INVESTIGATING COMPOSITE BEHAVIOR OF GEOSYNTHETIC- REINFORCED SOIL (GRS) MASS by Thang Quyet Pham B.S., Hanoi University of Civil Engineering, 1993 M.S., Hanoi University of Civil Engineering, 2001 A thesis submitted to the University of Colorado Denver In partial fulfillment of the requirements for the degree of Doctoral of Philosophy Civil Engineering 2009
378
Embed
Pham Thang Investigating Composite Behavior of Geosynthetic Reinforced Soil Mass Thesis
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
INVESTIGATING COMPOSITE BEHAVIOR OF GEOSYNTHETIC-
REINFORCED SOIL (GRS) MASS
by
Thang Quyet Pham
B.S., Hanoi University of Civil Engineering, 1993
M.S., Hanoi University of Civil Engineering, 2001
A thesis submitted to the
University of Colorado Denver
In partial fulfillment
of the requirements for the degree of
Doctoral of Philosophy
Civil Engineering
2009
Thang Quyet Pham (Ph.D., Civil Engineering)
“Investigating Composite Behavior of Geosynthetic-Reinforced Soil (GRS) Mass”
Thesis directed by Professor Jonathan T. H. Wu
ABSTRACT
A study was undertaken to investigate the composite behavior of a Geosynthetic
Reinforced Soil (GRS) mass. Many studies have been conducted on the behavior
of GRS structures; however, the interactive behavior between the soil and
geosynthetic reinforcement in a GRS mass has not been fully elucidated. Current
design methods consider the reinforcement in a GRS structure as “tiebacks” and
adopt a design concept the reinforcement strength, Tf, and reinforcement spacing,
Sv, have the same effects on the performance of a GRS structure. This has
encouraged the designers to use stronger reinforcement at larger spacing, as the
use of larger spacing will generally reduce time and effort in construction.
A series of large-size Generic Soil-Geosynthetic Composite (GSGC) tests were
designed and conducted in the course of this study to examine the behavior of
GRS mass under well-controlled conditions. The tests clearly demonstrated that
reinforcement spacing has a much stronger effect on the performance of GRS
mass than reinforcement strength. An analytical model was established to
describe the relative contribution of reinforcement strength and reinforcement
spacing. Based on the analytical model, equations for calculating the apparent
cohesion of a GRS composite, the ultimate load carrying capacity of a reinforced
DEDICATION
This thesis is dedicated to my loving parents, Lam Van Pham and Khoi
Thi Pham, who have continuously given me unlimited support in achieving all my
life goals.
ACKNOWLEDGMENTS
I would like to express my most sincere gratitude to my thesis advisor,
Professor Jonathan T.H. Wu, for his dedicated support and guidance throughout
the course of this study. His clear insight of the subject has made my study both a
great learning experience and a joy. I also wish to thank members of my thesis
committee, Professors Hon-Yim Ko, John McCartney, Brian Brady, and Ronald
Rorrer for their helpful comments.
A special thank-you is extended to Michael Adams of the Federal
Highway Administration for his enthusiastic assistance and expert technical
support of the GSGC tests. My gratitude also goes to Jane Li and Thomas Stabile
for their help with the GSGC tests. Without their help, I could not have
conducted five successful tests during my three-month stay at the Turner-Fairbank
Highway Research Center in McLean, Virginia.
I truly appreciate the help of a dear brother and a loyal partner, Dr. Sang
Ho Lee, a visiting professor from Kyungpook National University, South Korea,
who helped me with all my experiments, including those I conducted at the
Turner-Fairbank Highway Research Center.
Last but not least, I would like to thank my wife Thuy Vu and our three
young daughters for standing by me and for encouraging me every step of the
way.
I feel blessed to have all these nice people around me in the course of this
study. Without them, this thesis would not be a reality.
CONTENTS
Figures.......................................................................................................................... xi
Tables ...................................................................................................................... xviii
1.1 Problem Statement .............................................................................................1
1.2 Research Objectives ...........................................................................................4
1.3 Tasks of research ................................................................................................5
2. Literature Review...............................................................................................9
2.1 Mechanics of Reinforced Soil ............................................................................9
2.2 Composite Behavior of GRS Mass ..................................................................15
2.3 Compaction-Induced Stresses in an Unreinforced Soil Mass ..........................27
2.3.1 Lateral Earth Pressure Estimation by Rowe (1954) .........................................27
2.3.2 Stress Path Theory by Broms (1971) and Extension of Broms’ Work by Ingold (1979) ...................................................................................................31
2.3.3 Finite Element Analysis by Aggour and Brown (1974) ..................................36
2.3.4 Compaction-Induced Stress Models by Seed (1983) .......................................40
2.4 Compaction-Induced Stresses in a Reinforced Soil Mass ...............................62
2.4.1 Ehrlich and Mitchell (1994) .............................................................................62
2.4.2 Hatami and Bathurst (2006) .............................................................................65
2.4.3 Morrison et al. (2006) ......................................................................................66
2.5 Highlights on Compaction-Induced Stresses ...................................................68
3. Analytical Model for Calculating Lateral Displacement of a GRS Wall with Modular Block Facing .....................................................................................71
3.1 Review of Existing Methods for Estimating Maximum Wall Movement .......73
3.1.1 The FHWA Method (Christopher et al., 1989) ................................................74
3.1.2 The Geoservices Method (Giroud, 1989) ........................................................76
vii
3.1.3 The CTI Method (Wu, 1994) ...........................................................................77
3.1.4 The Jewell-Milligan Method ............................................................................78
3.2 Developing an Analytical Model for Calculating Lateral Movement and Connection Forces of a GRS Wall ...................................................................84
3.2.1 Lateral Movement of GRS Walls with Negligible Facing Rigidity ................85
3.2.2 Connection Forces for GRS walls with Modular Block Facing ......................87
3.2.3 Lateral Movement of GRS Walls with Modular Block Facing .......................92
3.3 Verification of Analytical Model .....................................................................93
3.3.1 Comparisons with the Jewell-Milligan Method for Lateral Wall Movement ..93
3.3.2 Comparisons of with Measured Data of Full-Scale Experiment by Hatami and Bathurst (2005 and 2006).................................................................................99
4.5.4 Preparation of Test Specimen for GSGC Tests .............................................143
4.6 Test Results ....................................................................................................163
4.6.1 Test 1-Unreinforced Soil ................................................................................163
4.6.2 Test 2-GSGC Test (T, Sv) ..............................................................................169
viii
4.6.3 Test 3-GSGC Test (2T, 2Sv) ..........................................................................187
4.6.4 Test 4-GSGC Test (T, 2Sv) ............................................................................199
4.6.5 Test 5-GSGC Test (unconfined with T, Sv) ...................................................211
4.7 Discussion of the Results ...............................................................................224
4.7.1 Effects of Geosynthetic Inclusion (Comparison between Tests 1 and 2) ......224
4.7.2 Relationship between Reinforcement Spacing and Reinforcement Strength (Comparison between Tests 2 and 3) .............................................................226
4.7.3 Effects of Reinforcement Spacing (Comparison between Tests 2 and 4) .....228
4.7.4 Effects of Reinforcement Strength (Comparison between Tests 3 and 4) .....230
4.7.5 Effects of Confining Pressure (Comparison between Tests 2 and 5) ............231
5. Analytical Models for Evaluating CIS, Composite Strength Properties of a GRS Composite, and Required Reinforcement Strength ...............................235
5.1 Evaluating CIS in a GRS Mass ......................................................................236
5.1.1 Conceptual Model for Simulation of Fill Compaction of a GRS Mass .........236
5.1.2 A Simplified Model to Simulate Fill Compaction of a GRS Mass ................237
5.1.3 Model Parameters of the Proposed Compaction Simulation Model ..............239
5.1.4 Simulation of Fill Compaction Operation ......................................................241
5.1.5 Estimation of K2,c ...........................................................................................246
5.2 Strength Properties of GRS Composite .........................................................250
5.2.2 Apparent Cohesion and Ultimate Pressure Carrying Capacity of a GRS Mass ...............................................................................................................257
5.3 Verification of the Analytical Model with Measured Data ...........................258
5.3.1 Comparison between the Analytical Model and GSGC Test Results............258
5.3.2 Comparison between the Analytical Model and Elton and Patawaran’s Test Data ...............................................................................................................261
5.3.3 Comparison of the Results between the Analytical Model and Finite Element Results ............................................................................................................266
5.4 Required Reinforcement Strength in Design .................................................268
ix
5.4.1 Proposed Model for Determining Reinforcement Force ...............................268
5.4.2 Comparison of Reinforcement Strength between the Analytical Model and Current Design Equation................................................................................270
5.4.3 Verification of the Analytical Model for Determining Reinforcement Strength ..........................................................................................................271
6 Finite Element Analyses ................................................................................275
6.1 Brief Description of Plaxis 8.2 .......................................................................275
6.2 Compaction-Induced Stress in a GRS Mass ..................................................278
6.3 Finite Element Simulation of the GSGC Tests ..............................................280
6.3.1 Simulation of GSGC Test 1 ...........................................................................287
6.3.2 Simulation of GSGC Test 2 ...........................................................................290
6.3.3 Simulation of GSGC Test 3 ...........................................................................295
6.4 FE Analysis of GSGC Test 2 under Different Confining Pressures and Dilation Angle of Soil-Geosynthetic Composites ..........................................300
6.5 Verification of Compaction-Induced Stress Model .......................................302
7. Summary, Conclusions and Recommendations .............................................308
1.1 Typical Cross-Section of a GRS Wall with Modular Block Facing ..................8
2.1 Concept of Apparent Cohesion due to the Presence of Reinforcement (Scholosser and Long, 1972) .........................................................................10
2.2 Concept of Apparent Confining Pressure due to the Presence of Reinforcement (Yang, 1972) ...........................................................................11
2.3 Strength Envelopes for Sand and Reinforced Sand (Mitchell and Villet, 1987) ................................................................................................................14
2.5 Reinforced Triaxial Test Specimen (Elton and Patawaran, 2005) ...................17
2.6 Stress-Strain Curves of Samples Reinforced at Spacing of 12 in. and 6 in. in Large-Size Unconfined Compression Tests (Elton and Patawaran, 2005) ................................................................................................................18
2.7 Mini Pier Experiments (Adams, 1997) ............................................................19
2.8 Stress-Strain Curve (Adams, et al., 2007)........................................................20
2.9 Test Set-up of Large Triaxial Tests with 1,100 mm High and 500 mm in Diameter (Ziegler, et al., 2008) ........................................................................21
2.10 Large-Size Triaxial Test Results (Ziegler, et al., 2008) ...................................22
2.11 Vertical Stress Distribution at 6-kN Vertical Load of the GRS Masses with and without Reinforcement (Ketchart and Wu, 2001) .....................................24
2.12 Horizontal Stress Distribution at 6-kN Vertical Load of the GRS Masses with and without Reinforcement (Ketchart and Wu, 2001) .............................25
2.13 Shear Stress Distribution at 6-kN Vertical Load the GRS Masses with and without Reinforcement (Ketchart and Wu, 2001) ...........................................26
2.14 Schematic Illustration of Rowe’s Theory (Rowe, 1954) .................................29
2.15 Results of the Two-Directional Direct Shear Tests (Rowe, 1954) ..................30
2.16 Hypothetical Stress Path during Compaction (Broms, 1971) ..........................32
2.17 Residual Lateral Earth Pressure Distribution (Broms, 1971) ..........................34
xi
2.18 Hypothetical Stress Path of Shallow and Deep Soil Elements (Broms, 1971) ................................................................................................................35
2.19 A Sample Problem Analyzed by Aggour and Brown (1974) ..........................39
2.10 The First-Cycle K0-Reloading Model (Seed, 1983) ........................................41
2.21 Suggested Relationship between sinφ’ and α (Seed, 1983) .............................42
2.23 K0-Unloading following Reloading (Seed, 1983) ............................................45
2.24 Unloading after Moderate Reloading (Seed, 1983) .........................................47
2.25 Basic Components of the Non-Linear K0-Loading/Unloading Model (Seed, 1983) ................................................................................................................49
2.26 Profile of against a Vertical Wall for a Single Drum Roller (Seed, 1983) ................................................................................................................50
',, pvchσΔ
2.27 Stress Path Associated with Placement and Compaction of a Typical Layer of Fill (Seed, 1983) ..........................................................................................51
2.28 Bi-Linear Approximation of Non-Linear K0-Unloading Model (Seed, 1983) ................................................................................................................52
2.29 Relationship between K2 and F in the Bi-Linear Unloading Model (Seed, 1983) ................................................................................................................52
2.30 Relationship between K3 and β 3 in the Bi-Linear Model (Seed, 1983) .........53
2.31 Basic Components of the Bi-Linear Model (Seed, 1983) ................................55
2.32 Compaction Loading/Unloading Cycles in the Bi-Linear Model (Seed, 1983) ................................................................................................................57
2.33 An Example Problem for Hand Calculation of Peak Vertical Compaction Profile (Seed, 1983) .........................................................................................60
2.34 Solution Results from the Bi-Linear Model and Non-Linear Model (Seed, 1983) ................................................................................................................61
2.35 Assumed Stress Path (Ehrlich and Mitchell, 1994) .........................................63
2.36 Compaction and Reinforcement Stiffness Typical Influence ..........................65
2.37 FE Model for FE Analysis (Morrison, et al., 2006) .........................................67
3.1 Basic Components of a GRS Wall with a Modular Block Facing ...................73
xii
3.2 Empirical Curve for Estimating Maximum Wall Movement during Construction in the FHWA Method (Christopher, et al., 1989) ......................75
3.3 Assumed Strain Distribution in the Geoservices Method ................................77
3.4 Stress Characteristics and Velocity Characteristics behind a Smooth Retaining Wall Rotating around the Toe (Jewell and Milligan, 1989) ............79
3.5 Major Zones of Reinforcement Forces in a GRS Wall and the Force Distribution along reinforcement with Ideal Length (Jewell and Milligan, 1989) ................................................................................................................80
3.6 Charts for Estimating Lateral Displacement of GRS Walls with the Ideal Layout (Jewell and Milligan, 1989) .................................................................83
3.7 Major Zones of Reinforcement Forces in a Reinforces Soil Wall (Jewell and Milligan, 1989) ..........................................................................................85
3.8 Forces acting on Two Facing Blocks at Depth zi .............................................88
3.9 Connection Forces in Reinforcement (q = 0) ...................................................91
3.10 Connection Forces in Reinforcement (q = 50) .................................................91
3.11 Comparison of Lateral Displacement Calculated by Jewell-Milligan Method and the Analytical Model, 0=bγ ......................................................95
3.12 Comparison of Lateral Displacement Calculated by Jewell-Milligan Method and the Analytical Model, 10=bγ ....................................................96
3.13 Comparison of Lateral Displacement Calculated by Jewell-Milligan Method and the Analytical Model, 20=bγ ....................................................97
3.14 Comparison of Lateral Displacement Calculated by Jewell-Milligan Method and the Analytical Model, 30=bγ ....................................................98
3.15 Configuration of a Full-Scale Experiment of a GRS Wall with Modular Block Facing (Hatami and Bathurst, 2005 and 2006)....................................100
3.16 Comparisons of Measured Lateral Displacement with Jewell-Milligan Method and the Analytical Model .................................................................102
4.1 Typical Geometric and Loading Conditions of a GRS Composite................107
4.2 Global Stress-Strain Curves for Soil-Geosynthetic Composites of Different Dimensions under a Confining Pressure of 0 kPa .........................................109
4.3 Global Volume Change Curves for Soil-Geosynthetic Composites of Different Dimensions under a Confining Pressure of 0 kPa ..........................110
xiii
4.4 Global Stress-Strain Curves for Soil-Geosynthetic Composites of Different Dimensions under a Confining Pressure of 30 kPa .......................................111
4.5 Global Volume Change Curves for Soil-Geosynthetic Composites of Different Dimensions under a Confining Pressure of 30 kPa ........................112
4.6 Global Stress-Strain Curves of the Unreinforced Soil under a Confining Pressure of 30 kPa ..........................................................................................114
4.7 Global Volume Change Curves of the Unreinforced Soil under a Confining Pressure of 30 kPa ..........................................................................................115
4.8 Specimen Dimensions for the GSGC Tests ...................................................116
4.9 Front View of the Test Setup .........................................................................117
4.10 Plan View of the Test Setup...........................................................................118
4.11 The Test Bin ...................................................................................................120
4.12 Grain Size Distribution of Backfill ................................................................124
4.13 Typical Triaxial Test Specimen before and after Test ...................................125
4.14 Triaxial Test Results ......................................................................................126
4.15 Mohr-Coulomb Failure Envelops of Backfill ................................................127
4.16 Uni-Axial Tension Test of Geotex 4x4 ..........................................................130
4.17 Load-Deformation Curves of the Geosynthetics ...........................................132
4.18 Locations of LVDTs and Digital Dial Indicator ............................................136
4.19 Strain Gauges on Geotext 4x4 Geotextile ......................................................138
4.20 Strain Gauges Mounted on Geotex 4x4 Geotextile .......................................139
4.21 Calibration Curve for Single-Sheet Geotex 4x4 ............................................141
4.22 Calibration Curve for Double-Sheet Geotex 4x4 ...........................................142
4.23 Applying Grease on Plexiglass Surfaces .......................................................145
4.25 Placement of the First Course of Facing Block .............................................147
4.26 Compaction of the First Lift of Backfill ........................................................148
4.27 Placement of Backfill for the Second Lift .....................................................149
4.28 Placement of a Reinforcement Sheet .............................................................150
xiv
4.29 Completion of Compaction of the Composite Mass and Leveling the Top Surface with 5 mm-thick Sand Layer ............................................................151
4.30 Completed Composite Mass with a Geotextile Sheet on the Top Surface ....152
4.31 Covering the Top Surface of the Composite Mass with a Sheet of Membrane ......................................................................................................153
4.32 Removing Facing Blocks and Trimming off Excess Geosynthetic Reinforcement ................................................................................................154
4.33 Insertion of the Strain Gauge Cables though Membrane Sheet .....................155
4.34 Vacuuming the Composite Mass with a Low Pressure .................................156
4.35 Sealing the Connection between Cable and Membrane with Epoxy to Prevent Air Leaks ..........................................................................................157
4.36 Checking Air Leaks under Vacuuming ..........................................................158
4.37 The LVDTs on an Open Side of Test Specimen ............................................159
4.38 Location of Selected Points to Trace Internal Movements of Tests ..............160
4.39 Soil Dry Unit Weight Results during Specimen Preparation of Five GSGC Tests ...............................................................................................................161
4.40 Soil Mass at Failure of Test 1 ........................................................................164
4.41 Results of Test 1-Unreinforced Soil Mass .....................................................165
4.42 Lateral Displacements on the Open Face of Test 1 .......................................166
4.43 Internal Displacements of Test 1 ...................................................................167
4.44 Composite Mass at Failure of Test 2 .............................................................170
4.45 Close-up of Shear Bands at Failure of Area A in Figure 4.44 .......................171
4.46 Failure Planes of the Composite Mass after Testing in Test 2 ......................172
4.47 Results of Test 2-Reinforced Soil Mass .........................................................173
4.48 Lateral Displacements on the Open Face of Test 2 .......................................174
4.49 Internal Displacements of Test 2 ...................................................................177
4.50 Locations of Strain Gauges Geosynthetic Sheets in Test 2 ...........................178
4.51 Reinforcement Strain Distribution of the Composite Mass in Test 2 ............179
4.52 Aerial View of the Reinforcement Sheets Exhumed from the Composite Mass after Test 2 ............................................................................................184
4.53 Location of Rupture Lines of Reinforcement in Test 2 .................................185
xv
4.54 Composite Mass after Testing of Test 3 ........................................................189
4.55 Global Stress-Strain Relationship of Test 3 ...................................................190
4.56 Lateral Displacements on the Open Face of Test 3 .......................................191
4.57 Internal Displacements of Test 3 ...................................................................192
4.58 Locations of Strain Gauges Geosynthetic Sheets in Test 3 ...........................193
4.59 Reinforcement Strain Distribution of the Composite Mass in Test 3 ............194
4.60 Aerial View of the Reinforcement Sheets Exhumed from the Composite Mass after Test 3 ............................................................................................196
4.61 Location of Rupture Lines of Reinforcement in Test 3 .................................197
4.62 Failure Planes of the Composite Mass after Testing in Test 4 ......................201
4.63 Global Stress-Strain Relationship of Test 4 ...................................................202
4.64 Lateral Displacements on the Open Face of Test 4 .......................................203
4.65 Internal Displacements of Test 4 ...................................................................204
4.66 Locations of Strain Gauges Geosynthetic Sheets in Test 4 ...........................205
4.67 Reinforcement Strain Distribution of the Composite Mass in Test 4 ............206
4.68 Aerial View of the Reinforcement Sheets Exhumed from the Composite Mass after Test 4 ............................................................................................208
4.69 Location of Rupture Lines of Reinforcement in Test 4 .................................209
4.70 Composite Mass after Failure of Test 5 .........................................................213
4.71 Failure Planes of the Composite Mass after Testing in Test 5 ......................214
4.72 Global Stress-Strain Relationship of Test 5 ...................................................215
4.73 Lateral Displacements on the Open Face of Test ..........................................216
4.74 Internal Displacements of Test 5 ...................................................................217
4.75 Locations of Strain Gauges Geosynthetic Sheets in Test 5 ...........................218
4.76 Reinforcement Strain Distribution of the Composite Mass in Test 5 ............219
4.77 Aerial View of the Reinforcement Sheets Exhumed from the Composite Mass after Test 5 ............................................................................................221
4.78 Location of Rupture Lines of Reinforcement in Test 5 .................................222
5.1 Conceptual Stress Path for Compaction of a GRS Mass ...............................237
xvi
5.2 Stress Path of the Proposed Simplified Model for Compaction of a GRS Mass ...............................................................................................................238
5.3 Locations of Compaction Loads and Stress Paths during Compaction at Depth a along Section I-I as Compaction Loads Moving toward Section I-I....................................................................................................................243
5.4 Locations of Compaction Loads and Stress Paths during Compaction at Depth a along Section I-I as Compaction Loads Moving away from Section I-I....................................................................................................................244
5.5 Stress Path at Depth a when Subject to Multiple Compaction Passes ...........245
5.6 Stress Path of the Proposed Model for Fill Compaction of a GRS Mass ......246
5.7 Concept of Apparent Confining Pressure and Apparent Cohesion of a GRS Composite ......................................................................................................250
5.8 An Ideal Plane-Strain GRS Mass for the SPR Model....................................254
5.9 Equilibrium of Differential Soil and reinforcement Elements .......................254
5.10 Reinforced Soil Test Specimen before Testing (Elton and Patawanran, 2005) ..............................................................................................................262
5.11 Backfill Grain Size Distribution before and after Large-Size Triaxial Tests (Elton and Patawanran, 2005) ........................................................................263
5.12 Large-Size Triaxial Test Results (Elton and Patawanran, 2005) ...................263
6.1 Distribution of Residual Lateral Stresses of a GRS mass with Depth due to Fill Compaction .............................................................................................279
6.2 Comparison of Results for GSGC Test 1 .......................................................288
6.3 Comparison of Lateral Displacement at Open Face of GSGC Test 1 ...........289
6.4 Comparison of Global Stress-Strain Relationship of GSGC Test 2 ..............291
6.5 Comparison of Lateral Displacement at Open Face of GSGC Test 2 ...........292
6.6 Comparison of Internal Displacements of GSGC Test 2 ...............................293
6.7 Comparison of Reinforcement Strains of GSGC Test 2 ................................294
6.8 Comparison of Global Stress-Strain Relationship of GSGC Test 3 ..............296
6.9 Comparison of Lateral Displacement at Open Face of GSGC Test 3 ...........297
6.10 Comparison of Internal Displacements of GSGC Test 3 ...............................298
6.11 Comparison of Reinforcement Strains of GSGC Test 3 ................................299
6.12 FE analyses of Test 2 with Different Confining Pressures ............................301
xvii
6.13 FE Mesh to Simulate CIS of a Reinforced Soil Mass ....................................304
6.14 Lateral Stress Distribution of a GRS Mass from FE Analyses ......................305
6.15 Comparison of Residual Lateral Stresses of a GRS Mass due to Fill Compaction between FE Analyses and Analytical Model ............................306
6.16 Comparison of Residual Lateral Stresses of a GRS Mass due to Fill Compaction between FE Analyses with Coarse Mesh and Analytical Model .............................................................................................................307
xviii
LIST OF TABLES
Table
2.1 Properties of material for the mini pier experiments (Adams, et al., 2007) ....19
2.2 None-Linear K0-loading/unloading model parameters (Seed, 1983) ..............48
2.3 Bi-Linear K0-loading/unloading model parameters (Seed, 1983) ...................54
4.1 Conditions and properties of material used in FE analyses ...........................108
4.2 Summary of some index properties of backfill ..............................................123
4.3 Summary of Geotex 4x4 properties ...............................................................128
4.4 Properties of Geotex 4x4 in fill-direction ......................................................131
4.5 Test program form the GSGC Tests ..............................................................133
4.6 Dimensions of the GSGC Test Specimens before Testing ............................162
4.7 Some Test Results for Test 1 .........................................................................168
4.8 Some Test Results for Test 2 .........................................................................186
4.9 Some Test Results for Test 3 .........................................................................198
4.10 Some Test Results for Test 4 .........................................................................210
4.11 Some Test Results for Test 5 .........................................................................223
4.12 Comparison between Test 1 and Test 2 .........................................................225
4.13 Comparison between Test 2 and Test 3 with the same Tf/Sv ratio .................227
4.14 Comparison between Test 2 and Test 4 .........................................................229
4.15 Comparison between Test 3 and Test 4 .........................................................230
4.16 Comparison between Test 2 and Test 5 .........................................................232
4.17 Comparison of strength properties of five GSGC Tests ...............................234
5.1 Model parameters for the proposed compaction simulation model ...............240
5.2 Values of factor r under different applied pressure and reinforcement lengths ............................................................................................................256
5.3 Comparison of the results between the analytical model and the GSGC tests ................................................................................................................259
xix
xx
5.4 Comparison of the results between Schlosser and Long’s method and GSGC tests ................................................................................................................260
5.5 Comparison of the results between the analytical model and Elton and Patawaran’s tests (2005) ................................................................................264
5.6 Comparison of the results between Schlosser and Long’s method and Elton and Patawaran’s tests (2005) .........................................................................265
5.7 Comparison of the results between the analytical model and the FE results for GSGC Test 2 ..................................................................................................267
5.8 Comparison of reinforcement forces between proposed model and current design equation for a GRS wall .....................................................................272
5.9 Comparison of reinforcement forces between proposed model and the GSGC tests ................................................................................................................273
5.10 Comparison of reinforcement forces between proposed model and test data from Elton and Patawaran (2005) ..................................................................274
6.1 Parameters and properties of the GSGC Tests used in analyses ....................282
6.2 The steps of analysis for the GSGC Tests .....................................................284
1. INTRODUCTION
1.1 Problem Statement
Over the past two decades, Geosynthetic-Reinforced Soil (GRS) structures, including
retaining walls, slopes, embankments, roadways, and load-bearing foundations, have
gained increasing popularity in the U.S. and abroad. In actual construction, GRS
structures have demonstrated a number of distinct advantages over their conventional
counterparts. GRS structures are generally more ductile, more flexible (hence more
tolerant to differential settlement and to seismic loading), more adaptable to low-
permeability backfill, easier to construct, require less over-excavation, and more
economical than conventional earth structures (Wu, 1994; Holtz, et al., 1997;
Bathurst, et al., 1997).
Among the various types of GRS structures, GRS walls have seen far more
applications than other types of reinforced soil structures. A GRS wall comprises two
major components: a facing element and a geosynthetic-reinforced soil mass. Figure
1.1 shows the schematic diagram of a typical GRS wall with modular block facing.
The facing of a GRS wall may take various shapes and sizes. It may also be made of
different materials. The other component of a GRS wall, a geosynthetic-reinforced
soil mass, however, is always a compacted soil mass reinforced by layers of
geosynthetic reinforcement.
It is a well-known fact that soil is weak in tension and relatively strong in
compression and shear. In a reinforced soil, the soil mass is reinforced by
1
incorporating an inclusion (or reinforcement) that is strong in tensile resistance.
Through soil-reinforcement interface bonding, the reinforcement restrains lateral
deformation of the surrounding soil, increases its confinement, reduces its tendency
for dilation, and consequently increases the stiffness and strength of the soil mass.
Many studies have been conducted on the behavior of GRS structures; however, the
interactive behavior between soil and reinforcement in a GRS mass has not been fully
elucidated. This has resulted in design methods that are fundamentally deficient in a
number of aspects (Wu, 2001). Perhaps the most serious deficiency with the current
design methods is that they ignore the composite nature of the reinforced soil mass,
and simply consider the reinforcement as “tiebacks” that are being added to the soil
mass. In current design methods, the reinforcement strength is determined by
requiring that the reinforcement be sufficiently strong to resist Rankine, Coulomb or
at-rest pressure that is assumed to be unaffected by the configuration of the
reinforcement. Specifically, the design strength of the reinforcement, Τrequired, has
been determined by multiplying an assumed lateral earth pressure at a given depth,
σh, by a prescribed value of reinforcement spacing, Sv, and a safety factor, Fs, i.e.,
svhrequired FST ∗∗= σ (1.1)
Equation 1.1 implies that, as along as the reinforcement strength is kept linearly
proportional to the reinforcement spacing, all walls with the same σh (i.e., walls of a
given height with the same backfill that is compacted to the same density) will behave
the same. In other words, a GRS wall with reinforcement strength of T at spacing Sv
will behave the same as one with reinforcement strength of 2*T at twice the spacing
2*Sv. Note that Equation 1.1 has very important practical significance. It has
encouraged designers to use stronger reinforcement at larger spacing, because the use
2
of larger spacing will generally reduce time and effort in construction.
A handful of engineers, however, have learned from actual construction that Equation
1.1 cannot be true. They realized that reinforcement spacing appears to play a much
greater role than reinforcement strength in the performance of a GRS wall.
Researchers at the Turner-Fairbank Highway Research Center have conducted a
series of full-scale experiments (Adams, 1997; Adams, et al., 2007) in which a weak
reinforcement at small spacing and a strong reinforcement (with several times the
strength of the weak reinforcement) at twice the spacing were load-tested. The
former was found to be much stronger than the latter. An in-depth study on the
relationship between reinforcement spacing and reinforcement stiffness/strength
regarding their effects on the behavior of a GRS mass is of critical importance to the
design of GRS structures and is urgently needed.
The effects of CIS in unreinforced soil masses and earth structures have been the
subject of study by many researchers, including Rowe (1954), Broms (1971), Aggour
and Brown (1974), Seed (1983), and Duncan, et al. (1984, 1986, 1991, and 1993).
These studies indicated that the CIS would increase significantly the lateral stresses in
soil (also known as the “locked-in” lateral stresses or “residual” lateral stresses),
provided that there was sufficient constraint to lateral movement of the soil during
compaction. The increase in lateral stresses will increase the stiffness and strength of
the compacted soil mass.
The effect of CIS is likely to be more significant in a soil mass reinforced with layers
of geosynthetics than in an unreinforced soil mass. This is because the interface
bonding between the soil and reinforcement will increase the degree of restraint to
lateral movement of the soil mass during fill compaction. With greater restraint to
lateral movement, the resulting locked-in lateral stresses are likely to become larger.
3
In numerical analysis of earth structures, the effects of CIS has either been overly
simplified (e.g., Katona, 1978; Hatami and Bathurst, 2005 and 2006; Morrison, et al.,
2006), or in most other studies, totally neglected. In the case of GRS walls, failure to
account for the CIS may be a critical culprit that has lead to the erroneous conclusion
by many numerical studies that Equation 1.1 is valid. Evaluation of compaction-
induced stresses in GRS structures is considered a very important issue in the study of
GRS structures.
In addition, GRS walls with modular block facing is rather “flexible”, hence the
design of these structures should consider not only the stresses in the GRS mass, but
also the deformation. Jewell-Milligan method (1989), recognized as the best
available method for estimating lateral movement of GRS walls applies only to walls
with little or no facing resistance. With increasing popularity of GRS walls with
modular block facing where facing rigidity should not be ignored, an improvement
over the Jewell-Milligan method for calculating lateral wall movement is needed.
1.2 Research Objectives
The objectives of this study were four-fold. The first objective was to investigate the
composite behavior of GRS masses with different reinforcing configurations. The
second objective was to examine the relationship between reinforcement strength and
reinforcement spacing regarding their effects on the behavior of a GRS mass. The
third objective was to develop an analytical model for evaluation of compaction-
induced stresses in a GRS mass. The fourth objective was to develop an analytical
model for predicting lateral movement of a GRS wall with modular block facing.
4
1.3 Tasks of Research
To achieve the research objectives outlined above, the following tasks were carried
out in this study:
Task 1: Reviewed previous studies on: (a) composite behavior of a GRS mass, (b)
compaction-induced stresses in a soil mass, and (c) reinforcing mechanism
of GRS structures.
Previous studies on composite behavior of a GRS mass were reviewed. The
review included theoretical analyses and experimental tests. Compaction-
induced stresses in an unreinforced soil mass that have been undertaken by
different researchers were also reviewed, including simulation models for
fill compaction. In addition, a literature study on reinforcing mechanisms of
GRS structures was conducted.
Task 2: Developed a hand-computation analytical model for simulation of
compaction-induced stresses in a GRS mass.
An analytical model for simulation of Compaction-Induced Stresses (CIS) in
a GRS mass was developed. The compaction model was developed by
modifying an existing fill compaction simulation model for unreinforced
soil. The model allows compaction-induced stress in a GRS mass to be
evaluated by hand computations. The CIS was implemented into a finite
element computer code for investigating performance of GRS structures.
Task 3: Developed an analytical model for the relationship between reinforcement
strength and reinforcement spacing, and derived an equation for calculating
composite strength properties.
An analytical model for the relationship between reinforcement strength and
reinforcement spacing was developed. Based on the model and the average
5
stress concept for GRS mass (Ketchart and Wu, 2001), an equation for
calculating the composite strength properties of a GRS mass was derived.
The model represents a major improvement over the existing model that has
been used in current design methods, and more correctly reflects the role of
reinforcement spacing versus reinforcement strength on the behavior of a
GRS mass. The equation allows the strength properties of a GRS mass to be
evaluated by a simple method.
Task 4: Designed and conducted laboratory experiments on a generic soil-
geosynthetic composite to investigate the performance of GRS masses with
different reinforcing conditions.
A generic soil-geosynthetic composite (GSGC) plane strain test was
designed by considering a number of factors learned from previous studies.
A series of finite element analyses were performed to determine the
dimensions of the test specimen that would yield stress-strain and volume
change behavior representative of a very large soil-geosynthetic composite
mass. Five GSGC tests with different reinforcement strength, reinforcement
spacing, and confining pressure were conducted. These tests allow direct
observation of the composite behavior of GRS mass in various reinforcing
conditions. They also provide measured data for verification of analytical
and numerical models, including the models developed in Tasks 2 and 3, for
investigating the behavior of a GRS mass.
Task 5: Performed finite element analyses to simulate the GSGC tests and analyze
the behavior of GRS mass.
Finite element analyses were performed to simulate the GSGC tests
conducted in Task 4. The analyses allowed stresses in the soil and forces in
the reinforcement to be determined. They also allowed the behavior of GRS
6
composites under conditions different from those employed in the GSGC
tests of Task 4 to be investigated.
Task 6: Verified the analytical models developed in Tasks 2 and 3 by using the
measured data from the GSGC tests and relevant test data available in the
literature.
The compaction model developed in Task 2 was employed to determine the
CIS for the GSGC tests; the results were then incorporated into a finite
element analysis to calculate the global stress-strain relationship and
compared to measured results. The measured data from the GSGC tests,
relevant test data available in the literature, and results from FE analyses
were also used to verify the analytical models developed in Task 3 for
calculating composite strength properties of a GRS mass and for calculating
required tensile strength of reinforcement based on the forces induced in the
reinforcement.
Task 7: Developed an analytical model for predicting lateral movement of GRS
walls with modular block facing.
An analytical model was developed for predicting the lateral movement of
GRS walls with modular block facing. The model was based on an existing
model for reinforced soil walls without facing (Jewell and Milligan, 1989).
The results obtained from the model were compared with measured data
from a full-scale experiment of a GRS wall with modular block facing.
7
8
Figure 1-1: Typical Cross-Section of a GRS Wall with Modular Block Facing
2. LITERATURE REVIEW
A GRS mass is a soil mass that is embedded with layers of geosynthetic
reinforcement. These layers are typically placed in the horizontal direction at vertical
spacing of 8 in. to 12 in. Under vertical loads, a GRS mass exhibits significantly
higher stiffness and strength than an unreinforced soil mass. This Chapter presents a
review of previous studies on the mechanics of reinforced soil, the composite
behavior of a Geosynthetic-Reinforced Soil (GRS) mass, and Compaction-Induced
Stresses (CIS) in a reinforced soil mass.
2.1 Mechanics of Reinforced Soil
In thee literature, three concepts have been proposed to explain the mechanical
behavior of a GRS mass: (1) the concept of enhanced confining pressure (Yang,
1972; Yang and Singh, 1974; Ingold, 1982; Athanasopoulos, 1994), (2) the concept of
enhanced material properties (Scholosser and Long, 1972; Hausmann, 1976; Ingold,
1982; Gray and Ohashi, 1983; Maher and Woods, 1990; Athanasopoulos, 1993; Elton
and Patawaran, 2004 and 2005), and (3) the concept of reduced normal strains (Basset
and Last, 1978).
The mechanics of a GRS mass has been explained by Schlosser and Long (1972) and
Yang (1972) by two concepts: (a) concept of apparent cohesion, and (b) concept of
apparent confining pressure.
9
a) Concept of apparent cohesion
In this concept, a reinforced soil is said to increase the major principle stress at failure
from σ1 to σ1R (with an apparent cohesion cR’) due to the presence of the
reinforcement, as shown by the Mohr stress diagram in Figure 2.1. If a series of
triaxial tests on unreinforced and reinforced soil elements are conducted, the failure
envelops of the unreinforced and reinforced soils shall allow the apparent cohesion
cR’ to be determined. Yang (1972) indicated that the φ value for unreinforced sand
and reinforced sand were about the same as long as slippage at the soil-reinforcement
interface did not occur.
Figure 2.1: Concept of Apparent Cohesion due to the Presence of Reinforcement (Scholosser and Long, 1972)
10
b) Concept of increase of apparent confining pressure
In this concept, a reinforced soil is said to increase its axial strength from σ1 to σ1R
(with an increase of confining pressure Δσ3R), as shown in Figure 2.2, due to the
tensile inclusion. The value of Δσ3R can also be determined from a series of triaxial
tests, again by assuming that φ will remain the same.
Figure 2.2: Concept of Apparent Confining Pressure due to the Presence of Reinforcement (Yang, 1972)
Note that the concept of apparent confining pressure allows the apparent cohesion to
be determined with only the strength data for the unreinforced soil as follows
(Schlosser and Long, 1972):
11
(1) Consider a reinforced soil mass with equally spaced reinforcement of strength
Tf (vertical spacing = Sv), it is assumed that the increase in confining pressure
due to the tensile inclusion Δσ3R is:
3f
Rv
TS
σΔ = (2.1)
(2) From Figure 2.1 and Figure 2.2 and using Rankine’s earth pressure theory to
equate the principal stress at failure σ1R,
Referring to Figure 2.1, '
1 3 2R C P RK c Kσ σ= + P (2.2)
Referring to Figure 2.2,
PRR K31 σσ = (2.3)
Knowing
RCR 333 σσσ Δ+= (2.4)
Equation 2.3 can be written as:
( PRCPRR KK 3331 )σσσσ Δ+== (2.5)
Equating Equations 2.2 to Equation 2.5, we obtain
2' 3' PR
R
Kc
σΔ= (2.6)
(3) Substituting equation 2.1 into equation 2.6,
v
PfR S
KTc
2'= (2.7)
Equation 2.7 may be very useful for evaluating the stability of a reinforced soil mass.
Given a granular soil with strength parameters c (c = 0) and φ, Equation 2.7 allows
the strength parameters of a reinforced soil mass (cR’ and φR) to be determined as a
function of T and Sv.
12
It should be noted that the validity of Equation 2.7 is rather questionable. There is a
key assumption involved in the derivation -- the assumption of v
fR S
T=Δ 3σ (this
expression implies that an increase in “Tf” has the same effect as a proportional
decrease in “Sv”.) Figure 2.3 shows the strength envelopes for sand and reinforced
sand based on the studies of Schlosser and Long (1972), Yang (1972), and Hausmann
(1976). Note that the increase in confining pressure was vT SR /3 =Δσ (see Figure
2.2) based on the explanation given by Schlosser and Long (1972) and Yang (1972);
and it was vT SR /3 ≤Δσ in Hausmann’s study (1976), where . fT TR ≡
13
Figure 2.3: Strength Envelopes for Sand and Reinforced Sand (Mitchell and Villet, 1987)
14
2.2 Composite Behavior of GRS Mass
The behavior of soil-geosynthetic composites have been investigated through
different types of laboratory experiments, including: small-size triaxial compression
tests with the specimen diameter no greater than 6 in. (Broms, 1977; Gray and Al-
Refeai, 1986; Haeri et al., 2000; etc), large-size triaxial compression tests (Ziegler et
al., 2008), large-size unconfined compression tests (Elton and Patawaran, 2005),
unconfined compression tests with cubical specimens (Adams, 1997 and Adams et
al., 2007), and plane strain tests (Ketchart and Wu, 2001).
Figure 2.4 shows the effects of reinforcement layers on the stiffness and strength of
soil-geosynthetic composites conducted by Broms (1977). For unreinforced soil
specimen (number 1 in Figure 2.4) and the specimen with reinforcement at the top
and bottom (number 2 in Figure 2.4), the stress-strain curves are nearly the same.
This suggests that unless the reinforcement is placed at locations where lateral
deformation of the soil occurs, there will not be any reinforcing effect. For the
specimens with 3 and 4 layers, the stiffness and strength of the composites are
significantly higher as the reinforcement effectively restrains lateral deformation of
- Fraction of peak lateral compaction stress retained as residual stress for virgin soil
- Incremental coefficient of at-rest lateral earth pressure for reloading
10 ≤≤ F
020 ≥≥ KK
- F (or K2) should be chosen such that the bi-linear unloading stress path intersects the α-type non-linear unloading stress path at a suitable OCR. Recommended:
( )( )1
1−
−−=
OCROCROCRF
α
Where a suitable OCR for “matching” the bi-linear and non-linear unloading curves is typical . 5≅OCR
K3 Incremental coefficient of at-rest lateral earth pressure for reloading
030 KK ≤≤ ( )( )FKKK −=≅ 1023
Note: vh K '' 1lim, σσ = , BB
B KcKK ,',13
,',11 ''2
φφ σ+= , ',1,',1 3
2φφ KK B ≅ , and . PKK =',1 φ
54
Figure 2.31 shows the basic components of the bi-linear model. Virgin loading is
assumed to follow the K0-line in the same manner as in the non-linear model.
Unloading initially follows a linear stress path according '' 2 vh K σσ Δ=Δ until a K1-
type of limiting condition is reached, at which point further unloading follows a linear
stress path according to '' 1 vh K σσ Δ= . Reloading follows a linear stress path
according to 'v' 3h K σσ Δ=Δ until the virgin K0-loading stress path is regained, after
that further reloading follows the virgin stress path.
Figure 2.31: Basic Components of the Bi-linear Model (Seed,1983)
From Figure 2.31, the parameters of bi-linear model can be calculated as:
'.' ,, phrh F σσ Δ=Δ (2.15)
( )( )1
110
2
−−
−=−=OCR
OCROCRKKF
α
(2.16)
55
( )30
2
233
11 β
β
−−=
KK
KK (2.17)
56
Figure 2.32: Compaction Loading/Unloading Cycles in the Bi-Linear Model (Seed, 1983)
57
Seed (1983) also developed a simplified hand calculation procedure for computing
the CIS. The procedure can be described by the following steps:
1. Calculate the peak lateral compaction pressure profile (i.e., Δσ'h,vc,p vs. depth
relationship) by the method described in Section c, as shown in Figure
2.32(b).
2. Multiply the Δσ'h,vc,p values with the bi-linear model parameter F.
3. Calculate the lateral residual stress as ',, . 0, '' pvchvrh FK σσσ Δ+=
4. Reduce the near-the-surface portion of the σ’h,r distribution with
'' ,',1, vBrh K σσ φ≤ at all depths.
5. Increase the residual effective stress distribution such that '' 0, vrh K σσ ≥ at all
depths.
Figure 2.33 shows an example problem given by Seed (1983) to show how to
determine the compaction-induced lateral pressure on a vertical non-deflecting
structure using bi-linear model and non-linear model. The material parameters of the
models based on the given angle of friction (φ = 35o) are shown in Table 2.4.
Table 2.4: Material parameters for non-linear and bi-linear models
Non-linear model parameters Bi-linear model parameters
K0 = 1 - sinφ' = 0.43
21, '
'tan 45 3.692
oK φφ⎛ ⎞= − =⎜ ⎟
⎝ ⎠
63.0=α (from Figure 2.20)
6.0=β (assumed)
c' = 0
K0 = 1 - sinφ' = 0.43
46.232
',1,',1 == φφ KK B
44.015
551 =−−
−=α
F
24.023 == KK
c'B = 0
58
Figure 2.34 shows the residual lateral stresses as a result of unloading due to fill
compaction and the at-rest earth pressure. The results of the bi-linear model (the solid
line) and the non-linear model (the bold dots) are approximately the same.
59
Figure 2.33: An Example Problem for Hand Calculation of Peak Vertical Compaction Profile (Seed, 1983)
60
Figure 2.34: Solution Results from the Bi-Linear Model and the Nonlinear Model (Seed, 1983)
The compaction models developed by Seed (1983) has been used to determine the
CIS for full-scale experiments and reported by Duncan and Seed (1986) and Duncan
et al. (1986). The papers showed that, the CIS could be calculated based on either the
simplified method (i.e., the bi-linear model) or the non-linear model with the aid of
finite element analysis. They have shown that the resulting lateral earth pressures
determined by the models are in good agreement with measured data.
61
2.4 Compaction-Induced Stresses in a Reinforced Soil Mass
of previous studies on GRS masses
cluding the effects of CIS is presented below.
4.1 Ehrlich and Mitchell (1994)
fecting the
the procedure are:
ion lift was subjected to only one cycle of loading, as shown
Figure 2.35.
stress-state condition at the end of construction. Ehrlich and Mitchell (1994) noted
Many researchers and practicing engineers have suggested that if a granular backfill
is well compacted, a GRS mass can usually carry a great deal of loads and experience
little movement. A handful of studies on performance of GRS structures have used
simplified, and somewhat arbitrary, procedures to simulate the effects of fill
compaction. In all other studies, the compaction-induced stresses (CIS) in the fills
have been ignored completely. The CIS in a GRS mass is likely to be more
pronounced than those induced in an unreinforced soil mass because soil-
reinforcement interface friction tends to restrain lateral deformation of the soil mass
and results in greater values of CIS. A review
in
2.
Ehrlich and Mitchell (1994) presented a procedure to include CIS in the analysis of
reinforced soil walls and noted that CIS was a major factor af
reinforcement tensions. The assumptions involved in
• The stress path was as shown in Figure 2.35.
• With the multi-cycle operations of soil placement and compaction during
construction, the soil surrounding the reinforcement maximum tension point
in each compact
in
Loading due to the weight of overlying soil layers plus some equivalent increase in
the stress state induced by the compaction operations is shown as paths 1 to 3 in
Figure 2.35. This is followed by unloading along paths 3 to 5 to the final residual
62
that by following this procedure, the stresses in each layer were calculated only once
and that each layer calculation was independent of the others.
The specific values of σ’z and σ’zc at point 3 in Figure 2.35 represent the maximum
stress applied to the soil at a given depth during the construction process. The
maximum past equivalent vertical stress, including compaction at the end of
construction (σ’zc), can be estimated using a new procedure, based on the method
given by Duncan and Seed (1986) for conventional retaining walls.
Figure 2.35: Assumed Stress Path (Ehrlich and Mitchell, 1994)
In Figure 2.35, the value of σ’zc can be estimated as the following:
0
''
Kxp
zc
σσ = (2.18)
where:
'sin10 φ−=K (Jaky, 1944) (2.19)
63
( )LQN
K Axpγγ
νσ'5.0
1' 0 += (2.20)
and Q = maximum vertical operating force of the roller drum
L = length of the roller drum
γ = effective soil unit weight.
0
00 1 K
K−
=υ (Poisson’s ratio under K0-condition) (2.21)
⎥⎦
⎤⎢⎣
⎡−⎟
⎠⎞
⎜⎝⎛ +⎟
⎠⎞
⎜⎝⎛ += 1
2'45tan
2'45tan 4 φφ
γooN (bearing capacity factor) (2.22)
Figure 2.36 shows the effects of CIS on compaction and reinforcement stiffness in
GRS walls. The conclusions drawn by Ehrlich and Mitchell from their study are:
- The soil shearing resistance parameters, the soil unit weight, the depth, the
relative soil-reinforcements stiffness index, Si, and compaction are the
major factors determining reinforcement tensions (typical Si for metallic
reinforcement: 0.500-3.200; plastic reinforcement: 0.030-0.120 and
geotextile reinforcement: 0.003-0.012);
- Increasing Si, usually means increased lateral earth pressure and
reinforcement tension, but at shallow depths the opposite effect can occur
depending on compaction conditions;
- The coefficient of horizontal earth pressure, K, can be greater than K0 at
the top of the wall and be greater than KA to depths of more than 6.1 m (20
ft) depending on the relative soil-reinforcements stiffness index, and the
compaction load; and
- K0 is the upper limit for the coefficient of horizontal earth pressure, K, if
there is no compaction of the backfill.
64
Figure 2.36: Compaction and Reinforcement Stiffness Typical Influence (Ehrlich and Mitchell, 1994)
2.4.2 Hatami and Bathurst (2006)
Hatami and Bathurst (2006) noted that fill compaction has two effects on the soil: (1)
increase the lateral earth pressure, (2) reduce the effective Poisson’s ratio. They
suggested that the first effect can be modeled in a numerical analysis by applying a
uniform vertical stress (8 kPa, and 16 kPa depending on compaction load) to entire
surface of each newly placed soil layer before analysis and removed it afterwards.
This procedure was based on a recommendation by Gotteland et al. (1997).
Gotteland et al. simulated the compacting effect by loading and unloading of a
uniform surcharge of 50 kPa and 100 kPa on the top of the wall.
65
For the second effect of compaction on the reduction of Poisson’s ratio, Hatami and
Bathurst used the numerical simulation to find νmin from matching measured and
analysis data. The results (wall lateral movement and reinforcement forces) obtained
from the numerical analysis including compaction effect were in a very good
agreement with the measured data.
In their numerical analyses, the compaction effects were also account for by
increasing the elastic modulus number, Ke value from triaxial test results by a factor
of 2.25 for Walls 1 and 2. In other words, the elastic modulus was increased by the
factor of 2.25 for Walls 1 and 2.
2.4.3 Morrison et al. (2006)
Morrison et al. (2006) simulated the effects of fill compaction of shored mechanically
stability earth (SMSE) walls. A 50 kPa inward pressure was applied to the top,
bottom and exposed faces of each lift to simulate the effects of fill compaction. The
inward pressure was then reduced to 10 kPa on the top and bottom of a soil lift prior
to placement of the next lift to simulate vertical relaxation or unloading following
compaction. The inward pressure acting on the exposed face was maintained at 50
kPa as this produced the most reasonable model deformation behavior compared with
that observed in the field-scale test. They considered that the inward maintained
pressures are "locking-in" stresses in soil due to compaction.
The stiffness of soil was increased by the factor of ten (10) to consider the
compaction-induced stresses in the GRS mass. This factor in Hatami and Bathurst
(2006) was about 2.25.
66
Figure 2.37 shows the model for finite element analysis by Morrison et al. (2006).
The figure shows the simulation of fill compaction of lift 5 by the applying uniform
pressures.
Figure 2.37: Finite Element Model for Finite Element Analysis (Morrison et al., 2006)
67
The results of the lateral movement and reinforcement forces showed a good
qualitative agreement with the measured data. However, general application of the
procedure may be questionable because neither the method of analysis nor the
magnitude of the applied inward pressure was properly justified.
2.5 Highlights on Compaction-Induced Earth Pressures in the Literature
A number of important highlights regarding compaction-induced lateral earth
pressures in the literature are summarized below:
1. Compaction of soil against a rigid, vertical, non-yielding structure appears to
result in the following residual lateral pressure distribution: (a) the lateral
pressures near the surface increase rapidly with depth, exceeding the at-rest
value, but limited a passive failure, (b) at intermediate depths, the lateral
pressures exceed the at-rest value, increase less rapidly with depth or remain
fairly constant with depth, and (c) at greater depths, the lateral pressures
appear to be the simple at-rest pressures, showing no affects of compaction
(Broms, 1971, Seed, 1983).
2. Compaction of soil against deflecting structures appears to increase near
surface.
3. The compaction-induced residual earth pressures are significant affected by
the compaction equipments. For compaction by small hand-operated rollers,
the increase in the lateral pressure occurs within a depth of about 3 to 4m, but
for very large rollers, the effect of compaction can be up to 15 to 25 m
(Duncan and Seed, 1986).
4. Structural deflections away from the soil, which occur during fill placement
and compaction, will reduce the residual lateral earth pressures. Reduction in
pressures appears to occur more rapidly in heavily compacted cohesionless
soil (Seed, 1983).
68
5. Compaction-induced residual lateral earth pressures in cohesive soils appear
to dissipate with time, even against non-deflecting structures, and eventually
approach at-rest values.
6. There is some evidence suggesting that the direction of rolling with the
compactor can have a significant effect in compaction-induced earth pressures
(Erhlich and Mitchell, 1993).
7. Field observations indicate that available overburden pressures are sufficient
that possible passive failure does not limit residual lateral earth pressure, a
high percentage of the peak lateral earth pressures induced during compaction
may be retained as residual pressures. In previously compacted soil, however,
additional compaction can result in only small increases in peak pressures, and
a negligible fraction of this (Aggour and Brown, 1974).
8. A number of simulation models have been proposed to explain and to evaluate
the residual lateral earth pressures induced by compaction. Common to all of
these theories is the idea that compaction represents a form of over-
consolidation wherein stresses resulting from a temporary or transient loading
condition are retained to some extent following removal of this peak load.
9. Many researchers, including Rowe (1954), Broms (1971), Gotteland et al.
(1997), and Hatami and Bathurst (2006), have simulated fill compaction by
application and removal of a surficial surcharge pressure.
10. Broms (1971) proposed a theory to calculate compaction-induced residual
lateral earth pressures against a rigid, vertical, frictionless, non-yielding wall.
The simulation results somewhat agree with available field data for walls
sustaining minimal deflections. Broms assumed that: (a) unloading results in
no decrease in lateral stress until a limiting passive failure-type condition is
reached, and (b) reloading results in no increase in lateral stress until the
virgin K0-loading stress path is regained. This type of model does not predict
well the peak lateral stresses induced by fill compaction, and is not suited for
69
70
computing lateral stresses induced by a surficial compaction plant of finite
lateral dimensions (not entire the surface). But Broms’ theory is very easy to
apply. Some researchers have adopted this theory for analysis of GRS
structures, including Gotteland et al. (1997), Hatami and Bathurst (2006), and
Morrison et al. (2006).
11. Seed (1983) developed two models for simulation of fill compaction: a non-
linear model and a bi-linear model. They are well suited for simulation of
compaction operation in GRS structures. The simulation results of the two
models are rather similar, and both agree well with measured data of
unreinforced earth retaining walls. The bi-linear model is easy to apply by
using hand calculation. Both models, bi-linear and non-linear models, are
based on the K0 condition. These are very useful to estimate CIS for soil only.
To use it for GRS structures, in-depth studies need to be carried out.
3. AN ANALYTICAL MODEL FOR CALCULATING LATERAL
DISPLACEMENT OF A GRS WALL WITH MODULAR BLOCK
FACING
Over the past two decades, Geosynthetic-Reinforced Soil (GRS) walls have gained
increasing popularity in the U.S. and abroad. In actual construction, GRS walls have
demonstrated a number of distinct advantages over the conventional cantilever and
gravity retaining walls. GRS walls are generally more ductile, more flexible (hence
more tolerant to differential settlement and to seismic loading), more adaptable to
low-permeability backfill, easier to construct, require less over-excavation, and
significantly more economical than conventional earth structures (Wu, 1994; Holtz
et al., 1997; Bathurst et al., 1997).
A GRS wall comprises two major components: a facing element and a GRS mass.
The facing element of a GRS wall have been constructed with different types of
material and in different forms, including wrapped geotextile facing, timber facing,
facing. Among the various facing types, modular concrete block facing has been
most popular in North America, mainly because of its ease of construction, ready
availability, and lower costs. The other component of a GRS wall, a GRS mass,
however, is always a compacted soil mass reinforced with layers of geosynthetic
reinforcement. Figure 3.1 shows the schematic diagram of a typical GRS wall with
modular block facing.
Current design methods for GRS walls consider only the stresses and forces in the
wall system. Even though a GRS wall with modular block facing is a fairly “flexible”
71
wall system, movement of the wall is not accounted for in current designs. A number
of empirical and analytical methods have been proposed for estimating lateral
movement of GRS walls. Most these methods, however, do not address the rigidity
of the facing although many full-scale experiments, numerical analysis, and field
experience have clearly indicated the importance of facing rigidity on wall movement
(e.g., Tatsuoka, et al., 1993; Rowe and Ho, 1993; Helwany et al., 1996; Bathurst et
al., 2006).
The prevailing methods for estimating the maximum lateral displacement of GRS
walls include: the FHWA method (Christopher, et al., 1989), the Geoservices method
(Giroud, 1989), the CTI method (Wu, 1994), and the Jewell-Milligan method (1989).
Among these methods, the Jewell-Milligan method has been found to give the closest
agreement with finite element analysis (Macklin, 1994). The Jewell-Milligan
method, however, ignores the effect of facing rigidity. Strictly speaking, the method
is only applicable to reinforced soil walls where there is little facing rigidity, such as a
wrapped-faced GRS wall.
A study aiming at developing an analytical model for calculating lateral movement of
a GRS wall with modular block facing was undertaken. The analytical model
modifies the Jewell-Milligan method (1989) to include the rigidity of facing element.
To verify the analytical model, the lateral wall displacements calculated by the
analytical model were compared with the results of the Jewell-Milligan method
(1989) for GRS walls with negligible facing rigidity. In addition, the lateral wall
displacements obtained from the analytical model were compared with the measured
data of a full-scale experiment of GRS wall with modular block facing (Hatami and
Bathurst, 2005 & 2006).
72
In addition to lateral displacement profiles, an equation for determining facing
connection forces (i.e., the forces in reinforcement immediately behind the facing) is
introduced.
Soil
Reinforcement
H
q
.L
Facing block
Sv
Figure 3.1: Basic Components of a GRS Wall with a Modular Block Facing.
3.1 Review of Existing Methods for Estimating Maximum Wall Movement
The most prevalent methods for estimating the maximum lateral displacement of
GRS walls are the FHWA method (Christopher, et al., 1989), the Geoservices method
(Giroud, 1989), the CTI method (Wu, 1994), and the Jewell-Milligan method (1989).
A summary of each method is presented below.
73
3.1.1 The FHWA Method (Christopher, et al., 1989)
The FHWA method correlates L/H ratio (L = reinforcement length, H = wall height)
with the lateral displacement of a reinforced soil wall during construction. Figure 3.2
shows the relationship between L/H and δR, the empirically derived relative
displacement coefficient. Based on 6 m high walls, the δR value is to increase 25%
for every 20 kPa of surcharge. For the higher walls, the surcharge effect may be
greater. The curve in Figure 3.2 has been approximated by a fourth-order polynomial
as:
For 0.3 1.175LH
≤ ≤ ,
471.945.3516.5725.4281.11234
+⎟⎠⎞
⎜⎝⎛−⎟
⎠⎞
⎜⎝⎛+⎟
⎠⎞
⎜⎝⎛−⎟
⎠⎞
⎜⎝⎛=
HL
HL
HL
HL
Rδ (3.1)
For extensible reinforcement, the maximum lateral wall displacement, δmax, can be
calculated from by the following equation (δmax is in units of H):
⎟⎠⎞
⎜⎝⎛=
75maxH
Rδδ (3.2)
74
Figure 3.2: Empirical Curve for Estimating Maximum Wall Movement During Construction in the FHWA Method (Christopher, et al., 1989)
75
The FHWA method was developed empirically by determining a displacement
“trend” from numerical analysis and adjusting the curve to fit with field measured
data. The method provides a quick estimate of the maximum lateral displacement.
Note that the maximum lateral displacement, δR, as obtained from Figure 3.2 has been
corrected for the wall with different height and surcharge.
3.1.2 The Geoservices Method (Giroud, 1989)
The Geoservices method relies on limit-equilibrium analyses to calculate the length
of the required reinforcement to satisfy a suggested factor of safety with regard to
three presumed external failure modes (e.g., bearing capacity failure, sliding and
overturning). The method provides a procedure for calculating the lateral wall
displacement.
The lateral displacement is calculated by first choosing a strain limit for the
reinforcement. This strain limit is usually less than 10 % and will depend on a
number of factors such as the type of wall facing, the displacement tolerances and the
type of geosynthetic to be used as reinforcement. Concrete facing panels, for
example, would not allow much lateral displacement without showing the signs of
distress. Therefore a low strain limit (1 to 3 %) should be selected.
Geosynthetics have a wide range of material properties depending on, among other
factors, the way they are manufactured. Non-woven geotextile exhibits low modulus
characteristics and if chosen as reinforcement for a wall, design would necessarily
imply that a large design strain is to be considered.
76
Once the strain limit has been selected, the method then assumes a distribution of
strain in the reinforcement, as shown in Figure 3.3 for calculating wall movement.
The horizontal displacement, δh, then becomes:
2Ld
hε
δ = (3.3)
where εd = strain limit (εmax), and L = reinforcement length.
Φ/2
Rankine Surface
Strain Distribution
Reinforcement
εmax
Figure 3.3: Assumed Strain Distribution in the Geoservices Method
3.1.3 The CTI Method (Wu, 1994)
Differing from all other design methods based on ultimate-strength of the
geosynthetic reinforcement, the CTI method is a service-load based design method.
77
The requirements of reinforcement are made in terms of stiffness at a design limit
strain as well as the ultimate strength.
In most cases, the designer will select a design limit strain of 1% to 3% for the
reinforcement. The maximum lateral displacement of a wall, δmax, can be estimated
by the following empirical equation:
⎟⎠⎞
⎜⎝⎛=
25.1maxH
dεδ (3.4)
where εd = design limit strain (typically 1 % to 3 % for H ≤ 30 ft) and H =
wall height.
If the maximum wall displacement exceeds a prescribed tolerance for the wall, a
smaller design limit strain should be selected so that the maximum lateral
displacement of the wall will satisfy the performance requirement. Equation 3.4
applies only to walls with very small facing rigidity, such as wrapped-faced walls.
Walls with significant facing rigidity will have smaller maximum lateral
displacement. For example, a modular block GRS walls will have δmax about 15%
smaller than that calculated Equation 3.4.
3.1.4 The Jewell-Milligan Method
Jewell (1988) and Jewell and Milligan (1989) proposed a procedure for calculating
wall displacement based on analysis of stresses and displacements in a reinforced soil
mass. The method describes a link between soil stresses (stress fields) in a reinforced
soil mass in which a constant mobilized angle of friction is assumed with the resulting
displacements (velocity fields). There are two parameters for plane-strain plastic
deformation of soil: the plane strain angle of friction, φps, and the angle of dilation, ψ.
78
The planes on which the maximum shearing resistance φps is mobilized are called the
“stress characteristics” and are inclined at ( )2/450psφ+ to the direction of major
principal stress, as shown in Figure 3.4(a). The directions along which there is no
linear extension strain in the soil are called the “velocity characteristics” and are
inclined at ( )2/450 ψ+ to the direction of major principal stress, as shown in Figure
3.4(b).
Figure 3.4: (a) Stress Characteristics and (b) Velocity Characteristics behind a Smooth Retaining Wall Rotating about the Toe (Jewell and Milligan, 1989).
Jewell and Milligan (1989) noted from limiting equilibrium analyses that there are
three important zones in a reinforced soil wall, as illustrated in Figure 3.5(a). The
boundary between zone 1 and 2 is at an angle ( )2/450 ψ+ to the horizontal, and
between zone 2 and 3 at an angle φds. Large reinforcement forces are required in zone
1 to maintain stability across a series of critically inclined planes. In zone 2, the
The assumptions of the Jewell-Milligan method for "ideal length" of reinforcement
are:
• The reinforcement length at every layer extends to the back of zone 2, so
called "ideal length".
• The horizontal movement of the facing may be calculated by assuming the
horizontal deflections starting at the fixed boundary between zones 2 and 3
and working to the face of the wall.
• The stability on the stress characteristics and the velocity characteristics is
equally critical in soil and hence reinforcement must provide equilibrium for
both. The consequence is that behind the Rankine active zone in a reinforced
soil wall, the equilibrium is governed by φds mobilized on the velocity
characteristics.
Figure 3.5: Major Zones of Reinforcement Forces in a GRS Wall and the Force Distribution along Reinforcement with Ideal Length (Jewell and Milligan, 1989).
80
In Figure 3.5, the maximum horizontal resultant force required for equilibrium, Prm, is
equal to the active force Pa:
⎟⎟⎠
⎞⎜⎜⎝
⎛+== HqHKPP saarm 2
2γ (3.5)
in which, γ = unit weight of the soil; H = wall height; qs = uniform surcharge;
Ka = active earth pressure coefficient that can be expressed as:
⎟⎠⎞
⎜⎝⎛ +
⎟⎠⎞
⎜⎝⎛ −+
=+−
=
245tan
245tan
)sin1()sin1(
ψ
φψ
φφ ds
ps
psaK (3.6)
The required reinforcement force Pr in zone 2 at an angle θ, as shown in Figure
3.5(b), can be estimated from the maximum reinforcement force Prm as:
θ
φθtan
)tan(
a
ds
rm
r
KPP −
= (3.7)
The results of the displacement analyses have been presented in the form of design
charts as shown in Figure 3.6. The charts can be used to determine the distribution of
lateral wall displacement along the wall face for different values of mobilized internal
friction φps and angles of dilation ψ.
To estimate the horizontal deflection at the GRS wall face for reinforcement with the
ideal length and uniform spacing, the charts as shown in Figure 3.6 can be used. The
horizontal deflection at the wall face depends on the wall height H, the mobilized soil
shearing resistance φds, the reinforcement force Pr, and the reinforcement stiffness K.
81
The charts in Figure 3.6 can be used to obtain a dimensionless factor,base
h
HPKδ , and
then the horizontal displacement δh can be calculated from this factor. The
reinforcement occurs at the base of the wall, Pbase, in the dimensionless factor above,
can be calculated by
( )svabase qHsKP += γ (3.8)
82
Figure 3.6: Charts for Estimating Lateral Displacement of GRS Walls with the Ideal Length Layout (Jewell and Milligan, 1989)
83
3.2 Developing an Analytical Model for Calculating Lateral Movement and
Connection Forces of a GRS Wall
Jewell and Milligan (1989) have presented design charts for estimating deformation
of reinforced soil walls where the rigidity of the facing can be ignored. Ho and Rowe
(1997) and Rowe and Ho (1993, 1998) pointed out that there is little variation in the
reinforcement forces and the lateral wall deformation when the reinforcement length
to wall height ratio, L/H, is equal to or greater than 0.7 (Note: L/H = 0.7 is commonly
used in practice and it is also suggested by the AASHTO). Based on a series of
numerical analyses of GRS walls, Rowe and Ho (1998) also showed that the
maximum lateral deformation obtained by the Jewell-Milligan method with an ideal
reinforcement length, as defined by Jewell and Milligan (1989), is generally in good
agreement with the numerical results for L/H = 0.7. For this reason, the Jewell-
Milligan method, with the ideal reinforcement length, can be used to estimate lateral
movement of a reinforced soil wall with L/H ≥ 0.7. The analytical model developed
in this study for GRS walls with modular block facing was base on Jewell-Milligan
method. The method is applicable to GRS walls with L/H ≥ 0.7.
The derivation of the analytical model is given below. It begins with the derivation of
the equations in the Jewell-Milligan method for predicting deformation of a
reinforced soil wall with negligible facing rigidity, followed by the derivation of the
equations for determining connection forces in the reinforcement for walls with
modular block facing, and ends with the equations for calculating lateral movement
of GRS walls with modular block facing.
84
3.2.1 Lateral Movement of GRS Walls with Negligible Facing Rigidity
Figure 3.7 shows the three major zones in a GRS wall and the force distribution in the
reinforcement at depth zi, used by Jewell and Milligan (1989) to develop an analytical
model for determination of wall deformation. Jewell and Milligan (1989) have
presented design charts based on the analytical model (without giving the derivation).
The following derivation is presented for completeness and for easier reference when
showing the derivation of the analytical model developed for this study.
Pr m
H
z
(H - z )
i
i
Reinforcement
Zone 3
Zone 1Zone 2
x
z
L Lzone-1 zone-2
L
45 + ψ/2 φds
i
Figure 3.7: Major Zones of the Reinforcement Force in a Reinforced Soil Wall (Jewell and Milligan, 1989)
The horizontal movement, Δh, of the wall face at depth zi can be evaluated as:
(3.9) 21 −− Δ+Δ=Δ zonezoneh
1inf0 inf
1
1
−− ∫−
==Δ zonere
rmL
re
rmzone L
KP
dxKPzone
(3.10)
85
∫−−
−
+
−− ⎟⎠⎞
⎜⎝⎛≈=Δ
21
1
2infinf
2 21zonezone
zone
LL
Lzone
re
rm
re
rzone L
KP
dxK
P (3.11)
where Kreinf stiffness of the reinforcement
Prm maximum reinforcement force at depth zi
Lzone-1 reinforcement length in Zone 1 at depth zi
Lzone-2 reinforcement length in Zone 2 at depth zi
Substituting Equation 2.10 and Equation 3.11 into Equation 3.9, we get
⎟⎠⎞
⎜⎝⎛ +=Δ −− 21
inf 21
zonezonere
rmh LL
KP
(3.12)
Since
( ) ⎟⎠⎞
⎜⎝⎛ −−=− 245tan1
ψoizone zHL (3.13)
and
( ) ( ) ⎥⎦
⎤⎢⎣
⎡⎟⎠⎞
⎜⎝⎛ −−−−=− 245tan90tan2
ψφ ods
oizone zHL (3.14)
Therefore, substituting Equation 3.13 and Equation 3.14 into Equation 3.12 leads to
( ) ( ) ( )⎭⎬⎫
⎩⎨⎧
⎥⎦
⎤⎢⎣
⎡⎟⎠⎞
⎜⎝⎛ −−−−+⎟
⎠⎞
⎜⎝⎛ −−=Δ
245tan90tan
21
245tan
inf
ψφψ ods
oi
oi
re
rmh zHzH
KP
(3.15)
Rearranging Equation 3.15, we have
( ) ( )⎭⎬⎫
⎩⎨⎧
⎥⎦
⎤⎢⎣
⎡⎟⎠⎞
⎜⎝⎛ −−−+⎟
⎠⎞
⎜⎝⎛ −−=Δ
245tan90tan
21
245tan
inf
ψφψ ods
ooi
re
rmh zH
KP
(3.16)
or
( ) ( )⎭⎬⎫
⎩⎨⎧
⎥⎦
⎤⎢⎣
⎡−+⎟
⎠⎞
⎜⎝⎛ −−=Δ ds
ooi
re
rmh zH
KP
φψ 90tan2
45tan21
inf
(3.17)
86
The value of Δh, lateral displacement of a GRS wall at depth zi, can be calculated
directly from the following equation:
( ) ⎥⎦
⎤⎢⎣
⎡−+⎟
⎠⎞
⎜⎝⎛ −−⎟⎟
⎠
⎞⎜⎜⎝
⎛⎟⎠⎞
⎜⎝⎛=Δ )90tan(
245tan
21 00
infdsi
re
rmh zH
KP
φψ (3.18)
where Kreinf stiffness of the reinforcement
Prm maximum reinforcement force at depth zi
H wall height
φds effective direct shear friction angle of soil
ψ angle of dilation of soil
Prm reinforcement force in zone 1 or connection force (will be
discussed in Section 2.2)
3.2.2 Connection Forces for GRS walls with Modular Block Facing
The connection forces, Prm in Equation 3.18, are defined as the forces in the
reinforcement at the back face of the wall facing. The assumptions made for the
determination of the connection forces include:
• The wall face is vertical or nearly vertical.
• There is only friction connection between adjacent facing blocks (i.e., there is
no additional mechanical connection elements, such as lips, keys, or pins).
• A uniform surcharge is being applied over the entire horizontal crest of the
wall.
• Each facing block is a rigid body, i.e., movement is allowed, but not the
deformation. As a facing block moves, the frictional resistance, F, between
two adjacent blocks will reach Fmax.
87
Consider the reinforcement at depth zi, sandwiched between two adjacent facing
blocks, as shown in Figure 3.8. The frictional forces above and below these blocks
are Fi-1 and Fi+1. The horizontal resultant force of lateral earth pressure acting on the
two facing blocks is Pi. The tensile connection force in the reinforcement at depth zi
is Ti.
Sv
± Fi+1
b
± Fi-1
Ti
reinforcement
Ni-1
Wi
Ni+1
Ffi
Pi
Top of the wall
Zi
Figure 3.8: Forces Acting on Two Facing Blocks at Depth zi
The tensile connection force Ti in the reinforcement will be:
) (3.19) ( 11 −+ −±= iiii FFPT
where ( vishi SqzKP += )γ (3.20)
If Ti should be set equal to 0, as geosynthetic reinforcement can
resist only tensile forces, i.e., Ti is always ≥ 0.
,0)( 11 <−± −+ iii FFP
From Figure 3.8,
δtan)( 11 −+ −±= iiii NNPT ; 0=iT if 0≤iT (3.21)
where Ni+1 and Ni-1 are normal forces on the top and the bottom of the two adjacent
blocks.
δtan)( fiiii FWPT +±= (3.22)
88
or,
δβγ tan)tan( vvbii SpbSPT +±= (3.23)
where
Ffi frictional resultant force between wall facing and soil
γb unit weight of facing block
b width of facing block
Sv reinforcement spacing
δ friction angle between modular block facing elements (δ can be the
friction angle between facing blocks if there is no reinforcement
between the blocks, or it can be the friction angle between facing
block and geosynthetic if there is reinforcement sandwiched between
blocks)
β friction angle between back face of wall and soil
p average net earth pressure acting on the facing, due to earth pressure
on the facing and the pressure caused by the reinforcement force. The
value of p can be estimated as:
δγ tan11 bS
FFp bv
ii =−
= −+ (3.24)
Substituting Equation 3.24 into Equation 3.23, Ti , the connection force at depth zi,
• Facing: solid masonry concrete blocks (300 mm wide by 150 mm high by 200
mm deep) with a shear key on the top surface of block, and γb = 20 kN/m3.
• Interface between facing blocks: δb-b = 57o and cb-b = 46 kPa.
Because the analytical model requires that the direct shear friction angle be used in
model calculations, and it also assumes a vertical wall face, the direct shear friction
angle of the soil and the facing batter factor were determined before using the
analytical model to evaluate the lateral movement of the facing.
(a) Direct shear friction angle:
ooo
oo
ps
psds 40
11sin44sin111cos44sin
sinsin1cossin
tan =−
=−
=ψφ
ψφφ (3.34)
99
(b) Empirical facing batter factor, Φ , from Allen and Bathurst (2001), with facing
batter of 8o: fb
88.0=⎟⎠
⎜⎝ avh
fb K
In the above equation, Kabh is the horizontal component of the active earth pressure
coefficient accounting for wall face batter, Kavh is the horizontal component of the
active earth pressure coefficient for a vertical wall, and d is a constant coef
⎟⎞
⎜⎛
=Φd
abhK (3.35)
ficient.
Allen and Bathurst (2001) found that the value of d = 0.5 would yield the best fit for
available Tmax data, and recommended using d = 0.5 for determination of Φfb.
Figure 3.15: Configuration of a Full-Scale Experiment of GRS Wall with Modular
Block Facing (Hatami and Bathurst, 2005 and 2006)
100
Figure 3.16 shows lateral movement of the GRS wall under surcharge pressures of 50
kN/m and 70 kN/m for: (a) mean value of measured displacement (Hatami and
Bathurst, 2006), (b) the Jewell-Milligan method, and (c) the analytical model (present
study). The Figure shows that the lateral movement calculated by the analytical
model is in very good agreement with the measured values. It is seen that that lateral
wall movement given by the analytical model agrees much better with the measured
values than that calculated by the Jewell-Milligan method for both surcharge
pressures. The lateral movement obtained by the Jewell-Milligan method was as
much as 3.5 times as large as the measured movement. Note that the analytical
odel, not unlike the Jewell-Milligan method, produced a displacement profile that
had lower bending stiffness than the measured profile. However, considering that the
analytical model is a simplified model, the simulation is considered adequate.
m
101
q = 50 kN/m
0
0.4
0.8
1.2
1.6
2
2.4
2.8
3.2
3.6
0 30 60 90 Lateral Displacement (mm)
Dep
th fr
om th
e To
p (m
)
Analyticalmodel
Jew ell-Milliganmethod
Averagemeasured
q = 70 kN/m
0
0.4
0.8
1.2
1.6
2
2.4
2.8
3.2
3.6
0 30 60 90 Lateral Displacement (mm)
Dep
th fr
om th
e To
p (m
)
Analyticalmodel
Jew ell-Milliganmethod
Averagemeasured
Figure 3.16: Comparisons of Measured Lateral Displacements with Jewell-Milligan Method and the Analytical Model
102
103
3.4 Summary
An analytical model has been developed to predict the profile of lateral movement of
a GRS wall with modular block facing. The connection forces in the reinforcement
can also be determined by a simple equation. The analytical model has been verified
through comparisons with Jewell-Milligan’s method. Jewell-Milligan’s method is
only a special case of the analytical model for GRS walls with negligible facing
rigidity. Comparisons were also made with a full-scale experiment of GRS wall with
modular block facing. It is shown that the analytical model offers a simple and
improved tool for predicting lateral movement of a GRS wall with modular block
facing.
4. THE GENERIC SOIL-GEOSYNTHETIC COMPOSITE (GSGC) TESTS
The understanding of soil-geosynthetic composite behavior in reinforced soil
structures has been lacking. As a result, current design methods have simply
considered the geosynthetics as added tensile elements, and have failed to account for
the interaction between soil and geosynthetics. A series of laboratory tests, referred
to as the “Generic Soil-Geosynthetic Composite” tests, or GSGC tests, were designed
and conducted to (a) examine the behavior of a generic soil-geosynthetic composite
with varying spacing and strength of reinforcement, (b) provide test data for verifying
the analytical model for calculating strength properties of a GRS composite as
described in Chapter 5, and (c) provide test data for calibration of Finite Element (FE)
model for a GRS mass. The GSGC tests were conducted at the Turner-Fairbank
Highway Research Center (TFHRC), Federal Highway Administration (FHWA) in
Mclean, Virginia.
4.1 Dimensions of the Plane Strain GSGC Test Specimen
A soil mass reinforced by layers of geosynthetic reinforcement is not a uniform mass.
To investigate the behavior of soil-geosynthetic composites by conducting laboratory
tests, it is necessary to determine the proper dimensions of the test specimen so that
the test will provide an adequate representation of soil-geosynthetic composite
behavior. Also, since most GRS structures resemble a plane strain condition, the test
specimen needs to be tested under a plane strain condition.
A number of factors were considered prior to determining the test specimen dimensions
of the GSGC test, including:
104
- Plane Strain Condition: As most GRS structures resemble a plane strain
condition, the test should be conducted in a plane strain condition.
- Backfill Particle Size: To alleviate the particle size effects on the test
specimen, the dimensions of a generic GRS mass should be at least 6 times as
large as the maximum particle size of the soil specimen, as suggested by the
U.S. Army Corps of Engineers, and 15 times larger than the average
particle size (D50) (Jewell, 1993). The recommended maximum particle
size for the backfill of GRS structures is 19 mm or ¾ in. (Elias and
Christopher, 1996). The specimen dimension, therefore, should be at least
120 mm.
- Reinforcement Spacing: The reinforcement spacing plays an important role
in the deformation behavior of GRS structures (Adams, 1999) and the load-
transfer mechanism of reinforced-soil masses (Abramento and Whittle,
1993). The height (H) of a generic GRS mass should be able to accommodate
the typical reinforcement spacing of 200 mm to 300 mm for GRS walls.
- Size of Reinforcement Sheet: The specimen dimensions in the plane strain
direction, referred to as the “width” (W) and in the longitudinal direction,
referred to as Length (L) should be sufficiently large to provide adequate
representation of the geosynthetic reinforcement. For polymer grids, enough
grid “cells” need to be included for a good representation of the polymer grid.
For nonwoven geotextiles, the aspect ratio of the reinforcement specimen
(i.e., the ratio of width to length) should be sufficiently large (say, greater
than 4) to alleviate significant “necking” effect. There will be little “necking”
effect for woven geotextiles, regardless of the aspect ratio.
A series of finite element analyses, using the computer code PLAXIS 8.2, were
conducted to examine the effect of specimen dimensions on the resulting global
stress-strain and volume change relationships of the composites. The objective of the
105
finite element analyses was to determine proper dimensions of a generic soil-
geosynthetic composite that will produce load-deformation behavior sufficiently close
to that of a large mass of soil-geosynthetic composite, referred to as the reference
composite.
Figure 4.1 shows the typical geometric and loading conditions of the GSGC tests.
The reference soil-geosynthetic composite is taken as a reinforced soil mass of
dimensions 7.0 m (23 ft) high and 4.9 m wide in a plane strain condition. Four
different dimensions of GRS composites were analyzed: specimen heights, H = 7.0
m, 2.0 m, 1.0 m and 0.5 m, while the width, W, of the test specimen was kept as
0.7*H. In these analyses, the soil was a dense sand. The sand was reinforced by a
medium-strength woven geotextile (Geotex 4x4) at 0.2 m vertical spacing. Table 4.1
lists the conditions and properties of the soil and reinforcement used in the analyses.
The global stress-strain curves obtained from the analyses are shown in Figures 4.2
and 4.4 for confining pressures, σc, of 0 and 30 kPa, respectively (note: a confining
pressure of 30 kPa is representative of the lateral stress at the mid-height of a 7.0 m
high wall). The corresponding global volume change curves are shown in Figures 4.3
and 4.5. The global vertical strain, εv, was calculated by the following equation:
%100⎟⎠⎞
⎜⎝⎛ Δ
=HH
vε (4.1)
where HΔ = total vertical displacement of the specimen; H = initial height of the specimen.
Figures 4.2 and 4.4 indicate that the composite with height H = 2.0 m, width W = 1.4
m, and under a confining pressure of 30 kPa yields stress-strain and volume change
relationships that are sufficiently close to those of the reference composite. Specimen
106
sizes with heights H = 1.0 m and 0.5 m appear too small for providing an adequate
representation of the reference composite.
Soil
Reinforcement
Sv = 0.2 m
H
σv
W = 0.7 H
σcσc
Figure 4.1: Typical Geometric and Loading Conditions of a GRS Composite
107
Table 4.1: Conditions and properties of the backfill and reinforcement used in F.E. analyses
Description
Soil
A Dense sand: unit weight = 17 kN/m3; cohesion = 5 kPa;
angle of internal friction, φ = 38o; angle of dilation, ψ = 8o; soil
modulus, E50 = 40,000 kPa; Poisson’s ratio = 0.3.
Reinforcement Geotex 4x4: axial stiffness, EA = 1,000 kN/m; ultimate
strength, Tult = 70 kN/m; reinforcement spacing = 0.2 m.
Confining
Pressure
Constant confining pressures of 0 and 30 kPa.
108
0
500
1000
1500
2000
0 5 10 15 20 25
Global Vertical Strain εv (%)
σv- σ
h (kP
a)
7.0 m x 4.9 m
2.0 m x 1.4 m
1.0 m x 0.7 m
0.5 m x 0.35 m
Specimen Dimensions:
Figure 4.2: Global Stress-Strain Curves for Soil-Geosynthetic Composites of Different Dimensions under a Confining Pressure of 0 kPa
109
-2
-1
0
1
2
3
4
5
6
0 2 4 6 8 10 12 14 16
Global Vertical Strain εv (%)
Glo
bal V
olum
etric
Str
ain
ΔV/
V (%
)
7.0 m x 4.9 m
2.0 m x 1.4 m
1.0 m x 0.7 m
0.5 m x 0.35 m
Specimen Dimensions:
Figure 4.3: Global Volume Change Curves for Soil-Geosynthetic Composites of Different Dimensions under a Confining Pressure of 0 kPa
110
0
500
1000
1500
2000
0 5 10 15 20 25
Global Vertical Strain εv (%)
σv- σ
h (kP
a)
7.0 m x 4.9 m
2.0 m x 1.4 m
1.0 m x 0.7 m
0.5 m x 0.35 m
Specimen Dimensions:
Figure 4.4: Global Stress-Strain Curves for Soil-Geosynthetic Composites of Different Dimensions under a Confining Pressure of 30 kPa
111
-1.5
-1
-0.5
0
0.5
1
1.5
2
2.5
0 2 4 6 8 10 12 14 16
Global Vertical Strain εv (%)
Glo
bal V
olum
etric
Str
ain
ΔV/
V (%
)
7.0 m x 4.9 m
2.0 m x 1.4 m
1.0 m x 0.7 m
0.5 m x 0.35 m
Specimen Dimensions:
Figure 4.5: Global Volume Change Curves for Soil-Geosynthetic Composites of Different Dimensions under a Confining Pressure of 30 kPa
112
For comparison purposes, additional analyses were conducted on unreinforced soil.
Figures 4.6 and 4.7 show, respectively, the global stress-strain curves and global
volume change curves of the soil masses without any reinforcement for the different
specimen dimensions. The results indicate that the specimen height as small as 0.5 m
will yield nearly the same stress-strain and volume change relationships as the
reference soil mass of height = 7.0 m when reinforcement is not present.
113
0.0
20.0
40.0
60.0
80.0
100.0
120.0
140.0
0 1 2 3 4 5 6
Global Vertical Strain εv (%)
σv- σ
h (k
Pa)
7.0 m x 4.9 m
2.0 m x 1.4 m
1.0 m x 0.7 m
0.5 m x 0.35 m
Specimen Dimensions:
Figure 4.6: Global Stress-Strain Curves of the Unreinforced Soil Under a Confining Pressure of 30 kPa
114
-0.4
-0.2
0
0.2
0.4
0.6
0.8
1
0 1 2 3 4 5 6
Global Vertical Strain εv (%)
Glo
bal V
olum
etric
Str
ain
ΔV/V
(%)
7m x 4.9m
2m x 1.4m
1m x 0.7m
0.5m x 0.35m
Specimen Dimensions
Figure 4.7: Global Volume Change Curves of the Unreinforced Soil Under a Confining Pressure of 30 kPa
115
Based on the results of the finite element analyses, a specimen height of 2.0 m, and
depth of 1.4 m, with 0.2 m reinforcement spacing, was selected as the specimen
dimensions for the GSGC tests. The actual specimen dimensions for the GSGC tests
are shown in Figures 4.8. Figures 4.9 and 4.10 show, respectively, the front and plan
views of the GSGC test setup.
Reinforcement
Sv = 0.2 m
H =
2.0
m
L = 1.2 m
W = 1.4 m
Figure 4.8: Specimen Dimensions for the GSGC Tests
116
160.00
81.6
5
Load cell LC1 (1000 kip)
11.38
108.00
2.625'' steel plate
70.0
0
36.00
1
3.00
1.50
6.00
6.00
5.0
2.6 6.0
76.2
5
134.
63
33.0
0
30.0
0
2.87
70.00
74.25
53.375
74.00
A 2
7.63
36.00
3
11.38
30.7
5
51.9
8
(Note: All Dimensions in Inches)
49.00
jack
12.2
5
Load cell LC2 (300 kip)
2'' steel plate
12.00
25.1
9
1.94 4.00
12.0
0
1.12
5.622.63
16.93
12.25
16.93
34.00
26.00
2.00
A
2.00
steel plate (26''x32.5''x2'')
concrete pad (49''x41 3/8''x12'')
concrete block (7.625''x7.625''x15.625'')
9.63
steel plate (43''x34''x2'')
steel plate (43''x34''x2'')
D=1.125''
FRONT VIEW
Figure 4.9: Front View of the Test Setup
117
36.00 108.00 36.00
PLAN
75.8
7
36.9
436
.94
74.0017.00 17.00
36.0
0
Open Face: Concrete blocks (7.625''x7.625''x15.625'') would be removed before testing
26.63
1 3
A
2
23.25
21.5
0
19.50
47.00
47.0
0
47.00
53.375 26.63
47.0
0
7.63
1.25
Rigid frame
Anchor
Specimen
(Note: All Dimensions in Inches)
Figure 4.10: Plan View of the Test Setup
118
4.2 Apparatus for Plane Strain Test
To maintain a plane strain condition for the GSGC specimens throughout the tests,
two major factors were considered: (1) the test bin needs to be sufficiently rigid to
have negligible lateral deformation in the longitudinal direction (i.e. the length
direction, L), and (2) the friction between the backfill and the side panels of the test
bin needs to be minimized to nearly zero.
4.2.1 Lateral deformation
Five GSGC masses were tested inside a test bin. The test bin was designed to
experience little deformation for a surcharge pressure of up to 2,800 kPa. The test bin
is shown in Figure 4.1. The details of the test bin and the design calculations are
presented in Appendix B.
4.2.2 Friction
Two transparent plexiglass panels were attached inside steel tubing frame to form the
side surfaces of the test bin. To minimize the friction between the plexiglass and the
backfill in these surfaces, a lubrication layer was created inside surfaces of the
plexiglass panels. The lubrication layer consisted of a 0.5 mm-thick latex membrane
and an approximate 1 mm-thick lubrication agent (Dow Corning 4 Electrical
Insulating Compound NSF 6). This procedure has been successful in many plane
strain tests conducted by Professor Tatsuoka at University of Tokyo and Professor
Wu at University of Colorado Denver. The friction angle between the lubricant layer
and the plexiglass as determined by direct shear test was less than one (1) degree.
119
Figure 4.11: The Test Bin
120
4.3 Test Material
The backfill and geotextile reinforcement employed in the tests are described as
follows.
4.3.1 Backfill
The backfill was a crushed Diabase from a source near Washington D.C. Before
conducting the GSGC tests, a series of laboratory tests were performed to determine
the properties of the backfill, including:
• Gradation test
• Specific gravity and absorption test of the coarse aggregates
• Moisture-Density tests (Proctor compaction) with rock correction
• Large-size triaxial tests with specimen diameter of 152 mm (6 in.)
The details of these tests are presented in Appendix A. A summary of some index
properties is given in Table 4.2. The grain size distribution of the soil is shown in
Figure 4.12. Two gradation tests were performed. The results agree well with each
other.
Four triaxial tests were conducted at confining pressures of 5 psi, 15 psi, 30 psi, and
70 psi, and the results were compared with those performed by Ketchart et al. on the
same soil. The soil specimen was approximately 6 in. in diameter and 12 in. in
height. The shapes of a typical specimen before and after failure are shown in Figure
4.13. Figure 4.14 presents the stress-strain curves and volume change curves of the
tests. The stress-strain curves obtained by Ketchart et al. are also included for
comparisons and for furnishing a more complete set of data. The stress-strain
relationships agree well in trend with those by Ketchart et al. The Mohr-Coulomb
121
failure envelops of the backfill are shown in Figure 4.15. For confining pressures
between 0 and 30 psi, the strength parameters are: c = 10.3 psi, φ = 50o. For
confining pressures between 30 and 110 psi, the strength parameters are: c = 35.1 psi,
φ = 38o.
122
Table 4.2: Summary of some index properties of backfill
Classification Well graded gravel: A-1a, per AASHTO M-15; and
GW-GM, per ASTM D2487.
Maximum Dry Unit Weight 24.1 kN/m3 (153.7 lb/ft3)
Optimum Moisture Content 5.2 %
Specific Gravity of Soil
Solids
3.03
123
0
10
20
30
40
50
60
70
80
0.010.1110100
Grain size (mm)
Perc
ent f
iner
(%)
Test 1
Test 2
Figure 4.12: Grain Size Distribution of Backfill
124
(a) (b)
Figure 4.13: Typical Triaxial Test Specimen (a) before and (b) after Test
125
Deviatoric Stress versus Axial Strain Relationships
0
50
100
150
200
250
300
350
400
450
500
550
600
0 1 2 3 4 5 6 7 8 9 10 11
Axial Strain (%)
Dev
iato
ric
Stre
ss (p
si)
5 psi
15 psi
30 psi
70 psi
70 psi - Ketchart
110 psi - Ketchart
(a)
Volumetric Strain versus Axial Strain
-0.4
-0.2
0
0.2
0.4
0.6
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0
Axial Strain (%)
Volu
met
ric
Stra
in (%
)
5 psi
30 psi
(b)
Figure 4.14: Triaxial Test Results:
(a) Stress-Strain Curves of Backfill at 24.1 kN/m3 Dry Density and 5.2% Moisture (b) Volume Change Curves of Backfill at Confining Pressures of 5 psi and 30 psi
126
σ3 = 15 psi
σ3 = 30 psi
Shea
r St
rees
s (ps
i)
Normal Stress (psi)
σ3 = 5 psi
σ3 = 70 psi
σ3 = 110 psi
433265172 61698300
C2 = 35.1 psi
C1 = 10.3 psi
Figure 4.15: Mohr-Coulomb Failure Envelops of Backfill
127
4.3.2 Geosynthetics
The geosynthetic used in the experiments was Geotex 4x4 manufactured by Propex,
Inc. (formally known as Amoco 2044). This geosynthetic is a woven polypropylene
geotextile, with its strength properties provided by the manufacture shown in Table
4.3.
Table 4.3: Summary of Geotex 4x4 properties provided by the manufacture
Property Test Method Machine Direction
(Wrap Direction)
Cross-Direction
(Fill Direction)
Tensile Strength
(Grab)
ASTM D-4632 2.67 kN
(0.6 kips)
2.22 kN
(0.5 kips)
Wide-Width
Tensile Ultimate
Strength
ASTM D-4595 70 kN/m
(400 lb/in.)
70 kN/m
(400 lb/in.)
Wide-Width
Strength at 5%
Strain
ASTM D-4595 21 kN/m
(121 lb/in.)
38 kN/m
(217 lb/in.)
Wide-Width
Ultimate
Elongation
ASTM D-4595 10 % 10 %
Puncture ASTM D-4833 0.8 kN (170 lb)
Trapezoid
Tearing Strength
ASTM D-4533 1.11 kN (250 lb)
128
Two types of geosynthetics were used for the experiments: a single-sheet of Geotex
4x4, and a double-sheet Geotex 4x4 (by gluing two sheets of Geotex 4x4 together
using 3M Super 77 spray adhesive). The use of the double-sheet was to create a
geosynthetic that is approximately twice as stiff (and as strong), yet maintaining the
same interface condition as that of the single-sheet geosynthetic. Geotex 4x4
geotextile has been used in the construction of hundreds of production GRS walls and
in many full-scale experiments, including the FHWA GRS pier (Adams, 1997),
Havana Yard Test abutment and pier (Ketchart and Wu, 1997), Blackhawk preloaded
GRS bridge abutment (Ketchart and Wu, 1998), and NCHRP test abutments (Wu, et
al., 2006).
Uniaxial tension tests were performed on both types of geosynthetic to determine the
load-deformation behavior using specimen dimensions of Width = 305 mm (12 in.)
and Length = 152 mm (6 in.), see Figure 4.16. The stiffness and strength of the two
geosynthetics are shown in Table 4.4 and load-deformation curves are shown in
Figure 4.17. It is seen that the stiffness and the strength of the double-sheet Geotex
4x4 are approximately twice as much as those of the single-sheet Geotex 4x4, with
the breakage strain being almost the same.
129
Figure 4.16: Uni-Axial Tension Test of Geotex 4x4
130
Table 4.4: Properties of Geotex 4x4 in fill-direction
Geosynthetics Wide-Width Tensile Strength,
per ASTM 4595
Stiffness (kN/m) at
1% Strain
Ultimate Strength (kN/m)
(% at break)
Single-Sheet
Geotex 4x4
1,000 70 (12 %)
Double-Sheet
Geotex 4x4
1,960 138 (12 %)
131
0
20
40
60
80
100
120
140
160
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Axial Strain (%)
Tens
ile L
oad
(kN
/m)
Single-Sheet
Double-Sheet
Figure 4.17: Load-Deformation Curves of the Geosynthetics
4.3.3 Facing Block
Blocks used for the facing of the GSGC mass during specimen preparation were
hollow concrete blocks with the dimensions of 397 mm x 194 mm x 194 mm (15.625
in. x 7.625 in. x 7.625 in.) and with the average weight of 18.1 kg/block (40
lbs/block).
132
4.4 Test Program
The test program comprises five GSGC tests, with their test conditions shown in
Table 4.5. The plate compactor, MBW - GP1200, used for the tests has the following
specifications: weight = 120 lbs; plate dimensions = 12 in. x 21 in.; centrifugal force
The vertical loads were applied to the test specimens by using a 1,000,000-pound
capacity loading frame with a 1,000,000-pound hydraulic jack. Loads were measured
by load cells and by hydraulic jack pressure gauges. For Test 1, two load cells of
100,000 pounds and 300,000 pounds were used to measure the loads. For Tests 2 to
5, a 1,000,000 pound load cell was used to measure the loads. A 12 in.-thick concrete
pad was placed on top of the specimen before loading. Vertical loads were applied in
equal increments with ten-minute elapsed time between increments to allow time for
equilibrium. The elapsed time also allow manual recording of displacements of the
test specimen. The vertical loads were applied until a failure condition was reached
to determine the strength of the composite specimen. The applied pressures on the
composite specimens were determined from the applied vertical loads divided by the
surface areas of the composite specimens.
4.5.2 Confining Pressure
The confining pressure on the test specimens was applied by vacuuming. The entire
surface area of the test specimen was vacuum-sealed with a 0.5-mm thick latex
membrane. A prescribed confining pressure of 34 kPa (5 psi) for tests 1 to 4 was applied
by connecting the latex membrane to a suction device through two 6-mm diameter
flexible plastic tubes. Only Test 5 was conducted without confining pressure.
4.5.3 Instrumentation
The specimens were instrumented to monitor their performance during tests. The
instruments used include:
134
a. Vertical Movement
Three Linear Variable Displacement Transducers (LVDT) and two digital dial
indicators were installed on the top of the concrete pad to measure the vertical
movement of the specimen during loading. The vertical movement was measured
along the top surface of the concrete pad.
b. Lateral Movement
Ten LVDTs and two digital dial indicators were installed along the height of the
specimen (two open sides of the specimen) to measure the lateral movement of the
specimen. The location of the LVDTs and digital dial indicators are shown in Figure
4.18. Figure 4.37 shows the LVDTs on a test specimen to monitor the lateral
movement.
c. Internal Movement
The internal movement of the soil at selected points in the soil mass was traced by
marking the locations of pre-selected points on a 2 in. x 2 in. grid system drawn on
the membrane. The locations of the selected points for the tests are depicted in Figure
4.38.
135
FRONT VIEWSIDE VIEW
Cross Section A-A
TOP VIEW
72.25'' (1835.15) 47.37'' (1203.32)
1-1/4''
73.72 (1872.49)
55.12'' (1400)
5.35''
11.37
(288
.92)
10.4
(264
.16)
11.37
(288
.92)
11.37
(288
.92)
11.37
(288
.92)
11.37
(288
.92)
11.37
(288
.92)
76.3'
' (193
8)
81.65
'' (20
73.91
)
304.8
13.0
(332
.1)13
.0(3
32.1)
11.0
(297
.4)13
.0(3
32.1)
2.34
(59.4
9)
LVDT
Dial Indicator
13.0
(332
.1)30
4.82.3
4(5
9.49)
11.0
(297
.4)13
.0(3
32.1)
13.0
(332
.1)
Dial Indicator
LVDT
13.0
(332
.1)
81.65
'' (20
73.91
)
76.3'
' (193
8)
13.0
(332
.1)
304.8
304.8
Dial Indicator
13.0
(332
.1)13
.0(3
32.1)
11.0
(297
.4)2.3
4(5
9.49)
5.35'' 55.12'' (1400)
72.25'' (1835.15)
2.34
(59.4
9)11
.0(2
97.4)
13.0
(332
.1)13
.0(3
32.1)
LVDT LVDT
Dial Indicator
GSGC MASS
13.0
(332
.1)
11.37
(288
.92)
81.65
'' (20
73.91
)
76.3'
' (193
8)
304.8
10.4
(264
.16)
13.0
(332
.1)13
.0(3
32.1)
11.0
(297
.4)2.3
4(5
9.49)
11.37
(288
.92)
11.37
(288
.92)
11.37
(288
.92)
11.37
(288
.92)
11.37
(288
.92)
Dial Indicator
2''
LVDT
LVDTs on the open side
Dial Indicator on the open side
6''
Dial Indicator on the top surface
LVDT on the top surface
LVDT on the top surface (the location can be adjusted dependent on the size of the loading system)
Footing on the top of GRS specimen
AJack
Center line
6''
Dial Indicator
4''4''
Dial Indicator on the top surface
Location of LVDTs and Dial Indicators
Figure 4.18: Locations of LVDTs and Digital Dial Indicators
136
d. Reinforcement Strain
To measure the strains in the geotextile, a number of high elongation strain gauge
(type EP-08-250BG-120), manufactured by Measurements Group, Inc., were used.
Each strain gauge was glued to the geotextile only at two ends to avoid inconsistent
local stiffening of geotextile due to the adhesive. The strain gauge attachment
technique was developed at the University of Colorado Denver. The gauge was first
mounted on a 25 mm by 76 mm patch of a lightweight nonwoven geotextile (see
Figure 4.19). A Microcystalline wax and a rubber coating (M-Coat B, Nitrile Rubber
coating) were used to protect the gauges from moisture. To check the effectiveness
of the moisture-protection technique, the geotextile specimens with the strain gauges
were tested after immersing in water for 24 hours. Before placing the reinforcement
sheet in the test specimen, an M-Coat FB-2, 6694, Butyl Rubber Tape was used to
cover the gauges to protect the gauges during compaction (see Figure 4.19). To
measure the strain distribution of the reinforcement, six strain gauges were mounted
on each Geotex 4x4 sheet (see Figure 4.20).
137
(a) (b)
Figure 4.19: Strain Gauge on Geotex 4x4 Geotextile: (a) Before Applying Protection Tape (b) After Applying Protection Tape (M-Coat FB-2, 6694, Butyl Rubber Tape).
138
Figure 4.20: Strain Gauges Mounted on Geotex 4x4 Geotextile
139
Due to the presence of the light-weight geotextile patch, calibration of the strain
gauge is needed. The calibration tests were performed to relate the strain obtained
from the strain gauge to the actual strain of the reinforcement. Figures 4.21 and 4.22
show the calibration curves along the fill direction of Geotex 4x4 geotextile for the
single-sheet and the double-sheet specimens, respectively.
140
y = 1.172xR2 = 0.9913
0
1
2
3
4
5
6
0 1 2 3 4 5Strain from Strain Gage (%)
Stra
in fr
om In
stro
n M
achi
ne (%
)
Figure 4.21: Calibration Curve for Single-Sheet Geotex 4x4
141
y = 1.078xR2 = 0.9986
0
1
2
3
4
5
6
0 1 2 3 4 5
Strain from Strain Gauge (%)
Stra
in fr
om In
stro
n M
achi
ne (%
)
Figure 4.22: Calibration Curve for Double-Sheet Geotex 4x4
142
4.5.4 Preparation of Test Specimen for GSGC Tests
The preparation procedure of a typical composite mass with the dimensions of 2.0 m
(H) x 1.4 m (D) x 1.2 m (L) is described as follows:
1. Mark the anticipated location of the GSGC mass on the plexiglass;
2. Apply approximately 1-mm thick lubricating agent (Dow Corning 4
Electrical Insulating Compound NSF 61) evenly on the inside surfaces
of the plexiglass (see Figure 4.23);
3. Attach a sheet of membrane (with 51 mm x 51 mm grid system pre-
drawn on membrane) over each plexiglass and at the bottom of the
specimen (see Figure 4.24);
4. Lay a course of the facing blocks on the open sides of the specimen (see
Figure 4.25);
5. Place the backfill in the test bin and compact in 0.2 m lifts (see Figures
4.26 and 4.27); check and adjust (if needed) the backfill moisture before
compaction to achieve the target moisture of 5.2%;
6. Check the water content and dry unit weight of each lift by using a
nuclear density gauge, Troxler 3440, by the direct transmission method
(note: the measured dry unit weights of five tests are shown in Figure
4.39);
7. Place the next layer of geosynthetic reinforcement (with strain gauges
already mounted) covering the entire top surface area of compacted fill
and the facing blocks (see Figure 4.28);
8. Repeat steps 4 to 8 until the full height of the composite mass is reached;
9. Sprinkle a 5 mm-thick fine sand layer over the top surface of the
completed composite mass to level the surface and protect the
membrane from being punctured by gravels in the backfill (see Figure
4.29);
143
10. Place a geosynthetic sheet on top of the composite mass (see Figure
4.30);
11. Glue a sheet of membrane to the top edge of the side membrane sheets
(see Figure 4.31);
12. Remove all facing blocks and trim off the excess geotextile (see Figure
4.32);
13. Insert strain gauge cables through the plastic openings that were already
attached on the membrane sheets at prescribed locations (see Figure
4.33);
14. Glue membrane sheets to enclose entire composite mass;
15. Apply vacuum to the composite mass at a low pressure of 14 kPa (see
Figure 4.34);
16. Seal the connection between cables and membrane with epoxy to
prevent air leaks (see Figure 4.35). The low vacuum pressure allows the
epoxy to seal the connection well;
17. Raise the vacuum pressure to 34 kPa and check air leaks under
vacuuming (see Figure 4.36) and measure the specimen dimensions (see
Table 4.6 for specimen dimensions of five tests).
144
Figure 4.23: Applying Grease on Plexiglass Surfaces
145
Figure 4.24: Attaching Membrane
146
Figure 4.25: Placement of the first Course of Facing Block
147
Figure 4.26: Compaction of the First Lift of Backfill
148
Figure 4.27: Placement of Backfill for the Second Lift
149
Figure 4.28: Placement of a Reinforcement Sheet
150
Figure 4.29: Completion of Compaction of the Composite Mass and Leveling the Top Surface with 5 mm-thick Sand Layer
151
Figure 4.30: Completed Composite Mass with a Geotextile Sheet on the Top Surface
152
Figure 4.31: Covering the Top Surface of the Composite Mass with a Sheet of Membrane
153
Figure 4.32: Removing Facing Blocks and Trimming off Excess Geosynthetic Reinforcement
154
Figure 4.33: Insertion of Strain Gauge Cables through the Membrane Sheet
155
Figure 4.34: Vacuuming the Composite Mass with a Low Pressure
156
Figure 4.35: Sealing the Connection between Cable and Membrane with Epoxy to Prevent Air Leaks
157
Figure 4.36: Checking Air Leaks under Vacuuming
158
Figure 4.37: The LVDT’s on an Open Side of Test Specimen
159
2.50
2.5024.50
42.0
0
66.0
0
16.0
0
1 2 3
4 5 6
7 8 9
2x2'' Grid System
10.2
5
60.2
5
34.2
5
32.50
54.0
0
16.0
0
42.0
0
31.383.38
3.38
4
7
10.3
5
10
1
65
34.3
5
8 9
2x2'' Grid System
11
2 3
(b): Tests 3,4 and 5
Width
Width
Hei
gth
= 76
.25
in. (
Test
1),
76.3
5 in
. (Te
st 2
)H
eigh
t = 7
6.30
in.
15.38
(a): Tests 1 and 2
Figure 4.38: Locations of Selected Points to Trace Internal Movement of (a) Tests 1 and 2 and (b) Tests 3, 4 and 5
160
0.00
0.20
0.40
0.60
0.80
1.00
1.20
1.40
1.60
1.80
2.00
15 18 21 24 27 30
Dry Unit Weight (kN/m3)
Spec
imen
Hei
ght (
m)
Test 1
Test 2
Test 3
Test 4
Test 5
Figure 4.39: Soil Dry Unit Weight Results during Specimen Preparation of the Five
Tests
161
Table 4.6: Dimensions of the GSGC specimens before loading
Test Height, m (in.)
Width, m (in.)
Length, m (in.)
Test 1 1.937 (76.25) 1.448 (57.00) 1.194 (47.00)
Test 2 1.939 (76.35) 1.372 (54.00) 1.187 (46.75)
Test 3 1.939 (76.35) 1.346 (53.00) 1.187 (46.75)
Test 4 1.938 (76.30) 1.492 (58.75) 1.187 (46.75)
Test 5 1.939 (76.35) 1.245 (49.00) 1.187 (46.75)
162
4.6 Test Results
4.6.1 Test 1 – Unreinforced Soil
This test is perhaps the largest plane strain test for soil with a confining pressure.
Test 1 was conducted as the base line for the other four GSGC tests.
The loading sequence of the soil mass was:
• Loading up to a vertical pressure of 250 kPa (nearly 1% vertical strain)
• Unloading to zero
• Reloading until a failure pressure of 770 kPa was reached.
The soil mass at failure is shown in Figure 4.40. Figure 4.41 shows the global
vertical stress-strain and volume change relationships of the soil mass. The average
lateral displacements, measured by LVDT’s, on the open faces of the soil mass under
different vertical stresses are presented in Figure 4.42. The internal displacements of
the soil at selected points under vertical applied pressures of 190 kPa, 310 kPa, 620
kPa, and 770 kPa are shown in Figure 4.43. The test results of Test 1 for
unreinforced soil are summarized in Table 4.7.
163
Figure 4.40: Soil Mass at Failure of Test 1
164
0
100
200
300
400
500
600
700
800
900
0 1 2 3 4 5Global Vertical Strain (%)
App
lied
Vert
ical
Str
ess
(kPa
)
6
(a)
-0.4
-0.2
0
0.2
0.4
0.6
0.8
1
1.2
0 0.5 1 1.5 2 2.5 3 3.5
Global Vertical Strain (%)
Volu
met
ric S
train
(%)
(b)
Figure 4.41: Test 1-Unreinforced Soil Mass:
(a) Global Vertical Stress-Vertical Strain Relationship (b) Global Volume Change Strain Relationship
165
0.00
0.25
0.50
0.75
1.00
1.25
1.50
1.75
2.00
0 10 20 30 40 50
Lateral Displacement (mm)
Spec
imen
Hei
ght (
m)
200 kPa
400 kPa
600 kPa
770 kPa
700 kPa
Applied Pressure:
Figure 4.42: Lateral Displacements on the Open Face of Test 1
166
76.2
5
Test 1
57.00
2.502.50
42.0
0
66.0
0
16.0
0
1
2'' x 2'' Grid System
10.2
5
34.2
5
32.50
4 5 6
2 3
8 97
Before Loading
190 kPa (75 kips)
310 kPa (120 kips)
620 kPa (240 kips)
770 kPa (300 kips)
Legend:
(All Displacements in inches, Drawn to Scale of the Soil Mass)
Note:
Figure 4.43: Internal Displacements of Test 1
167
Table 4.7: Some test results of Test 1
Test 1
Test Conditions
Geosynthetic Reinforcement None
Wide-Width Strength of Reinforcement None
Reinforcement Spacing No Reinforcement
Confining Pressure 34 kPa
Test Results
Applied Stress at Vertical Strain of 1% 335 kPa (48.6 psi)
Ultimate Applied Pressure 770 kPa (112 psi)
Vertical Strain at Failure 3 %
Maximum Lateral Displacement of the Open Face at Failure 47 mm
Stiffness at 1% vertical strain (Eat 1%) 33,500 kPa
Stiffness for Unloading-Reloading (Eur) 87,100 kPa
168
4.6.2 Test 2 – GSGC Test (T, Sv)
In this test, the GSGC mass was reinforced by nine sheet of single-sheet Geotex 4x4
with spacing of 0.2 m. The soil layer was compacted at 0.2 m-thick lifts. Each
reinforcement sheet was mounted with 54 strain gauges.
The failure load in this test was 1,000,000 pounds. All nine reinforcement sheets
were ruptured after testing. The composite mass after testing is shown in Figure 4.44.
The shear bands of the composite mass after testing are visible through the diagonal
lines of the mass (see Figures 4.44 and 4.45). Along the shear bands, the square grids
of 51 mm by 51 mm (2 in. by 2 in.) were severely distorted after testing (Figure 4.45).
These shear bands correspond exactly with the failure surfaces seen in Figure 4.46.
The location of rupture lines of all reinforcement sheets in the GSGC mass can be
seen in Figures 4.52 and 4.53.
The measured data of Test 2 are highlighted as bellows:
• Global stress-strain relationship: Figure 4.47 shows the global stress-strain
relationship of the composite up to and post failure. The maximum applied
vertical pressure was about 2700 kPa, where the corresponding vertical
displacement was 125 mm (6.5% vertical strain).
• Lateral displacement: The average lateral displacement profiles are on the
open faces of the composite under different vertical pressures shown in Figure
4.48. The lateral displacements were nearly uniform along the height of the
composite up to a pressure of about 600 kPa. At vertical pressures between
770 kPa and 1,500 kPa, the maximum lateral displacement occurred at about
3/8 H (H = the height of the composite mass) from the base. The locations of
the maximum displacements were about the same as those of Test 1
(unreinforced).
169
Figure 4.44: Composite Mass at Failure of Test 2
170
Figure 4.45: Close-Up of Shear Bands at Failure of Area A in Figure 4.44
171
Figure 4.46: Failure Planes of the Composite Mass after Testing in Test 2
172
0
500
1,000
1,500
2,000
2,500
3,000
0 1 2 3 4 5 6 7 8 9
Global Vertical Strain (%)
App
lied
Verti
cal S
tress
(kPa
)
(a)
-1.8
-1.6
-1.4
-1.2
-1
-0.8
-0.6
-0.4
-0.2
00 1 2 3 4 5 6
Global Vertical Strain (%)
Vol
umet
ric
Stra
in (%
)
7
(b)
Figure 4.47: Test 2-Reinforced Soil Mass:
(a) Global Vertical Stress-Vertical Strain Relationship (b) Global Volume Change Strain Relationship
173
0
0.25
0.5
0.75
1
1.25
1.5
1.75
2
0 10 20 30 40 50 60 70
Lateral Displacement (mm)
Spec
imen
Hei
ght (
m)
200 kPa
400 kPa
600 kPa
770 kPa
1000 kPa
1250 kPa
1500 kPa
1750 kPa
2000 kPa
2250 kPa
2500 kPa
2700 kPa
Applied Pressure:
Figure 4.48: Lateral Displacements on the Open Face of Test 2
174
The maximum lateral displacement at the mid-height of the composite under
the applied pressure of 2,700 kPa was 60 mm.
• Internal displacement: The internal displacements of the composite mass at
selected points under vertical applied pressures of 270 kPa to 2,700 kPa, at an
increment of 270 kPa are shown in Figure 4.49. At points 1 (and 3), 4 (and 6),
and 7 (and 9) near the open faces, the displacements moved downward and
outward with an angle, measured after testing, of about 67o, 47o, and 31o,
respectively, to the horizontal. The vertical displacements of the points at the
upper part of the soil mass were greater than those at the lower part. Along
the center line, there were only vertical displacements (Points 2, 5 and 8).
There was almost no displacement at Point 8 near the bottom and at the center
line. The maximum lateral displacement in the soil body was 60 mm at the
mid height on the open sides, and the maximum vertical displacement was
125 mm at the top of the specimen.
• Reinforcement strain: Figure 4.50 shows the locations of the strain gauges on
the geosynthetic sheets. The strain in the reinforcement of the GSGC mass is
shown in Figure 4.51. Most of the strain gauges performed well at strains less
than 4%. All reinforcement layers were found ruptured after the test
completed. The locations of the rupture lines can be seen from the aerial view
of the reinforcement sheets exhumed from the composite after testing (see
Figure 4.52). Based on the locations of the rupture lines, the rupture planes
can be constructed as shown in Figure 4.53. Note that this agrees perfectly
with the shear bands in Figure 4.46. The maximum strain in reinforcement at
ruptured was about 12 %, while the measured data from strain gauges were
only less than 4 %. From the strain distribution in Figure 4.51, the locations
of the maximum strain in reinforcement were different between layers. In
reinforcement layers near at the mid height of the GSGC mass (0.8 m and 1.0
175
m from the base), the maximum reinforcement strains were close to the
centerline, while the reinforcement layers at near the top and the base of the
GSGC mass (0,2 m and 1.8 m from the base), the maximum reinforcement
strains were at about 0.3 m from the edge of the composite mass. The
maximum strain locations in all reinforcement layers were at the ruptured
lines of reinforcement that can be seen in Figures 4.52 and 4.53.
• The test results of Test 2 are summarized in Table 4.8.
176
2'' x 2'' Grid System
76.3
5
Test 2
54.00
3.00
3.00
38.0
0
66.0
0
16.0
0
1 2 3
4 5 6
7 8 9
10.3
5
38.3
5
31.00
Before Loading
270 kPa
540 kPa
2,700 kPa
...
Legend:
(All Displacements in inches, Drawn to Scale of the Composite Mass)
Note:
Figure 4.49: Internal Displacements of Test 2
177
Strain Gages
11.00 8.00 8.00 8.00 8.00 11.00
54
7.63
7.63
7.63
7.63
7.63
7.63
7.63
7.63
7.63
7.63
Sheet No 9
Sheet No 8
Sheet No 7
Sheet No 6
Sheet No 5
Sheet No 4
Sheet No 3
Sheet No 2
Sheet No 1
Reinforcement
(Note: All Dimensions in inches)
Figure 4.50: Locations of Strain Gauges on Geosynthetic sheets in Test 2
178
0
0.5
1
1.5
2
2.5
0.0 0.1 0.2 0.3 0.4 0.5 0.6
Distance from the Edge of the Composite Mass (m)
Stra
in (%
)
200 kPa400 kPa
600 kPa800 kPa
1000 kPa1250 kPa
1500 kPa
Applied Pressure:
(a)
0.00
0.50
1.00
1.50
2.00
2.50
3.00
0.0 0.1 0.2 0.3 0.4 0.5 0.6
Distance from the Edge of the Composite Mass (m)
Stra
in (%
)
200 kPa
400 kPa
600 kPa
800 kPa
1000 kPa
1250 kPa
1500 kPa
Applied Pressure:
(b)
Figure 4.51: Reinforcement Strain Distribution of the Composite Mass in Test 2:
(a) Layer 1, at 0.2 m from the Base (b) Layer 2, at 0.4 m from the Base
179
00.5
11.5
22.5
33.5
44.5
0.0 0.1 0.2 0.3 0.4 0.5 0.6
Distance from the Edge of the Composite Mass (m)
Stra
in (%
)
200 kPa
400 kPa
600 kPa
800 kPa
1000 kPa
1250 kPa
1500 kPa
Applied Pressure:
(c)
0
0.5
1
1.5
2
2.5
3
3.5
0.0 0.1 0.2 0.3 0.4 0.5 0.6
Distance from the Edge of the Composite Mass (m)
Stra
in (%
)
200 kPa
400 kPa
600 kPa
800 kPa
1000 kPa
Applied Pressure:
(d)
Figure 4.51 (continued): Reinforcement Strain Distribution of the Composite
Mass in Test 2: (c) Layer 3, at 0.6 m from the Base (d) Layer 4, at 0.8 m from the Base
180
0
0.5
1
1.5
2
2.5
3
0.0 0.1 0.2 0.3 0.4 0.5 0.6
Distance from the Edge of the Composite Mass (m)
Stra
in (%
) 200 kPa
400 kPa
600 kPa
800 kPa
1000 kPa
Applied Pressure:
(e)
0
0.5
1
1.5
2
2.5
3
0.0 0.1 0.2 0.3 0.4 0.5 0.6
Distance from the Edge of the Composite Mass (m)
Stra
in (%
) 200 kPa
400 kPa
600 kPa
800 kPa
1000 kPa
Applied Pressure:
(f)
Figure 4.51 (continued): Reinforcement Strain Distribution of the Composite Mass in Test 2:
(e) Layer 5, at 1.0 m from the Base (f) Layer 6, at 1.2 m from the Base
181
00.5
11.5
22.5
33.5
4
0.0 0.1 0.2 0.3 0.4 0.5 0.6
Distance from the Edge of the Composite Mass (m)
Stra
in (%
)
200 kPa
400 kPa
600 kPa
800 kPa
1000 kPa
1250 kPa
Applied Pressure:
(g)
0
0.5
1
1.5
2
2.5
3
0.0 0.1 0.2 0.3 0.4 0.5 0.6
Distance from the Edge of the Composite Mass (m)
Stra
in (%
)
200 kPa
400 kPa
600 kPa
800 kPa
1000 kPa
1250 kPa
Applied Pressure:
(h)
Figure 4.51 (continued): Reinforcement Strain Distribution of the Composite
Mass in Test 2: (g) Layer 7, at 1.4 m from the Base (h) Layer 8, at 1.6 m from the Base
182
00.20.40.60.8
11.21.41.6
0.0 0.1 0.2 0.3 0.4 0.5 0.6
Distance from the Edge of the Composite Mass (m)
Stra
in (%
) 200 kPa
400 kPa
600 kPa
800 kPa
Applied Pressure:
(i)
Figure 4.51 (continued): Reinforcement Strain Distribution of the Composite Mass in Test 2:
(i) Layer 9, at 1.8 m from the Base
183
Figure 4.52: Aerial View of the Reinforcement Sheets Exhumed from the Composite Mass after Test 2
(The numbers next to each sheet indicate the sheet number in Figure 4.53)
184
Rupture Lines
547.
637.
637.
637.
637.
637.
637.
637.
637.
637.
63
1
2
3
4
5
6
7
8
9
ReinforcementTest 2
(Note: All Dimensions in inches, Drawn to Scale)
Figure 4.53: Locations of Rupture Lines of Reinforcement in Test 2; Constructed based on Figure 4.52
185
186
Table 4.8: Some test results of Test 2
Test 2
Test Conditions
Geosynthetic Reinforcement Geotex 4x4
Wide-Width Strength of Reinforcement 70 kN/m
Reinforcement Spacing 0.2 m
Confining Pressure 34 kPa
Test Results
Ultimate Applied Pressure 2700 kPa
Vertical Strain at Failure 6.5 %
Maximum Lateral Displacement of the Open Face at Failure 60 mm
Stiffness at 1% vertical strain (Eat 1%) 61,600 kPa
Maximum strain in reinforcement at ruptured 12 %
Maximum measured strain in reinforcement 4.0 %
4.6.3 Test 3 – GSGC Test (2T, 2Sv)
In this test, the GSGC mass was reinforced by four double-sheet Geotex 4x4 at 0.4 m
spacing. The strength and stiffness of the double-sheet reinforcement were nearly
doubled compared to those of the single-sheet reinforcement used in Test 2. The
GSGC mass after testing is shown in Figure 4.54.
The measured data of Test 3 are highlighted below:
• Global stress-strain relationship (see Figure 4.55): The maximum applied
vertical pressure was about 1,750 kPa. The vertical displacement at the failure
pressure was 118 mm (6.1% global vertical strain).
• Lateral displacement: The average lateral displacement profiles on the open
faces of the composite under different vertical pressures shown in Figure 4.56.
The maximum lateral displacement under the failure pressure 1750 kPa was
54 mm.
• Internal displacements at selected points are shown in Figure 4.57. The trend
of the internal movements in Test 3 was nearly the same as that in Test 2. The
points 1, 3, 4, 5, 6, 7 and 9 near the open faces, the displacements move
downward and outward with angles from 49o to 71o to the horizontal. The
points 2, 5 and 8 along the center line, the displacements were almost vertical.
The maximum vertical displacement at the top of the specimen was 118 mm.
• Reinforcement strain: Figure 4.58 shows the locations of the strain gauges on
the geosynthetic sheets. The strain in the reinforcement of the GSGC mass is
shown in Figure 4.59. Three reinforcement layers near the top of the
composite were ruptured after testing. The locations of the rupture lines can
be seen from the aerial view of the reinforcement sheets exhumed from the
composite after testing (see Figure 4.60). Based on the locations of the
187
rupture line, the rupture planes can be constructed as shown in Figure 4.61.
This rupture line agrees perfectly with the failure line in Figure 4.54. The
maximum strain in reinforcement measured was 4 % and located at near the
ruptured line as shown in Figure 4.61.
• The test results of Test 3 are summarized in Table 4.9.
188
Figure 4.54: Composite Mass after Testing of Test 3
189
0
200
400
600
800
1,000
1,200
1,400
1,600
1,800
2,000
0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00
Global Vertical Strain (%)
App
lied
Verti
cal S
tress
(kPa
)
Figure 4.55: Global Stress-Strain Relationship of Test 3
190
0
0.25
0.5
0.75
1
1.25
1.5
1.75
2
0.0 10.0 20.0 30.0 40.0 50.0 60.0
Lateral Displacement (mm)
Spe
cim
en H
eigh
t (m
)
260 kPa
400 kPa
600 kPa
800 kPa
1000 kPa
1250 kPa
1500 kPa
1750 kPa
Applied Pressure:
Figure 4.56: Lateral Displacements on the Open Face of Test 3
191
76.3
5
Test 3
53.00
2.502.50
42.0
0
66.0
0
16.0
0
2x2'' Grid System
10.3
5
60.3
6
34.3
5
54.0
0
1 2
4
10 11
65
987
2'' x 2'' Grid System
3
30.50
Before Loading
280 kPa
560 kPa
840 kPa
1120 kPa
1400 kPa
1680 kPa
1750 kPa
14.5
Legend:
(All Displacements in inches, Drawn to Scale of the Composite Mass)
Note:
Figure 4.57: Internal Displacements of Test 3
192
Strain Gages
10.50 8.00 8.00 8.00 8.00 10.50
53.00
15.2
515
.25
15.2
515
.25
15.2
5
Sheet No 4
Sheet No 3
Sheet No 2
Sheet No 1
Reinforcement
(Note: All Dimensions in inches)
Figure 4.58: Location of Strain Gauges on Geosynthetic Sheets in Test 3
193
00.5
11.5
22.5
33.5
44.5
0.0 0.1 0.2 0.3 0.4 0.5 0.6
Distance from the Edge of the Composite Mass (m)
Stra
in (%
)
260 kPa
400 kPa
600 kPa
800 kPa
1000 kPa
1250 kPa
`
Applied Pressure:
(a)
00.5
11.5
22.5
33.5
44.5
0.0 0.1 0.2 0.3 0.4 0.5 0.6
Distance from the Edge of the Composite Mass (m)
Stra
in (%
) 260 kPa
400 kPa
600 kPa
800 kPa
1000 kPa
Applied Pressure:
(b)
Figure 4.59: Reinforcement Strain Distribution of the Composite Mass in Test 3:
(a) Layer 1, at 0.4 m from the Base (b) Layer 2, at 0.8 m from the Base
194
0.00.51.01.52.02.53.03.54.04.5
0.0 0.1 0.2 0.3 0.4 0.5 0.6
Distance from the Edge of the Composite Mass (m)
Stra
in (%
) 260 kPa
400 kPa
600 kPa
800 kPa
1000 kPa
Applied Pressure:
(c)
0.00.51.01.52.02.53.03.54.04.5
0.0 0.1 0.2 0.3 0.4 0.5 0.6
Distance from the Edge of the Composite Mass (m)
Stra
in (%
) 260 kPa
400 kPa
600 kPa
800 kPa
1000 kPa
Applied Pressure:
(d)
Figure 4.59 (continued): Reinforcement Strain Distribution in Test 3:
(c) Layer 3, at 1.2 m from the Base (d) Layer 4, at 1.6 m from the Base
195
Figure 4.60: Aerial View of the Reinforcement Sheets Exhumed from the Composite Mass after Test 3 (The numbers on each sheet indicate the sheet number in
Figure 4.61)
196
53
Test 3Rupture Lines
15.2
515
.25
15.2
515
.25
15.2
5
1
2
3
4
Reinforcement
(Note: All Dimensions in inches, Drawn to Scale)
Figure 4.61: Locations of Rupture Lines of Reinforcement in Test 3; Constructed based on Figure 4.60.
197
Table 4.9: Some test results of Test 3
Test 3
Test Conditions
Geosynthetic Reinforcement Geotex 4x4
Wide-Width Strength of Reinforcement 140 kN/m
Reinforcement Spacing 0.4 m
Confining Pressure 34 kPa
Test Results
Ultimate Applied Pressure 1,750 kPa
Vertical Strain at Failure 6.1 %
Maximum Lateral Displacement of the Open Face at Failure 54 mm
Stiffness at 1% vertical strain (Eat 1%) 48,900 kPa
Maximum strain in reinforcement at ruptured 12 %
Maximum measured strain in reinforcement 4.0 %
198
4.6.4 Test 4 – GSGC Test (T, 2 Sv)
The reinforcement used in this test was single-sheet of Geotex 4x4 at spacing of 0.4
m. The composite mass after testing is shown in Figure 4.62. The failure surfaces of
the composite mass after testing can be seen clearly in this Figure.
The measured data for of Test 4 are highlighted below:
• Global stress-strain relationship (see Figure 4.63): The maximum applied
vertical pressure was about 1,300 kPa. The vertical displacement at the failure
pressure was 77 mm (4.0% global vertical strain).
• Lateral displacement: The average lateral displacements on the open faces of
the composite mass under different vertical pressures as shown in Figure 4.64.
The maximum lateral displacement under the failure pressure was 52 mm.
• Internal displacements at selected points are shown in Figure 4.65. The trends
of the internal movements in Tests 2, 3 and 4 were identical. The points 1, 3,
4, 5, 6, 7 and 9 near the open faces, the displacements move downward and
outward with angles from 30o to 63o to the horizontal. The points 2, 5 and 8
along the center line, the displacements were almost vertical. The maximum
vertical displacement at the top of the specimen was 77 mm.
• Reinforcement strain: Figure 4.66 shows the locations of the strain gauges on
the geosynthetic sheets. The strain in the reinforcement of the GSGC mass is
shown in Figure 4.67. All reinforcement layers near the top of the composite
were ruptured after the test was completed. The locations of the rupture lines
can be seen from the aerial view of the reinforcement sheets exhumed from
the composite after testing (see Figure 4.68). Based on the locations of the
rupture line, the rupture planes can be constructed as shown in Figure 4.69.
This rupture line agrees perfectly with the failure line in Figure 4.62. The
199
maximum strain in reinforcement at ruptured was about 12 % and located at
near the ruptured line as shown in Figure 4.69.
• The test results of Test 4 are summarized in Table 4.10.
200
(a) (b)
Figure 4.62: Failure Planes of the Composite Mass after Testing in Test 4: (a) Front View at the South (b) Back View at the North.
201
0
200
400
600
800
1,000
1,200
1,400
0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00
Global Vertical Strain (%)
Appl
ied
Verti
cal S
tress
(kPa
)
Figure 4.63: Global Stress-Strain Relationship of Test 4
202
0
0.25
0.5
0.75
1
1.25
1.5
1.75
2
0.0 10.0 20.0 30.0 40.0 50.0 60.0
Lateral Displacement (mm)
Spec
imen
Hei
ght (
m)
200 kPa
400 kPa
600 kPa
800 kPa
1000 kPa
1250 kPa
1300 kPa
Applied Pressure:
Figure 4.64: Lateral Displacements on the Open Face of Test 4
203
76.3
0
Test 4
58.00
3.00
3.00
42.0
0
16.0
0
10.3
5
34.3
5
54.0
0
15.00
31.00
1 2
10
4 5
11
6
987
2'' x 2'' Grid System
3
Before Loading
250 kPa
500 kPa
750 kPa
1000 kPa
1250 kPa
1300 kPa
Legend:
Note:
(All Displacements in inches, Drawn to Scale of the Composite Mass)
Figure 4.65: Internal Displacements of Test 4
204
Strain Gages
13.35 8.00 8.00 8.00 8.00 13.35
58.75
15.2
515
.25
15.2
515
.25
15.2
5
Sheet No 4
Sheet No 3
Sheet No 2
Sheet No 1
Reinforcement
(Note: All Dimensions in inches)
Figure 4.66: Location of Strain Gauges on Geosynthetic Sheets in Test 4
205
00.20.40.60.8
11.21.41.61.8
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
Distance from the Edge of the Composite Mass (m)
Stra
in (%
)
100 kPa200 kPa
300 kPa400 kPa500 kPa600 kPa
800 kPa
Applied Pressure:
(a)
00.20.40.60.8
11.21.41.61.8
2
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
Distance from the Edge of the Composite Mass (m)
Stra
in (%
)
100 kPa
200 kPa
300 kPa
400 kPa
500 kPa
600 kPa
Applied Pressure:
(b)
Figure 4.67: Reinforcement Strain Distribution of the Composite Mass in Test 4:
(a) Layer 1, at 0.4 m from the Base (b) Layer 2, at 0.8 m from the Base
206
0
0.2
0.4
0.6
0.8
1
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
Distance from the Edge of the Composite Mass (m)
Stra
in (%
) 100 kPa
200 kPa
300 kPa
400 kPa
Applied Pressure:
(c)
00.20.40.60.8
11.21.41.61.8
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
Distance from the Edge of the Composite Mass (m)
Stra
in (%
) 100 kPa
200 kPa
300 kPa
400 kPa
500 kPa
Applied Pressure:
(d)
Figure 4.67 (continued): Reinforcement Strain Distribution of the Composite Mass in
Test 4: (c) Layer 3, at 1.2 m from the Base (d) Layer 4, at 1.6 m from the Base
207
Figure 4.68: Aerial View of the Reinforcement Sheets Exhumed from the Composite Mass after Test 4 (The numbers on each sheet indicate the sheet number in
Figure 4.69)
208
Rupture Lines
58
15.2
515
.25
15.2
515
.25
15.2
5
1
2
3
4
ReinforcementTest 4
(Note: All Dimensions in inches, Drawn to Scale)
Figure 4.69: Locations of Rupture Lines of Reinforcement in Test 4; Constructed based on Figure 4.68.
209
Table 4.10: Some test results of Test 4
Test 4
Test Conditions
Geosynthetic Reinforcement Geotex 4x4
Wide-Width Strength of Reinforcement 70 kN/m
Reinforcement Spacing 0.4 m
Confining Pressure 34 kPa
Test Results
Ultimate Applied Pressure 1,300 kPa
Vertical Strain at Failure 4.0 %
Maximum Lateral Displacement of the Open Face at Failure 53 mm
Stiffness at 1% vertical strain (Eat 1%) 46,600 kPa
Maximum strain in reinforcement at ruptured 12 %
Maximum measured strain in reinforcement 2.0 %
210
4.6.5 Test 5 – GSGC Test (unconfined with T, Sv)
The configuration of this test was the same as Test 2. The reinforcement was the
single-sheet Geotex 4x4 at spacing of 0.2 m. The confining pressure was not applied
for this test. Without applying confining pressure, the soil on the open faces fell of
continuously with increasing applied pressure. The composite mass and failure
surfaces after testing are shown in Figures 4.70 and 4.71.
The measured data of Test 5 highlighted below:
• The global stress-strain relationship (see Figure 4.72): The maximum applied
vertical pressure was about 1,900 kPa. The vertical displacement at the failure
pressure was 111 mm (6.0 % global vertical strain).
• Lateral displacement: The average lateral displacements on the open faces of
the composite mass under different vertical pressures as shown in Figure 4.73.
The maximum lateral displacement at the open faces under the failure
pressure could not been measured because the soil at these faces dropped
during testing under the high applied pressures.
• Internal displacements at selected points are shown in Figure 4.74. The trend
of the internal movements in Test 5 was nearly the same as that in the other
Tests 2, 3 and 4. The points 1, 3, 4, 5, 6, 7 and 9 near the open faces, the
displacements move downward and outward with angles from 35o to 63o to the
horizontal. The points 2, 5 and 8 along the center line, the displacements were
almost vertical. The maximum vertical displacement at the top of the
specimen was 111 mm.
• Reinforcement strain: Figure 4.75 shows the locations of the strain gauges on
the geosynthetic sheets. The strain in the reinforcement of the GSGC mass is
shown in Figure 4.76. Eight reinforcement layers near the top of the
211
composite were ruptured after testing. The locations of the rupture lines can
be seen from the aerial view of the reinforcement sheets exhumed from the
composite after testing (see Figure 4.77). Based on the locations of the
rupture line, the rupture planes can be constructed as shown in Figure 4.78.
This rupture line agrees perfectly with the failure line in Figure 4.70. The
maximum strain in reinforcement measured was 3.2 % and located at near the
ruptured line as shown in Figure 4.78.
• The test results of Test 5 are summarized in Table 4.11.
212
Figure 4.70: Composite Mass at Failure of Test 5
213
(a) (b)
Figure 4.71: Failure Planes of the Composite Mass after Testing in Test 5: (a) Front View (South), (b) Back View (North)
214
0
200
400
600
800
1,000
1,200
1,400
1,600
1,800
2,000
2,200
0 1 2 3 4 5 6 7 8
Global Vetical Strain (%)
App
lied
Vert
ical
Str
ess
(kPa
)
Figure 4.72: Global Stress-Strain Relationship of Test 5
215
0.000
0.250
0.500
0.750
1.000
1.250
1.500
1.750
2.000
0.0 10.0 20.0 30.0 40.0
Lateral Displacement (mm)
Spec
imen
Hei
ght (
m)
200 kPa
400 kPa
600 kPa
1500 kPa
Applied Pressure:
Figure 4.73: Lateral Displacements on the Open Face of Test 5
216
76.3
5
Test 5
49.00
2.50
2.50
42.0
0
16.0
0
10.3
5
60.3
5
34.3
5
12.50
28.50
54.0
0
1
10
2 3
11
4 5 6
7 8 9
2'' x 2'' Grid System
Before Loading
300 kPa
600 kPa
900 kPa
1200 kPa
1500 kPa
1800 kPa
Legend:
(All Displacements in inches, Drawn to Scale of the Composite Mass)
Note:
Figure 4.74: Internal Displacements of Test 5
217
Strain Gages
8.50 8.00 8.00 8.00 8.00 8.50
49.0015
.25
15.2
515
.25
15.2
515
.25
Sheet No 9
Sheet No 8
Sheet No 7
Sheet No 6
Sheet No 5
Sheet No 4
Sheet No 3
Sheet No 2
Sheet No 1
Reinforcement
(Note: All Dimensions in inches)
Figure 4.75: Location of Strain Gauges on Geosynthetic Sheets in Test 5
218
0
0.5
1
1.5
2
2.5
3
3.5
0.0 0.1 0.2 0.3 0.4 0.5 0.6Distance from the Edge of the Composite Mass (m)
Stra
in (%
)
100 kPa
200 kPa
300 kPa
400 kPa500 kPa
600 kPa
750 kPa
Applied Pressure:
(a)
0
0.5
1
1.5
2
2.5
3
0.0 0.1 0.2 0.3 0.4 0.5 0.6Distance from the Edge of the Composite Mass (m)
Stra
in (%
)
100 kPa200 kPa
300 kPa400 kPa500 kPa600 kPa
750 kPa
Applied Pressure:
(b)
Figure 4.76: Reinforcement Strain Distribution of the Composite Mass in Test 5:
(a) Layer 1, at 0.4 m from the Base (b) Layer 2, at 0.8 m from the Base
219
0
0.2
0.4
0.6
0.8
1
1.2
0.0 0.1 0.2 0.3 0.4 0.5 0.6
Distance from the Edge of the Composite Mass (m)
Stra
in (%
)
100 kPa
200 kPa
300 kPa
400 kPa
525 kPa
Applied Pressure:
(c)
Figure 4.76 (continued): Reinforcement Strain Distribution of the Composite Mass in
Test 5: (c) Layer 3, at 1.2 m from the Base
220
Figure 4.77: Aerial View of the Reinforcement Sheets Exhumed from the Composite Mass after Test 5 (The number on each sheet indicate the sheet number in
Figure 4.78)
221
Rupture Lines
497.
637.
637.
637.
637.
637.
637.
637.
637.
637.
63
1
2
3
4
5
6
7
8
9
Test 5Reinforcement
(Note: All Dimensions in inches, Drawn to Scale)
Figure 4.78: Locations of Rupture Lines of Reinforcement in Test 5; Constructed based on Figure 4.77.
222
Table 4.11: Some test results of Test 5
Test 5
Test Conditions
Geosynthetic Reinforcement Geotex 4x4
Wide-Width Strength of Reinforcement 70 kN/m
Reinforcement Spacing 0.2 m
Confining Pressure 0
Test Results
Ultimate Applied Pressure 1,900 kPa
Vertical Strain at Failure 6.0 %
Maximum Lateral Displacement of the Open Face at Failure Not Measured
Stiffness at 1% vertical strain (Eat 1%) 52,900 kPa
Maximum strain in reinforcement at ruptured 12 %
Maximum measured strain in reinforcement 3.2 %
223
4.7 Discussion of the Results
The results of the GSGC Tests are discussed in term of the following:
• Effects of geosynthetic inclusion (comparison between Tests 1 and 2).
• Relationship between reinforcement spacing and reinforcement strength
(comparison between Tests 2 and 3).
• Effects of reinforcement spacing (comparison between Tests 2 and 4).
• Effects of reinforcement stiffness (comparison between Tests 3 and 4).
• Effects of confining pressure (comparison between Tests 2 and 5)
• Composite strength properties.
4.7.1 Effects of Geosynthetic Inclusion (Comparison between Tests 1 and 2)
Table 4.12 shows the result comparisons between an unreinforced soil mass (Test 1)
and a soil mass reinforced by Geotex 4x4 at 0.2 m spacing (Test 2). With the
presence of the reinforcement, the reinforced soil was much stronger than the
unreinforced soil. The ultimate applied pressure for the GSGC mass was about 3.5
times as large as the strength of the soil mass without reinforcement. The stiffness of
the unreinforced soil mass was 50 % of that for the reinforced soil mass. In addition,
the reinforced soil mass was much more ductile than the unreinforced soil mass. The
global vertical strain was 6.5 % at failure for Test 2; whereas it was only 3.0 % for
Test 1.
224
Table 4.12: Comparison between Test 1 and Test 2
Test 1 Test 2 (T, Sv)
Geosynthetic Reinforcement None Geotex 4x4
Wide-Width Strength of Reinforcement None Tf = 70 kN/m
Reinforcement Spacing No Reinforcement Sv = 0.2 m
Confining Pressure 34 kPa 34 kPa
Ultimate Applied Pressure 770 kPa 2,700 kPa
Vertical Strain at Failure 3 % 6.5 %
Maximum Lateral Displacement of the Open Face at Failure 47 mm 60 mm
This Chapter described three analytical models. The models are for evaluating (1)
Compaction-Induced Stresses (CIS) in a GRS mass, (2) strength properties of a GRS
mass, and (3) required tensile strength of reinforcement in the design of GRS
structures. The first analytical model is a simple compaction model that is capable of
estimating the compaction-induced stresses, or the increase of the horizontal stresses,
in a GRS mass due to fill compaction. The model was developed by combining a
compaction model developed by Seed (1983) and the companion hand-calculation
procedure by Duncan and Seed (1986) for an unreinforced soil mass, and (b) the
theory of GRS composite behavior proposed by Ketchart and Wu (2001).
The second analytical model is for determination of the strength properties of a GRS
composite. With the analytical model, a new relationship between reinforcement
strength and reinforcement spacing is introduced to reflect an observation made in
actual construction and in controlled experiments regarding the relative effects of
reinforcement spacing and reinforcement strength on the performance of GRS
structures.
The third analytical model is for determination of reinforcement strength in design.
This model was also developed based on the new relationship between reinforcement
strength and reinforcement spacing.
235
5.1 Evaluating Compaction-Induced Stress in a GRS Mass
5.1.1 Conceptual Model for Simulation of Fill Compaction of a GRS Mass
Based on previous studies regarding CIS for unreinforced soil masses, and very
limited study for reinforced soil masses, a conceptual stress path for loading-
unloading-reloading of a GRS mass is shown in Figure 5.1. An explanation of Figure
5.1 is given below:
- A GRS mass is loaded (due to application of compaction loads) from an initial
state (point A) following the Ki,c-line (with horizontal stress, vcih K '' , σσ = ,
where Ki,c = coefficient of lateral earth pressure of the GRS mass for initial
loading) up to point B. At point B, the GRS mass reaches a maximum stress
state with the vertical stress of max,,max, ''' cvvv σσσ Δ+= (σ'v = vertical stress at
the initial stress; Δσ'v,c,max = maximum increase in vertical stress due to
compaction loading).
- Upon unloading (i.e., upon removal of the compaction loads), the stresses in
the soil are reduced by following a non-linear path from point B to point C.
- In cases of “significant” unloading, i.e., during unloading, the unloading-path
reaches the limiting line (the “K1,c-line”) at point E, further unloading stress
path will follow line EF (with vch K '' ,1 σσ = ).
- Upon reloading due to the next cycle of compaction load application, the
stresses are to follow a K3,c-line (with vch K '' ,3 σσ = ) from either point C or
pint F until it meets the initial loading path (line AB).
- The subsequent cycles of unloading and reloading shall not deviate much
from the K3,c-line, as suggested by Broms (1971), Seed (1983), and Erlich and
Mitchell (1993). Therefore, the same K3-line can be used for all subsequent
cycles of reloading and unloading.
236
σ'h = Ki,c σ'v
σv
σhK1,c-line
K1,c
1B
σ'v + Δσ'v,c,max
G
Unloading-curve
E
F
A
D
C
σ'v
Ki,c-line
K3,c-line
K3,c-line Δσ'h,c,r
Figure 5.1: Conceptual Stress Path for Compaction of a GRS Mass
5.1.2 A Simplified Model to Simulate Fill Compaction of a GRS Mass
A simplified compaction simulation model, as depicted in Figure 5.2, is proposed for
simulation of fill compaction of a GRS mass. By using the proposed model, the
increase of the horizontal stresses in a GRS mass due to compaction can be estimated.
These stresses, namely CIS, are represented by the horizontal residual stresses,
Δσ'h,c,r, in Figure 5.2. The proposed model is based on the bi-linear compaction
model suggested by Seed (1983) for an unreinforced soil mass, and the companion
hand-calculation procedure suggested by Duncan and Seed (1986). The proposed
model considers the presence of geosynthetic inclusions.
The stress path for fill compaction of a GRS mass in the proposed model can be
considered as a simplified form of the conceptual model described in Section 5.1.
The differences between the simplified model and the conceptual model are:
237
- Upon the removal of the compaction loads, the stresses in the soil are
reduced by following the K2,c-line from point B to point C, of which the
vertical stress, σ’v, equals to that of point A. The horizontal residual stress
due to the compaction loading is Δσ'h,c,r.
- In case of an unreinforced soil mass (or the reinforcement stiffness is
negligible), the stress in the soil in response to removal of compaction
loads will reduce to point S (instead of point C). The horizontal residual
stress due to compaction loading will be Δσ'h,s,r with Δσ'h,s,r ≤ Δσ'h,c,r.
- Upon application of the next cycle of compaction loads or placement of
new fill layers, the reloading path will follow the K3,c-line
(with vch K '' ,3 σσ = , and cc KK ,2,3 ≤ ). The stress path is to follow K3,c-line
from point C or point F until it meets the initial loading path, then follow
Ki,c-line to a new stress state.
A Unreinforced soil
σ'h = Ki,c σ'vσ'h,c
σ'h,s
σ'v
Ki,c-line
σv
σh
1
KA1
KA-line
K2,c-line
S
C
Δσ'h,s,r
B
Δσ'h,c,max
Δσ'h,c,r
σ'v + Δσ'v,c,max
D
σ'h = K3,c σ'v
σ'h = K2,c σ'v
K3,c-line 1
K1,c
K1,c-line
E
F
K0
K0-line
Figure 5.2: Stress Path of the Proposed Simplified Model for Fill Compaction of a GRS Mass
238
5.1.3 Model Parameters of the Proposed Compaction Simulation Model
Four model parameters are needed for the proposed compaction simulation model,
including: Ki,c, K1,c, K2,c, and K3,c. These parameters can be estimated from soil and
reinforcement properties using correlations shown in Table 5.1, of which the
empirical coefficients (e.g., α and F) are to be calibrated by the measured data from
the Generic Soil-Geosynthetic Composite (GSGC) tests (to be described in Chapter
4). Recommended values of the empirical coefficients are to be given for routine
applications. Alternatively, the model parameters can be obtained directly from the
results of GSGC tests. Note that the term ⎟⎟⎠
⎞⎜⎜⎝
⎛− rvs
r
JSEE
7.07.0 , for the estimation of
K2, c, is to account for the presence of reinforcement. For hand calculations, the
maximum increase of "vertical" stress, max,,' cvσΔ as shown in Figure 5.2, due to
compaction can be estimated simply by using the Westergaard's solution (1938).
239
Table 5.1 : Model parameters for the proposed compaction simulation model
Parameter Name Range of the Parameter
Values
Preliminary Estimation Based on Soil and Reinforcement Properties
Ki,c Coefficient of lateral earth pressure of a GRS mass for initial loading
0, KKK ciA ≤≤ Aci KK β≅,
1.0 1.5β≤ ≤
⎟⎠⎞
⎜⎝⎛ −≅
2'45tan 2 φ
AK
'sin10 φ−≅K
K1,c Limiting coefficient of lateral earth pressure for unloading
Pc KK ≅,1 ⎟⎠⎞
⎜⎝⎛ +≅≅
2'45tan 2
,1φ
Pc KK
K2,c Coefficient of lateral earth pressure for unloading
0,20 KK c ≤≤ ci
rvs
rc K
ESEEFK ,,2 7.0
7.011 ⎥⎦
⎤⎢⎣
⎡⎟⎟⎠
⎞⎜⎜⎝
⎛−
+−≅
where: ( )( )1
1−
−−=
OCROCROCRF
α
(Seed,
1983); 'sin φα ≅
K3,c Coefficient of lateral earth pressure for reloading
0,30 KK c ≤≤ cc KK ,2,3 ≅
Note: K0 = coefficient of at-rest lateral earth pressure KA = coefficient of active lateral earth pressure KP = coefficient of passive lateral earth pressure OCR = over-consolidation ratio Es = soil stiffness (kPa) Er = reinforcement stiffness (kN/m) Sv = reinforcement spacing (m).
240
5.1.4 Simulation of Fill Compaction Operation
Fill compaction is a complex operation in term of change in stresses. This section
describes the maximum increase of vertical stress at depth z along a given section in a
soil mass due to typical compaction operation. Let us consider the change of vertical
stress at depth z in section I-I due to moving compaction loads, as shown in Figures
5.3 and 5.4. The directions of the moving compaction plant may be: (1) coming
toward section I-I, and (2) going away from section I-I. The compaction loads are
simulated by loading and unloading at different locations (locations 1, 2, and i), as
shown in Figures 5.3(a) and 5.4(a). Figure 5.3(b) and Figure 5.4(b) show the stress
path of the stresses at depth z along section I-I.
An explanation of Figure 5.3(b) is given below:
1. The initial stress condition at depth z along section I-I is denoted by
point A, with the initial vertical stress being σv;
2. With the compaction loads at location 1, the stresses are increased by
following the Ki,c-line to point B;
3. As the compaction loads are removed from location 1, the stresses will
reduce from point B to point C by following the K2,c-line;
4. When compaction loads move to a new location (location 2), the
stresses will increase from point C through point B to point D;
5. As the compaction loads are removed from location 2, the stresses will
reduce from point D to point E by following again the K2,c-line.
6. Steps 1 through 5 are repeated for all subsequent new locations as the
compaction plant moves toward section I-I. Note that as the
compaction plant moves closer to section I-I, the vertical stress at
depth z will become larger.
7. The maximum vertical stress condition will be reached when the
compaction plant is directly above section I-I. The corresponding
241
stress condition is represented by point F. Upon removal of the
compaction loads, the stress path will follow K2-line to point G.
In Figure 5.4, the compaction plant moves away from section I-I. Initially the
compaction is located directly above section I-I, which causes the stresses to increase
from point A to point F in Figure 5.4(b) due to the compaction loads. As the
compaction loads are removed from section I-I, the stresses are reduced from point F
to point G, following the K2,c-line. As the compaction plant moves away from
section I-I, the stress conditions will move to points D and B, along line FG. As the
compaction plant is finally removed, the stress condition will be at point G.
From Figures 5.3 and 5.4, it is noted that the residual stresses, as denoted by the
vertical distance AG, are the same for the two cases. It indicates that to determine the
compaction-induced stresses at a certain section due to a moving compaction plant,
one only need to determine the residual lateral stresses as the compaction loads are
directly above the section under consideration.
Figure 5.5 shows the conceptual stress path on the effect of the number of compaction
passes. For the first pass, the residual stress at point A is represented by point G.
With the subsequent pass of the compaction plant, the slope of the K2-line will
increase, and point G becomes point G'. As the number of compaction pass increases,
the final residual stresses will move from G’ to G’’, then to G'''.
242
A
σ'vσv
σh
GF
C
D
B
E
compaction load
location 1location 2location i
I
I
Δσ'vΖ
(a)
(b)
1
2i
changing K2,c-line due to passes of compaction
Ki,c-line
K2,c and K3,c-line(Assume: K2,c = K3,c)
Δσ'h,c,r
σ'v + Δσ'v,c,max
Figure 5.3: (a) Locations of Compaction Loads, and (b) Stress Paths during Compaction at Depth z along Section I-I, as Compaction Loads Moving toward
Section I-I.
243
A
σ'vσv
σh
GF
compaction load
location ilocation 2location 1
I
I
Δσ'vΖ
BD
(a)
(b)
1
2i
P
changing K2,c-line due to passes of compaction
Ki,c-line
K2,c and K3,c-line(Assume: K2,c = K3,c)
Δσ'h,c,r
σ'v + Δσ'v,c,max
Figure 5.4: (a) Locations of Compaction Loads, and (b) Stress Paths during Compaction at Depth z along Section I-I, as Compaction Loads Moving away from
Section I-I.
244
A
σ'vσv
σh
G
FG'''
G'G''
Δσ'h,c,r
changing K2,c-line due to a number of compaction passes
K2,c and K3,c-line(Assume: K2,c = K3,c)
Ki,c-line
σ'v + Δσ'v,c,max
Figure 5.5: Stress Path at Depth z when Subject to Multiple Compaction Passes
245
5.1.5 Estimation of K2,c
The estimation of the K2,c in the proposed model for determining the compaction-
induced stresses in a GRS mass is presented in this section. Two stress reduction
factors “F” and “A” (as shown in Figure 5.6) are introduced. Factor F represents the
compaction-induced stresses for unreinforced soil in Seed’s model (1983), while
factor A is considered the presence of the reinforcement in the GRS mass.
A
K2, s for soil
σ'v
Ki, c - line
σv
σh
K2, c for composite
S
C
B
Δσ
σ'v + Δσ'v, c, max
G*(Δσ)
F∗(Δσ)
Figure 5.6: Stress Path of the Proposed Model for Fill Compaction of a GRS Mass
(a) For Compacted Soil
For an unreinforced soil fill, the coefficient of lateral earth pressure for unloading in
Figure 5.2 can be estimated by using the expression suggested by Seed (1983):
( ) cis KFK ,,2 1−= (5.1)
246
where ( )1
1−
−−=
OCROCROCRF
α
; 'sinφα ≅ ; and 5≅OCR for typical
compacted sand (Seed, 1983).
(b) For GRS Composite
For a GRS mass, total residual strain in the soil can be determined as:
( ) ( )s
s EGF σε Δ+
= (5.2)
where F and G = stress reduction factors shown in figure 5.6; σΔ = increase in horizontal stress due to compaction; Es = soil stiffness.
From Equation 5.2, the reinforcement force, T, due to residual strain in the soil can be
determined as:
( )r
s
EEGFT ⎟⎟
⎠
⎞⎜⎜⎝
⎛ Δ+=
σ (5.3)
where Er = reinforcement stiffness.
The average residual stress in the soil due to compaction, )(* σΔG , is:
( )v
r
s SE
EGFG ⎟⎟
⎠
⎞⎜⎜⎝
⎛ Δ+=Δ
σσ 7.0* (5.4)
where Sv = reinforcement spacing.
or rrvs EGEFSEG 7.07.0 += (5.5)
Thus,
rvs
r
ESEEFG7.0
7.0−
= (5.6)
Since ( )[ cic KGFK ,,2 1 +−= ] (5.7)
Substituting (5.6) into (5.7), we have
247
cirvs
rc K
ESEE
FK ,,2 7.07.0
11 ⎥⎦
⎤⎢⎣
⎡⎟⎟⎠
⎞⎜⎜⎝
⎛−
+−≅ (5.8)
The Factor 0.7 in Equation 5.4 will be explained in the section 5.2.1. The increased
horizontal stress in a GRS mass due to compaction can be estimated as:
( ) ( )ccicvch KKFG ,2,max,,, ' −Δ=+Δ=Δ σσσ (5.9)
Substituting (5.8) into (5.9), we have
⎪⎭
⎪⎬⎫
⎪⎩
⎪⎨⎧
⎥⎦
⎤⎢⎣
⎡⎟⎟⎠
⎞⎜⎜⎝
⎛−
++−Δ=Δrvs
rcicvch ESE
EFK7.0
7.0111' ,max,,, σσ (5.10)
or
⎟⎟⎠
⎞⎜⎜⎝
⎛−
+Δ=Δrvs
rcicvch ESE
EFK7.0
7.01' ,max,,, σσ (5.11)
Equation 5.11 shows calculation of the residual lateral stress in a GRS mass due to
compaction. The effect of compaction-induced stress in a GRS mass can be also seen
from Equation 5.11.
Using Equation 5.11, the increase of lateral stress in a GRS mass can be estimated
and the increase in soil stiffness can be evaluated. For example, the stiffness of a soil
can be evaluated as:
( ) ( )( )
( )n
a
cha
ch
chf
PPK
cR
tE ⎟⎟⎠
⎞⎜⎜⎝
⎛ Δ+
⎥⎥⎦
⎤
⎢⎢⎣
⎡
Δ++
Δ+−−−= ,3
2
,3
,31
sin2cos2sin1
1σσ
φσσφσσσφ
(5.12)
where
248
Et = tangent modulus
Rf = ratio of ultimate deviator stress to the failure deviator stress
c = cohesive strength
φ = angle of friction
Pa = the atmospheric pressure
K and n = material parameters.
249
5.2 Strength Properties of GRS Composite
Schlosser and Long (1972) proposed the concept of increase of apparent confining
pressure and concept of apparent cohesion of a GRS composite. The Mohr circles of
an unreinforced cohesive soil and a reinforced cohesive soil at failure are shown in
Figure 5.7.
Shea
r Stre
ess,
τ
σ
C
Normal Stress, σ
φ
φ
CR
3σ
3Rσ
1σ
1RΔσ3R
Unreinforced Soil
Reinforced Soil
Reinforced Soil
Figure 5.7: Concept of Apparent Confining Pressure and Apparent Cohesion of a GRS Composite
The apparent cohesion of a GRS composite can be determined as:
cK
c PRR +
Δ=
23σ
(5.13)
where = apparent cohesion of a GRS composite Rc
c = cohesion of soil
250
Kp = coefficient of passive earth pressure
R3σΔ = increase of confining pressure due to reinforcement
Schlosser and Long (1972) also proposed an equation to calculate increased confining
pressure as:
v
f
ST
=Δ 3σ (5.14)
This expression implies that an increase in reinforcement strength, Tf, has the same
effect as a proportional decrease in reinforcement spacing, Sv. Many experimental
test results have shown that Equation 5.14 is not correct. Reinforcement spacing
plays a far more important role than reinforcement strength (Adams, 1997 and 2007;
Elton and Patawaran, 2004 and 2005; Ziegler et al., 2008). This point is supported by
the experiments conducted as a part of this study, as presented in Chapter 4.
5.2.1 Increased Confining Pressure
A new method to estimate the increased confining pressure in soil due to the presence
of reinforcement is presented. The proposed equation for the increased confining
pressure can be expressed as:
⎟⎟⎠
⎞⎜⎜⎝
⎛=Δ
v
f
ST
W3σ (5.15)
where the factor, W, can be estimated as:
⎟⎟⎠
⎞⎜⎜⎝
⎛
= ref
v
SS
rW (5.16)
where Tf = extensile strength of reinforcement
r = a dimesionless factor (will be discussed later in this Section)
251
Sv = vertical spacing of reinforcement
Sref = the reference spacing (will be discussed further later)
To estimate the factor r in Equation 5.16, the concept of “average stresses” proposed
by Ketchart and Wu (2001) was employed. Instead of using average stresses,
however, average reinforcement forces were used.
a. Average Stress in GRS Mass by Ketchart and Wu (2001)
Ketchart and Wu (2001) developed a concept of “average stress” to determine the
behavior of a GRS composite based on a load-transfer analysis. From a simplified
preloading-reloading model for GRS mass, the equations to calculate stresses and
displacements of a GRS mass were developed using the idealized geometry of plain-
strain GRS mass and differential elements of the soil and reinforcement for
equilibrium equations (Hermann and Al-Yassin, 1978) as shown in Figures 5.8 and
5.9.
From the equilibrium equations and a number of assumptions, the stresses in soil and
the force in the reinforcement can be calculated as:
The force in the reinforcement:
( )( )⎟⎟
⎠
⎞⎜⎜⎝
⎛−⎟
⎠⎞
⎜⎝⎛ −⎟
⎟⎠
⎞⎜⎜⎝
⎛ −⎥⎦
⎤⎢⎣
⎡−⎟⎟
⎠
⎞⎜⎜⎝
⎛−
=Lx
EPPEAF
s
shv
s
srrx α
ααβυ
υυ
coshcosh11
11 2
2
(5.17)
The horizontal stress in the soil:
252
( )( ) ⎟⎟
⎠
⎞⎜⎜⎝
⎛⎟⎠⎞
⎜⎝⎛ −+⎟
⎟⎠
⎞⎜⎜⎝
⎛−⎟⎟
⎠
⎞⎜⎜⎝
⎛−
+⎟⎟⎠
⎞⎜⎜⎝
⎛−
−= 22 1coshcosh
11 αβ
αα
αβ
υυ
υυ
σLxPPP hv
s
sv
s
sx (5.18)
Based on a load-transfer analysis, the “average stresses” was determined.
The average vertical stress, vσ , is assumed to be equal to the boundary vertical
pressure, i.e.,
vv P=σ (5.19)
The average horizontal stress, hσ , is:
L
dxxL
x
h
∫= 0
)(σσ (5.20)
or
( )⎟⎟⎠
⎞⎜⎜⎝
⎛⎟⎠⎞
⎜⎝⎛ −+⎟
⎟⎠
⎞⎜⎜⎝
⎛−⎟⎟
⎠
⎞⎜⎜⎝
⎛−
−⎟⎟⎠
⎞⎜⎜⎝
⎛−
= 1.
tanh111 2 L
LPPP hvs
sv
s
sh α
ααβ
υυ
υυ
σ (5.21)
253
Figure 5.8: An Idealized Plane-Strain GRS Mass for the SPR Model
Figure 5.9: Equilibrium of Differential Soil and Reinforcement Elements (Reproduced from Hermann and Al-Yassin, 1978)
254
b. Average Reinforcement Forces in a GRS Mass
The equations of the forces in the reinforcement and the maximum force in the
reinforcement could be expresses as:
( )( )⎟⎟
⎠
⎞⎜⎜⎝
⎛−⎟
⎠⎞
⎜⎝⎛ −⎟
⎟⎠
⎞⎜⎜⎝
⎛ −⎥⎦
⎤⎢⎣
⎡−⎟⎟
⎠
⎞⎜⎜⎝
⎛−
=Lx
EPPEAF
s
shv
s
srrx α
ααβυ
υυ
coshcosh11
11 2
2
(5.22)
and
( )⎟⎟⎠
⎞⎜⎜⎝
⎛−⎟
⎠⎞
⎜⎝⎛ −⎟
⎟⎠
⎞⎜⎜⎝
⎛ −⎥⎦
⎤⎢⎣
⎡−⎟⎟
⎠
⎞⎜⎜⎝
⎛−
=LE
PPEAFs
shv
s
srr αα
βυυ
υcosh
1111
1 2
2
max (5.23)
The average force in the reinforcement may be calculated as:
L
dxFF
L
x∫= 0 (5.24)
Substituting Equation (5.22) into Equation (5.24), we have:
( )( )∫ ⎟⎟
⎠
⎞⎜⎜⎝
⎛−⎟
⎠⎞
⎜⎝⎛ −⎟
⎟⎠
⎞⎜⎜⎝
⎛ −⎥⎦
⎤⎢⎣
⎡−⎟⎟
⎠
⎞⎜⎜⎝
⎛−
=L
s
shv
s
srr dxLx
EPP
LEAF
02
2
coshcosh11
11 α
ααβυ
υυ
(5.25)
or
( )( )∫ ⎟⎟
⎠
⎞⎜⎜⎝
⎛−⎟
⎠⎞
⎜⎝⎛ −⎟
⎟⎠
⎞⎜⎜⎝
⎛ −⎥⎦
⎤⎢⎣
⎡−⎟⎟
⎠
⎞⎜⎜⎝
⎛−
=L
s
shv
s
srr dxLx
EPP
LEA
F0
2
2
coshcosh11
11
.αα
αβυ
υυ
(5.26)
Thus,
( )( )⎟⎟
⎠
⎞⎜⎜⎝
⎛−⎟
⎠⎞
⎜⎝⎛ −⎟
⎟⎠
⎞⎜⎜⎝
⎛ −⎥⎦
⎤⎢⎣
⎡−⎟⎟
⎠
⎞⎜⎜⎝
⎛−
=L
xLE
PPLEA
Fs
shv
s
srr
ααα
αβυ
υυ
coshsinh1
11
.2
2
(5.27)
255
Introducing a factor maxFFr = , and from Equations (5.23) and (5.27), r can be
determined as:
( ) ( )( )( )1cosh
sinhcosh−
−=
LLLLLr
ααααα (5.28)
Using the data from the calculation example in the SPR model (Ketchart and Wu,
2001), the values of factor r for different applied pressures and reinforcement lengths
are presented in Table 5.2.
Table 5.2: Values of factor r under different applied pressure and reinforcement lengths
Increment of
Vertical Pressure
Reinforcement
Length, L (m)
α r
(maxFFr = )
Pv = 9.0 kPa 0.127 13.875 0.698
Pv = 18.0 kPa 0.127 14.616 0.701
Pv = 9.0 kPa 0.225 6.851 0.691
Pv = 18.0 kPa 0.225 6.966 0.692
It can be seen from Table 5.2, the average reinforcement forces are about 70% of the
maximum reinforcement force. The highest value of the maximum reinforcement
force, Fmax, can not exceed the tensile strength of reinforcement, Tf.
256
(5.29) fTF =max
and the average reinforcement force favg TT 7.0= .
Equation 5.16 becomes:
⎟⎟⎠
⎞⎜⎜⎝
⎛
= ref
v
SS
W 7.0 (5.30)
5.2.2 Apparent Cohesion and Ultimate Pressure Carrying Capacity of a GRS
Mass
Substituting Equation 5.30 into Equation 5.15, the increased confining pressure in a
GRS mass becomes:
⎟⎟⎠
⎞⎜⎜⎝
⎛
⎥⎥⎥
⎦
⎤
⎢⎢⎢
⎣
⎡=⎟⎟
⎠
⎞⎜⎜⎝
⎛=Δ
⎟⎟⎠
⎞⎜⎜⎝
⎛
v
fSS
v
f
ST
ST
W ref
v
7.03σ (5.31)
Therefore, the apparent cohesion, CR, of a GRS composite can be evaluated as:
cKS
TcKC p
v
fSS
pRref
v
+⎥⎥⎥
⎦
⎤
⎢⎢⎢
⎣
⎡=+
Δ=
⎟⎟⎠
⎞⎜⎜⎝
⎛
27.0
23σ
(5.32)
where c = cohesion of soil Kp = coefficient of passive earth pressure Tf = extension strength of reinforcement Sv = vertical spacing of reinforcement Sref = the reference spacing and can be calculated by: (5.33) max6 dSref = (dmax = the maximum particle/grain size of soil)
Therefore, the ultimate pressure carrying capacity, R1σ , of a soil-geosynthetic
composite mass is:
257
1 3 2fR p
v
TW K c K
Sσ σ
⎛ ⎞= + +⎜ ⎟⎝ ⎠
p (5.34)
where 3σ = confining pressure.
5.1 Verification of the Analytical Model with Measurement Data
Verification of the proposed analytical model for GRS composite strength properties
is made by comparing the model calculation results with the measured data from
GSGC tests (as presented in Chapter 4), and with the measured data by Elton and
Patawaran (2005).
5.1.1 Comparison between the Analytical Model and GSGC Test Results
The results of the GSGC tests have been reported in section 4.6. The dimensions of
the GSGC tests are 2 m high and 1.4 m wide in a plane strain condition. The soil
mass in the tests were reinforced with Geotext 4x4 geotextile at 0.2 m and 0.4 m
spacing. For the Diabase soil used in GSGC Tests, the maximum particle size was
about 1.3 in.; therefore, .8.76 max indSref == (or 0.2 m). Comparisons of the results
between the analytical model and the GSGC tests are presented in Table 5.3. The
deviatoric stresses at failure calculated from the analytical model are in good
agreement with those of the GSGC tests. The differences between them are less than
10 %.
For reference purposes, comparisons of the results between Schlosser and Long’s
method and the GSGC tests are also presented (see Table 5.4). The deviatoric
stresses at failure calculated from the Schlosser and Long’s method are about 20 % to
86 % larger than the measured values.
258
Table 5.3: Comparison of the results between the analytical model and the GSGC tests
Parameter Test 2 (T, S) Test 3 (2T, 2S) Test 4 (T, 2S)
Tf (kN/m) 70 140 70
Sv (m) 0.2 0.4 0.4
3σΔ (kN/m2)
by the Analytical Model
245
172
86
CR (kN/m2)
by the Analytical Model
407
305
188
( 31 )σσ −R (kN/m2)
from Measured Data
2,700 1,750 1,300
( 31 )σσ −R (kN/m2)
by the Analytical Model
2,460
1,900
1,250
Difference between
the Analytical Model and
Measured Data
- 9 %
+ 8 %
- 4 %
Note: Internal friction angle of soil, φ = 50o; cohesion of soil, c = 70 kPa.
259
Table 5.4: Comparison of the results between Schlosser and Long’s method and the GSGC tests
Parameter Test 2 (T, S) Test 3 (2T, 2S) Test 4 (T, 2S)
Tf (kN/m) 70 140 70
Sv (m) 0.2 0.4 0.4
3σΔ (kN/m2)
by Schlosser & Long’s Method
350 350 175
CR (kN/m2)
by Schlosser & Long’s Method
550 550 310
( 31 )σσ −R (kN/m2)
from Measured Data
2,700 1,750 1,300
( 31 )σσ −R (kN/m2)
by Schlosser & Long’s Method
3,250 3,250 1,930
Difference between
Schlosser & Long’s Method
and Measured Data
+ 20 % + 86 % + 48 %
Note: Internal friction angle of soil, φ = 50o; cohesion of soil, c = 70 kPa.
260
5.3.2 Comparison between the Analytical Model and Elton and Patawaran’s
Test Results
Elton and Patawaran (2005) conducted seven large-size triaxial tests for reinforced
soil with the dimensions of 5 ft in height and 2.5 ft in diameter (see Figure 5.10). All
the tests were conducted under an unconfined condition. The properties of the tests
are summarized as follows:
• Backfill: The soil used in the test was a poorly graded sand with the gradation
test results shown in Figure 5.11; maximum dry unit weight γdry = 121 pcf;
• Reinforcement: Six types of reinforcement (TG500, TG600, TG 700, TG800,
TG1000 and TG028) were used for the tests with the reinforcement spacing of
6 in. and 12 in. The strength, Tf, of the reinforcement and reinforcement
spacing, Sv, are shown in Table 5.5.
The maximum particle size (from gradation tests, Figure 5.11) of the backfill in the
large-size triaxial tests was .5.0max ind = (or 12.7 mm); therefore,
(or 0.08 m). The measured results are shown in Figure 5.12.
The comparisons of Elton and Patawaran’s tests results with the analytical model and
with Schlosser and Long’s Method are presented in Tables 5.5 and 5.6, respectively.
.36 max indSref ==
The differences in the deviatoric stresses at failure calculated from the analytical
model and Elton and Patawaran’s measured data are less than 18 %. Whereas, the
results calculated from Schlosser and Long’s method are 69 % to 97 % larger than the
measured values.
261
Figure 5.10: Reinforced Soil Test Specimen before Testing (Elton and Patawaran, 2005)
262
Figure 5.11: Backfill Grain Size Distribution before and after Large-Size Triaxial
Tests (Elton and Patawaran, 2005)
Figure 5.12: Large-Size Triaxial Test Results (Elton and Patawaran, 2005)
263
Table 5.5: Comparison of the results between the analytical model and Elton and Patawaran’s tests (2005)
Reinforcement
Type
TG
500
TG
500
TG
600
TG
700
TG
800
TG
1000
TG
028
Tf (kN/m) 9 9 14 15 19 20 25
Sv (m) 0.15 0.30 0.15 0.15 0.15 0.15 0.15
3σΔ (kN/m2) by the Analytical
Model
30 8 47 48 62 67 83
CR (kN/m2)
by the Analytical Model
60 36 78 79 94 99 116
( 31 )σσ −R (kN/m2) from Measured
Data
230 129 306 292 402 397 459
( 31 )σσ −R (kN/m2)
by the Analytical Model
256 153 333 341 402 426 498
Difference between the Analytical
Model and Measured Data
11 % 18 % 9 % 17 % 0 % 7 % 8 %
Note: Internal friction angle of soil, φ = 40o; cohesion of soil, c = 27.6 kPa.
264
Table 5.6: Comparison of the results between Schlosser and Long’s method and Elton and Patawaran’s tests (2005)
Reinforcement Type
TG
500
TG
500
TG
600
TG
700
TG
800
TG
1000
TG
028
Tf (kN/m) 9 9 14 15 19 20 25
Sv (m) 0.15 0.30 0.15 0.15 0.15 0.15 0.15
3σΔ (kN/m2) by Schlosser & Long’s Method
59 30 92 95 122 132 163
CR (kN/m2)
by Schlosser & Long’s Method
91 59 126 130 158 169 202
( 31 )σσ −R (kN/m2)
From Measured Data
230 129 306 292 402 397 459
( 31 )σσ −R (kN/m2)
by Schlosser & Long’s Method
390 254 541 557 678 726 868
Difference between Schlosser & Long’s
Method and Measured Data
70 % 97 % 77 % 91 % 69 % 83 % 89 %
Note: Internal friction angle of soil, φ = 40o; cohesion of soil, c = 27.6 kPa.
265
5.3.3 Comparison of the Results between the Analytical Model and Finite
Element Results
Finite element analyses were conducted to provide additional data for verifying the
analytical model. The test conditions and material properties used for the finite
element analyses were the same as those used in GSGC Test 2, but with confining
pressures of 34 kPa, 70 kPa, 100 kPa and 200 kPa. The confining pressure used in
GSGC Test 2 was 34 kPa. The comparison indicates that the results of the analytical
model are in good agreement with those obtained from the finite element analyses at
the different confining pressures. The largest difference in terms of the deviatoric
stress at failure is 9%.
266
Table 5.7: Comparison of the results between the analytical model and results for GSGC Test 2 with different confining pressures from FE analyses
Parameter S3 = 34 kPa
S3 = 70 kPa
S3 = 100 kPa
S3 = 200 kPa
3σΔ (kN/m2)
by the Analytical Model 245 245 245 245
CR (kN/m2)
by the Analytical Model 407 407 407 407
( 31 )σσ −R (kN/m2)
from FE Analysis 2,700 2,970 3,190 3,860
( 31 )σσ −R (kN/m2)
by the Analytical Model 2,490 2,760 2,990 3,740
Difference between
the Analytical Model and FE analyses
- 8 %
- 7 %
- 6 % - 3 %
Note: Internal friction angle of soil, φ = 50o; cohesion of soil, c = 70 kPa; reinforcement strength, Tf = 70 kN/m; reinforcement spacing, Sv = 0.2 m.
267
5.4 Required Reinforcement Strength in Design
In this Section, an analytical model for determining required tensile strength in
reinforcement is developed, a comparison of the analytical model with current design
equation is made, and verification of the analytical model is presented.
5.4.1 Proposed Model for Determining Reinforcement Force
In current design methods the following equation is used to determine the required
reinforcement strength, Trequired, for the design of GRS structures:
svhrequired FST ∗∗= σ (5.35)
where
Trequired = required strength for reinforcement at depth z
hσ = horizontal stress in a GRS mass at depth z
Fs = safety factor
Assuming Fs = 1, we have frequired TT = (ultimate strength of reinforcement) and
Equation 5.35 becomes
hv
f
ST
σ= (5.36)
Note that when the horizontal stress, hσ , is a constant, the ratio f
v
TS
becomes a
constant; i.e., Tf is linearly proportional to Sv. Using Equation 5.15 a new expression
for the increase of confining pressure due to tensile inclusion, a modified equation for
determination of required reinforcement strength can be obtained. The derivation of
268
the modified equation is described as follows. The horizontal stress, hσ , in a GRS
structure at depth z is:
33 σσσ Δ+=h (5.37)
or 33 σσσ −=Δ h (5.38)
Since,
v
f
ST
W=Δ 3σ (see Section 5.2.1) (5.39)
Substituting Equation 5.39 into Equation 5.36 leads to
v
fh S
TW=− 3σσ (5.40)
or ( )
WST h
v
f 3σσ −= (5.41)
Since,
⎟⎟⎠
⎞⎜⎜⎝
⎛
= max67.0 dSv
W (5.42)
Therefore,
( )
⎟⎟⎠
⎞⎜⎜⎝
⎛
−=
max6
3
7.0 dS
h
v
f
vST σσ
(5.43)
where
Tf = ultimate strength of reinforcement at depth z
hσ = horizontal stress in a GRS mass at depth z
3σ = lateral constraint pressure at depth z, lateral earth pressure exerted by
external constraint. For a GRS wall with modular block facing, 3σ
can be estimated as:
δγσ tan3 bb= (5.44)
269
where
γb = unit weight of facing block
b = width of facing block
δ = friction angle between modular block facing elements (δ can be
the friction angle between facing blocks if there is no
reinforcement between the blocks, or it can be the friction
angle between facing block and geosynthetic if there is
reinforcement sandwiched between blocks)
Sv = reinforcement spacing
dmax = maximum grain size of the backfill
The required tensile strength of the reinforcement in design can be expressed as:
sv
dS
hrequired FST
v**
7.0 max6
3
⎥⎥⎥
⎦
⎤
⎢⎢⎢
⎣
⎡−
=⎟⎟⎠
⎞⎜⎜⎝
⎛
σσ (5.45)
Note that Trequired is always equal or greater than zero. For a GRS mass without
lateral constraint (e.g., a wrapped wall), 3σ = 0, and Equation 5.45 becomes
sv
dS
hrequired FST
v**
7.0 max6 ⎥⎥⎥
⎦
⎤
⎢⎢⎢
⎣
⎡=
⎟⎟⎠
⎞⎜⎜⎝
⎛
σ (5.46)
5.4.2 Comparison of Reinforcement Strength between the Analytical Model
and Current Design Equation
A comparison of reinforcement forces is made between the proposed model
(Equation 5.45) and the current design equation (Equation 5.35). The reinforcement
270
forces in a 6.0 m high GRS wall, as determined by the two equations, are shown in
Table 5.8. Note that the facing stiffness is ignored in the current design method;
while the facing rigidity, as denoted by the lateral constraint pressure, σ3, can be
accounted for in the proposed model.
5.4.3 Verification of the Analytical Model for Determining Reinforcement
Strength
Verification of the analytical model for determination of reinforcement strength is
made by comparing the results with the forces in reinforcement at failure from the
GSGC tests (see Chapter 4), with the measured data by Elton and Patawaran (2005),
and with a typical GRS wall.
Table 5.9 shows a comparison of the results from the proposed model and measured
data from the GSGC tests. The largest difference in reinforcement forces between the
two is 16 %, whereas there is 47 % between the current design methods and the test
data.
A comparison of the reinforcement forces between the proposed model and the
measured data from Elton and Patawaran’s tests (2005) is shown in Table 5.10. The
largest difference in reinforcement forces between the two is 13 %, whereas it was as
high as 74 % between the current design method (Equation 5.35) and the test results.
The proposed model clearly gives a much improved value for estimating
reinforcement strength compared to the current design methods.
271
272
Table 5.8: Comparison of reinforcement forces between proposed model and current design equation for a GRS wall
z (m)
Trequired (kN/m), for Fs = 1
Current Design Equation
Proposed Model
No Facing Modular Block Facing with
bb = 35o
Modular Block Facing with
bb = 54o
0.4 0.3 1.1 0 0
0.8 0.6 1.5 0.1 0
1.2 1.0 2.0 0.5 0
1.6 1.3 2.4 1.0 0
2.0 1.6 2.9 1.4 0
2.4 1.9 3.3 1.9 0.1
2.8 2.3 3.7 2.3 0.6
3.2 2.6 4.2 2.8 1.0
3.6 2.9 4.6 3.2 1.5
4.0 3.2 5.1 3.6 1.9
4.4 3.6 5.5 4.1 2.4
4.8 3.9 6.0 4.5 2.8
5.2 4.2 6.4 5.0 3.2
5.6 4.5 6.8 5.4 3.7
6.0 4.9 7.3 5.8 4.1 Note: Internal friction angle of soil, = 38o; cohesion of soil, c = 0; reinforcement spacing, Sv = 0.2 m; maximum grain size of soil, mmd 38max ; unit weight of soil,
backfill 17 kN/m3; unit weight of facing block, block 25 kN/m3; bb = friction angle between facing blocks; width of blocks, b = 0.3 m.
Table 5.9: Comparison of reinforcement forces between proposed model and the GSGC tests
Parameter Test 2 (T, S)
Test 3 (2T, 2S)
Test 4 (T, 2S)
Test 5 (T, S)
Reinforcement Force at Failure Tf (kN/m) 70 140 70 70
Difference between Current Design Equation (Equation 5.35) and Tf
- 11 % - 47 % - 28 % - 41 %
Maximum Reinforcement Force from Proposed Model, Equation 5.45
(kN/m)
79.4 124.1 75.4 58.8
Difference between Proposed Model (Equation
5.45) and Tf
+ 13 % - 11 % + 8 % -16 %
Note: Internal friction angle of soil, φ = 50o; cohesion of soil, c = 70 kPa; unit weight of soil, =backfillγ 24 kM/m3; maximum grain size of soil, mmd 33max = .
273
274
Table 5.10: Comparison of reinforcement forces between proposed model and test data from Elton and Patawaran (2005)
Parameter TG 500
TG 500
TG 600
TG 700
TG 800
TG 1000
TG 028
Reinforcement Force at Failure Tf (kN/m) 9 9 14 15 19 20 25
• Interface: the interface between the soil and geosynthetic reinforcement is
fully bonded.
The very fine mesh of a FE analyses to simulate CIS in a GRS mass is shown in
Figure 6.13. Figure 6.14 shows the lateral stress distributions at the center line of the
GRS mass without considering CIS and with CIS under the maximum compaction
pressures of 200 kPa and 500 kPa. The compaction operation was simulated by
loading and unloading at different locations on the surface area of each lift. The
“residual lateral stresses” were the differences between the lateral stresses with
simulating CIS and those without CIS at the same location as shown in Figure 6.14.
Comparisons of residual lateral stresses distribution resulting from compaction
pressures of 200 kPa and 500 kPa between the CIS model and the FE analysis are
302
shown in Figures 6.15. It is seen that the compaction-induced stresses calculated
from the CIS hand-computation model are in very good agreement with the values
obtained from the FE analysis. Note that the residual lateral stresses in a GRS mass
are higher under a higher vertical compaction pressure. In actual construction, the
maximum vertical pressure of compaction is in the range of 200 kPa to 500 kPa. As
seen from Figure 6.15, the effects of CIS can be rather significant in actual
construction. The zigzag lines in FE analyses in Figure 6.15 were caused by the
thickness of compaction lift of 0.2 m. With the larger compaction lift, the amplitude
of the zigzag is larger and the effect of CIS is smaller.
The analyses form FE show more accurately with the finer meshes. But to get the
results for only one curve from the FE analysis as shown in Figure 6.14, it took more
than 20 hours for inputting data and running program with a strong PC configuration
e.g. Dual Core 1.86 GHz, 3 GB RAM, whereas, the analytical model is simple and
can use hand calculation for several minutes. To reduce the time consuming, a coarse
mesh can be used with somewhat tolerated error. Figure 6.16 shows the comparison
between FE results with the coarse mesh and the analytical model. The EF results in
Figure 6.16 obtained from simulating the compaction operation by applying the
compaction pressure over the entire surface area of the GRS mass at each compaction
lift. From Figures 6.15 and 6.16, it can be seen that the fine mesh should be used
when analyzing GRS structures.
303
Figure 6.13: FE Mesh to Simulate CIS in a Reinforced Soil Mass
304
-6
-5
-4
-3
-2
-1
00 25 50 75
Residual Lateral Stress (kPa)
Dep
th (m
)
Lateral StressDistributionw ithoutConsidering CIS
Lateral StressDistribution w ithConsidering CIS(200 kPa)
Residual LateralStress (due toMaximum VerticalCompactionPressure of 200kPa)
-6
-5
-4
-3
-2
-1
00 25 50 75 100 125
Residual Lateral Stress (kPa)
Dep
th (m
)
Lateral StressDistributionw ithoutConsidering CIS
Lateral StressDistribution w ithConsidering CIS(500 kPa)
Residual LateralStress (due toMaximum VerticalCompactionPressure of 500kPa)
(a) (b)
Figure 6.14: Lateral Stress Distribution of a GRS Mass from FE Analyses with:
(a) Maximum Vertical Compaction Pressures of 200 kPa (b) Maximum Vertical Compaction Pressures of 500 kPa
305
-6
-5
-4
-3
-2
-1
00 25 50 75 100
Residual Lateral Stress (kPa)
Dep
th (m
)
Maximum verticalpressure due tocompaction:
200 kPa (Model)
500 kPa (Model)
200 kPa (FE)
500 kPa (FE)
Figure 6.15: Comparison of Residual Lateral Stresses of a GRS Mass due to Fill Compaction between FE Analysis with Very Fine Mesh and the Analytical Model
306
307
-6
-5
-4
-3
-2
-1
00 25 50 75 100
Residual Lateral Stress (kPa)
Dep
th (m
)
Maximum verticalpressure due tocompaction:
200 kPa (Model)
300 kPa (Model)
500 kPa (Model)
200 kPa (FE)
300 kPa (FE)
500 kPa (FE)
Figure 6.16: Comparison of Residual Lateral Stresses of a GRS Mass due to Fill Compaction between FE Analysis with Coarse Mesh and the Analytical Model
7. SUMMARY AND CONCLUSIONS
7.1 Summary
A study was taken to investigate the composite behavior of a geosynthetic-reinforced
soil (GRS) mass. The study focused on the strength of a GRS mass, the compaction-
induced stresses in a GRS mass, and the lateral deformation of a GRS wall with
modular block facing.
The following tasks were carried out:
1. Reviewed previous studies on: (1) composite behavior of a GRS mass and (2)
CIS in an unreinforced soil mass and in a GRS mass.
2. Designed a generic soil-geosynthetic composite (GSGC) test for investigating
the composite behavior of GRS mass, and conducted five GSGC tests with
well-controlled condition with extensive instrumentation to monitor the
behavior under different reinforcement spacing, reinforcement strength, and
confining pressure.
3. Developed an analytical model for the relationship between reinforcement
strength and reinforcement spacing, and derived an equation for calculating
composite strength properties.
4. Developed a hand-computation analytical model for simulation of
compaction-induced stresses in a GRS mass.
5. Performed finite element analyses to simulate the GSGC tests, generate
additional data (with different confining pressures) for verifying the analytical
models in this study, and investigate the behavior of GRS composites.
308
6. Verify the analytical models using measured data from the GSGC tests,
relevant test data available in the literature, and FE analyses.
7. Developed an analytical model for predicting lateral movement of GRS walls
with modular block facing.
7.2 Findings and Conclusions
The findings and conclusions of this study can be summarized as follows:
1. The results of the GSGC tests are consistent and appear very reliable. The
tests provide direct observation on the behavior of a GRS mass as related to
reinforcement strength and spacing. The tests also provide better
understanding of the composite behavior of GRS mass and can be used for
validation of analytical models in this study and other models of GRS
structures in the future.
2. An equation describing the relative effects of reinforcement spacing and
reinforcement strength was developed and verified. Based on the equation,
the required reinforcement strength in a GRS wall can be determined, and so
as the composite strength properties and ultimate pressure carrying capacity of
a GRS mass.
3. An analytical model for calculating lateral deformation of a GRS wall with
modular block facing was developed and verified.
4. An analytical model for simulating compaction operation of a GRS mass was
developed. The model allows the compaction-induced stresses in the fill to be
determined.
5. The presence of geosynthetic reinforcement has a tendency to suppress
dilation the surrounding soil, and reduce the angle of dilation of the soil mass.
The dilation behavior offers a new explanation of the reinforcing mechanism,
309
310
and the angle of dilation provides a quantitative measure of the degree of
reinforcing effect of a GRS mass.
APPENDIX A
MATERIAL TESTS
A.1 Backfill
A.1.1 Specific Gravity and Absorption of Coarse Aggregate (per AASHTO T85
and ASTM C127)
• Test Date: April 26 and 27, 2008.
The results of the tests are shown in Table A1 – 1.
311
Table A.1 Specific Gravity determination
A = mass of oven-dry test sample in air, g 184.7
B = mass of saturated-surface-dry test sample in air, g 185.5
C = mass of saturated test sample in water, g 123.4
1. Bulk Specific Gravity = A/(B-C) 2.974
2. Bulk Specific Gravity (Saturated-Surface-Dry) = B/(B-C) 2.987
3. Apparent Specific Gravity = A/(A-C) 3.013
Average Specific Gravity:
3.03
G1 = specific gravity for size fraction passing sieve # 4: 3.038
G2 = apparent specific gravity for size fraction retained on sieve # 4: 3.013
P1 = mass percentage of size faction passing sieve # 4: 58.0
P2 = mass percentage of size faction retained on sieve # 4: 42.0
4. Absorption = [(B – A) / A] x 100, (%) 0.433
2
2
1
1
100100
1
GP
GP
G+
=
312
A.1.2 Moisture-Density (Compaction) Tests (per AASHTO T99 and ASTM
698, Method A)
• Test Date: from May 1st to May 3rd, 2008
A.1.2.1 Density
• Volume of mold = 944 cm3;
• Mass of mold = 4191.5 g.
Table A.2 Unit weight determination for size fraction passing sieve # 4
Compacted Soil number 1 2 3 4 5 6
Actual average water content, % 6.06 7.63 8.57 9.63 11.24 11.92
Mass of compacted soil and mold (g) 6396.9 6466.1 6543.9 6538.9 6521.6 6501.0
Wet mass of soil in mold (g) 2205.4 2274.6 2352.4 2347.4 2330.1 2309.5
Therefore, shear capacity of the tubings is acceptable.
B.2.3 Checking the Bolts
The maximum required shear force of one bolt is:
( ) ( ) ( ) 2.362/2
10*4.25'*'4.11*5002/2
3
=⎟⎟⎠
⎞⎜⎜⎝
⎛=⎟
⎠⎞
⎜⎝⎛=
− mkPaboltslqQreq kN
The required shear stress will be:
( )311,93
4/10*54.2'*'8/7*2.36
23===
−πσ
AQ
req kPa = 13,436 psi = 13.4 kpsi
The shear capacity of 7/8 A325 bolt is: 0.4 * Fy = 0.4 * 36 = 14.4 kpsi.
Therefore, the bolt capacity is acceptable.
347
B.2.4 Checking the Capacity of the Plexiglass
The maximum required moment acting on Plexiglass is:
( ) ( ) 66.312
7/074.2*50012
22
=⎟⎟⎠
⎞⎜⎜⎝
⎛=⎟⎟
⎠
⎞⎜⎜⎝
⎛=
mkPalqM req kN.m
The maximum required stress is:
( ) ( )784,21
61000/4.25'*'25.1*1
66.32 =
⎟⎟⎟⎟
⎠
⎞
⎜⎜⎜⎜
⎝
⎛
=⎟⎟⎠
⎞⎜⎜⎝
⎛=
mSM
x
reqreqσ kPa = 3,137 psi
With the tensile strength of the Plexiglass acrylic sheet of 10,500 psi, the moment
capacity of the plexiglass is acceptable.
348
REFERENCES
Adams, M. T. (1997). “Performance of a Prestrained Geosynthetic Reinforced Soil
Bridge Pier.” Proc. of the International Symposium on Mechanically Stabilized Backfill, J. T. H. Wu, ed., Balkema, Rotterdam, The Netherlands, 25-34.
Adams, M. T. (2007). “Mini Pier Experiments Geosynthetic Reinforcement Spacing
and Strength as Related to Performance.” Proc. of Geo-Denver 2007, ASCE, Denver.
Aggour, M. S., and Brown, C. B. (1974). “The Prediction of Earth Pressure on
Retaining Walls Due to Compaction.” Geotechnique, 24(4), 489-502. American Association of State Highway and Transportation Officials (AASHTO).
(2002). Standard Specifications for Highway Bridges, 17th Ed., Washington, D.C., 829 pages.
Allen, T. M., and Bathurst, R. J. (2001). "Application of the K0-stiffness Method to
Reinforced Soil Wall Limit State Design." Final Research Report to Washington State Department of Transportation, Seatle, Washington State.
Allen, T. M., and Bathurst, R. J. (2001). "Prediction of Soil Reinforcement Loads in
MSE Walls." Report WA-RD-522.1, Washington State Department of Transportation and Federal Highway Administration (FHWA).
Athanasopoulos, G.A. (1994). “On the Enhanced Confining Pressure Approach to the
Mechanics of Reinforced Soil.” Geotechnical and Geological Engineering, 12, 122-132.
Athanasopoulos, G.A. (1993). “Effect of Particle Size on the Mechanical Behavior of
Sand-Geotextile Composite.” Geotextiles and Geomembranes, 12, 255-273.
349
ASTM D4595 (1986). “Standard Test Method for Tensile Properties of Geotextile by the Wide-Width Strip Method.” American Society for Testing and Materials. USA.
Bassett, A.K. and Last, N.C. (1978). “Reinforcing Earth below Footings and
Embankments.” Proc. Of the ASCE Spring Convention and Exhibit, Pittsburgh, PA.
Bathurst, R. J. and Cai, Z. (1994). "In-isolation Cyclic Load-extension Behavior of
Two Geogrids." Geosynthetics International, 1(1), 1-19. Bathurst R. J., and Simac, M. R. (1997). “Design and Performance of the Facing
Column for Geosynthetic Reinforced Segmental Walls.” Proc. of the International Symposium on Mechanically Stabilized Backfill, J. T. H. Wu, ed., Balkema, Rotterdam, The Netherlands, 193-208.
Bathurst R. J., Cai, Z., Alfaro, M., and Pelletier, M. (1997). “Seismic Design Issues
for Geosynthetic Reinforced Segmental Retaining Walls.” Proc. of the International Symposium on Mechanically Stabilized Backfill, J. T. H. Wu, ed., Balkema, Rotterdam, The Netherlands, 79-97.
Bathurst, R. J., Vlachopoulos, N., Walters, D. L., Burgess, P. G., and Allen, T. M.
(2006). “The Influence of Facing Stiffness on the Performance of Two Geosynthetic Reinforced Soil Retaining Walls.” Canadian Geotechnical Journal, 43, 1225-1237.
Bolton, M. D. (1986). "The Strength and Dilatancy of Sands." Geotechnique, 36(1),
65-78. Broms, B. (1971). “Lateral Earth Pressures Due to Compaction of Cohesionless
Soils.” Geotechnical Engineering Report No. 83-1, Cornell University. Broms, B. (1971). "Lateral Earth Pressure due to Compaction of Cohesionless Soils."
Proc., 4th Budapest Conference on soil mechanics and foundation engineering, 373-384.
350
Broms, B. (1977). “Triaxial Tests with Fabric-Reinforced Soil.” Proc. of the International Conference on the use of Fabric in Geotechnics, Paris, 3, 129-134.
Broms, B., and Ingleson, I. (1971). "Earth Pressure against Abutment of Rigid Frame
Bridge." Geotechnique, 21(1), 15-28. Bueno, B. S., Vinicius, C., Benjamim, S., and Zornberg, J. G. (2005). “Field
Performance of a Full-scale Retaining Wall Reinforced with Nonwonven Geotextiles.” Geo-Frontiers, ASCE.
Carder, D. R., Pocock, R. G., and Murray, R. T. (1977). “Experimental Retaining
Wall Facility-Lateral Stress Measurements with Sand Backfill.” Report No. LR 766, Transport and Road Research Laboratory (TRRL), England.
Applicable to Reinforced Soil Analysis.” Geotextile and Geomembranes, 18, 349-366.
Chew, S.H., and Mitchell, J.K. (1994). “Deformation Evaluation Procedure for
Reinforced Walls.” Fifth International Conference on Geotextiles, Geomembranes and related Products, Singapore, 5-9 September 1994.
Collin, J. G. (2002). "Design Manual for Segmental Retaining Walls", NCMA, 2nd
Ed., National Concrete Masonry Association, Herndon, Va. Christopher, B. R., Gill, S. A., Giroud, J. P, Mitchell, J. K., Schlosser, F., and
Dunnicliff J. (1989). “Reinforced Soil Structures. Vol. 1: Design and Construction Guidelines", FHWA-RD-89-043, Federal Highway Administration, Washington, D.C.
Christopher, B.R., Leshchinsky, D., and Stulgis, R. (2005). “Geosynthetic-Reinforced
Soil Walls and Slopes: US Perspective.” GSP 141 International Perspectives on Soil Reinforcement Applications, ASCE.
351
Day, R. (1993). Discussion of "Estimation Earth Pressures due to Compaction" by Duncan, et al. (1991). Journal of Geotechnical Engineering, ASCE, 119(7), 1162.
Duncan, J. M., and Seed, R. M. (1986). "Compaction-Induced Earth Pressures Under
K0-Conditions." Journal of Geotechnical Engineering, ASCE, 112(1), 1-22.
Duncan, J. M., Williams, G. W., Sehn, A. L., and Seed R. M. (1991). "Estimation
Earth Pressures due to Compaction." Journal of Geotechnical Engineering, ASCE, 117(12), 1833-1847.
Duncan, J. M., Williams, G. W., Sehn, A. L., and Seed R. M. (1993). Closure of
"Estimation Earth Pressures due to Compaction." Journal of Geotechnical Engineering, ASCE, 119(7), 1172-1177.
Ehrlich, M., and Mitchell, J. K. (1994). “Working Stress Design Method for
Reinforced Soil Walls.” Journal of Geotechnical Engineering, ASCE, 120(4), 625-645.
Elias, V., and Christopher, B. R. (1996). "Mechanically Stabilized Earth Walls and
Reinforced Soil Slope Design and Construction Guidelines." Report FHWA-SA-96-071, Federal Highway Administration, Washington, D.C.
Elias, V., Christopher, B. R., and Berg, R. R. (2001). "Mechanically Stabilized Earth
Walls and Reinforced Soil Slopes, Design and Construction Guidelines." Report FHWA-NHI-00-043, Federal Highway Administration, Washington, D.C.
Elton, D. J. and Patawaran, M. A. B. (2004). "Mechanically Stabilized Earth
Reinforcement Tensile Strength from Tests of Geotextile-Reinforced Soil." Journal of the Transportation Research Board, No. 1868, TRB, National Research Council, Washington, D.C., 81-88.
352
Elton, D. J. and Patawaran, M. A. B. (2005). "Mechanically Stabilized Earth (MSE) Reinforcement Tensile Strength from Tests of Geotextile Reinforced Soil." A Report to the Alabama Highway Research Center, Auburn University, 77 pages.
Giroud, J. P. (1989). “Geotextile Engineering Workshop-Design Examples.” Report
FHWA-HI-89-002, Federal Highway Administration, Washington, D.C. Gray, D.H. and Ohashi, H. (1983). “Mechanics of fiber reinforcement in sand.”
ASCE, Journal of Geotechnical Engineering, 109, 335-353. Hatami, K., and Bathurst, R. J. (2005). "Development and Verification of a
Numerical Model for the Analysis of Geosynthetic Reinforced-soil Segment Walls." Canadian Geotechnical Journal, 42(4), 1066-1085.
Hatami K. and Bathurst R. J. (2006). "Numerical Model for Reinforced Soil
Segmental Walls under Surcharge Loading." Journal of Geotechnical and Geoenvironental Engineering, ASCE, 132(6), 673-684.
Hausmann, M.R. (1976). “Strength of Reinforced Earth.” ARRB Proc., Vol. 8. Helwany, M. B., Tatsuoka, F., Tateyama, M., and Kojima, K. (1996). “Effects of
Facing Rigidity on the Performance of Geosynthetic-Reinforced Soil Retaining Walls.” Soil and Foundations, Japan, 36(1), 27-38.
Ho, S. K. and Rowe, R. K. (1997). “Effect of Wall Geometry on the Behavior of
Reinforced Soil Walls.” Geotextiles and Geomembranes, 14, 521–541. Holtz, R. D., Christopher, B. R., and Berg, R. (1997). Geosynthetic Engineering,
BiTech Publishers, Vancouver, 452 pages. Holtz, R. D. and Lee, W. F. (2002). "Internal Stability Analyses of Geosynthetic
Reinforced Retaining Walls." Report WA-RD 532.1, Washington State Department of Transportation, Washington.
353
Ingold, T. S. (1979). “The Effect of Compaction on Retaining Walls.” Geotechnique, 29(3), 265-283.
Ingold, T.S. (1982). “Reinforced Earth.” Thomas Telford Ltd, London. Jewell, R. A. (1985). “Limit Equilibrium Analysis of Reinforced Soil Walls.” Proc. of
11th International Conference of Soil Mechanics and Foundation Engineering, San Francisco, 3, 1705-1708.
Jewell, R. A. and Wroth, C. P. (1987). “Direct Shear Tests on Reinforced Sand.”
Geotechnique, 37(1), 53-68. Jewell, R. A. (1988). "Reinforced Soil Wall Analysis and Behavior." The Application
of Polimeric Reinforcement in Soil Retaining Structures, NATO Advanced Research Workshop, 365-408.
Jewell, R. A., and Milligan, G. W. (1989). "Deformation Calculation for Reinforced
Soil Walls." International Conference on Soil Mechanics and Foundation Engineering, 1259-1262.
Jewell, R. A. (1990). "Strength and Deformation in Reinforced Soil Design." The
fourth International Conference on Geotextiles, Geomembranes and Related Products, The Hague, Netherlands.
Katona, M. G. (1978). "The Analysis of Long Span Culverts by Finite Element
Method." Paper submitted to Transportation Research Board, Washington D.C.
Ketchart, K. (2000). "Behavior of Preloaded Geosynthetic-Reinforced Soil Mass."
PhD. Thesis, University of Colorado at Denver, 347 pages. Ketchart, K., and Wu, J. T. H. (2001). "Performance Test for Geosynthetic
Reinforced Soil Including Effects of Preloading", Report FHWA-RD-01-018, Federal Highway Administration, Washington, D.C.
354
Lee, W. F. (2000). "Internal Stability Analyses of Geosynthetic Reinforced Retaining Walls." Ph.D. Thesis, Dept. of Civil and Environmental Engineering, University of Washington, 379 pages.
Long N. T., Legeay, G., and Madani, C. (1983). “Soil-Reinforcement Friction in a
Triaxial Test.” Proc. of the 8th European Conference on Soil Mechanics and Foundation Engineering, Hesinki, Finland, 1, 381-384.
Macklin, P. M. (1994). "A Comparison of Five Methods for Calculation Lateral
Deformation of Geosynthetically Reinforced Soil (GRS) walls." Master Thesis, University of Colorado at Denver, Colorado, 70 pages.
Maher, M.H. and Woods, R.D. (1990). “Dynamic Response of Sand Reinforced with
Randomly Distributed Fibers.” ASCE, Journal of Geotechnical Engineering, 116, 1116-1131.
Mitchell, J.K., Villet, W.C.B. (1987). “Reinforcement of Earth Slopes and
Embankments.” National Cooperative Highway Research Program, Report 290, TRB, 323 pages.
Morrison, K. F., Harrison, F. E., Collin, J. G., Dodds, A., and Arndt, B. (2006).
"Shored Mechanically Stabilized Earth (SMSE) Wall Systems Design Guidelines." Report FHWA-CFL/TD-06-001, Federal Highway Administration, Lakewood, Colorado, 210 pages.
Morrison, K. F., Harrison, F. E., Collin, J. G., and Anderson, S. A. (2007). "Full-
Scale Testing of a Shored Mechanically-Stabilized Earth (SMSE) Wall Employing Short Reinforcements." Proc. of Geo-Denver 2007, ASCE, Denver.
Peck, R. B. (1993). Discussion of "Estimation Earth Pressures due to Compaction "
by Duncan et al. (1991), 117(12). Journal Geotechnical Engineering, ASCE, 119(7), 1171-1172.
Plaxis B.V. (2002). “Plaxis 2D - Version 8 Manual.” Balkema, Rotterdam, The
Netherlands.
355
Rowe, P.W. (1954). “A Stress-Strain Theory for Cohesionless Soil with Applications to Earth Pressures at Rest and Moving Walls.” Geotechnique, 4(2), 70-88.
Rowe, R. K. and Ho S. K. (1993). “Some Insights into Reinforced Wall Behaviour on
Finite Element Analysis.” Earth Reinforcement Practice, Ochiai, Hayashi, and Otani, eds., Balkema, Rotterdam, The Netherlands, 485-490.
Rowe, R. K. and Ho S. K. (1998). “Horizontal Deformation in Reinforced Soil
Walls.” Canadian Geotechnical Journal, 35, 312-327. Schlosser, F., Long, N. (1972). “Comportement de al terre armée dans les ouvrages
de soutènement”. In: Proceedings of the European Conference on Soil Mechanics and Foundation Engineering, vol. 1. Madrid, pp. 299–306.
Schlosser, F. and Long, N.T. (1974). ”Recent Results in French Research on
Reinforced Earth.” ASCE, Journal of Construction Division, 100, 223-237. Seed, R. M. (1983). "Compaction-induced Stresses and Deflections on Earth
Structure." PhD. Thesis, University of California, Berkeley, CA, 447 pages. Seed, R. M., Duncan, J. M. (1986), "FE Analyses: Compaction-Induced Stresses and
Deformations." Journal of Geotechnical Engineering, ASCE, 112(1), 23-43.
Sower, G. F., Robb, A. D., Mullis, C. H. and Glenn, A. J. (1957). “The Residual Lateral Pressures Produced by Compacting Soils.” Proc. of 4th International Conference on Soil Mechanics and Foundation Engineering, London.
Tatsuoka, F., Molenkamp, F., Torii, T., and Hino, T. (1997). "Preloaded and
Prestressed Reinforced Soil." Soil and Foundations, Japan, 37(3), 79-94. Tatsuoka, F. (1993). “Roles of Facing Rigidity in Soil Reinforcing.” Earth
Reinforcement Practice, Ochiai, Hayashi, and Otani, eds., Balkema, Rotterdam, The Netherlands, 831-867.
Terzaghi, K. (1943). “Large Retaining Wall Tests (I): Pressure of Dry Sand.”
Engineering News Record, 112, 136-140.
356
Vulova, C. and Leshchinsky, D. (2003). “Effect of Geosynthetic Reinforcement Spacing on the Performance of Mechanically Stabilized Earth Walls.” Publication No. FHWA-RD-03-048, Federal Highway Administration, McLean, VA, USA.
Westergaard, C. M. (1938). “A Problem of Elasticity Suggested by a Problem in Soil
Mechanics: A Soft Material Reinforced by Numerous Strong Horizontal Sheet,” in Contributions to the Mechanics of Solids, Stephen Timoshenko 60th Anniversary Volume, Macmillan, New York, 268-277.
Wu, J.T.H. (1989), “Behavior of Soil-Geotextile Composites and Its Application to
Finite Element Analysis,” Proc. of Geosynthetics’ 89, 365-372. Wu, J.T.H. (1994). "Design and Construction of Simple, Easy and Low cost
Retaining Walls." Colorado Transportation Institute, Report CTI-UCD-1-94. Wu, J. T. H. (2001). “Revising the AASHTO Guidelines for Design and Construction
of GRS Walls.” Report CDOT-DTD-R-2001-6, Colorado Department of Transportation, University of Colorado at Denver, 148 pages.
Wu, J. T. H. "Design and Construction of Geosynthetic-Reinforced Soil Retaining
Walls." Draft of Textbook (unpublished). Wu, J. T. H., Lee, K. Z. Z., Helwany, S. B., and Ketchart, K. (2006). "Design and
Construction Guidelines for GRS Bridge Abutment with a Flexible Facing." NCHRP Report 556, National Cooperative Highway Research Program, Washington, D.C.
Wu, J.T.H., Lee, K.Z.Z., and Pham, T. (2006). “Allowable Bearing Pressure of
Bridge Sills on GRS Abutments with Flexible Facing.” J. Geotechnical and Geoenvironmental Engineering, ASCE, 132(7), 836-841.
Wu, J. T. H. (2007). "Lateral Earth Pressure against the Facing of Segmental GRS
Walls", Proc. of Geo-Denver 2007, ASCE, Denver.
357
358
Yang, Z. (1972). "Strength and Deformation Characteristics of Reinforced Sand." PhD. Thesis, University of California at Los Angeles, CA, 236 pages.
Yang, Z. and Singh, A. (1974). “Strength and Deformation Characteristics of
Reinforced Sand.” International Meeting on Water Resources Engineering, Los Angeles, CA.
Zhang, M.X., Javadi, A.A., Lai, Y.M., and Sun, J. (2006). “Analysis of Geosynthetic
Reinforced Soil Structures with Orthogonal Anisotropy.” Geotechnical and Geological Engineering, 24, 903-917.
Ziegler, M., Heerten, G., and Ruiken. G. (2008). “Progress in the Understanding of
Geosynthetic/Soil Composite Material Behaviour in Geosynthetic Reinforced Earth Structures.” The First Pan American Geosynthetics Conference & Exhibition, 2-5 March 2008, Cancun, Mexico.