
Shear modulus (Gref) (𝑘𝑃𝑎) 22250
Poisson’s coefficient (υ) 0.25
Cohesion (c) (𝑘𝑃𝑎) 0
Friction angle (φ) ( 0 ) 35
Dilatancy angle (ψ) ( 0 ) 5
Initial stress field coefficient (K0) 0.4265

86
 Figure 4.6 2D axisymmetric model and associated mesh (Levasseur et al. (2009))

d/2=2.5cm

5m

Pressure L=40cm

D/2 =3 m

Figure 4.7 Geometry of the PMT model

87
Figure 4.8 Deformed mesh diagram

88
Figure 4.9 Zoomed view of the deformed mesh diagram

89
Figure 4.10 Horizontal displacement diagram

90
Figure 4.11 Zoomed view of the horizontal displacement diagram

91
 1200

1000
Pressure(kPa)

800

600

400
Levasseur et al.(2009)
200 Current study model

0
0 0.2 0.4 0.6 0.8 1 1.2 1.4
Volumetric strain
𝛥𝑣
Figure 4.12 Pressure (p) vs volumetric strain ( 𝑉 ) curves for verification of the model

4.5 SENSITIVITY STUDY

This study is performed to investigate the influence of mesh coarseness, horizontal and vertical
boundary conditions in the PMT model. Each case is elaborated below.

4.5.1 MESH COARSENESS

A mesh is a composition of finite elements that can be created in mesh mode in Plaxis 2D. In
Plaxis, the mesh coarseness provides a significant influence on the calculation results. The model
is implemented with five types of mesh coarseness such as very coarse, coarse, medium, fine and
very fine mesh.

In order to evaluate the mesh influence in the Plaxis model, the Levasseur et al. (2009) is used as
a bench mark problem to develop the horizontal displacements at the midpoint of the probe for
different types of mesh coarseness with some applied pressures which are shown in Table 4.2.

92
The difference in the horizontal displacement is about 1.27 times higher when the mesh changes
from “very course” into “very fine”. It shows that the very fine mesh gives more accurate
results compare to other mesh coarseness, but it consumes more time during the simulation
process. Due to that, fine mesh coarseness is adopted in this study. Mean time the differences
between very fine to fine mesh coarseness is very small, nearly 0.85% different from very fine to
fine.

Table 4.2 Comparison of horizontal displacement related to mesh coarseness

Pressure Horizontal displacement x 10-3 (m) for mesh coarseness
(kPa) Very fine Fine Medium Course Very course
500 3.53 3.50 3.49 3.37 2.78
1000 10.98 10.90 10.85 10.83 8.78
1500 21.90 21.41 20.67 20.78 16.81
2000 37.18 35.76 33.87 33.63 26.42

4.5.2 HORIZONTAL BOUNDARY CONDITIONS

The influence of horizontal distance from probe is investigated in this study. The right side
boundary from the probe is not clearly stated in the literature. In order to evaluate the horizontal
distance’s influence in the Plaxis model, the Levasseur et al. (2009) is used as a bench mark
problem to develop the pressure vs volumetric strain curves for different distances from the
center of probe. The developed volumetric strain curves are shown in Figure 4.13 with the
Levasseur et al. (2009) curve. The values of the volumetric strains with relation to the horizontal
distances are shown in Table 4.3 with the Levasseur et al. (2009) value. Concluded from this
study, the horizontal distances have no significant influence on the volumetric strain curves other
than the 1 m distances.

93
 1200

1000
Pressure (kPa)

800
1m
600 2m
3m
400 4m
5m
200 10 m
Levassuer et al. (2009)
0
0 0.2 0.4 0.6 0.8 1 1.2 1.4
Volumetric strain
𝛥𝑣
Figure 4.13 Pressure (p) vs volumetric strain ( 𝑉 ) curves for horizontal boundary conditions

Table 4.3 The values of volumetric strain related to horizontal distances

Horizontal boundary condition (m) 𝛥𝑣
Volumetric strain ( ) 𝑉

1 1.0183
2 1.2620
3 1.1943
4 1.2290
5 1.1894
10 1.1785
Levassuer et al. (2009) 1.2000

4.5.3 VERTICAL BOUNDARY CONDITIONS

The influence of vertical distance below the bottom of the probe is investigated in this study. The
vertical distance from bottom of the probe is not clearly stated in the literature. In order to

94
evaluate the vertical distance’s influence in the Plaxis model, the Levasseur et al. (2009) is used
as a bench mark problem to develop the pressure vs volumetric strain curves for different
distances from the bottom of the probe. The developed volumetric strain curves are shown in
Figure 4.14 with the Levasseur et al. (2009) curve. The values of the volumetric strain with
relation to the vertical boundary conditions are shown in Table 4.4 with the Levasseur et al.
(2009) value. Concluded from this study, the vertical boundaries have no significant influence on
the volumetric strain curves other than the 0 m distance.

1200

1000
Pressure (kPa)

800 0m
1m
600 2m
3m
400 4m
5m
200 10 m
Levassuer et al. (2009)
0
0 0.2 0.4 0.6 0.8 1 1.2 1.4
Volumetric strain
𝛥𝑣
Figure 4.14 Pressure (p) vs volumetric strain ( 𝑉 ) curves for vertical boundary conditions

Table 4.4 The values of volumetric strain related to vertical boundary conditions
Vertical distance below the bottom of 𝛥𝑣
Volumetric strain ( )
𝑉
the probe (m)
0 0.9414
1 1.1756
2 1.2013
3 1.2033

95
 4 1.1776
5 1.1737
10 1.1815
Levassuer et al.(2009) 1.2000

4.6 CASE STUDY AT MOUNT DENNIS STATION

The main goal of this study is to back calculate the Young’s modulus (E) of different types of
glacial tills from PMT results using FEM. It was made from an extensive research on PMT
results conducted in MD station in the ECLRT project in Toronto. It was analyzed from eleven
(11) PMT results. The PM tests depth varies from 3.8 m to 35 m. It was concluded, with TTC
geotechnical standard (2014), most of the PMT results were from sand to sandy silt from 3 m to
21.3 m and clayey silt till deposited interbedded between silty clay in the depth which varies
from 24.5 m to 35 m. These types of materials are very heterogeneous and mixture of gravel,
sand, silt and clay size particle in varying proportions (Ng and Xue (2011)). The water table is
observed a depth of 4.6 m below the ground surface. The processes of back calculating the E for
different types of glacial tills are very complex and arduous task. Therefore in this study, back
calculating EPMT for different types of glacial tills with knowing values of E with other soil
parameters (c, , ψ and υ) are keeping constant. The EPMT values are computed from the quasi-
linear portion of the pressure vs radial strain curves. The EPMT is correlated with E value for
various types of glacial tills. Then the linear correlation equations between EPMT and E are
established for different types of glacial tills. The E values are predicted to the field measured
EPMT using the established correlation equations. The predicted E values are used as an input
values in the simulation and again the EPMT values are calculated from the quasi-linear portion of
the pressure vs radial strain curves. The calculated EPMT values have good agreement with field
measured EPMT. The Menard “” factors are developed for various types of glacial tills.

96
4.6.1 FINITE ELEMENT ANALYSES

The FEM model geometry is created according to soil profile in the MD Station borehole
number 101C3. The borehole report and PMT results are attached in Appendix 4.1 and 4.2
respectively. The soil profile has many layers which are shown in Figure 4.15. The width(x axis)
and depth (y axis) of the soil profiles are pre-defined in the model tab on the project properties
window. The limit of the soil contour is defined as the xmin = 0, xmax = 40 m and ymin = -40 m,
ymax = 0. The top boundary of the soil layer is at y = 0 at grade level and the bottom boundary of
the soil layer is y = -40 m at bed rock. Once the soil layers are drawn, the soil properties can be
assigned according to values shown in Table 4.5. These parameters are grasped from ECLRT
geoengineering factual data reports for different type’s glacial tills. (cohesionless glacial tills
such as sand and sandy silt, cohesive glacial tills such as silty clay and clayey silt till). In
addition, a small amount of cohesion (c = 0.1 kPa) is assigned for sand to prevent soil failure
upon unloading, which the soil may experience near the borehole wall during drilling or pre-
boring (Sedran et al. (2013)). A small value of cohesion (c) is adopted to avoid complication
while performing the simulation (Plaxis (2012)).

97
Table 4.5 Summary of soil parameters used in the FEM analysis
Soil type Depth Plaxis 2D (MCM )
(m) γUnsaturated γSaturated c  Ψ υ Initial E
(kg/m3) (kg/m3) (kN/m2) (0) (0) (kN/m2)
Fill 0.41 14 16 15 0 0 0.25 10000
Sand 3.8 17 21 0.1 41 11 0.33 25000
Sand 6.0 19 22 0.1 42 12 0.33 25000
Sandy silt 9.4 17 20.4 10 43 13 0.33 25000
Sandy silt 13.9 17 20.4 10 39 9 0.33 25000
Sandy silt 15.2 17 20.4 10 42.5 12.5 0.33 25000
Sandy silt 18.3 17 20.4 10 39 9 0.33 25000
Sand 21.3 19 22 0.1 45 15 0.33 25000
Silty clay 24.5 17 20.4 100 32 2 0.33 20000
Clayey silt till 27.3 17 21.8 50 35 5 0.33 20000
Clayey silt till 30.4 17 22.1 50 35 5 0.33 20000
Silty clay 35.0 17 20.4 100 32 2 0.33 20000
Bentonite 12 14 10 0 0 0.10 6000

The soil is modelled with the M-C model since its limited number of input parameters and its
popularity in the practice. Due to the granular nature of the soils such as sand and sandy silt, all
calculations are made in drained condition. For the silty clay and clayey silt till soils, all
calculations are made in undrained (A) condition according to Plaxis 2D material model (2012).
The analysis is performed as an axisymmetric and the mesh elements are 15 nods triangles. The
standard fixity boundary condition is applied for the soil profile. As a result Plaxis will
automatically generate full fixity at the base of the geometry and roller boundaries at the vertical
sides (Ux =0; Uy = free). The soil is free on the vertical walls of drilling and vertical movement is
possible on the two vertical borders of the soil profile (Houari and Abdeldjalil (2015)). The
boundary conditions of the MD Station soil profile are shown in Figure 4.15.

The PMT geometry is discretized using a 2D axisymmetric configuration for PMT probe with a
length (460mm)-to-diameter (76mm) ratio of 6.05, typical of the Roctest NX-sized PMT probe.

98
The recommended length to diameter ratio is 6 or more (Briaud (1992)). Positive interface is
introduced on the probe and vertical surface of the PMT borehole. A “Type B” loading is applied
on the probe as shown in Figure 4.15 at the test depth during the simulation stages. This loading
is applied radially on a length equal to the length of the probe, in downhole (Husein (2001)). The
loading condition is shown in Figure 4.15.

Once the geometry modelling process is complete, calculations are proceeded which consists of
the generation of meshes and definition of the construction stages. The defined geometry has to
be divided into finite elements in order to perform a FEM calculation. A mesh is a composition
of finite elements that can be created in mesh mode in Plaxis 2D. In Plaxis, the mesh coarseness
provides a significant influence on the calculation results. The model is implemented with five
types of mesh coarseness such as very coarse, coarse, medium, fine and very fine mesh.

At the end of the analysis is performed in the section 4.5.1, the fine mesh density is selected due
to its accuracy and speed of calculations. A fine mesh is used for the Plaxis 2D models analysis
(Khanal (2013)). In addition to that extra geometry lines are created around the probe to locally
generate a finer mesh. A typical FE mesh of MD Station is shown in Figure 4.16.

The PMT is simulated in the following three stages.

(i) Generation of in-situ initial stress condition by imposed by K0 value as shown in
Figure 4.17.
(ii) Borehole drilling and filled the borehole with mud(bentonite)
(iii) Applied pressure at the probe borehole interface incrementally.

During the above procedures the probe’s volume increases due to the pressure applied to the
probe and therefore the soil around it will deform. Then the horizontal displacement is recorded
from Plaxis output to calculate the radial strain. The pressure vs radial strain curve is plotted for
each Young’s modulus (E) in each depth to calculate the PMT modulus (EPMT). In this
calculation a special attention is paid to that the two slopes of the experimental and numerical
curves in the elastic phase should be similar.

99
 Fill
0.41m

Depth 6 m
Sand
Pressure

Sandy silt 8.69m

Sand 10.21m
11.73m

Sandy silt
40 m

20.88m
Sand
23.93m
Silty clay
26.97m

Clayey silt till

31.55m

Silty clay

40.0m
Bed rock
40 m
Fihure 4.15 10 Soil profile at Mount Dennis Station according to borehole MD101-PMT and test
@ 6.0m depth

100
Fihure 4.16 Typical FE mesh for numerical simulation at MD Station

Figure 4.17 Water table diagram for soil profile at MD Station @ test depth 6m after generate the
in-situ initial stress condition.

101
4.6.2 2D FINITE ELEMENT RESULTS AND ANALYSIS

The numerical simulation is performed at different depths from 3.8 m to 35 m according to MD

Station borehole MD101-PMT. A range of simulations are completed for different values of
Young’s modulus in each depth. The pressures are applied on the probe incrementally during the
simulation. The displacements are measured at the mid-point of the probe for each pressure
increments. The results can be viewed in the output mode and the most notable results are the
deformed mesh, total displacement, lateral (horizontal) displacement and cross section of the
lateral displacement (Ux).

The typical deformed mesh with the total displacement diagram at 6 m depth is shown in Figure
4.18. The zoomed view of the deformed mesh diagram is shown in Figure 4.19. The x direction
(horizontal) displacement at the mid-point of the probe after FEM analysis is shown in Figure
4.20. The zoomed view of the horizontal displacement at the mid-point of the probe is shown in
Figure 4.21. The horizontal displacement shaded diagram and the zoomed view of the shaded
diagram are shown in Figure 4.22 and Figure 4.23 respectively. A typical cross section of
horizontal displacement diagram is shown in Figure 4.24. The zoomed view of the cross section
of horizontal displacement diagram is shown in Figure 4.25.

102
Figure 4.18 Typical deformed meshes @ 6.0 m depth at MD Station

103
Figure 4.19 Zoomed views of the deformed meshes diagram@ 6.0 m depth at MD station

104
Figure 4.20 Typical horizontal displacement arrow diagram @ 6.0m depth at MD Station

105
Figure 4.21 Zoomed view of the horizontal displacement arrow diagram @ 6.0m depth at MD
Station

106
Figure 4.22 Typical horizontal displacement shaded diagram @ 6.0m depth at MD Station

107
Figure 4.23 Zoomed view of horizontal displacement shaded diagram @ 6.0 m depth at MD
Station

108
Figure 4.24 Typical horizontal displacement cross section @ 6.0 m depth at MD Station

Figure 4.25 Zoomed view of horizontal displacement cross section (along the probe) @ 6.0 m
depth at MD Station

109
4.6.3 COMPARISON OF PRESSURE VS RADIAL STRAIN CURVES FROM PMT AND
PLAXIS

It is instructive to compare the results of PMT with those obtained from our numerical results.
Figure 4.26 shows the typical pressure vs radial strain graphs for field PMT measurement and
simulated PMT curves at 6.0 m depth with different values of Young’s modulus. It can be seen
that the measured PMT curves and the numerical curves obtained using the proposed numerical
model are more or less similar for each depth that was investigated. During the simulation, it is
seen that the curves obtained from simulations are not best fit with the field curves. There should
be a many reasons for that, those are listed below:

(i) The diameter of the probe has an impact on the quality of the test. That is a diameter
of drilling bit should be equal to the diameter of the probe.
(ii) Rotation should be slow to minimize enlargement of borehole.
(iii) Mud circulation should be slow to minimize erosion.
(iv) Borehole walls left behind the bit may be disturbed.
(v) The Poisson’s ratio was taken as 0.33 for whole soil profile but not at all. For
saturated soils Poisson’s ratio vary from 0.33 to 0.45. The Menard PM modulus
means Poisson’s ratio is 0.33. In this project whole calculations were performed for
Poisson’s ratio 0.33.
(vi) The soil mass was assumed to be continuum, uniform and isotropic.
(vii) Special training is required for drillers to prepare a good PMT borehole as drilling for
PMT.

In addition to the reasons above the field curve initially starts horizontally then increased, but
numerical simulation curves increased vertically. The reason is that in the field PMT the volume
is increased then the pressure was measured, but in the simulation the pressure is applied then the
deformation is measured. Due to that, in the initial the field PMT curve is going horizontally
until it touches the borehole wall and after it has touched the borehole wall, it will increase. But
in the simulation, the pressure vs radial curve is going vertically until it touches the borehole
wall and after it has touched the wall, it will increase as well.

110
For each simulation, back-calculation is done for the values of EPMT by using Equation 2.8 in the
section 2.2.2.3, in Chapter 2, for each Young modulus and tabulated in Table 4.6. In this
calculation a special attention is paid to that the two slopes of the experimental and numerical
curves in the elastic phase should be similar. The portion of the curves that are used for the
calculation is shown by an arrow in Figure 4.26. Then the calculated PMT modulus (EPMT) vs
Young modulus (E) graphs that are plotted for various depths are shown in Figure 4.27 for
cohesionless glacial tills and Figure 4.28 for cohesive glacial tills respectively. The correlation
equations between the modulus above with their coefficients are also tabulated in Table 4.6. The
Menard “α” factors are calculated at each depth for different types of glacial tills at MD Station
are also shown in Table 4.6. This study shows a very good agreement with Menard “α” factors.

Figure 4.26 Pressure vs radial strain graphs at 6m depth for MD Station

111
Table 4.6 Linear correlation equations for glacial tills at MD Station
Depth Soil type Young’s PMT modulus Correlation Correlation
(m) modulus (E) (EPMT) (Mpa) equation coefficient 𝐸𝑃𝑀𝑇
α= 𝐸
(MPa) (R2 )
3.8 Sand 25 7.95 EPMT = 0.34E 1.0 0.32
50 18.77 0.38
75 25.91 0.35
100 34.68 0.35
125 43.26 0.35
150 50.24 0.34
6 Sand 25 4 EPMT = 0.15E 1.0 0.16
50 8 0.16
75 12 0.16
100 15 0.15
125 19 0.15
150 22 0.15
9.4 Sandy 25 7.4 EPMT = 0.28E 1.0 0.30
silt 50 14.4 0.29
75 21.4 0.29
100 28.4 0.28
125 35.2 0.28
150 42.2 0.28
13.9 Sandy 25 9.44 EPMT = 0.35E 1.0 0.38
silt 50 19.32 0.39
75 29.43 0.39
100 36.18 0.36
125 43.16 0.35
150 50.85 0.34
15.2 Sandy 25 8.22 EPMT = 0.31E 1.0 0.33
silt 50 15.84 0.32

112
 75 23.70 0.32
100 31.44 0.31
125 39.01 0.31
150 46.74 0.31
18.3 Sandy 25 9.03 EPMT = 0.35E 1.0 0.36
silt 50 17.78 0.36
75 26.77 0.36
100 35.47 0.36
125 43.90 0.35
150 52.61 0.35
21.3 Sand 25 10 EPMT = 0.39E 1.0 0.40
50 20 0.40
75 29 0.39
100 39 0.39
125 49 0.39
150 58 0.39
24.5 Silty 20 9.06 EPMT = 0.43E 1.0 0.45
clay 40 17.80 0.45
60 26.36 0.44
80 34.51 0.43
100 43.04 0.43
27.3 Clayey 20 7.12 EPMT = 0.34E 1.0 0.36
silt till 40 13.76 0.34
60 20.23 0.34
80 27.12 0.34
100 33.32 0.33
30.4 Clayey 20 5.13 EPMT = 0.23E 1.0 0.26
silt till 40 9.64 0.24
60 14.16 0.24
80 18.80 0.24

113
 100 23.10 0.23
35 Silty 20 8.93 EPMT = 0.43E 1.0 0.45
clay 40 17.30 0.43
60 25.76 0.43
80 34.21 0.43
100 42.70 0.43

60
Epmt @3.8m for Sand
50 Epmt @ 6m for Sand
Epmt @ 9.4m for Sandy silt
40 Epmt @ 13.9m for Sandy silt
(MPa)

Epmt @ 15.2m for Sandy silt

30 Epmt @ 18.3m for Sandy silt
Epmt @ 21.3m for Sand
PMT
E

20

10

0
0 20 40 60 80 100 120 140 160
E(MPa)

Fig 4.27 Linear relationships for EPMT vs E for cohesionless glacial tills at MD Station

114
 50
Epmt @ 24.5m for Silty clay
Epmt @ 27.3m for Clayey silt till
40
Epmt @ 30.4m for Clayey silt till
(MPa)

Epmt @ 35m for Silty clay

30
PMT

20
E

10

0
0 20 40 60 80 100 120
E(MPa)

Fig 4.28 Linear relationships for EPMT vs E for cohesive glacial tills at MD Station

In order to compare with Sedran et al. (2013), the linear relationship with intercept correlation
equations are developed between PMT and Young modulus. The Sedran et al. (2013) suggested
that, if a relation between E and E PMT exists, it would be similar to E = a + b EPMT. In this study,
the same format of the equation is achieved, and is tabulated in Table 4.7.

Table 4.7 Linear with intercept correlation equations for glacial tills at MD Station
Depth(m) Soil type Correlation equation R2
3.8 Sand E = 2.97 EPMT - 1.99 1.0
6.0 Sand E = 6.85 EPMT – 4.96 1.0
9.4 Sandy silt E = 3.60 EPMT – 1.83 1.0
13.9 Sandy silt E = 3.07 EPMT – 8.79 1.0
15.2 Sandy silt E = 3.24 EPMT - 1.63 1.0
18.3 Sandy silt E = 2.87 EPMT – 1.24 1.0
21.3 Sand E = 2.62 EPMT – 1.29 1.0
24.5 Silty clay E = 2.36 EPMT – 1.75 1.0

115
 27.3 Clayey silt till E = 3.04 EPMT – 1.77 1.0
30.4 Clayey silt till E = 4.44 EPMT – 2.82 1.0
35 Silty clay E = 2.37 EPMT – 1.07 1.0

4.6.4 VALIDATE THE RESULTS

To validate the results the developed correlation equations are used to predict the Young’s
modulus for the field PMT modulus (EPMT) which is conducted in the MD Station in the ECLRT
project. Then these predicted Young’s modulus are used as input parameters in the simulation
model in the Plaxis 2D. These simulations are carried out at each depth according to MD Station
borehole #101 soil profiles. Then PMT modules (EPMT) are calculated by using the Equation 2.8
in the Chapter 2(Section 2.2.2.3) from the linear portion of the pressure vs radial strain curves.
The calculated PMT modules (EPMT) are same as the field values. In addition, the E values are
calculated analytically by using Pasturel’s formula which is shown in the Chapter 2 (Section
2.3.8). Analytical E values seem to be similar to the predicted values for cohesionless glacial soil
but not the same for cohesive glacial soil. Analytical and predicted E values are both tabulated in
Table 4.8. The predicted E values for sand vary from 75 to 172 MPa. According to Bowles
(1996), E value for dense sand is 50 to 81 MPa. But E values vary for loose glacial tills from 10
to 150 MPa and dense glacial tills from 150 to 720 MPa. From this study the predicted E values
vary from 53 to 234 MPa. The studied values are within the Bowles’ (1996) range.

Table 4.8 Summary of predicted and analytical E and calculated EPMT at MD Station
Depth Soil type Field Predicted Calculated EPMT Analytical E 𝐸𝑃𝑀𝑇
α= 𝐸
(m) PMT EPMT E (MPa) from the FEM from Pasturel’s
(MPa) Simulation (MPa) formula (MPa)
3.8 Sand 0.2 0.60 0.24 1.05 0..34
6.0 Sand 25.8 172 25.90 200 0.15
9.4 Sandy silt 40.8 146 41.06 144 0.28
13.9 Sandy silt 38.6 110 39.03 104 0.35

116
15.2 Sandy silt 27.8 90 28.30 52 0.31
18.3 Sandy silt 28.9 83 29.50 23 0.35
21.3 Sand 29.1 75 29.20 28 0.39
24.5 Silty clay 36.2 84 36.20 8 0.43
27.3 Clayey silt till 46.2 136 45.30 3 0.34
30.4 Clayey silt till 53.9 234 53.30 3 0.23
35.0 Silty clay 22.7 53 22.50 5 0.43

4.7 SUMMARY

In this chapter, the PMT modulus (EPMT) is back calculated for glacial tills at MD Station in the
ECLRT project in Toronto. The PMT is investigated using FEM analysis. The FEM analysis is
performed with Mohr-Coulomb models which is linear elastic perfectly plastic constitutive
model. This model requires five parameters (E, c, , ψ and υ). These parameters are used from
ECLRT geoengineering factual data reports to simulate the model for each types of soil in the
MD Station. The site contains glacial tills which consist of cohesionless glacial till such as sand
and sandy silt from 3.8m to 21.3m and cohesive glacial tills such as silty clay and clayey silt till
from 21.3m to 35m.

First the model is created by using Plaxis 2D then validated by using one of the literature model
𝛥𝑣
Levasseur et al. (2009). The simulation is done to get the pressure (p) vs volumetric strain ( 𝑉 )

curve which show that curve is well best fit to the literature curve. Further to evaluate the mesh
dependency, the same model is used. The fine mesh coarseness gave exactly the same curve
which is compared with the literature model Levasseur et al. (2009).

Then the PMTs are analyzed numerically using Plaxis 2D. After the simulations are executed,
pressure (p) vs radial strain curves are obtained to calculate the PMT modulus (EPMT) for
different values of Young’s modulus in each depth according to MD borehole 101-PMT soil
profile. The correlation equations are developed between PMT (EPMT) and Young’s (E) modulus

117
in each depth for different types of glacial tills. Then the Young’s modulus (E) is predicted by
using the correlation equation for the field PMT modulus (EPMT). The predicted E values for the
glacial till vary from 53 to 243 MPa. These values are very good agreements with Bowles’
(1992) ranges. The Bowles (1992) suggested E values for glacial tills that vary from 10 to 720
MPa.

Then the Menard “α” factors are suggested for different types of glacial tills at MD Station.
According to Menard, the “α” factors are between 0 and 1. In this study, these factors are
retrieved by the results of the numerical simulations of the PMT. There is a good agreement with
the Menard “α” factors. The summary of the EPMT, E and Menard “α” factors for glacial tills at
MD Station are shown in Table 4.9.

Further the linear relationship with intercept correlation equations are developed between PMT
and Young modulus. There is a very good agreement with Sedran et al. (2013) due to the similar
equation format of E = a + b EPMT.

Table 4.9 Summary of the EPMT, E and Menard “α” factors for MD Station
Depth(m) Soil types PMT modulus (EPMT) Young’s modulus (E) α = 𝐸𝑃𝑀𝑇
𝐸
(MPa) (MPa)
3.8 Sand 0.20 0.60 0.34
6.0 Sand 25.8 172 0.15
9.4 Sandy silt 40.8 146 0.28
13.9 Sandy silt 38.6 110 0.35
15.2 Sandy silt 27.8 90 0.31
18.3 Sandy silt 28.9 83 0.35
21.3 Sand 29.1 75 0.39
24.5 Silty clay 36.2 84 0.43
27.3 Clayey silt till 46.2 136 0.34
30.4 Clayey silt till 53.9 234 0.23
35.0 Silty clay 22.7 53 0.43

118
CHAPTER 5: CONCLUSIONS AND RECOMMENDATIONS FOR
FUTURE RESEARCH

5.1 MAIN CONCLUSIONS

This study is performed based on a comprehensive geotechnical investigation program for a light
rail transit (LRT) project in the City of Toronto. The following main conclusions are made from
this investigation.

(1) The field SPT-N values are corrected and the ratio of corrected SPT-N to field measured
(𝑁1 )60
SPT-N which is ( ) is recommended for glacial tills.
𝑁𝐹

(2) The ranges of SPT-N, EPMT and PL with all data and corrected and filtered data format for
glacial tills are suggested.

(3) The statistical correlation equations between SPT-N values with PMT parameters such as
PM modulus (EPMT) and pressure limit (PL) with whole data as well as the corrected and
filtered data format for glacial tills are developed.

(4) The EPMT/PL ratios for sand and glacial till are also recommended.

(5) The correlation equations between PM modulus with Young’s modulus for glacial tills
are also suggested.

(6) The Menard “α” factors for glacial tills are also recommended.

119
5.2 RECOMMENDATIONS FOR FUTURE RESEARCH

The following recommendations are made for future research:

(1) There are many possible applications to correct the field measured SPT-N. Since there is
not any general agreement on these application of correction of field measured SPT-N. In
contrast to heavy criticisms about the SPT-N correction, there is strong need and
necessity to carry out a suitable research on correction methods which are more suitable
for local conditions.

(2) This study also proves once more, the correlation between in-situ test parameters still
involves a large amount of uncertainties as presented by many researchers and they
should not be preferred unless there is not any other data available. Therefore it is
recommended to carry out another study in the glacial tills in these areas in the future in
order to provide a good relationship.

(3) The correlation between SPT-N and PMT has been investigated by many researchers in
the past. They mentioned that the scatter in the data is considerably large which makes
the correlation essentially useless in design. Due to that more theoretical study is needed
to develop a sound rationale to filter the data to minimize the scatters.

(4) The future study was recommended to predict the undrained shear strength from net limit
pressure of PMT and suggest the β factor for glacial tills. Then develop the correlation
equations between undrained shear strength and SPT-N for glacial tills.

(5) In the case of FEM simulation of PMT, the FEM provides efficient results. Even though
the program needs many input parameters and may be complicated to use. Due to that the
research is recommended, in order to develop a best curve fitting methodology to derive
the soil strength parameters quickly and more accurately.

120
(6) Model the PMT in glacial tills by using other constitutive models such as hardening soil
model for sand and soft soil creep model for silty clay. But it is more time consuming to
compare the results from different constitutive models.

121
REFERENCES

ASTM D 4719–00 (2000). Standard tests method for pre-bored pressuremeter testing in soils.
Annual book of ASTM standards, vol 04.08.

ASTM D 1586–11 (2014). Standard test method for standard penetration test (SPT) and split –
barrel sampling of soils. Annual book of ASTM standards.

Baguelin, F., Jezequel, J. F. and Shields, D.H. (1978). The Pressuremeter and foundation
engineering, Trans Tech. Publications, Clausthal, Germany. 617p.

Baker, C.L., Lahti, L.R., and Roumbanis, D.C. (1998). Urban Geology of Toronto and
surrounding area. Urban Geology of Canadian Cities. Edited by: P.F. Karrow, 42, 323-
352.

Bolton, M.D. and Whittle, R. W. (1999). A nonlinear elastic/perfectly plastic analysis for plane
strain undrained expansion tests. Geotechnique, 49(1), 133-141.

Boone, S. J., Shirlaw, J.N. et al. (1996). Boulder assessment report for TTC Sheppard Subway,
Golder associates report no: S-GIR3-R. Toronto.

Boone, S. J. and Westland, J. (2008). Geotechnical summary report for tunnel boring machine
procurement for the Toronto–York Spadina Subway extension (TYSSE), Golder
associates report no: 08-111-0039-R01.

Bowles, J.E. (1997). Foundation analysis and design. The McGraw–Hill Co., Inc., Singapore, 5th
edition.

Bozbey, I. and Togrol, E. (2010). Correlation of standard penetration test and pressuremeter data.
A case study from Istanbul, Turkey. Bull eng geol environ.

Braja, M. Das. (1990). Principles of Foundation Engineering, 5th ed., Cole engineering division.

Briaud, J.L. (1992). The pressuremeter, A. A. Balkema, Rotterdam, Netherlands.

Brinkgreve, R.B.J. and Vermeer, P.A. (1998). Plaxis. Finite element code for soil and rock
analysis, Balkema, Rotterdam, The Netherlands.

Burd, H. (1999). The history of PLAXIS. Beyond 2000 in computational geotechnics: 10 years of
PLAXIS international; Proceedings of the international symposium beyond 2000 in
computational geotechnics, Amsterdam, The Netherlands, 18-20 March 1999, Taylor &
Francis group.

Cambou, B. and Bahar, R. (1993). Utilisation de I’essai pressiometrique pour I’ identification de

parametres intrinseques du comportement dusol. Revue Franqaise de Geotechnique, 63,
39-50.

122
Campanella, R.G., Berzins, W.E. and Shields, D.H. (1979). A preliminary evaluation of Menard
pressuremeter, cone penetrometer and standard penetration test in the lower main land,
British Columbia. Soil mechanics series no: 40.

Canadian Geotechnical Society, (2006). Canadian Foundation Engineering Manual. 4th ed., The
Canadian Geotechnical Society Co & Bi Tech, publishers Ltd. Canada.

Cao, L., Peaker, S. and Ahmad, S. (2015). Pressuremeter tests in glacial tills in Toronto,
Symposium International ISP7/Pressio.

Clarke, B.G. (1995). Pressuremeter in Geotechnical Design. 1st ed.,Chapman & Hall, Glasgow.

Eglinton Cross Town (LRT), Geoengineering factual data report.

Fawaz, A., Hagechehade, F. and Farah, E. (2014). A study of the pressuremeter modulus and its
comparison to the elastic modulus of soil. Study of civil engineering and architecture
(SCEA).

Finn, P.S., Nisbet, R.M. and Hawkins, P.G. (1984). Guidance on pressuremeter, flat dilatometer
and cone penetration tests in sand, Geotechnique, vol.34, no:1, pp.81-97.

Gibson, R.E. and Anderson, W.F. (1961). In – situ measurement of soil properties with the
pressuremeter. Civil Eng.

Houari, O. and Abdeldjalil, Z. (2015). Numerical modeling of the test on slope pressuremeter.
International journal of emerging technology and advanced engineering. vol 5, issue5.

Jacques, M. (2007). Numerical validation of an elastoplastic formulation of the conventional

limit pressure measured with the pressuremeter test in cohesive soil. Journal of
Geotechnical and Geoenvironmental Engineering. ASCE. September.

Karrow, P.F. (1967). Pleistocene geology of the Scarborough area. Ontario department of mines,
Geological reports 46.

Karrow, P.F. and White, O.L. (1998). Urban geology of Canadian cities, Geological association
of Canada, GAC special paper 42.

Kovacs, W.D. and Salomone, L.A. (1982). SPT hammer energy measurements. American Society
of Civil Engineers, ASCE, Journal of the Geotechnical Engineering Division, vol.108,
GT4, pp.599-620.

Kovacs, W.D., Yokel, F.Y., Salomone, L.A and Holtz, R.D. (1984). Liquefaction potential and the
international SPT; Proceeding of the 8th world conference on earthquake engineering,
San Francisco, CA.

Kulhawy and Mayne (1990). Manual on estimating soil properties for foundation design, Electric
power research institute, Palo Alto, CA

123
Levasseur, S., Malecot, Y., Boulon, M. and Flavigny, E. (2009). Statistical inverse analysis based
on genetic algorithm and principal component analysis. Application to excavation
problems and pessuremeter tests. International journal for numerical and analytical
methods in geomechanics. 34: 471-491.

Levasseur, S., Malecot, Y., Boulon, M. and Flavigny, E. (2008). Statistical inverse analysis based
on genetic algorithm and principal component analysis. Methods and developments
using synthetic data. International journal for numerical and analytical methods in
geomechanics. 33: 1485-1511.

Manzari, M., Drevininkas, A., Olshansky, D. and Galaa, A. (2014). Behavioral modelling of
Toronto glacial soils and implementation in numerical modeling. 67th Canadian
Geotechnical Conference, Geo Regina.

Menard, L. (1965). Regle pour le Calcul de la Force Portante et du Tassement des Foundation en
Fonction des Resultats Pressiometriques, Proceedings 6th ICSMFE, Montreal, vol., pp.
295- 299.

Michel, G., Armando, A. and Antonio, G.C. (2008) Using a nonlinear constitutive law to
compare Menard PMT and PLT E-moduli.

Monnet, J. (2007). Numerical validation of an elastoplastic formulation of the conventional limit

pressure measured with the pressuremeter test in cohesive soil. Journal of geotechnical
and geoenvironmental engineering. September.

Monnet, J. (2012). Elasto-plastic analysis of the pressuremeter test in granular soil part 2.
Numerical study. European journal of environmental and civil engineering. vol 16, no.6
715 – 729.

Nahra, R. and Frank, R. (1986). Contributions numeriques et analytiques a I’ etude de la

consolidation autour du pressiometre. Research Rep. LPC No: 137, Laboratoire Central
des Ponts et Chaussees, Paris.

Ng, R., Xue, T., Wheeler, C. and Campo, D. (2011). The Eglinton cross town light rail transit.14
th
Pan-Am conference on Soil Mechanics and Geotechnical Engineering. 64th Canadian
Geotechnical conference.

Ohya, H., Kazama, E. and Negishi, Y. (1982). Reverse osmotic concentration of aqueous ethyl-
alcohol solutions. Analysis of data obtained with composite membranes (PEC), Kagaku
Kogaku Ronbunshu, 8 (2): 144-149.

Owen, D.R.J. and Hinton, E. (1980). Finite elements in plasticity. Theory and practice.
Pinerridge press limited, Swansea, U.K.

Phoon, K.K. and Kulhawy, F.H. (1999). Evaluation of geotechnical variability. Canadian geotech
J 36:625-639.

Plaxis 2D. (2012). Material Model Manual.

124
Plaxis 2D. (2012). Reference Manual.

Plaxis 2D. (2012). Tutorial Manual.

Potts, D. M. and Zdravkovic, L. (2001). Finite Element Analysis in Geotechnical Engineering.

Thomas Telford, London.

Prapaharan, S., Chameau, J. L. and Holtz, R.D.(1989). Effect of strain rate on undrained strength
derived from presssuremeter tests. Geotechnique, 39(4), 615-624.

Raquel, R. (2008). Characterization of Material Behavior by the pressuremeter test.

Robertson, P. K., Campanella, R. C. and Wightman, A.(1983). SPT-CPT correlations, American

society of civil engineers, ASCE, Journal of the geotechnical engineering division,
vol.109, GT11, pp. 1449-1459.

Schanz, T., Vermeer, P.A. and Bonnier, P.G. (2000). The hardening soil model. Formulation and
verification. Beyond 2000 in computational geotechnics. 10 years of Plaxis @ Balkema,
Rotterdam, ISBN 90 5809 040 X.

Schmertmann, J.H. and Palacios, A. (1979), Energy dynamics of SPT, American society of civil
engineers, ASCE, Journal of the geotechnical engineering division, vol. 105, GT8, pp.
909-926.

Sedran, G., Failmezger, R.A. and Drevininkas, A. (2013). Relationship between Menard EM and
Young’s E moduli for cohesionless soils. Proceeding of the 18th International
conference on soil mechanics and geotechnical engineering, Paris 2013.

Seed, H.B., Tokimatsu, K., Harder, L.F. and Chung, R.M. (1984). Influence of SPT procedures in
soil liquefaction resistance evaluations. Report no: UCB/EERC-84/15, Berkeley.
Reprinted in journal of geotechnical engineering, ASCE, vol. 111, no.12, pp 1425-1445.

Sharp, D.R. (1980). Quaternary geology of Toronto and surrounding area. Ontario geological
survey, Geological series preliminary map, p 2204.

Sharpe, D.R., Barnett, P. J. et al. (1999). Regional Geological Mapping of Oak Ridges Moraine –
Greater Toronto Area – Southern Ontario, in current research 1999– E, Geological
Survey of Canada, pp 123-136.

Silvestri, V. (2003). Assessment of self – boring pressuremeter tests in sensitive clay. Can.
Geotech. J., 40, 362-387.

Silvestri, V. (2004). Disturbance effects in pressuremeter tests in clay. Can. Geotech. J., 41, 738-
759.

Sivrikaya, O., Togrol, E., (2007), Turkiye de SPT N Degeri ile Dnce Daneli Zeminlerin Drenajsiz
Kayma Mukavemeti A rasindaki Dliskiler, Technical magazine of chamber of civil
engineers, pp 4229 – 4246, paper no: 279.

125
Skempton, A.W. (1986). Standard penetration test, procedures and effects in sands of
overburden, relative density, particle size, aging and over- consolidation. Geotechnique,
vol.36, no. 3, pp. 425-447.

Terzaghi, K. and Peck, R.B. (1967). Soil mechanics in enginering practice, second edition, John
Wiley and Sons, New York, 729p.

Thorburn, S. (1986). Field testing: standard penetration test, engineering geology special
publication, no: 2, Geological society.

Timoshenko, S and Goodier, J.N. (1951). Theory of elasticity. McGraw Hill, New York.

Toronto transit commission geotechnical standards. (2014). version 8

Yagiz, S., Akyol, E. and Sen, G. (2008). Relationship between the standard penetration test and
the pressuremeter test on sandy silty clays. A case study from Denizli. Bulletin of
engineering geology and the environment, 67(3), 405-410.

126
APPENDICES

127
APPENDIX 3.1

128
129
130
 APPENDIX 3.2
SPT-N CORRECTION

Sample calculation for SPT-N correction for Bathurst Station:

First correction is performed according to Cao et al. (2015) for field measured SPT-N for
penetration depth.

305 𝑁
𝑁𝐹 = 𝛥𝑠

Where 𝑁𝐹 - Corrected SPT-N value

N - Field recorded SPT-N value
Δs - Measured penetration depth in mm

Table A3.1 Summary of NC calculation

Depth(m) Soil types Field measured Penetration depth 305𝑁

NF = 𝛥𝑠
SPT-N value (ΔS) in mm
3.7 Sandy silt 43 305 305𝑥43
= = 43
305

7.1 Silty clay till 91 305 305𝑥91

= = 91
305

10.5 Silty clay till 98 200 305𝑥98

= = 150
200

13.6 Clayey silt till 68 305 305𝑥68

= = 68
305

16.4 Clayey silt till 50 100 305𝑥50

= = 153
100

19.8 Silty clay 50 125 305𝑥50

= = 122
125

22.8 Sand 50 125 305𝑥50

= = 122
122

25.6 Silty sand 68 150 305𝑥68

= = 138
150

131
Second correction is performed according to the CFEM (2006).
(𝑁1 )60 = CECNCRCBCS𝑁𝐹

(𝑁1 )60 = CN 𝑁60

𝑁60 = CE 𝑁𝐹

𝑃 0.5
CN = (𝜎1 )

𝐸𝑅𝑅
CE = 60

Where CE - Hammer energy correction factor

𝐸𝑅𝑅 – Rod energy ratio
CN - Overburden pressure correction factor
P - Atmospheric pressure
σ' - Effective overburden pressure
CR - Rod length correction factor
CB - Borehole diameter correction factor
CS - Sampler correction factor
NF - Corrected SPT-N value for penetration depth
N60 - SPT-N value corrected to 60% of theoretical free fall hammer energy
(𝑁1 )60 - SPT-N value correctd for both vertical effective stress and input energy

Calculation procedure for overburden pressure correction factor (CN):

Water table is located at 2.8 m below the grade level in this borehole. Density of water (γw) is
9.81 kN/m3 and Atmospheric pressure (P) 100Kpa. Dry density of sandy silt is 17kN/m3

132
Table A3.2 Summary of CN calculations
Depth(m) γ (kN/m3 ) σ' (kPa) 𝑃 0.5
CN = (𝜎′ )

3.7 21 2.8*17+(3.7-2.8)*(21-9.81) = 57.671 1.316804

7.1 23.6 57.671 + (7.1-3.7)(23.6-9.81) = 104.557 0.977965
10.5 22.8 104.557 + (10.5-7.1)(22.8-9.81) =148.723 0.819994
13.6 22.9 148.723 + (13.6-10.5)(22.9-9.81) = 189.302 0.726813
16.4 23 189.302 + (16.4-13.6)(23-9.81) = 226.234 0.664846
19.8 23.2 226.234 + (19.8-16.4)(23.2-9.81) = 271.76 0.606607
22.8 23.2 271.76 + (22.8-19.8)(23.2-9.81) = 311.93 0.566202
25.6 23.2 311.93 + (25.6-22.8)(23.2-9.81) = 349.422 0.534964

Table A3.3 Summary of correction factors and (𝑁1 )60 calculations

Depth(m) CE CN CR CB CS NF (𝑁1 )60
3.7 1 1.316804 0.7 1 1 43 39.63581
7.1 1 0.977965 0.95 1 1 91 84.5451
10.5 1 0.819994 1 1 1 150 122.5482
13.6 1 0.726813 1 1 1 68 49.42325
16.4 1 0.664846 1 1 1 153 101.3890
19.8 1 0.606607 1 1 1 122 74.00602
22.8 1 0.566202 1 1 1 122 69.07665
25.6 1 0.534964 1 1 1 138 73.96774

(𝑁1 )60
Table A3.4 Summary of the corrected SPT-N values and ratios of for Bathurst Station
𝑁𝐹

Depth (m) Soil types Field measured NF (𝑁1 )60 (𝑁1 )60
Ratio 𝑁𝐹
SPT-N
3.7 Sandy silt 43 43 40 0.9
7.1 Silty clay till 91 91 85 0.9

133
 10.5 Silty clay till 98 150 123 0.8
13.6 Clayey silt till 68 68 49 0.7
16.4 Clayey silt till 50 153 101 0.7
19.8 Silty clay 50 122 74 0.6
22.8 Sand 50 122 69 0.6
25.6 Silty sand 68 138 74 0.5

(1) Allen Station

(𝑁1 )60
Table A3.5 Summary of the corrected SPT-N values and ratios of for Allen Station from
𝑁𝐹

Borehole AL12-PMT
Depth (m) Soil types Field measured NF (𝑁1 )60 (𝑁1 )60
Ratio 𝑁𝐹
SPT-N
4.95 Clayey silt till 33 33 26 0.8
7.62 Clayey silt till 51 51 37 0.7
10.87 Silty clay 58 58 37 0.6
13.92 Silty clay 89/250mm 109 62 0.6
16.97 Silty clay 50/75mm 203 110 0.5
20.02 Sand 83/250mm 101 52 0.5
23.06 Sand 57/150mm 116 57 0.5
26.01 Silty sand 51/150mm 104 48 0.5
30.68 Sand 50/150mm 102 45 0.4

(𝑁1 )60
Table A3.6 Summary of the corrected SPT-N values and ratios of for Allen Station from
𝑁𝐹

Borehole AL20-PMT

Depth (m) Soil types Field measured NF (𝑁1 )60 (𝑁1 )60
Ratio 𝑁𝐹

134
 SPT-N
4.72 Clayey silt till 152 152 123 0.8
7.77 Silty clay till 41 41 29 0.7
10.82 Silty clay 49 49 31 0.6
13.36 Clayey silt till 55/150mm 112 67 0.6
14.63 Clayey silt till 72/150mm 146 85 0.6
17.09 Sand 88/10mm 2684 1476 0.6
19.71 Sand 104/225mm 141 74 0.5
22.81 Sand 150/150mm 305 152 0.5
25.98 Sand 50/75mm 203 97 0.5
30.02 Silty clay 84 84 38 0.5

(2) Avenue Station

(𝑁1 )60
Table A3.7 Summary of the corrected SPT-N values and ratios of for Avenue Station from
𝑁𝐹

Borehole MD101-PMT (i.e AV101A)

Depth (m) Soil types Field measured NF (𝑁1 )60 (𝑁1 )60
Ratio 𝑁𝐹
SPT-N
5.33 Silty clay till 37 37 31 0.8
8.31 Silty clay till 38 38 28 0.7
13.11 Silt 56 56 34 0.6
14.76 Silty clay 38 38 22 0.6
17.6 Sandy silt 50/100mm 153 80 0.5
20.6 Sand/Silty sand 50/100mm 153 73 0.5
23.93 Sand 50/130mm 117 54 0.5
26.9 Sand 50/280mm 55 24 0.4
30.02 Silty sand 50/130mm 117 50 0.4
32.72 Silty sand 50/100mm 153 63 0.4
34.82 Sandy silt/Silt 82 82 33 0.4

135
 37.44 Silty clay 47 47 19 0.4
40.31 Clayey silt till 50/150mm 102 39 0.4
43.23 Clayey silt till 50/100mm 153 57 0.4
47.93 Silty clay 91/230mm 121 44 0.4

(3) Bathurst Station

(𝑁1 )60
Table A3.8 Summary of the corrected SPT-N values and ratios of for Bathurst Station
𝑁𝐹

Borehole 103-PMT
Depth (m) Soil types Field measured NF (𝑁1 )60 (𝑁1 )60
Ratio 𝑁𝐹
SPT-N
3.7 Sandy silt 43 43 47 1.1
7.1 Silty clay till 91 91 92 1.0
10.5 Silty clay till 98/200mm 149 130 0.9
13.6 Clayey silt till 68 68 52 0.8
16.4 Clayey silt till 50/100mm 153 105 0.7
19.8 Silty clay 50/125mm 122 76 0.6
22.8 Sand 50/125mm 122 71 0.6
25.6 Silty sand 68/150mm 138 76 0.6

(4) Bayview Station

(𝑁1 )60
Table A3.9 Summary of the corrected SPT-N values and ratios of for Bayview Station
𝑁𝐹

Borehole BV100-PMT (i.e BV100B)

Depth (m) Soil types Field measured NF (𝑁1 )60 (𝑁1 )60
Ratio 𝑁𝐹
SPT-N

136
 7.16 Sandy silt 57 57 58 1.0
10.36 Silt 43 43 38 0.9
13.31 Sandy silt 58 58 46 0.8
16.36 Sand/Sandy silt 74 74 52 0.7
19.33 Sandy silt/Silty clay 75 75 49 0.7
22.86 Silty clay 36 36 22 0.6
25.78 Clayey silt till 58 58 33 0.6
28.58 Clayey silt till 85 85 46 0.5
31.8 Silt 70/150mm 142 72 0.5
34.7 Silt 78 78 38 0.5

(5) Bermondsey Station

(𝑁1 )60
Table A3.10 Summary of the corrected SPT-N values and ratios of for Bermondsey Station
𝑁𝐹

Borehole BE05-PMT
Depth (m) Soil types Field measured NF (𝑁1 )60 (𝑁1 )60
Ratio 𝑁𝐹
SPT-N
3.2 Silty clay till 20 20 17 0.9
5.97 Silty clay till 13 13 10 0.8
9.3 Silty clay till 15 15 10 0.7
11.89 Silty clay till 10 10 6 0.6
15.6 Silty clay till 08 08 4 0.5
18.62 Silty clay till 12 12 6 0.5
21.59 Silty clay till 13 13 6 0.5
24.41 Clayey silt till 15 15 7 0.5
27.69 Silty clay till 26 26 12 0.5
30.68 Silty clay 14 14 6 0.4
(6) Blackcreek Station

137
 (𝑁1 )60
Table A3.11 Summary of the corrected SPT-N values and ratios of for Blckcreek Station
𝑁𝐹

Borehole BH7 – PMT

Depth (m) Soil types Field measured NF (𝑁1 )60 (𝑁1 )60
Ratio 𝑁𝐹
SPT-N
6.05 Gravelly sand 7 7 9 1.3
8.69 Silty clay 14 14 15 1.0
11.84 Silty clay 9 9 8 0.9
15.09 Silty clay till 14 14 11 0.8
18.06 Silty clay 8 8 6 0.8
20.85 Clayey silt till 6 6 4 0.7
24.03 Clayey silt till 20 20 12 0.6
27.53 Clayey silt till 13 13 7 0.5
30.53 Silt 8 8 4 0.5
33.60 Sandy silt 8 8 4 0.5
36.17 Sandy silt till 34 34 16 0.5

(7) Birchmount Station

(𝑁1 )60
Table A3.12 Summary of the corrected SPT-N values and ratios of for Birchmount Station
𝑁𝐹

Borehole BM01 – PMT

Depth (m) Soil types Field measured NF (𝑁1 )60 (𝑁1 )60
Ratio 𝑁𝐹
SPT-N
3.78 Sandy silt till 46 46 38 0.8
6.55 Sandy silt till 86/275mm 95 79 0.8
8.99 Sandy silt till 53/150mm 108 80 0.7
11.94 Clayey silt till 50/100mm 153 106 0.7
15.06 Sandy silt till 51/150mm 104 66 0.6
18.14 Sandy silt till 54/150mm 110 64 0.6

138
 21.34 Sandy silt/Sand 72/150mm 146 80 0.6
24.41 Sandy silt 61/150mm 124 64 0.5
27.69 Silt 50/75mm 203 98 0.5

(8) Caledonia Station

(𝑁1 )60
Table A3.13 Summary of the corrected SPT-N values and ratios of for Caledonia Station
𝑁𝐹

Borehole CA203 – PMT

Depth (m) Soil types Field measured NF (𝑁1 )60 (𝑁1 )60
Ratio 𝑁𝐹
SPT-N
3.91 Sand and gravel fill 18 18 15 0.8
6.76 Clayey silt fill 4 4 3 0.8
9.37 Silty clay fill 10 10 7 0.7
12.14 Silty sand 38 38 26 0.7
15.34 Sand 27 27 17 0.6
18.44 Silty sand, silt to sandy 86 86 51 0.6
21.59 silt 67 67 38 0.6
24.64 46 46 25 0.5
27.71 52 52 27 0.5
30.68 Silty clay 31 31 15 0.5
33.68 25 25 12 0.5
36.75 20 20 9 0.5
39.88 40 40 18 0.5

(𝑁1 )60
Table A3.14 Summary of the corrected SPT-N values and ratios of for Caledonia Station
𝑁𝐹

Borehole CA205A – PMT

139
Depth (m) Soil types Field measured NF (𝑁1 )60 (𝑁1 )60
Ratio 𝑁𝐹
SPT-N
4.01 Silty clay fill 03 3 3 1.0
5.74 Silty clay fill 06 6 5 0.8
8.92 Silty clay fill 09 9 7 0.8
11.79 Silty sand 55 55 40 0.7
14.86 Silty sand 85/250mm 104 70 0.7
19.30 Sandy silt 50/100mm 153 91 0.6
21.82 Sandy silt 85 85 48 0.6
24.77 Silty sand 52 52 28 0.5
27.89 Silty sand 58 58 30 0.5
30.89 Silty clay 38 38 19 0.5
33.68 Silty clay 17 17 8 0.5
36.81 Silty clay 16 16 7 0.4

(9) Don Mills Station

(𝑁1 )60
Table A3.15 Summary of the corrected SPT-N values and ratios of for Don Mills Station
𝑁𝐹

Borehole DM06 – PMT

Depth (m) Soil types Field measured NF (𝑁1 )60 (𝑁1 )60
Ratio 𝑁𝐹
SPT-N
4.22 Silty clay till 50/100mm 153 174 1.1
7.92 Silty clay 42 42 40 1.0
9.42 Silty clay 35 35 31 0.9
11.91 Sand 66 66 56 0.9
14.99 Sand 54 54 41 0.8
17.58 Silty clay 27 27 19 0.7
21.01 Silty clay 32 32 21 0.7
24.44 Silty clay 28 28 17 0.6

140
 27.58 Silty clay 30 30 17 0.6
30.18 Silty clay 28 28 16 0.6

(10) Kennedy Station

(𝑁1 )60
Table A3.16 Summary of the corrected SPT-N values and ratios of for Kennedy Station
𝑁𝐹

Borehole BH7 – PMT

Depth (m) Soil types Field measured NF (𝑁1 )60 (𝑁1 )60
Ratio 𝑁𝐹
SPT-N
4.11 Sandy silt till Disturbed - - -
6.65 Sandy silt till 93/275mm 103 79 0.8
9.80 Silty sand till 46 46 29 0.6
12.07 Silty sand 127/200mm 194 116 0.6
15.52 Clayey silt 115/250mm 140 78 0.6
19.2 Silt 123/200mm 188 98 0.5
21.39 Silty clay 59 59 30 0.5
24.79 Silty sand 79 79 38 0.5
27.74 Sand 50/125mm 122 57 0.5
30.89 Silty sand 50/100mm 153 69 0.5

(11) Lesile Station

(𝑁1 )60
Table A3.17 Summary of the corrected SPT-N values and ratios of for Lesile Station
𝑁𝐹

Borehole LE02 – PMT

141
Depth (m) Soil types Field measured NF (𝑁1 )60 (𝑁1 )60
Ratio 𝑁𝐹
SPT-N
3.51 Silty clay (Fill) 9 9 8 0.9
6.40 Silty clay (Fill) 8 8 7 0.9
12.9 Silty clay 25 25 16 0.6
15.7 Silty clay 24 24 14 0.6
18.75 Silty clay 35 35 18 0.5
21.79 Silty clay 43 43 21 0.5
24.84 Silty sand 93/250mm 114 54 0.5
27.89 Sand 50/125mm 122 56 0.5
30.94 Sand 98/250mm 120 53 0.4

(12) Mount Dennis Station

(𝑁1 )60
Table A3.18 Summary of the corrected SPT-N values and ratios of for Mount Dennis
𝑁𝐹

Station Borehole MD101 – PMT (i.e MD 101C1)

Depth (m) Soil types Field measured NF (𝑁1 )60 (𝑁1 )60
Ratio 𝑁𝐹
SPT-N
3.8 Sand 33 33 27 0.8
6.0 Sand 35 35 30 0.9
9.4 Sand silt 43 43 35 0.8
13.9 Sand silt 25 25 18 0.7
15.2 Sand silt 39 39 28 0.7
18.3 Sand silt 26 26 17 0.7
21.3 Sand 51 51 31 0.6
24.5 Silty clay 12 12 7 0.6
27.3 Clayey silt till 29 29 16 0.6
30.4 Clayey silt till 45 45 23 0.5
35 Silty clay 38 38 19 0.5

142
(13) Victoria Station

(𝑁1 )60
Table A3.19 Summary of the corrected SPT-N values and ratios of for Victoria Station
𝑁𝐹

Borehole VP01 – PMT

Depth (m) Soil types Field measured NF (𝑁1 )60 (𝑁1 )60
Ratio 𝑁𝐹
SPT-N
4.37 Silty clay till 37 37 31 0.8
6.83 Clayey silt till 14 14 11 0.8
9.83 Clayey silt till 24 24 15 0.6
13.61 Sand 21 21 13 0.6
16.18 Silt 123 123 69 0.6
19.46 Silt 79/150mm 161 85 0.5
22.25 Silt 70 70 35 0.5
25.86 Silt 15 15 7 0.5
28.88 Silt 92 92 42 0.5
31.06 Silt 34 34 15 0.4
35 Silt 88 88 37 0.4

(14) Warden Station

(𝑁1 )60
Table A3.20 Summary of the corrected SPT-N values and ratios of for Warden Station
𝑁𝐹

Borehole WA04 – PMT

Depth (m) Soil types Field measured NF (𝑁1 )60 (𝑁1 )60
Ratio 𝑁𝐹
SPT-N
3.63 Silty clay till 29 29 22 0.8
5.82 Silty clay till 17 17 13 0.8
8.97 Gravelly sand 23 23 17 0.7

143
 11.71 Silty clay till 53/150mm 108 74 0.7
15.67 Sandy silt till 50/75mm 203 126 0.6
18.44 Silty sand 86/250mm 105 61 0.6
21.01 Sandy silt 50/100mm 153 85 0.6
24.08 Silt 50/100mm 153 80 0.5
27.03 Silty sand 50/75mm 203 102 0.5
30.3 Silty sand 50/125mm 122 58 0.5

(15) West Portal Station

(𝑁1 )60
Table A3.21 Summary of the corrected SPT-N values and ratios of for West Portal Station
𝑁𝐹

Borehole C3 – PMT
Depth (m) Soil types Field measured NF (𝑁1 )60 (𝑁1 )60
Ratio 𝑁𝐹
SPT-N
1.7 Sand to silty sand fill 28 28 32 1.1
3.84 Sand to silty sand fill 32 32 24 0.8
9.19 Sandy silt 06 06 4 0.7
13.77 Silty clay till 15 15 9 0.6
15.42 Silty clay till 05 05 3 0.6

(16) Wynford Station

(𝑁1 )60
Table A3.22 Summary of the corrected SPT-N values and ratios of for Wynford Station
𝑁𝐹

Borehole WY03 – PMT

144
Depth (m) Soil types Field measured NF (𝑁1 )60 (𝑁1 )60
Ratio 𝑁𝐹
SPT-N
3.68 Silty clay till 28 28 21 0.8
5.84 Silty clay till 23 23 17 0.7
8.79 Silty clay 10 10 7 0.7
12.14 Clayey silt till 32 32 19 0.6
14.71 Silty clay 83/275mm 92 52 0.6
18.14 Silty clay till 50/125mm 122 64 0.5
20.88 Silty clay till 50/250mm 61 31 0.5
24.38 Silty sand 50/100mm 153 73 0.5
27.64 Silty sand 50/100mm 153 70 0.5
30.2 Inferred sandy soil 50/100mm 153 67 0.4

145
APPENDIX 4.1

146
147
148
149
APPENDIX 4.2

150