Soil Mechanics and Foundation Engineering

SOIL MECHANICS AND
FOUNDATION ENGINEERING
SECOND EDITION
A01_PURU1773_01_SE_PREL.indd 1 2/12/2013 5:21:17 PM

SOIL MECHANICS AND
FOUNDATION ENGINEERING
SECOND EDITION
P. Purushothama Raj
Director
Adhiparasakthi Engineering College, Melmaruvathur
Kancheepuram District, Tamil Nadu

&RS\ULJKW'RUOLQJ.LQGHUVOH\,QGLD3YW/WG
/LFHQVHHVRI3HDUVRQ(GXFDWLRQLQ6RXWK$VLD

1RSDUWRIWKLVH%RRNPD\EHXVHGRUUHSURGXFHGLQDQ\PDQQHUZKDWVRHYHUZLWKRXWWKHSXEOLVKHU¶V
SULRUZULWWHQFRQVHQW

7KLVH%RRNPD\RUPD\QRWLQFOXGHDOODVVHWVWKDWZHUHSDUWRIWKHSULQWYHUVLRQ7KHSXEOLVKHU
UHVHUYHVWKHULJKWWRUHPRYHDQ\PDWHULDOLQWKLVH%RRNDWDQ\WLPH

,6%1
H,6%19789332514119

+HDG2IILFH$$6HFWRU.QRZOHGJH%RXOHYDUGWK)ORRU12,'$,QGLD
5HJLVWHUHG2IILFH/RFDO6KRSSLQJ&HQWUH3DQFKVKHHO3DUN1HZ'HOKL,QGLD

Brief Contents
Preface xvii
1. Soil Formation and Composition 1

2. Index Properties of Soils 25
3. Identification and Classification of Soils 65
4. Compaction of Soils 81
5. Permeability and Capillarity 103
6. Seepage 137
7. Stress and Stress Distribution in Soil 167
8. Consolidation and Consolidation Settlement 215
9. Shear Strength of Soils 261
10. Laboratory Measurement of Soil Properties 309
11. Lateral Earth Pressure 383
12. Earth-Retaining Structures 425
13. Stability of Slopes 471
14. Bearing Capacity of Soils 515
15. Shallow Foundations 567
16. Pile Foundations 599
17. Drilled Piers and Caisson Foundations 641
18. Ground Investigation 671
19. Soil Improvement 699
20. Embankment Dams 729
21. Dynamic Loading of Soil 759
22. Environmental Geotechnology 777
23. Introductory Rock Mechanics 787
24. Pavements 815
List of Symbols 837

Bibliography 843
Index 857

Contents
Preface xvii
1. Soil Formation and Composition 1

Chapter Highlights
1.1 Introduction 1
1.2 Origin of Soils 2
1.3 Types of Weathering 3
1.4 Soil Formation 4
1.5 Major Soil Deposits of India 7
1.6 Components of Soils 9
1.7 Particle Sizes and Shapes 11
1.8 Inter-Particle Forces 12
1.9 Soil Minerals 13
1.10 Soil–Water System 16
1.11 Physico-Chemical Behaviour of Clays 16
1.12 Soil Structure 19
Points to Remember 22
Questions 22
2. Index Properties of Soils 25

Chapter Highlights
2.1 Introduction 25
2.2 Three-Phase System 25
2.3 Particle-Size Analysis 34
2.4 Consistency of Soils 40
Worked Examples 45
Questions 60
Exercise Problems 61
3. Identification and Classification of Soils 65

Chapter Highlights
3.1 Introduction 65
3.2 Field Identification of Soils 65
3.3 Engineering Classification of Soils 67
Worked Examples 75

viii Contents
Questions 77
4. Compaction of Soils 81
Chapter Highlights
4.1 Introduction 81
4.2 Principles of Compaction 81
4.3 Compactive Effort 82
4.4 Laboratory compaction 82
4.5 Field Compaction and Equipment 84
4.6 Compaction Specification and Control 87
4.7 Factors Affecting Compaction 89
4.8 Effect of Compaction on Soil Structure 92
4.9 Compaction Behaviour of Sand 93
4.10 California Bearing Ratio Test 93
Worked Examples 94
Questions 100
5. Permeability and Capillarity 103

Chapter Highlights
5.1 Introduction 103
5.2 Water Flow 103
5.3 Darcy’s Law 104
5.4 Range of Validity of Darcian Flow 105
5.5 Laboratory Permeability Tests 106
5.6 Field Permeability Tests 108
5.7 Permeability of Stratified Soils 112
5.8 Values of Permeability 115
5.9 Factors Affecting Permeability 115
5.10 Surface Tension 117
5.11 Capillary Phenomenon in Soils 118
5.12 Shrinkage and Swelling of Soils 120
Worked Examples 121
Questions 132
6. Seepage 137
Chapter Highlights
6.2 Seepage Forces 137
6.3 General Flow Equation 139

Contents ix
6.4 Significance of Laplace Equation 141

6.5 Properties and Applications of Flow Nets 142
6.6 Construction of Flow Net 144
6.7 Anisotropic Soil Conditions 149
6.8 Non-Homogeneous Soil Conditions 152
6.9 Piping 153
6.10 Design of Filters 154
Worked Examples 155
Questions 161
7. Stress and Stress Distribution in Soil 167

Chapter Highlights
7.2 Stresses at a Point 167
7.3 Mohr’s Circle 169
7.4 Stress Paths 170
7.5 Effective Stress Concept 171
7.6 Geostatic Stresses 175
7.7 Stresses due to Surface Loads 179
Worked Examples 198
Questions 209
8. Consolidation and Consolidation Settlement 215

Chapter Highlights
8.2 Rheological Models of soils 216
8.3 Compressibility of soils 217
8.4 One-Dimensional Consolidation 217
8.5 Consolidation Test 219
8.6 Compressibility Characteristics 221
8.7 Types of Clay Deposits 223
8.8 Prediction of Pre-consolidation Pressure 226
8.9 Rate of Consolidation 226
8.10 Secondary Compression 233
8.11 Consolidation Settlement and Its Rates 234
8.12 Acceleration of Consolidation by Sand Drains 239
8.13 Compressibility of Sands 241
Worked Examples 242
Questions 255

x Contents
9. Shear Strength of Soils 261

Chapter Highlights
9.2 Stress–Strain Curve 261
9.3 Mohr–Coulomb Failure Criterion 262
9.4 Peak and Residual Shear Strengths 263
9.5 Laboratory Measurement of Shear Strength 266
9.6 Field Measurement of Shear Strength 278
9.7 Shear Strength of Saturated Cohesive Soils 281
9.8 Pore Pressure Coefficients 286
9.9 Sensitivity of Cohesive Soils 290
9.10 Thixotrophy of Clays 291
9.11 Shear Strength of Granular Soils 292
Worked Examples 294
Questions 304
10. Laboratory Measurement of Soil Properties 309

Chapter Highlights
10.2 Test No. 1: Preparation of Dry Soil Samples for Various Tests 310
10.3 Test No. 2: Specific Gravity of Soil Solids 311
10.4 Test No. 3: Water Content Determination by Oven-Drying Method 314
10.5 Test No. 4: In-Place Dry Density of Soil by Core-Cutter Method 315
10.6 Test No. 5: In-Place Dry Density of Soil by the Sand Replacement Method 317
10.7 Test No. 6: Grain-Size Distribution by Sieve Analysis 320
10.8 Test No. 7: Grain-Size Distribution by Pipette Method 322
10.9 Test No. 8: Grain-Size Distribution by the Hydrometer Method 329
10.10 Test No. 9: Liquid Limit of Soil 334
10.11 Test No. 10: Plastic Limit of Soil 337
10.12 Test No. 11: Shrinkage Factors of Soil 338
10.13 Test No. 12: Linear Shrinkage of Soil 341
10.14 Test No. 13: Permeability Test 342
10.15 Test No. 14: Free Swell Index of Soils 346
10.16 Test No. 15: Moisture Content – Dry Density Relationship (Standard
Proctor Compaction Test) 347
10.17 Test No. 16: Density Index of Non-Cohesive Soils 351
10.18 Test No. 17: Consolidation Test 353
10.19 Test No. 18: Unconfined Compression Test 359
10.20 Test No. 19: Direct Shear Test 362
10.21 Test No. 20: Triaxial Shear Test 365
10.22 Test No. 21: California Bearing Ratio (CBR) Test 373
Questions 381

Contents xi
11. Lateral Earth Pressure 383

Chapter Highlights
11.2 Limit Analysis and Limit Equilibrium Methods 384
11.3 Earth Pressure at Rest 384
11.4 Rankine’s States of Plastic Equilibrium 386
11.5 Rankine’s Earth Pressure Theory 389
11.6 Coulomb’s Earth Pressure Theory 400
11.7 Culmann’s Graphical Method 402
11.8 Poncelet’s Graphical Method 404
11.9 Arching of Soils 405
Worked Examples 407
Questions 419
12. Earth-Retaining Structures 425

Chapter Highlights
12.2 Gravity-Type Retaining Walls 425
12.3 Sheet Pile Walls 432
12.4 Braced Excavations 445
Worked Examples 450
Questions 464
13. Stability of Slopes 471

Chapter Highlights
13.2 Causes of Slope Failures 471
13.3 Short- and Long-Term Failures 473
13.4 Types of Landslides and Slope Movements 473
13.5 Factor of Safety 476
13.6 Basic Concepts of Slope Stability Analysis 478
13.7 Infinite and Finite Slopes 478
13.8 Analysis of Infinite Slopes 480
13.9 Analysis of Finite Slopes 483
13.10 Selection of Shear Strength Parameters and Stability Analysis 500
13.11 Slope Protection Measures 501
Worked Examples 503
Questions 509

xii Contents
14. Bearing Capacity of Soils 515

Chapter Highlights
14.2 Bearing Capacity 516
14.3 Modes of Shear Failure 516
14.4 Terzaghi’s Bearing Capacity Theory 519
14.5 Foundation Pressures 525
14.6 Special Loading and Ground Conditions 526
14.7 Other Bearing Capacity Theories 533
14.8 Bearing Capacity of Soils from Building Code 540
14.9 Permissible Settlements 542
14.10 Allowable Bearing Pressure 544
14.11 Estimation of Bearing Capacity from Field Tests 546
Worked Examples 551
Questions 561
15. Shallow Foundations 567

Chapter Highlights
15.2 Design Criteria 567
15.3 Types of Shallow Foundations 568
15.4 Selection of the Type of Foundation 570
15.5 Location and Depth of the Foundation 571
15.6 Causes of Settlement 573
15.7 Settlement of Shallow Foundations 573
15.8 Design Steps for a Shallow Foundation 583
15.9 Proportionating Footing Size 584
15.10 Design of Combined Footings 585
15.11 Mat Foundation 589
Worked Examples 591
Questions 595
16. Pile Foundations 599

Chapter Highlights
16.2 Classification of Piles 599
16.3 Pile-Driving Equipment 605
16.4 Bearing Capacity of Single Pile 607
16.5 Under-Reamed Piles 618
16.6 Pile Groups 621

Contents xiii
16.7 Uplift Resistance of Piles 626

16.8 Lateral Resistance of Piles 627
16.9 Inclined Loading of Vertical Piles 629
16.10 Pile Cap 629
Worked Examples 630
Questions 636
17. Drilled Piers and Caisson Foundations 641

Chapter Highlights
17.2 Drilled Piers 641
17.3 Caissons 650
17.4 Well Foundations 655
Questions 668
18. Ground Investigation 671

Chapter Highlights
18.2 Planning the Ground Investigation Programme 672
18.3 Types of Soil and Rock Samples 672
18.4 Indirect Methods of Sub-Surface Exploration 673
18.5 Semi-Direct Methods of Sub-Surface Exploration 677
18.6 Direct Methods of Sub-Surface Exploration 679
18.7 Routine Field Tests 688
18.8 Recording of Field Data 693
18.9 Location, Spacing, and Depth of Borings 694
Questions 695
19. Soil Improvement 699

Chapter Highlights
19.2 Improvement Techniques 700
19.3 Surface Compaction 700
19.4 Drainage Methods 701
19.5 Vibration Methods 705
19.6 Pre-Compression and Consolidation 712
19.7 Grouting and Injection 715
19.8 Chemical Stabilization 719

xiv Contents
19.9 Soil Reinforcement 720

19.10 Geotextiles and Geomembranes 721
19.11 Other Methods 724
Questions 726
20. Embankment Dams 729

Chapter Highlights
20.2 Types of Embankment Dams 729
20.3 Components of Embankment Dams 731
20.4 Design Criteria for Earth Dams 734
20.5 Selection of Dam Section 751
Worked Examples 751
Questions 754
21. Dynamic Loading of Soil 759

Chapter Highlights
21.2 Earthquakes 759
21.3 Other Dynamic Loads 761
21.4 Theory of Vibrations 761
21.5 Types of Machines and Machine Foundations 766
21.6 Dynamic Bearing Capacity of Shallow Foundations 767
21.7 Design Requirements 768
21.8 Methods of Analysis for Block Foundation 771
21.9 Liquefaction of Soils 773
Questions 775
22. Environmental Geotechnology 777

Chapter Highlights
22.2 Environmental Cycles 777
22.3 Natural Cycles 778
22.4 Environmental Imbalance 779
22.5 Birth of Environmental Geotechnology 781
22.6 Contaminated Soils 781
22.7 Applications 782
22.8 Load–Environment Design Criteria 783
Questions 784

Contents xv
23. Introductory Rock Mechanics 787

Chapter Highlights
23.2 Index Properties of Rocks 787
23.3 Classification of Rocks 793
23.4 In Situ State of Stress 797
23.5 Mechanical Properties of Rocks 800
Questions 812
24. Pavements 815

Chapter Highlights
24.2 Components of Pavement 815
24.3 Types of Pavement 816
24.4 Requirements of Pavement Components 816
24.5 Subgrade 817
24.6 Pavement Design 818
24.7 Design of Flexible Pavements 820
24.8 Design of Rigid Pavements 828
Worked Examples 831
Questions 835
List of Symbols 837

Bibliography 843
Index 857

Preface
Popularity of this book amongst the undergraduate students and practising engineers has
made the author to revive the book with updated materials.
The second edition of the book comprises 24 chapters dealing with four components,
viz. Basic Principles of Soil Mechanics, Laboratory Determination of Soil Parameters Under
Different Field Conditions, Earth Pressure Problems Including Foundations and Advanced
Topics on Soil Mechanics Applications.
Chapters 1–9 deal with Basic Principles of Soil Mechanics. Chapter 1 deals with soil
formation and composition, highlighting the types of weathering, soil formation in nature
and major soil deposits of India. Chapters 2 and 3 explain the methods of identification
and classification of soil including Bureau of Indian Standards, with the background
knowledge of the index properties of soils. The basic properties of soils are compaction,
permeability, consolidation and shear strength. Chapters 4–9 detail these properties. Prin-
ciples of compaction and field compactions are explained in Chapter 4. Flow through
porous medium and its applications are dealt within Chapters 5 and 6. Stresses in nature
and applied stresses cause consolidation and failure due to shear. These aspects are dealt
with at length in Chapters 7, 8 and 9. Chapter 7 details the different types of loading and
the methods of computing stresses. Chapter 8 explains the basic theory of consolidation
followed by computation of settlements while Chapter 9 explains the methods of deter-
mining shear strength of different types of soils under different loading and drainage
conditions.
Chapter 10 discusses the basic techniques of testing of soils as per the Bureau of Indian
Standards. Further, it includes methods of material collection, data presentation, com-
putation and presentation of results and discussion. Necessary diagrams and standard
values are included in the chapter.
Chapters 11–19 explain the principles of earth pressure and its applications in the
design of earth-retaining structures, stability of slopes and foundations. Principles of
earth pressure theories, in particular, the classical earth pressure theories of Coulomb
and Rankine, and other modern theories are dealt with in Chapter 11. Chapter 12 gives
the design concepts of retaining walls, including sheet piles and cuts, for both active
and passive cases with different backfill conditions. Stability of slopes for different soil
conditions, seepage conditions and pure pressure conditions are dealt with in detail in
Chapter 13 with different methods of analysis.
Chapter 14 deals with the bearing capacity of soils and the connected theories for various
ground conditions. Determination of safe bearing capacity and allowable soil pressure for
different loading conditions are presented in Chapter 14. Chapters 15–17 cover the subjects
on foundation engineering viz. shallow foundations, pile foundations drilled and caisson
foundations. The subject matter has been dealt with in depth so as to introduce the student

xviii Preface
to the field of foundation engineering such that he/she will gain the ability to investigate
and select the most suitable type of foundation. A foundation engineer or a student can select
the best foundation, provided he/she knows the subsoil conditions and methods of ground
improvements. These aspects are dealt with in Chapters 18 and 19. The latest methods of
ground investigation and soil improvement techniques are explained in these chapters.
Advanced topics on Soil Mechanics Applications viz. Embankment Dams, Dynamic
Loading of Soils, Environmental Geotechnology and Introductory Rock Mechanics are
explained in Chapters 20–23. Chapter 24 on Highway Pavements, an application of soil
mechanics, has been added in this second edition. Chapter 20 on Embankment Dams covers
both homogeneous and non-homogeneous dams, including rock-fill dams. Theory of vibra-
tions, theory of machine and machine foundations and design requirements are explained
in Chapter 21. Chapter 22 gives a brief account of Environmental Geotechnology. An intro-
duction on rock mechanics has been given in Chapter 23 which explains the index proper-
ties of rocks, classification and in situ stresses.
The second edition would not have been possible but for the excellent encouragement
given by M/S Pearson Education for which the author expresses his gratitude.
Pondicherry
January 2013 P. Purushothama Raj

1
Soil Formation and
Composition
CHAPTER HIGHLIGHTS
Origin of soils – Types of rocks – Effects of weathering – Soil formation –
Major soil deposits of India – Components of soils – Size and shape of
soil particles – Inter-particle forces – Soil minerals – Soil–water system – Soil
structure
1.1 INTRODUCTION
Soil is an unconsolidated material that has resulted from the disintegration of rocks. It
includes sediments and deposits beneath rivers and seas and on land along with all
organic and inorganic materials overlying the bedrock. It, thus, constitutes the earth’s
surface both on land and beneath water.
The type and characteristics of soil depend largely on its origin. Transportation
causes the sizes and shapes of the particles to alter and sort into sizes. Cementation due
to carbonates, oxides, or organic matter provides additional particle binding. Thus, the
engineering properties, viz., permeability, consolidation, and shear strength, of a soil
deposit are governed by the mode of formation, stress history, groundwater condition,
and physico-chemical characteristics of the parent material.
Soil deposits constitute an assemblage of solid particles resulting in the formation of
certain voids or pore spaces. These voids are in turn filled with a gas or liquid or both.
These components, viz., solid particle, gas, and liquid, play a significant role in the fun-
damental behaviour of soil. Thus, we can visualize the soil deposit as a particulate system
comprising three phases, viz., the solid phase, the liquid phase, and the gaseous phase.
This chapter deals with the geological aspects of the formation of several types of
soil deposits. The composition of such soil deposits has been treated as a three-phase
system. The factors contributing to the behaviour of each phase and to the soil
structure formation are dealt with in this chapter.
M01_PURU1773_01_SE_C01.indd 1 2/12/2013 7:50:54 AM

2 Soil Mechanics and Foundation Engineering
1.2 ORIGIN OF SOILS

The earth’s crust consists of both rock and weathered rock (as soil) up to an approximate
depth of 20 km. Soils (weathered rocks) originated from the rocks and minerals of the earth’s
crust. The principal minerals subjected to weathering to produce soil at or near the earth’s
surface and available in the order of abundance are quartz, feldspar, pyroxene, amphibole, etc.
Continuous slow weathering processes, aided by crustal deformities in the past, are believed
to have decomposed the solid rock to fragments, creating soils. The type of soil developed
depends on the rock type, its mineral constituents, and the climatic regime of the area. Rock
types are grouped into three major classes: igneous, sedimentary, and metamorphic.
Cooling and hardening of molten magma resulted in the formation of igneous rocks. Slow
cooling of molten magma yields large crystals, while rapid cooling results in small crystals.
Granite, syenites, diorites, and gabbros have large crystals, while basalts, rhyolites, and
andesites have small crystals.
Rocks containing quartz or orthoclase minerals with high silica content (e.g., granite and
rhyolite) mostly decompose into sands* or gravelly** soils with a little clay.*** On the other
hand, rocks (e.g., gabbros and basalt) whose minerals contain iron, magnesium, calcium, or
sodium with a little silica decompose to yield fine-textured silty+ and clayey soils. Clays are
not fragments of primary minerals from the parent rock but secondary minerals formed by
the decomposition of primary minerals. Thus, the behaviour of clay is different from that of
gravel and sand as the latter are composed of primary minerals.
Transportation agents such as wind, water, and ice may move the loose weathered
rock materials and deposit them in layers called sediments. Such sediments, with the
cementing properties of fragments, when subjected under heavy pressure to compaction
and cementation, result in sedimentary rocks. Sedimentary type of rocks are classified as
chemical (e.g., limestones and dolomites), clastic (e.g., shale and sandstone), and biochemical
or organic (e.g., fossil limestone, chalk, coral, and coal in the form of peat, lignite, bitumen,
or anthracite).
Sedimentary rocks, and to a lesser extent igneous rocks, when subjected to metamor-
phism (changes brought about by combinations of heat, pressure, and plastic flow),
undergo changes in their texture, structure, and mineral composition, resulting in rocks
called metamorphic rocks. Metamorphism is of two types, thermal and dynamic. Thermal
metamorphism occurs primarily due to temperature increase and high hydrostatic pressure,
whereas dynamic metamorphism is due to differential pressure. Metamorphism changes
limestone to marble, sandstone to quartzite, and shale to slate.
Metamorphic rocks may be categorized as foliated or non-foliated. Foliation occurs during
the process of metamorphism when some metamorphic rocks reduce back to sedimentary
rocks. Gneiss and schist decompose into silt–sand mixtures with mica, slates and phyllites
to clays, marble to limestone, and quartzite to sands and gravels.
The cyclic process of transforming rock to soil and vice versa is a continuous process
occurring over millions of years through complex chemical and physical processes. This
phenomenon, referred to as the geological cycle, is schematically shown in Fig. 1.1.
Particle sizes: *sand, 0.075 to 2 mm; **gravel, 2 to 4.75 mm; ***clay, < 0.002 mm; +silt, 0.002 to 0.075 mm.
M01_PURU1773_01_SE_C01.indd 2 2/12/2013 7:50:54 AM

Soil Formation and Composition 3
sible melting of deep rocks

Pos
Heat, pressure, and Heat, pressure, and

Igneous solution Meta- solution Sedi-
Melting of deeply mentary
rocks morphic
buried rocks rocks Weathering, rocks
deposition, and
consolidation
d
weathering
Erosion and
W
ea an
n
th i o i ng nd
er os er a
in
g Er ath tion
we pac tion
m ta
Co men
ce
Gravel,
sand, mud,
and
other
sediments
Fig. 1.1 Geological cycle (Source: Bowles, 1984)
1.3 TYPES OF WEATHERING

Rock disintegration, also called weathering, is one of the important geological processes.
This disintegration of rock produces and deposits unconsolidated sediments as soils for
plant and animal life. Weathering may be either physical (or mechanical) or chemical.
1.3.1 Physical Weathering

The process by which rock disintegrates into smaller fragments due to factors like stress
changes, climatic changes, etc., without involving any change in its properties is called
physical or mechanical weathering.
The principal factor causing physical weathering is climatic change. Adverse tempera-
ture changes coupled with different thermal coefficients of rock minerals produce rock
fragments. The effect will be greater when temperature changes cause a freeze–thaw cycle.
Similarly, heavy rainfall also brings about physical weathering.
Stress readjustments during regional uplift, accompanied by water runoff, cause the
outer shell to separate from the main rock. This process is called exfoliation and may also be
caused by sudden temperature changes.
On a rugged topography, heavy wind and rain may cause erosion of the rock surface
and move disintegrated fragments. This is a continuing event and depends on the type of
topography and the velocity of wind and water. Abrasion of rock is caused by ice under
pressure or by the pushing of large unconsolidated materials. Mechanical weathering may
also be caused by organic activity, such as cracking forces exerted by plants growing in the
crevasses of rocks and the moving of fragments towards the surface by animals or insects.
M01_PURU1773_01_SE_C01.indd 3 2/12/2013 7:50:54 AM

1.3.2 Chemical Weathering

Due to alteration in the chemical properties of rock minerals, new compounds are formed.
This is referred to as chemical weathering. Rainwater coming in contact with rock surfaces
reacts forming hydrated iron oxide, carbonates, and sulphates. If there is a volume increase,
the disintegration continues further. Rainwater with pH < 7, or with carbonic acid, may react
chemically with some rock (e.g., limestone) and completely dissolve it. Further, during a
geological time period, even a weak acid solution may cause decomposition. Due to leaching,
sedimentary rocks may lose their cementing properties.
1.4 SOIL FORMATION

Based on the method of formation, soil may be categorized as residual and transported.
Residual soils are formed from the weathering of rocks and practically remain at the location
of origin with a little or no movement of individual soil particles. Transported soils are those
that have formed at one location (like residual soils) but are transported and deposited at
another location.
1.4.1 Residual Soils

Weathering (due to climate effects) and leaching of water-soluble materials in the rock are
the geological processes in the formation of these soils. The rate of rock decomposition is
greater than the rate of erosion or transportation of weathering material and results in the
accumulation of residual soils. As the leaching action decreases with depth, there is a
progressively lesser degree of rock weathering from the surface downwards, resulting in
reduced soil formation, until one finally encounters unaltered rock (Fig. 1.2). Residual soils
generally comprise a wide range of particle sizes, shapes, and composition.
In general, the rate of weathering is greater in warm, humid regions than in cold, dry
regions. Humid, warm regions are favourable to chemical weathering. Also, because of the
Top Soil – Humus Zone I
Surficial Soil Zone Zone II

(Oldest Soil Material)
Completely Weathered
Rock Zone (Virtually Zone III
all Soil Material)
Highly Weathered
Rock Zone (Mostly Zone IV
Soil Material)
Moderately Weathered
Rock Zone (Distintegrated Zone V
Rock)
Slightly Weathered
Rock Zone Zone VI
Unweathered Rock
with Fissures and Zone VII
Fractures in Upper
Zone
Fig. 1.2 Stages of formation of residual soil (Source: McCarthy, 1982)
M01_PURU1773_01_SE_C01.indd 4 2/12/2013 7:50:56 AM

presence of vegetation, there is less possibility of transportation of the decomposed materials

as sediments. Residual soils exist in different parts of the world, viz., Asia, Africa, south-east-
ern North America, Central America, and South America. Sowers (1963) reported that the
depth of residual soils varies from 6 to 25 m in general and from 7.5 to 15 m in South India.
1.4.2 Transported Soils

Weathered materials have been moved from their original location to new locations by one
or more of the transportation agencies, viz., water, glacier, wind, and gravity, and deposited
to form transported soils. Such deposits are further classified depending on the mode of
transportation causing the deposit.
Water-transported Soils. Swift-running water is capable of moving a considerable vol-
ume of soil. Soil may be transported in the form of suspended particles or by rolling and
sliding along the bottom of the stream. The size of the particle that can be in suspension
is related to the square of the velocity of the flowing water. Particles transported by water
range in size from boulders to clay.
Coarser particles are dropped when a decrease in water velocity occurs as the stream or
river deepens, widens, or changes direction. Fine particles still remain in suspension and get
deposited in quieter waters downstream. This is a typical case of a stream moving downhill,
passing over a valley, and ultimately reaching a large body of water. Soils that are carried
and deposited by rivers are called alluvial deposits. River deltas are formed in this manner.
Soils carried by rivers, while entering a lake, deposit all the coarse particles because of a
sudden decrease in velocity. Such coarse soil deposits are called lake deltas. But the fine-
grained particles move to the centre of the lake and settle when the water becomes quiet.
Alternate layers are formed with season, and such lake deposits are called lacustrine deposits.
These deposits are weak and compressible and pose problems for foundations.
If coarse and fine-grained deposits are formed in sea water areas, then they are called
marine deposits. Marine sediments are made up of terrestrial and marine contributions. The
terrestrial contribution consists of particulate material eroded from the shore, as well as
mineral matter, in true or colloidal solution, and this contribution decreases both in propor-
tion and in grain size with increasing distance from the shoreline. The marine contribution
is represented by the organic and inorganic remnants of dead marine life, and this normally
increases with time (Iyer, 1975).
In marine deposits, marine life and environment play a more significant role than the
salt concentration of the water. The clay particles absorb certain chemical elements from
the organisms, which in turn can extract mineral substances from sea water. Some acids
produced by the digestive tracts of marine organisms can alter the composition of the clay
minerals (Iyer, 1975).
Marine clay deposits (excluding the deep deposits which have been subjected to many
further changes and are overlain by other deposits) are generally weak, compressible, and
problematic for foundations. If such a deposit is exposed above sea level and experiences
leaching of sodium due to percolation of fresh water, it becomes very sensitive to disturbance.
Glacial soils transported by rivers from melting glacial water create deposits of stratified
glacial drift and are referred to as glacio fluvial deposit or stratified drift.
Glacial Deposits. Compaction and re-crystallization of snow leads to the formation of
glaciers. Glacier growth and movement depend on the formation of ice. Glacial deposits
M01_PURU1773_01_SE_C01.indd 5 2/12/2013 7:50:56 AM

form a very large group of transported soils. A glacier moves extremely slowly but deforms
and scours the surface and the bedrock over which it passes. Melting of a glacier causes
deposition of all the materials, and such a deposit is referred to as till. The land form or
topographic surface after a glacier has receded is called a ground moraine or till plain. Till
deposits which have been overrun by glaciers contain coarser particles and form good
construction material.
Soils deposited by the surface and sub-surface glacial rivers that remain in the form of
long-winding ridges are called eskers. They may vary from about 10 to 30 m in height and
about 0.5 km to several kilometres in length. Isolated mounds of glacial debris varying from
about 10 to 70 m in height and 200 to 800 m in length are called drumlins. Large boulders
picked up by a glacier, transported to a new location, and dropped are called erratics.
Glacial deposits provide a poor to excellent foundation. In many locations it is often found
that the material is dense and contains considerable sand and gravel. It is believed that glaciers
covered a large portion of the land during the ice age. Northern USA, Northern Europe, and
Canada were subjected to continent glaciers. Now glaciers cover approximately 10% of the
earth’s surface. Almost all glaciers are now concentrated in Greenland and Antarctica.
Wind-transported Soils. Like water, wind can erode, transport, and deposit fine-grained
soils. Soils carried by wind are subsequently deposited as aeolian deposits. Dunes are formed
due to the accumulation of such wind-deposited sands. Dunes are a rather common
occurrence in the desert areas of Africa, Asia, and the USA. Sands from dunes may be used
to a limited extent for construction purposes.
Fine-grained soils such as silts and clays can be transported by wind in arid regions.
Wind-blown silts and clays deposited with some cementing minerals in a loose, stable
condition are classified as loess. Loess deposits have low density, high compressibility, and
poor bearing resistance when wet. Loess is a clastic sediment comprising a uniformly sorted
mixture of silt, fine sand, and clay-size particles. The structure of a loess deposit is susceptible
to collapse on saturation.
Gravity Deposits. Gravity can transport materials only for a short distance. As the
movement is limited, there is no appreciable change in the materials moved. Gravity depos-
its are termed talus. They include the material at the base of cliff and landslide deposits. The
talus material at the cliff is formed due to the disintegration and subsequent failure of the
cliff face. These fragments are generally loose and porous.
Swamp and Marsh Deposits. In water-stagnated areas where the water table is fluctuating
and vegetational growth is possible, swamp and marsh deposits develop. Soils transported
and deposited under this environment are soft, high in organic content, and unpleasant in
odour. Accumulation of partially or fully decomposed aquatic plants in swamps or marshes
is termed muck or peat. Muck is a fully decomposed material, spongy, light in weight, highly
compressible, and not suitable for construction purposes.
1.4.3 Desiccated Soils

If a fine-grained soil is exposed to atmosphere, water is drawn from the interior to the surface.
From the surface, the water gets evaporated. This sort of drainage is referred to as drainage
by desiccation. During this process the soil becomes stiffer and ultimately becomes hard. The
point at which evaporation ceases depends on the relative humidity of the air around.
M01_PURU1773_01_SE_C01.indd 6 2/12/2013 7:50:56 AM

Under field conditions, desiccation may take place whenever the surface of the soil is not
permanently flooded. Due to periodic desiccation, even fine-grained silty sands show
apparent cohesion.* As the apparent cohesion is very large, even rains of long duration
cannot completely remove the cohesion. This phenomenon of desiccation is very much
pronounced in soils of semi-arid and arid regions. Such soils are quite often mistaken for soft
rocks. In the case of soft clays, the desiccation proceeds very slowly from the exposed surface
and forms a thick crust, and the thickness grows with age.
1.5 MAJOR SOIL DEPOSITS OF INDIA

Among different types of soils spread over the Indian Peninsula, only five major deposits
have been identified (Katti et al., 1975), viz., marine deposits, black cotton soils, laterite and
lateritic soils, alluvial deposits, and desert soils. Figure 1.3 shows the regions covered by
these soil deposits.
1.5.1 Marine Deposits

The marine deposits all along the Indian coast are generally derived from terrestrial sources.
These deposits cover a narrow belt of tidal flats all along the coast from Porbandar in the
west to Puri in the east. However, they are present over wide areas in places such as Rann of
Kutch. These tidal flats experience high tide inundation. The deposits are very soft to soft
clays, and the thickness varies from 5 to 20 m. The clay is medium sensitive and inorganic in
nature. These deposits generally need a pre-treatment before application of any external
load (Iyer, 1975). In order to prevent failures during construction, controlled loading should
be planned (Katti et al., 1975).
1.5.2 Black Cotton Soils

Black cotton soil is one of the major soil deposits of India and is spread over a wide area of
3,00,000 km2. The primary bed rock is basalt or trap, and in some locations, quartzites,
schists, and sedimentary rocks are also reported (Katti et al., 1975). The black cotton soil is
expansive in nature due to the presence of montmorillonite and illite clay minerals. The top
black subsoil varies in thickness up to a maximum of 20 m. Based on the pedological
conditions, crack depth and pattern vary. The soil surface is hard during summer and
becomes slushy during the rainy season. The effect of seasonal moisture change brings in
volume changes up to a maximum depth of 1.5 m. Because of the swelling and shrinking
nature of the soil, there is a necessity for treatment of the soil, and special foundations need
to be adopted in such soils to prevent failure of structures.
1.5.3 Laterites and Lateritic Soils

In tropical regions of high moisture and temperature, weathering activity is so intense that
a tremendously thick soil (exceeding 30 m) may be formed from the parent rock through
processes collectively termed laterization. Laterization is mainly due to the decomposition of
*A temporary shear strength gained by weathering.
M01_PURU1773_01_SE_C01.indd 7 2/12/2013 7:50:56 AM

68˚ 72˚ 76˚ 80˚ 84˚ 88˚ 92˚ 96˚

36˚N
36˚
xxx
x x xx
xxx
32˚
32˚
x x x xx x x x
x x x xx x x x x x x
x x x xx x x x xx x x
x x x xx x x x xx x x
x x x xx x x x x
28˚ x x x xx x x x xx x x x
xx x x x xx x x x xx x x x 28˚
xx x x x xx x x x xx x x x
x x x x xx x xx x x x xx x x x x xxxx
x x x x xx x x x x x xx x x x x x x x x xx x x x
x x x x xx x x x x x xx x x x x x x x x xx x x x
x x x x xx x x x x x xx x x x x x xx x x x xx x x x
x x x x xx x x x x x xx x x x xx x x x x x xx x x x xx x x x xx x x x xx x x x
x x x xx x x x xx x x x x x x x x xx x x x x x xx x x x xx x x x xx x x x xx x x x
x x x xx x x x xx x x x x x x x x xx x x x xx x x x x xx x x x xx x x x x
x x x xx x x x xx x x x x xxxx x xx x x x x
24˚ x x x xx x x x xx x
x x x xx x x x x
x x x x xx x x x xx
xx x x x xx x x x xx
xx x x x xx x x x xx
x xx
xx
24˚
xxx xx x x
xxx xx x x
x xx x x xx x x
x xx x x xxx
x x x xx x x
x x x xx x x
x x x xx x x
xxxx
20˚ 20˚
16˚ 16˚
Marine deposits
12˚ Black cotton soils

12˚
Laterites and lateritic soils
x x xx x x x
500 km x x xx x x x
x x xx x x x Alluvial deposits
x x xx x x x
8˚N Scale
Desert soils 8˚
72˚E 76˚ 80˚ 84˚ 88˚ 92˚ 96˚
Fig. 1.3 Map of India showing approximate extent of major regional deposits (Source:
Katti et al., 1975)
rock, removal of silica and bases, and accumulation of aluminium and iron sesquioxides.
The red, pink, or brown colour of laterites is essentially due to the presence of iron oxide. If
about 90% of the material contains coarse grains, then this is called laterite; instead, if rela-
tively fine grains are present, it is referred to as lateritic soil. In India, lateritic soils spread
over an area of 100,000 km2. Indian laterites are mostly residual soils.
M01_PURU1773_01_SE_C01.indd 8 2/12/2013 7:50:57 AM

The characteristic property of this type of soil is high strength when it is cut and dried in
the sun. The specific reason for such a behaviour has been attributed to the dehydration of
iron oxides and the presence of halloysite type of clay mineral. Some of the laterites show
extremely high strength comparable to that of burnt bricks. After hardening, the strength
gained is not affected when it comes in contact with water. Rao and Raymahashay (1981)
studied the mineralogy of Calicut and Rajahmundry laterites. Calicut laterites were found
to be rich in halloysite and crystalline goethite, whereas Rajahmundry laterites showed the
presence of crystalline kaolinite and metahalloysite. The reason for the difference has been
attributed to the geological environment of the areas.
The formation of sesquioxides in the top layers during laterization and weathering of
the bottom layers present serious problems for civil engineers in the assessment of lateral
stresses in lateritic profiles (Iyer and Pillai, 1972). Further road cuts in such deposits pose a
serious stability problem.
1.5.4 Alluvial Deposits

The well-known alluvial deposits of India are in the Indo-Gangetic and Brahmaputra flood
plains. Alluvial deposits exist up to a depth of 100 m. The north of the Vindhya Satpura range
is covered with river alluvium, and other alluvial deposits of deltaic type are also present. Allu-
vial deposits exhibit alternate layers of sandy silt and clay, and in some locations organic layers
are also encountered. The Bengal basin is another important alluvial deposit. The subsoil of the
upper strata (which is of immediate relevance to civil engineers) is of recent origin and is
believed to have been deposited by the Ganga river system. The soil around the Calcutta (now
Kolkata) region, usually referred to as the Calcutta deposit, consists primarily of desiccated
brownish grey silty clay up to a depth of about 15 m. Another deposit of the Bengal basin is the
river channel deposits, consisting of sandy silts to silty sands up to a depth of 30 m (Som, 1975).
Alluvial sands are used as fine aggregates in concrete, whereas alluvial clay is used for
manufacturing of bricks.
1.5.5 Desert Soils

The Thar Desert covers most of the area of the continent which forms the desert soil of India.
These are wind-blown deposits generally present in the form of sand dunes. These deposits
are formed under arid conditions and are predominantly of fine or silty sands. Scarcity of
water is a serious problem for any construction activity.
1.6 COMPONENTS OF SOILS

The composition of natural soils may include diverse components which may be classified
into three groups (Fig. 1.4): (i) solid phase (minerals, cementation, and organic material), (ii)
liquid phase (water with dissolved salts), and (iii) gaseous phase (air or some other gas with
water vapour). These are the components of a soil which affect its engineering properties.
1.6.1 Solid Phase

This consists of primary rock minerals, clay minerals, and cementing and organic materials.
One or all may be present in a soil.
M01_PURU1773_01_SE_C01.indd 9 2/12/2013 7:50:57 AM

Gaseous Gaseous
phase phase
Liquid
phase Liquid
phase
Solid
Solid phase
phase
(a) Elements of natural soil (b) Representation

of soil elements
Fig. 1.4 Components of natural soil
Primary Rock Minerals. These are rock fragments from the parent rock, formed due to
weathering. In general, they are relatively large in size and rounded or angular in shape.
When such particles form a major part of the soil minerals (as in gravels and sands), the
engineering properties will be governed by the gradation and packing of the grains. The
shape and texture of such particles (discussed in the next section) may have some bearing
on the properties.
Clay Minerals. These are secondary minerals formed by chemical weathering, and the
particle size is less than 2 μm. The particles commonly occur in the form of flat plates and
are flaky in shape. The main characteristic of such particles is their large surface areas. A
detailed treatment of clay minerals is presented in the next section.
Cementing and Organic Materials. Due to the decomposition of minerals by leaching
or due to the presence of dissolved salts, certain cementing materials (such as calcite, iron
oxide, or silica) may be deposited on the surface of the soil particles. Such materials improve
the engineering properties of soils.
Organic matter in the soil has originated from plant or animal remains. It generally occurs
in the top soil up to a depth of 0.5 m. Muck or peat deposits are primarily organic in nature
and occur at considerable depths. Organic matter absorbs more water, compresses consider-
ably under a load, fails due to low bearing resistance, and affects the setting of foundation
concrete. Thus, organic materials have many undesirable properties harmful for engineering
structures.
1.6.2 Liquid Phase

In soils of interest to the civil engineers, the only liquid phase is water. In geotechnical engi-
neering, water is the prime factor which governs the engineering properties of soils. It is an
incompressible fluid capable of taking normal stresses but not shear stresses.
Water can dissolve and transport, in solutions, various salts and compounds, some of
which may seriously affect the soil behaviour. Calcium sulphate occurs in many clays but is
only slightly soluble. Sulphate ion solutions in water have adverse effects on the properties
of concrete structures.
M01_PURU1773_01_SE_C01.indd 10 2/12/2013 7:50:58 AM

1.6.3 Gaseous Phase

Air is the gaseous phase found in soil. It is extremely difficult to get a perfectly dry soil or
fully saturated soil. The so-called saturated soils contain about 2% of air voids. Similarly, in
dry soil, water vapour may exist. In partially saturated soils, because of vapour pressure and
a continuous air path, there may be migration of water in the form of water vapour.
1.7 PARTICLE SIZES AND SHAPES

Naturally occurring soil deposits comprise soil particles of varying sizes and shapes. Size
and, to a lesser extent, shape are factors that affect the material behaviour of a soil. However,
most engineering properties are not controlled by particle size and shape but depend on soil
mineral composition, interaction with water, and soil structure.
1.7.1 Particle Size

A soil particle does not have a specific size and shape so that a unique linear dimension can
be assigned (as in a solid of regular geometrical shape). Thus, a representative size for the
particle has to be fixed, based on a certain analysis (as defined in sieve or hydrometer
analysis, discussed in the next chapter).
Soil may have particle sizes as big as several centimetres (pebbles) or as small as 10–6 mm
(colloid). It is reasonable to assign a name to a certain size range. Such names and their size
ranges are given below.
Cobbles or pebbles – rock fragments, size range 150 to 300 mm
Gravel – rock particles, size range 4.75 to 150 mm
Sand – rock particles, size range 0.075 to 4.75 mm
Silt – rock particles, size range 0.002 to 0.075 mm
Clay – mineral particles, size <0.002 mm (i.e., <2 μm)
The lowest particle size which could be identified without magnification is sand, and
hence particle sizes (sand and above) bigger than 0.075 mm are grouped under coarse-grained
soil. Particle sizes less than 0.075 mm are grouped under fine-grained soil because they are too
small to be seen without magnification.
1.7.2 Particle Shape

The shape of individual particles depends on the parent rocks, their age, degree of weathering,
and transporting processes. In general, all soil particles except clay-size particles which are
plate-like in shape are approximately equidimensional. Other shapes common to clay-size
particles are rods and laths.
Generally, the edges of a new coarse-grained particle are angular and rough surfaced, but
with time they become round and smooth. The sharpness of the edges and corners of a
particle is referred to as the degree of roundness. The degree of roundness in increasing
order may be given as angular, sub-angular, sub-rounded, rounded, and well rounded.
Figure 1.5 shows five levels of degree of roundness.
M01_PURU1773_01_SE_C01.indd 11 2/12/2013 7:50:58 AM

(a) Angular (b) Sub-angular (c) Sub-rounded
(d) Rounded (e) Well rounded
Fig. 1.5 Degree of particle roundness
The minor features of a particle surface are defined as the surface texture of a soil. This
is independent of the size, shape, or degree of roundness of a particle. Dull, polished,
smooth, rough, striated, frosted, etched, or pitted are terms used to define the surface texture
of a particle.
1.8 INTER-PARTICLE FORCES

The size of a particle is an indirect measure of the magnitude of forces between particles.
Inter-particle forces are broadly classified as gravitational forces or mass-derived forces and
surface-bonding forces or surface-derived forces. Gravitational forces are proportional to
mass or approximately to the volume of the soil particles. Thus, there is the same gravitational
interaction between particles of equal volume. In coarse-grained soils, mass-derived forces
are predominant. Surface-bonding forces between particles are dependent on the surface
area, its character, and its environment. Accordingly, the magnitude of bonding forces need
not be similar for particles of equal surface area. Surface-derived forces are predominant in
fine-grained soils.
For a complete understanding of the formation of soil minerals and the connected inter-
particle forces, a basic knowledge of atomic and molecular bonds is needed. There are two
main types of bonds of interest in the discussion of soil minerals: (i) primary or high-energy
bonds and (ii) secondary valence or weak bonds.
Primary or high-energy bonds are also referred to as electrostatic bonds and are of two
types, viz., ionic and covalent bonds. An ionic bond is formed when one atom joins with
another by adding electrons to its outer shell or shells or by losing them to arrive at a stable
configuration. This is the strongest and simplest of the bonds that hold atoms together. For
example, aluminium and oxygen combine to form aluminium oxide, and sodium and
chlorine combine to form sodium chloride. Sometimes, one electron can be shared by two
atoms and thereby create a bond called covalent bond (e.g., carbon). Primary valence bonds
are also termed intra-molecular bonds.
Secondary valence bonds are also known as inter-molecular bonds. They are of two types,
viz., hydrogen bonds and van der Waals forces. The hydrogen atom possesses only one
M01_PURU1773_01_SE_C01.indd 12 2/12/2013 7:50:58 AM

electron in its shell. It can take one electron from another element to form an anion or
can lose its electron to another element to form a proton. Thus, under certain conditions,
if a hydrogen atom is attracted by two atoms instead of one, then the bond formed is
called a hydrogen bond. The hydrogen bond is weaker than the ionic bond, and only two
atoms can be bonded by the hydrogen ion. Fluorine, oxygen, nitrogen, and chlorine form
hydrogen bonds.
Electrical or electromagnetic attraction between molecules is referred to as van der Waals
forces. In a molecule, the centre of gravity of the positive and negative charges may not
coincide, thus developing an electric moment. This system is referred to as polar. Thus, in
such a system van der Waals forces develop depending on the orientation of the particle.
The relative magnitudes of these bonds are given below (Leonards, 1962):
van der Waals forces 1 to 10

Hydrogen bonds 10 to 20
Primary valence bonds 40 to 400
1.9 SOIL MINERALS

The composition of a soil particle plays an important role in the fundamental understanding
of soil behaviour. Based on their origin, soils may be either organic or inorganic. Apart from
the fact that organic soils are mixtures of decayed plant life, skeletons or shells of small
organisms, and soil particles, not much is known about the composition of organic soils.
Inorganic soils are formed due to mechanical or chemical weathering. An inorganic soil
particle may be a rock fragment or a mineral. In the present context, a mineral is a chemical
element or a compound formed in nature during a geological process, and a rock fragment
is representative of the parent rock, which may have a combination of one or more minerals.
Based on the nature of atoms, minerals are classified as carbonates, phosphates, oxides,
and silicates. Of these, silicate minerals are important to civil engineers. Different
arrangements of atoms in the silicate minerals give rise to different silicate structures. The
most abundant soil minerals are the common rock-forming minerals, viz., sheet and
framework silicate minerals. These are discussed below.
1.9.1 Clay Minerals

Soil minerals are formed from two basic structural units: tetrahedral and octahedral. The
tetrahedral unit comprises a central silicon atom surrounded by four oxygen atoms positioned
at the vertices of the tetrahedron (the basic unit and its symbolic representation are shown in
Fig. 1.6). The tetra-hedrons are combined into hexagonal units called silica sheets. The
octahedral unit comprises a central ion of either aluminium or magnesium surrounded by six
hydroxyl ions (the basic unit and its symbolic representation are given in Fig. 1.6). Combining
aluminium–oxygen octahedrons gives a gibbsite sheet, and combining magnesium–oxygen
octahedrons gives a brucite sheet. Considering the valencies of atoms in the tetrahedral and
two octahedral units, it is clear that the units are not electrically neutral and do not exist as
single units. However, gibbsite and brucite sheets exist as they are electrically neutral.
The sheets may combine to form two-layer or three-layer sheet minerals or clay minerals.
The basic kaolinite unit (two-layer unit) is formed by stacking a gibbsite sheet on a silica
M01_PURU1773_01_SE_C01.indd 13 2/12/2013 7:50:58 AM

Silicon Aluminium
Oxygen Hydroxyl
Silica tetrahedron Alumina octehedron
(a) Silica sheet (b) Alumina sheet
Fig. 1.6 Basic units and symbolic blocks of clay minerals
sheet. These units may be stacked one on top of the other to form a lattice of the mineral
(Fig. 1.7a). The bondage between the basic kaolinite units is due to hydrogen bonding and
secondary valence forces. Kaolinite minerals are thus very stable and do not permit water to
enter the lattice and hence do not expand when saturated. Kaolinite is the most abundant
constituent of residual clay deposits. Isomorphous substitution (replacement of one kind of
atom with another) is quite common during the formation of sheet silicate minerals. In
kaolinite minerals, the amount of isomorphous substitution is negligible. Other clay minerals
with two-layer sheets are serpentine (brucite silica sheet combination, Fig. 1.7b) and hallosite
(similar to kaolinite except for the presence of water between the sheets).
Among three-layer sheet minerals, montmorillonite and illite clay minerals are the most
common. Basic three-layer sheet units are formed by keeping one silica sheet on the top and
one at the bottom of a gibbsite sheet. Figure 1.8a shows the unit and lattice of the
G B
G B
G B
G B
G – Gibbsite B – Brucite
sheet sheet
(a) Kaolinite mineral (b) Serpentine mineral
Fig. 1.7 Symbolic structure of two-layer sheet minerals
M01_PURU1773_01_SE_C01.indd 14 2/12/2013 7:50:59 AM

G G
+
K
Loosely held
G water and G K+ – Potassium ions
exchangeable
metallic ions
K+
G G G – Gibbsite
sheet
(a) Montmorillonite mineral (b) Illite mineral
Fig. 1.8 Symbolic structure of three-layer sheet minerals
montmorillonite mineral. Isomorphous substitution of magnesium or iron for the alumin-

ium in the gibbsite sheet is common. The bonding between the three-layer units is by
secondary valence forces and exchangeable ion linkage. The bonding of these sheets is very
weak, and water may enter between the units. The mineral has significant affinity for water
and can imbibe a large quantity of water, causing swelling.
Accordingly, there will be shrinkage during dry weather. Most of the Indian black cotton soils
contain this mineral, and sufficient care has to be exercised in designing structures on such soils.
Illite consists of the basic montmorillonite units but is bonded by secondary valence
bonds and potassium ions (K+), as shown in Fig. 1.8b. There is about 20% replacement of
aluminium with silicon in the gibbsite sheet due to isomorphous substitution. The mineral
is a very stable one and does not swell or shrink unlike montmorillonite.
As the basic units of clay particles are sheeted structures, the particle formed by
stacking such basic units ends up with a plate shape. The surface area of clay parti-
cles per unit mass is generally referred to as the specific surface. Clays with the smallest
particles have the largest surface area. The surface area of kaolinite is quite less compared
to that of montmorillonite. The specific surfaces of kaolinite, illite, and montmorillonite are
about 15, 90, and 800 m2/g, respectively.
1.9.2 Framework Silicate Minerals

Quartz and feldspar are very common rock-forming minerals, and their frameworks are
abundant in soils. Particles of framework minerals are approximately equidimensional
because of the nature of their structure. Quartz is one of the minerals which can resist
weathering the most. Feldspars can be weathered to form clay minerals. Thus, these
minerals sometimes find place in clay and silt-size particles.
M01_PURU1773_01_SE_C01.indd 15 2/12/2013 7:50:59 AM

1.10 SOIL–WATER SYSTEM

Mass-derived forces on coarse-grained soils may not have a significant effect when they
come in contact with water except for a reduction in voids due to submergence. But surface-
derived forces of fine-grained soils play a vital role when they come in contact with water.
The surfaces of clay particles carry a net negative charge. This results from any one of the
combinations of the following factors (Lambe and Whitman, 1979): (i) isomorphous substitution,
(ii) surface dissociation of hydroxyl ions, (iii) absence of cations in the crystal lattice, (iv)
adsorption of anions, and (v) presence of organic matter. Isomorphous substitution is the most
significant of all the five factors. The edges of a clay particle may have a positive or a negative
charge. Because of net negative surface charges, the clay mineral particles repel each other, but
edge-to-surface attraction is possible. However, the particles will attract cations (positive ions).
When water comes in contact with clay particles, the net negative charge of the mineral
attracts the cation in the water to the surface of the mineral. Additional water molecules are
also attracted towards the edges.
The nature of water immediately surrounding a clay particle is more intense and is
believed to possess properties different from liquid water. Further, to balance the negative
charges beyond the surface layer of molecules surrounding the particles, more cations are
attracted. As the effect of attraction decreases with distance from the surface, there is no
attraction and the pore water is normal. The distance from the surface of the particle to
the limit of attraction is termed the diffuse double layer (Fig. 1.9). The water contained in the
diffuse double layer and adjacent to the soil particle is termed the rigid layer.
The cations in the diffuse double layer are not permanently attached to the surface of clay
particles and can be replaced by other cations. For example, if a soil with sodium ions is
leached with potassium chloride solution, most of the sodium ions will be replaced by
potassium ions. This transformation is known as cation or base exchange, and the ions
involved in the replacement are called exchangeable ions. Such a transformation has a
significant effect on the behaviour of clayey soils.
1.11 PHYSICO-CHEMICAL BEHAVIOUR OF CLAYS

Physico-chemical contributions of soils and the intrinsic forces acting in the soil–water
system are of utmost importance for a better understanding of the deformation and strength
characteristics of the soil. This subject of physico-chemical properties is a common topic of
discussion in colloidal chemistry, mineralogy, soil physics, and soil mechanics.
1.11.1 Characteristics of Electrolytes

Substances that form ions in water solutions are called electrolytes. Substances which ionize
completely are called strong electrolytes, and others which ionize slightly are called weak
electrolytes. Substances which do not conduct electricity are called non-electrolytes. Most of
the salts and all soluble metal hydroxides are strong electrolytes. Bases are weak electrolytes.
Dielectric constant is a measure of a material’s ability to perform as an insulator. The
higher the dielectric constant of a material, the larger its insulating capacity. Compared to
other liquids, water has a high dielectric constant. This is the reason for water having great
solvent power for strong electrolytes.
M01_PURU1773_01_SE_C01.indd 16 2/12/2013 7:50:59 AM

Diffuse
double layer
Particle surface
+ – –
+
+ –
Net negative charge
+ + –
+
– + –
+
+ –
–
+ +
+ – +
+ –
+ –
–
Normal
water
Rigid Diffuse layer
layer
Ion concentration (+)
Distance from surface of particle (Å)
Fig. 1.9 Electric double layer
Zeta potential is the electric potential developed at the solid–liquid interface in response
to movement of colloidal particles. The magnitude of the zeta potential depends on the
thickness of the double layer. Zeta potential depends on ion types, temperature, adsorption,
particle size, shape of particles, and pH value.
1.11.2 Electrochemical Characteristics

Interaction with the pore fluid occurs on the solid’s surface. The surfaces of the soil solids
may affect the water structure due to the presence or absence of interaction. The essential
components of a soil–water system are as follows:
1. Surfaces of the soil mineral present with their characteristic geometry and electric patterns.
2. Location of cations so as to maintain equilibrium between the kinetic dispersive forces
and the Coulombic attractive forces.
M01_PURU1773_01_SE_C01.indd 17 2/12/2013 7:50:59 AM

3. Presence of water molecules with different densities from the solid surfaces.
4. Soil–water interaction as a complex phenomenon comprising ion-exchange, leaching,
osmotic hydration, adsorption, heat-of-wetting, and oxidation–reduction reactions.
Soil equilibrium conditions differ depending on liquid water and water vapour. Even at the
same water contact, the structure and physico-chemical properties of a soil–water system may
differ considerably. This may depend on the flow path from which this moisture content has
been attained.
Soil–water interaction is a complex phenomenon, which may be still more complicated
if polluted water is involved. Polluted water may be gravity water or environmental water.
Gravity water depends on the dissolved ions and molecules. Environmental water depends
on pore fluid character, temperature, and pressure.
1.11.3 Characteristics of Bacteria

Bacteria are single-cell forms. Most bacteria exist at a shallow depth from the ground
surface. The activity of bacteria depends on the availability of nutrients, microorganic types,
substrate composition, and temperature. Bacteria may cause a soil to change behaviour in
three possible ways, viz., change the soil–water structure, change the characteristic of the
pore fluid, or change through decomposition of soil constituents.
1.11.4 Electrokinetic Phenomena

In order to know the water and soil properties, electric current is passed into the soil mass.
The interaction of soil and electricity depends on size of particle, soil structure, soil surface
condition, pore fluid, and ion-exchange capacity. Soil–electricity interaction depends on
whether the soil is dry or moist.
In a soil–water system, fluid flow caused by electrical potential is called electric
conductivity. Such a movement of fluid under electric potential is called electrokinetic
phenomena. These phenomena may be categorized under two groups:
1. Phenomena between soil particles: In this case, the fluid flows from one electrode to another
electrode through the voids of the soil and the soil particles remain stagnant. This
phenomenon is referred to as electro-osmosis (dealt in detail elsewhere).
2. Phenomena in clay suspensions: In this case, soil particles in a clay suspension are attracted
or repelled from one another; e.g., negatively charged particles move toward an anode in
a soil suspension. This phenomenon is referred to a electrophoresis.
Both these methods are effectively used in de-watering, soil densification, etc., in the field
of ground improvement.
1.11.5 Characteristics of Heat

Heat is a form of energy which may be induced into the soil mass through man-made
activities or due to natural causes. The man-made sources are heat released from steam
pipes, electrical cable lines, and heat pump systems and heat caused due to environmental
wastes. Natural sources include solar and geothermal heat.
M01_PURU1773_01_SE_C01.indd 18 2/12/2013 7:50:59 AM

Thermal conductivity is synonymous to hydraulic conductivity. Thermal conductivity

of soil is defined as the quantity of heat flowing due to unit thermal gradient in unit time.
The reciprocal of thermal conductivity is called thermal resistivity, which is a measure of a
material’s ability to resist heat flow.
In a saturated soil–water system, increase in temperature decreases the bound water
and increases the free water and decreases the viscosity of the free water. This may be
true for a limited range of temperature in a clay–water system. Under high temperature
conditions in a clay–water system, dispersion or flocculation may occur depending on
exchangeable ions and type of clay mineral.
1.12 SOIL STRUCTURE

Soil structure is a more generalized term, applicable to all types of soils. This includes
gradation, compactness, geometric and skeletal arrangements of particles, inter-particle
forces, and bonding agents. The term soil fabric has been explicitly used for the structure of
clay-size particles.
1.12.1 Structure of Granular Soils

In granular soils, the ratio of the volume to the surface area is large, so that mass-derived
forces are dominant and surface-derived electrical forces are negligible. Single-grained
structures are formed when the soil grains settle out independently due to mass-derived
forces. The individual granular particle is strong enough to withstand the normal pressures
encountered in practice. Thus, the aggregate physical properties in coarse-grained soils are
a function of relative density and particle shape.
Packing is the term used for the arrangement of individual soil particles in granular soils.
But gradation and particle shape affect packing. For an idealized spherical particle, the
minimum and maximum void ratios* are 0.35 and 0.91, respectively. But such a condition
does not arise. However, ideal particle size distributions never exist in real soils. Possible
packings representing upper (well graded) and lower limits (poorly graded) are shown in
Fig. 1.10. Optimum packing gives rise to more particle contacts, more shear strength, less
settlement, and more stability.
(a) Upper limit (b) Lower limit

well-graded particles poorly graded particles
Fig. 1.10 Possible limits of packing
*Ratio of volume of voids to volume of soil solids.
M01_PURU1773_01_SE_C01.indd 19 2/12/2013 7:51:00 AM

1.12.2 Structure of Cohesive Soils

The structure of a fine-grained cohesive soil can be described fully with the understanding
of inter-particle forces and the geometrical arrangement, or fabric, of the particles. In a
cohesive soil deposit, the structural arrangement of individual particles or a group of
particles at a microsocopic scale is defined as the microstructure or microfabric of the deposit.
Miscrostructure is fundamentally important in the general understanding of soil behaviour.
The larger, usually visible, structural features of cohesive deposits such as joints, fissures,
rootlets, varves, silt and sand seams, and other discontinuities define the macrostructure or
macrofabric of the deposit. Macrostructure of cohesive deposits has an important practical
engineering influence on the entire soil behaviour.
The concepts of clay structure were proposed earlier by Terzaghi (1925) and Casagrande
(1932b). Terzaghi postulated that during the process of sedimentation, adsorbed water
surrounding the clay particles is squeezed out, thus making particles attain high attractive
forces. Application of external pressure breaks the contacts between the solid water shells,
leading to displacement and complete collapse of structure. Casagrande’s concept of clay
structure was that of a coarse-grained skeleton cemented together by a highly compressed
“bond clay.” The interstices are filled with partly consolidated high-water-content “matrix
clay.” Remoulding causes the mixing of matrix and bond clays and, thus, destroys the
primary load-carrying structure.
It was Lambe (1953, 1958) who fundamentally delineated structures into two extreme
groups – flocculated and dispersed. They are discussed below.
If two clay particles approach each other in a suspension, attractive forces due to van der
Waals forces and repulsive forces due to positively ionized adsorbed layers come into action.
At lesser separation, van der Waals forces dominate, and so particles adhere. However, there
is a decrease in van der Waals forces with an increase in separation. If the adsorbed layer is
thin, the repulsive force will be absent or negligible, and the random movement of particles
will bring them into contact. Further, the net repulsive force will be greatest when particle
faces approach closer. As a result, the flocculating particles form an edge-to-face arrangement.
A group of such particles settles down together rapidly. This process is called flocculation, and
the structure so formed is termed flocculated structure. Clays deposited under sea water have
a high cation concentration with thin adsorbed water, thus resulting in a flocculated structure.
On the other hand, if clay particles have a thick adsorbed layer, the repulsion is greater and
the particles are led to a slow independent settlement, after which they remain dispersed.
This process is known as dispersion, and it causes face-to-face orientation of particles, which
is known as dispersed structure.
The smallest clay-particle-forming structure, either flocculated or dispersed, is termed
the first-order fabric unit. These small particles are normally aggregated. Van Olphen
(1963) categorized such particle clusters as primarily aggregated and ultimately floccu-
lated or deflocculated based on the mode of formation, e.g., flocculated–aggregated or
deflocculated–aggregated. Ideal clay structures which are of interest to civil engineers are
shown in Fig. 1.11.
The structural arrangement in nature can be highly complicated because of the presence
of bulky particles along with clay particles. Interaction between single clay mineral
particles is rare. Individual clay particles aggregate or flocculate together to form submi-
crosopic fabric units called domains (Fig. 1.12). Domains, in turn, group together to form
submicrosopic groups called clusters. These submicroscopic groupings are attributed to
M01_PURU1773_01_SE_C01.indd 20 2/12/2013 7:51:00 AM

(a) Edge-to-face (b) Edge-to-edge (c) Deflocculated

flocculated flocculated
(d) Edge-to-face (e) Edge-to-edge (f) Deflocculated–

flocculated– flocculated– aggregated
aggregated aggregated
Fig. 1.11 Idealized clay structures
Pore Silt
grain
Silt
grain Pore
Domain
Pore
Ped
Cluster
Fig. 1.12 Structure of natural clay (Source: Bowles, 1984)
inter-particle forces acting between small basic units. Clusters group together to form peds,
which are macroscopic.
Clay deposits which exhibit no visible variation in macrostructure are said to be uni-
form or intact. Many clays are layered or stratified, with predominant bedding planes. Clay
deposits formed in glacial melt-water lakes exhibit thin layers of silt, fine sand and clay,
and such deposits are referred to as varved clays. Stiff clays exist in thin natural states with a
network of hair cracks, joints, or fissures; such clay deposits are called fissured. The fissures
are formed due to stress release of highly stressed clays, past earth movements, or volume
changes caused by desiccation.
A well-defined macrostructue significantly influences the engineering behaviour of clay
deposits. Presence of fissures in a clay indicate planes of weakness and hence low strength
along the planes, compared to a clay deposit which is intact. Apart from low strength, fis-
sured clays or laminated clays show higher seepage than an intact clay, due to the preferen-
tial drainage path formed along the silt-filled fissures or bedding planes.
M01_PURU1773_01_SE_C01.indd 21 2/12/2013 7:51:00 AM

POINTS TO REMEMBER
1.1 Rocks are of three types, viz., igneous, sedimentary, and metamorphic, which
constitute the earth’s crust along with weathered rock (as soil).
1.2 Soil is a particulate material resulting from disintegration of rocks and depends
largely on its origin.
1.3 Type of soil developed is based on the rock type, its mineral constituents, and the
climatic regime of the area.
1.4 Coarse-grained soils like gravelly and sandy soils are formed due to decomposition of
rocks containing quartz or orthoclause minerals with high silica content.
1.5 Fine-grained soils like silts and clays are formed from rocks which contain iron,
magnesium, calcium, or sodium minerals with little silica.
1.6 Clays are not fragments of primary minerals but secondary minerals formed by
decomposition of primary minerals.
1.7 Physical weathering and chemical weathering are the two types which cause
disintegration of rock.
1.8 Major formations of soils are residual soils, transported soils, and desiccated soils.
1.9 Major soil deposits of India are marine deposits, black cotton soils, laterite and lateritic
soils, alluvial deposits, and desert soils.
1.10 Natural soils may be broadly grouped into three components or phases, viz., solid
phase, liquid phase, and gaseous phase.
1.11 Inter-particle forces acting between particles depend on the surface area, its character
and environment.
1.12 In soil minerals, two types of bonds are of interest, viz., primary or high-energy bonds
and secondary valence or weak bonds.
1.13 Soil minerals may be a framework of silicate minerals or two-layer/three-layer sheet
minerals. The three important clay minerals are kaolinite (two-layer sheet mineral),
illite, and montmorillonite (three-layer sheet mineral).
1.14 Mass-derived forces are responsible for the formation of soil structure in coarse-
grained soils. Surface-derived forces contribute to the formation of soil structure in
1.15 Water in clay leads to the attractive and repulsive forces on the particles, resulting in
the formation of the electric double layer. Electric double layer (or diffused double
layer) contributes to the formation of a particular type of structure.
1.16 Flocculated structure and dispersed structure are the two basic structures in clays.
1.17 Natural clays may be uniform or varied with cracks, joints, or fissures.
QUESTIONS
Objective Questions
1.1 Glaciers are formed by
(a) Compaction and recrystallization of snow
(b) Continuous freezing of water
M01_PURU1773_01_SE_C01.indd 22 2/12/2013 7:51:00 AM

(c) A sudden drop in temperature below 0°C

(d) None of the above processes
1.2 When the products of rock weathering are not transported as sediments but remain in
place, the soil is known as
(a) Alluvial soil (b) Glacial soil
(c) Residual soil (d) Aeolian soil
1.3 Among the clay minerals, the one having the maximum swelling tendency is
(a) Kaolinite (b) Illite
(c) Montmorillonite (d) Halloysite
1.4 Cations from soil moisture are attracted to the surface of clay minerals to
(a) Balance the unsatisfied valence bonds
(b) Balance the negative electrical charge
(c) Form a diffuse double layer
(d) Replace the low-valence bonds
1.5 Identify the true statements from the following:
(a) A soil transported by gravitational force is called talus.
(b) Lateritic soil is a category of organic soil.
(c) Water held firmly to the clay particles has the same properties as ordinary water.
(d) A clay deposit which exhibits no evidence of fissuring is described as intact.
(e) Most clay crystals consists of atomic sheets, principally of two types: silica and
alumina.
1.6 Chemical weathering may include the following processes:
(1) Oxidation (2) Solution (3) Leaching (4) Hydrolysis
Of these statements
(a) All are correct (b) 1, 2, and 3 are correct
(c) 2, 3, and 4 are correct (d) 3, 4, and 1 are correct
1.7 Clays which are deposited in marine water and later leached by groundwater are
called
(a) Quick clays (b) Fat clays (c) Stiff clays (d) Fissure clays
1.8 Match list I with list II, and select the correct answer using the codes given below the lists:
List I (soil deposit) List II (soil name)

A. Gravity 1. Stratified drift
B. Lake 2. Talus
C. Glacier 3. Loess
D. Wind 4. Lacustrine
Codes:
A B C D
(a) 1 3 2 4
(b) 2 4 1 3
(c) 3 4 1 2
(d) 4 3 2 1
M01_PURU1773_01_SE_C01.indd 23 2/12/2013 7:51:00 AM

1.9 Indian marine deposits are

(a) Very soft to soft clays (b) Soft to medium stiff clays
(c) Stiff to very stiff clays (d) Hard clays
1.10 Calicut laterites are rich in
(a) Halloysite (b) Montmorillonite
(c) Crystalline kaolinite (d) Metalalloysite
Descriptive Questions
1.11 What are the different types of land forms associated with igneous rocks?
1.12 Distinguish between mechanical weathering and chemical weathering.
1.13 Which sedimentary deposits are of concern to civil engineers? Give examples.
1.14 What are loess soils, and what is the potential danger to loss of stability in loess soils?
1.15 Describe briefly the origin of soils and bring out the factors which control their formation.
1.16 Describe briefly the physical properties of residual and transported soils.
1.17 Name the parent rocks from which the following soils are derived: clay, sand, laterite,
black cotton soil, and loess.
1.18 Explain the significance of the structure of clay minerals as constituents of engineer-
ing soils.
1.19 Explain why electrical surface forces are important for fine-grained soils but have
little effect on coarse-grained soils.
1.20 What effect does the structure of clay have on the physical properties of soils?
1.21 From an engineering point of view, explain the properties of clay minerals.
1.22 Explain some of the stronger influences that are continually breaking rock into grains
of soil.
1.23 What is degradation? Explain.
1.24 What influences can produce chemical changes in rock rather than mechanical break-
age?
1.25 How does clay absorb water? Explain.
1.26 What soil types would be expected in a river or stream delta?
1.27 Explain the potential danger to stability in areas where the land is formed from marine
clay.
1.28 Distinguish between the shape and size of clay particles with silts and sands.
1.29 What is a hydrogen bond? Explain its role in the physico-chemical behaviour of clays.
1.30 How does the modern concept of clay structure contrast with that given by Casagrande?
M01_PURU1773_01_SE_C01.indd 24 2/12/2013 7:51:01 AM

2
Index Properties of Soils
CHAPTER HIGHLIGHTS
Three-phase system – Definition and applications of void ratio, porosity,
degree of saturation, air void ratio, specific gravity of soil solids, moisture
content, density, and density index – Particle-size analysis – Grain-size dis-
tribution curves – Consistency limits and indices of soils – Activity of clays
2.1 INTRODUCTION
Bulk soil, as it exists in nature, is a more or less random assembly of soil particles, water,
and air. The properties of soils are complex and variable. Every civil engineering work
involves the determination of soil type and its associated engineering application; certain
properties are more significant than others. The common problems faced by civil engineers
are related to the bearing capacity and compressibility of soil and the seepage through the
soil. The possible solution to these problems is arrived at based on a study of the physical
and index properties of the soil. A knowledge of these properties with value-oriented
judgement gives a complete solution to such problems. The phase concept for soils intro-
duced in the previous chapter is of great assistance in understanding these properties.
2.2 THREE-PHASE SYSTEM

Soil deposits are particulate systems containing three distinct phases. For the purpose
of engineering analysis and design, it is essential to express the relationships among
them in numerical terms.
As the relative proportions of these three phases, viz., soil solids, water, and air, are
intermixed, it is customary to consider a soil model which will represent these phases
distinctly and properly quantify the amount of each phase. A schematic diagram of the
three-phase system is shown in Fig. 2.1.
M02_PURU1773_01_SE_C02.indd 25 2/12/2013 7:49:12 AM

Fig. 2.1 Three-phase system
In dry and saturated soils, the three-phase system reduces to two phases, viz., soil solids
and air, and soil solids and water, respectively (Fig. 2.2). Thus, soil exists in either the two-
phase or three-phase state. The following are the basic relations.
2.2.1 Void Ratio (e)

Void ratio is the ratio of the volume of voids (Vv) to the volume of soil solids (Vs); that is,
Vv
e = (2.1)
Vs
It is expressed as a decimal and may take values greater than 1.0. Typical values of the
void ratio in soils may range from 0.50 to 1.50.
2.2.2 Porosity (n)

Porosity is the ratio of the volume of voids to the total volume of soil (V); that is,
Vv
n= (2.2)
V
Fig. 2.2 Two-phase systems
M02_PURU1773_01_SE_C02.indd 26 2/12/2013 7:49:13 AM

Index Properties of Soils 27
It is expressed as a percentage. The range of porosity is 0% ≤ n ≤ 100%. Void ratio and

porosity are related as
n
e= (2.3)
and 1− n
e
n= (2.4)
1+ e
2.2.3 Degree of Saturation (Sr)

Degree of saturation is the ratio of the volume of water (Vw) to the volume of voids; that is,
Vw
Sr = (2.5)
Vv
This is expressed as a percentage. Thus for dry soil, Sr = 0%; for saturated soil, Sr = 100%;
and for partially saturated soil, 0% < Sr < 100%.
2.2.4 Air Void Ratio (Av) or Air Content

Air void ratio is the ratio of the volume of air (Va) to the total volume of the soil; that is,
Av = Va V
Vv − Vw Vv Vv − Vw Vv 1 − Sr
= = =
Vv + Vs Vv Vv + Vs Vv 1+1 e
e
= (1 − Sr ) (2.6)
1+ e
Av = n (1 − Sr ) (2.7)
2.2.5 Specific Gravity of Soil Solids (G)

Specific gravity is the ratio of the mass of soil solids (Ms) to the mass of an equal volume of
water, given by
Ms (2.8)
G=
Vs ρw
ρ
G= s
ρw
where ρs = Ms/Vs is the density of soil solids and ρw is the density of water (1.0 Mg/m3 =
1,000 kg/m3).
G can be determined as follows (Fig. 2.3). An oven-dried sample of soil is placed in an
empty pycnometer of mass M1, and the combined mass M2 is found. Water is added to the
soil and agitated to remove the air. The mass of the pycnometer with soil and water, Mb, and
the mass of the pycnometer with water, Ma, are found. Then,
Ms
G=
Ms −( M b − Ma )
where Ms = M2 − M1 .
M02_PURU1773_01_SE_C02.indd 27 2/12/2013 7:49:14 AM

Fig. 2.3 Apparatus for specific gravity determination
The detailed procedure for determining specific gravity is given in Chapter 10. Typical
values of specific gravity are given in Table 2.1(a).
Specific gravity is relatively important as far as the qualitative behaviour of the soil is
concerned. It is determined in a majority of testing programmes. The specific gravity of most
soils falls within a narrow range and thus it is not used in the identification or classification
of soils as it is in the identification of minerals.
2.2.6 Water Content (w) or Moisture Content

Water content is the ratio of the mass of water (Mw) to the mass of soil solids; that is,
Mw
w=
Ms
Vw ρw
=
Vs ρs
Table 2.1(a) Typical values of specific gravity
Type of soil Specific gravity, G
Sand 2.65–2.67
Silty sand 2.67–2.70
Inorganic clay 2.70–2.80
Soil with mica or iron 2.75–3.00
Organic soils Variable but may be under 2.00
Source: Bowles (1978).
M02_PURU1773_01_SE_C02.indd 28 2/12/2013 7:49:15 AM

(Vw Vv ) ρw
= (2.9)
(Vs Vv ) ρs
Sr ρw
=
(1 / e )ρw G
Sr e
w= (2.10)
G
This is represented as a percentage. The water content for a dry soil is zero. The natural
water content for most soils is around 60%, but it can be as high as 400%.
The water content is obtained by weighing a moist sample of the soil, drying it in a tem-
perature-controlled oven at 105ºC for 24 hours, and then re-weighing. The detailed proce-
dure is given in Chapter 10.
The dry soil mass is a function of the absolute magnitude of certain soil parameters, such
as field moisture content and consistency limits (discussed elsewhere), and is used in the
determination of void ratio and the degree of saturation.
The dry mass of a soil is not a definite quantity. It can be arbitrarily, but conveniently,
defined as that mass which is obtained at 105ºC (Lambe, 1949). It is believed that drying at
105ºC for 24 hours completely removes the free water and some of the adsorbed water which
forms part of the effective soil grain. For certain organic soils (like peat), for soils containing
an appreciable amount of gypsum or some other mineral, and for some tropical soils, the
conventional oven temperature of 105ºC may be high. At this high temperature, the loosely
bound water of hydration or molecular water may be lost, resulting in a change in the soil
characteristics, such as the consistency limit. For such soils, the Indian Standards (IS: 2720 –
Part 2, 1973) recommends a low oven temperature of 60 to 80°C.
For certain field controls (such as compaction), moisture content results are needed imme-
diately. The conventional oven-drying method of measuring moisture content is so slow that
results will not be available in time to make corrections in the field where deficiencies exist.
Hence, rapid methods of testing are resorted to (this is discussed under compaction control
in Chapter 4).
2.2.7 Soil Densities

Dry Density (ρd). It is the ratio of the total (dry soil) mass to the total volume (Fig. 2.2a); that is,
M Ms + Ma
ρd = =
V V
Ms
= (mass of air Ma = 0)
V
Ms Ms Vs ρ
= = = s
Vs + Vv 1 + Vv Vs 1 + e
Gρw
ρd = (2.11)
1+ e
M02_PURU1773_01_SE_C02.indd 29 2/12/2013 7:49:15 AM

Saturated Density (ρsat). It is the ratio of the total saturated soil mass to the total volume
(Fig. 2.2b); that is,
M Ms + M w
ρsat = =
V Vs + Vv
Ms Vs + Mw Vs ρs + (Vw ρw ) Vs
= =
1 + Vv Vs 1+ e
⎛ V ⎞
ρsat = ⎜⎜Gρw + v ρw ⎟⎟⎟ (1 + e ) (Since VW = VV)
⎜⎝ Vs ⎟⎠
(G + e)
ρsat = ρw (2.12)
1+ e
Submerged Density or Bouyant Density (ρ′). It is the effective mass per unit volume
when submerged. When a unit volume of soil is submerged in water, it displaces an equal
volume of water, and the density is only (ρsat – ρw). Thus,
⎡G + e ⎤
ρ ′ = ρsat − ρw = ⎢ ⎥ ρ − ρw
⎢⎣ 1 + e ⎥⎦ w
⎡ G − 1⎤
ρ′ = ⎢ ⎥ρ (2.13)
⎢⎣ 1 + e ⎥⎦ w
Bulk Density or Total Density. It is the ratio of the total soil mass to the total volume:
M Ms + M w + Ma
ρ= =
V Vs + Vv
M V + Mw Vs
= s s (Since Ma = 0)
1 + Vv Vs
ρs + (Vw ρw Vs ) Gρw + (Vw Vs )×(Vv Vv ) ρw

ρ= =
1+ e 1+ e
Gρw + eSr ρw
=
1+ e
G + eSr
ρ= ρw (2.14)
1+ e
That is,
Gρw eSr ρw
ρ= +
1+ e 1+ e
(wG Sr ) Sr ρw
= ρd +
1+ e
ρ = ρd (1 + w)
(2.15)
M02_PURU1773_01_SE_C02.indd 30 2/12/2013 7:49:17 AM

For dry soil:

G ρw
Sr = 0, ρ = = ρd
1 +e
For saturated soil:

G+e
Sr = 1, ρ = ρw = ρsat
1+ e
Dry density of the in situ soil is usually measured using the core cutter and sand replace-
ment methods.
The core cutter method consists of a cylindrical cutter with dolly and a rammer to drive
the cutter into the soil. The cutter containing soil is then carefully dug out and excess soil
on the ends trimmed off. The mass and volume (from the dimensions of the core) of the soil
within are found and the wet density calculated. Dry density of the soil is found after deter-
mining the water content.
In the sand replacement method, a hole about 100 mm in diameter is dug with suitable
tools at the selected location. The mass of the soil removed is weighed and a specimen sam-
ple taken for determination of water content. The mass of sand required to fill the conical
portion, and hence of the sand filling the hole, is determined. The volume of sand required
to fill the hole is obtained from the mass of sand filling the hole and the bulk density of the
sand. The bulk density of sand is calibrated by filling a can of known volume under similar
conditions. From the mass of the soil and volume of the hole, the wet density and, hence, the
dry density are determined. Details of both procedures are given in Chapter 10.
Knowing the dry density and specific gravity of soil solids, the in situ void ratio is obtained
from the expression
Gρw
e= −1 (2.16)
ρd
2.2.8 Unit Weights

The unit weight (γ) of a soil is the ratio of the weight of the soil (force) to its total volume. The
units of unit weight are force per unit volume (kN/m3), whereas the units of density are
mass per unit volume (Mg/m3 or g/cm3).
1. Dry unit weight γd = ρd g (2.17)
2. Saturated unit weight γ sat = ρsat g (2.18)
3. Submerged unit weight γ ′ = ρ ′ g (2.19)
4. Bulk unit weight γ = ρ g (2.20)
5. Unit weight of water γ w = ρw g (2.21)
The value of gravitational acceleration, g, to be considered in Eqs. 2.17 to 2.21 is 9.807 m/

s2, and the unit weight of water at 4°C is 9.807 kN/m3. Table 2.1(b) shows how γ, γd and γsat
are related through various known parameters.
M02_PURU1773_01_SE_C02.indd 31 2/12/2013 7:49:19 AM

Table 2.1(b) Relationship for γ, γd, and γsat in various forms
Sl. No. Bulk unit weight (γ) Sl. No. Dry unit weight (γd) Sl. No. Saturated unit weight (γsat)
Known Relationship Known Relationship Known Relationship
1. w, G, e (1 + w)Gγ w 1. γ, w γ 1. G, e (G + e) γ w
1+ e 1+ w 1+ e
2. S, G, e (G + Se) γ w 2. G, e Gγ w 2. G, n ⎡(1− n)G + n⎤ γ w

⎣ ⎦
1+ e 1+ e
(1 + w)Gγ w ⎛ 1 + wsat ⎞⎟
3. w, G, s 3. G, n Gγ w (1− n) 3. G, wsat ⎜⎜ ⎟
wG ⎜⎜⎝ 1 + w G ⎟⎟⎠Gγ w
1+ sat
S
Gγ w ⎛ e ⎞⎟⎛ 1 + wsat ⎞
4. w, G, n Gγ w (1 − n)(1 + w) 4. G, w, s 4. e, wsat ⎜⎜ ⎟⎟⎜⎜ ⎟⎟ γ
⎛ wG ⎞⎟
1 + ⎜⎜⎜ ⎟ ⎜⎜⎝ w ⎟⎠⎝⎜ 1 + e ⎟⎠ w
⎝ s ⎟⎠ sat
5. S, G, n Gγ w (1− n) + nS γ w 5. e, w, s esγ w 5. n, wsat ⎛ 1 + wsat ⎞⎟

n ⎜⎜⎜ ⎟γ
(1 + e) w ⎜⎝ wsat ⎟⎟⎠ w
γsat, e eγ w γ d, e ⎛ e ⎞⎟
6. 6. γd + ⎜⎜ γ
γ sat −
1+ e ⎜⎝ 1 + e ⎟⎟⎠ w
7. γsat, n γ sat − nγ w 7. γ d, n γd + nγ w
(γ sat − γ w )G ⎛ ⎞
8. γsat, G 8. γ d, s ⎜⎜1 − 1 ⎟⎟ γd + γ w
(G − t) ⎜⎝ G ⎟⎠
9. γd, wsat γd (1 + wsat )
Source: Das (2002)
2.2.9 Density Index (ID)

Density index is the ratio of the difference between the void ratio of a cohesionless soil in the
loosest state and any given void ratio to the difference between its void ratios in the loosest
and the densest states (IS: 2720 – Part 14, 1983). That is,
(emax − e)
ID = × 100 (2.22)
(emax − emin )
or in terms of dry densities,
ρd max ρd − ρd min
ID = × × 100 (2.23)
ρd ρd max − ρd min
where emax is the void ratio of soil in the loosest state, emin is the void ratio of soil in the densest
state, e is the void ratio of soil in the in situ state, ρd max the dry density of soil in the densest state,
ρd min the dry density of soil in the loosest state, and ρd the in situ dry density of the deposit.
M02_PURU1773_01_SE_C02.indd 32 2/12/2013 7:49:23 AM

Fig. 2.4 Relative density conditions
Figure 2.4 shows the three density conditions. Density index is a measure of the degree of
compaction, and the stability of a stratum is indirectly reflected in the compactness. Table 2.2
presents designations associated with the density index for granular soils.
Simple practical methods may be adopted for determination of the maximum and mini-
mum void ratios of granular soils (Whitlow, 1983). Minimum void ratio is determined by
placing a soil in three layers in a mould of known volume, the mould being placed under
water. Each layer is compacted thoroughly using a vibratory hammer or tamping rod. The
collar of the mould is removed, the soil is levelled, and the mass of the soil, M, in the mould
is found. Then, maximum saturated density is given as
M G + e min
ρsat max = = ρw
V 1 + e min
Hence,
Gρw − ρsat max
emin =
ρsat max − ρw
Table 2.2 Density index of granular soils
Designation ID (%)
Very loose 0–15

Loose 15–35
Medium dense 35–65
Dense 65–85
Very dense 85–100
Source: Lambe and Whitman (1979).
M02_PURU1773_01_SE_C02.indd 33 2/12/2013 7:49:23 AM

To determine emax, the mould is placed under water, and the soil is poured quickly into it
just from the top. Knowing the mass and volume, the minimum saturated density is found
and, hence,
G ρw − ρsat min
e max =
ρsat min − ρw
The in situ void ratio is found as explained earlier. Hence, the density index can be com-
puted. A simple laboratory procedure for the determination of ID, for granular soils is given
in Chapter 10.
2.3 PARTICLE-SIZE ANALYSIS

In view of the great diversity of soil grain sizes, civil engineers have long recognized a need
for particle- or grain-size analysis. Particle-size analysis of soils determines the range of size
of particles and the percentage of particles in each of the sizes between the maximum and
the minimum. The designations given to different grain size ranges in soils by the Indian soil
classification system (IS: 1498, 1970) are gravel > 4.75 mm, sand > 0.075 mm, silt > 0.002 mm,
and clay < 0.002 mm.
The coarse particles (gravel and sand) may be separated by sieving, while a sedimenta-
tion procedure is used for analysing fine-grained soils (silt and clay). These are explained in
the subsequent paragraphs.
2.3.1 Sieve Analysis

This is the most direct method for determining particle sizes but has a lower limit with
respect to sieve opening. This lower limit corresponds to the fine-sand-size particles. In this
method, the soil sample is passed through a stack of standard sieves having successively
smaller mesh sizes. The percentage of weight retained, the cumulative percentage of weight
retained, and the percentage passing (by weight) in each sieve are calculated.
The resulting data are presented with grain size along the x-axis (long scale) and percent-
age passing or finer along the y-axis (arithmetic scale). All the points are connected by a
smooth curve which is referred to as a grain or particle size distribution curve. A detailed
procedure for conducting sieve analysis is given in Chapter 10.
In the case of clayey soils, the fine fraction cannot be easily passed through a 75 μm sieve
in the dry condition. In such cases, the materials are washed through with water (preferably
mixed with 2 gm of sodium hexametaphosphate per litre), until the wash water is clean. The
washed material is allowed to dry and then weighed. This is referred to as wet sieve analysis.
2.3.2 Sedimentation Analysis

The procedure commonly used for obtaining the particle-size distribution of fine-grained
soil or the fine-grained fraction of a coarse-grained soil is the sedimentation method. The
procedure is based on Stokes’ law, which states that in a suspension the velocity of a spherical
particle is governed by the diameter of the particle and the properties of the suspension.
Thus, the terminal velocity v (m/s) is given by
M02_PURU1773_01_SE_C02.indd 34 2/12/2013 7:49:24 AM

D2 (γ s γ w )
v= (2.24)
18ηw
where D is the diameter of the particle (m), γs the unit weight of grains of particles (kN/m3),
γw the unit weight of suspension fluid (usually water; kN/m3), and ηw the viscosity of the
suspension fluid (kN s/m2). Then,
18ηw He
D= (2.25)
( s w) t
γ − γ
where He is the height of the distance fallen (in metres) by the particles in time (seconds).
Equation 2.25 is valid for particles larger than 0.002 mm because smaller-sized particles will
be influenced by Brownian movement. The size of the particle is taken to be that of an equiv-
alent sphere which will have the same settling velocity as that of the particle.
In this method, the soil is placed as a suspension in distilled water. To ensure independent
settling of particles, a deflocculating agent is added to the suspension. The soil particles in the
suspension are allowed to settle out. A sample at a depth He below the surface, after allowing
the suspension to settle for time t, will contain no particles of size larger than D. All particles
smaller than D will be present in the sample in the same proportion as at the beginning of the
test. Thus, the effect is the same as if the sample had been separated on a sieve of mesh size D.
The concentration of particles remaining in the suspension at any level at any time may
be determined by adopting any one of the following methods.
Pipette Method. The sample of suspension is drawn off with a pipette (Fig. 2.5) at the
specified depth from the surface. The sample will contain only particles smaller than the
Fig. 2.5 Pipette
M02_PURU1773_01_SE_C02.indd 35 2/12/2013 7:49:24 AM

size D. Similarly, samples are taken at the specified depth at times corresponding to other
chosen particle sizes. The samples are dried, and the weight of the solid residue is recorded.
If Mb is the mass of the soil sample taken for sedimentation analysis after pre-treatment and
M′i is the mass of material (of specific sizes of particles) in the entire suspension from cor-
responding samplings, then the percentage of finer particles N is given as
Mi′
N= × 100
Mb
Now considering D (in mm), H (in cm), t (in min), and ηw (in poise) (1 poise = 10–4 kN s/m2)
and substituting the respective units in Eq. 2.25 we get
30ηw He
D= (2.26)
980 (ρs − ρw ) t
The diameter of the particle at every specified depth is obtained from Eq. 2.26. Then, the
grain-size distribution is obtained.
Hydrometer Method. This method measures the specific gravity of the suspension
using a special hydrometer (Fig. 2.6). The specific gravity of the solution decreases as the
Fig. 2.6 Hydrometer
M02_PURU1773_01_SE_C02.indd 36 2/12/2013 7:49:25 AM

settling starts. Specific gravity readings from the hydrometer at different time intervals
provide information about the size of the particles that have settled down and the mass of
soil remaining in solution.
A number of corrections are made for the hydrometer reading: (i) the meniscus correction
(cm), which is necessary as the suspension is opaque and the lower meniscus cannot be seen
clearly; (ii) the correction for the expansion of the hydrometer bulb due to increase in tem-
perature (°C); and (iii) the correction due to the addition of a dispersing agent. The first two
factors lead to a lower reading (note the hydrometer is graduated in increasing order from
top to bottom), and hence the corrections are positive. The third factor increases the density
of the suspension, and hence the correction is negative. The methods of determining these
corrections are explained in Chapter 10. A calibration curve is drawn between the hydrom-
eter reading corrected for meniscus correction (Rb) and the height of fall of the particle (H).
The calibration procedure is given in Chapter 10. Thus, the diameter of the particle at time t
after the starting of sedimentation is obtained from Eq. 2.26.
To obtain the percentage of finer particles corresponding to each hydrometer reading,
the density of suspension at that particular depth is required. Let M be the mass of the
pre-treated soil used in suspension of volume V. Before the start of the test; that is, at time
t = 0, the density of the suspension is uniform and the mass of solids in a unit volume of
the suspension is Mb/V.
Thus, the volume of solids Vs in unit volume of suspension at time t = 0 and at any depth
He is
⎛ ⎞
M
Vs = b =
Mb ⎜⎜Since G = ρs ⎟⎟
V ρs VG ρw ⎜
⎝ ρw ⎟⎟⎠
Therefore, the volume of water in unit volume of suspension (at t = 0) is
⎡ M b ⎤⎥
Vw = (1 − Vs ) = ⎢1 −
⎢ VGρ ⎥
⎣ w⎦
The initial density of the suspension is

M ⎡ M b ⎤⎥
ρi = b + ⎢1 − ρ
V ⎢ VGρ ⎥ wT
⎣ w⎦
where ρw is the density of water at 4°C and is 1 g/cm3, and ρwT is the density of water at test
temperature, T°C.
The density of suspension after time t and at temperature T°C can be written in a form
similar to the above expression for ρi as
MD ⎛⎜ MD ⎞⎟
ρf = + ⎜⎜1 − ⎟ρ
V ⎜⎝ VGρw ⎟⎟⎠ wT
where MD/V is the mass of particles of diameter smaller than D in unit volume of suspen-
sion at depth He, after time t. Substituing ρwT = GwT ρw, we have
MD ⎛⎜ MD ⎞⎟
ρf = + ⎜⎜1 − ⎟G ρ
V ⎜⎝ VGρw ⎟⎟⎠ wT w
M02_PURU1773_01_SE_C02.indd 37 2/12/2013 7:49:26 AM

Dividing both sides by ρw, we have

ρf M ⎛ MD ⎞⎟
= D + ⎜⎜⎜1 − ⎟⎟ GwT
ρw V ρw ⎜⎝ VGρw ⎟⎠
ρf/ρw is nothing but the specific gravity of suspension obtained after making the necessary
corrections of meniscus, temperature, and dispersion agent. Let the hydrometer reading be
r h. Then,
M ⎛ MD ⎞⎟
r h = D + ⎜⎜⎜1 − ⎟⎟ GwT
V ρw ⎜⎝ VGρw ⎟⎠
MD ⎛⎜ GwT ⎞⎟
rh = ⎜1 − ⎟ + GwT
V ρw ⎜⎝ G ⎟⎠
or
MD ⎛⎜ G − GwT ⎞⎟
(r h − GwT ) = ⎜ ⎟⎟
V ρw ⎜⎝ G ⎠
⎛ G ⎞⎟
MD = V ρw (r h − GwT )⎜⎜⎜ ⎟
⎜⎝ G − GwT ⎟⎟⎠
MD
N =
Mb
V ρw ⎛⎜ G ⎞⎟
N% = ⎜⎜ ⎟⎟(r h − GwT ) × 100
M b ⎜⎝ G − GwT ⎟⎠
Taking V = 1,000 ml, ρw = 1 g/cm3, and GwT = 1, we get

1000 ⎛⎜ G ⎞⎟
N% = ⎜ ⎟ (r h − 1) × 100
M b ⎜⎝ G − 1⎟⎠
Let r h 1000 = r h − 1. Therefore,

100G
N% = rh (2.27)
M b (G − 1)
From the data for D and the corresponding percentage of finer particles after each instant
of time, the grain-size distribution curve is obtained.
The sedimentation method is not absolutely correct as this is based on Stokes’ assump-
tion that (i) the particles are spherical, (ii) the flow around the particles is laminar, and
(iii) the particles are much larger than the molecular size. Assumptions (i) and (iii) are not
valid for fine-grained soils. Departure from spherical shape and molecular influence cause
the particles to settle slowly. The dispersion of particles may be incomplete, and viscosity is
not constant but varies due to changes in temperature.
However, this method has been in wide use and is more applicable to silts than to clays.
The gradation curve is not used for evaluation of the engineering properties of fine-grained
soils, and hence, a slight variation is insignificant.
Procedures for conducting pipette and hydrometer analyses are described in Chapter 10.
M02_PURU1773_01_SE_C02.indd 38 2/12/2013 7:49:28 AM

2.3.3 Grain-Size Distribution Curves

Grain-size distribution curves, as obtained from wet and dry methods, can be combined to form
one complete grain-size distribution curve. It has to be remembered that the particle sizes
defined in the two methods are different. In sieve analysis, a cylindrical particle of diameter D
and a spherical particle of the same diameter D would fit through the same sieve opening.
Though these two particles have different shapes, the sieve analysis identifies them as having
the same size. Further, particle size as measured in the sedimentation method assumes an
equivalent diameter of a spherical particle which would settle at the same rate. Thus, the accu-
racy of the gradation curve is questionable. In spite of serious limitations, particle-size curves of
sands and silts have some practical value in the design of filters and in the assessment of perme-
ability, capillarity, and frost susceptability, based on certain representative sizes of the particle.
However, very relevant and useful information may be obtained from a grain-sized
curve, such as (i) the total percentage of larger or finer particles than a given size (to identify
gravel-, sand-, silt-, and clay-size percentages) and (ii) the uniformity or the range in grain-
size distribution.
The range of particle sizes present in a soil is reflected in the flatness of the curve. The flat-
ter the curve, larger the range of size of particles and steeper the curve, smaller the range. The
effective particle size has been defined by Hazen (1892) as that for which 10% of the material
by weight is smaller than that size. Other particle sizes are also often used in describing or
classifying soils; e.g., D50 of a soil is used to represent the medium particle size, while D85
and D15 sizes are used to decide certain filter criteria.
A soil is said to be well graded or non-uniform if there is a distribution of particles over
a relatively wide range (Fig. 2.7). A soil is said to be poorly graded if the sample has a
Fig. 2.7 Typical particle-size distribution curves
M02_PURU1773_01_SE_C02.indd 39 2/12/2013 7:49:29 AM

very narrow range of particles (also called uniform soil) or the sample is deficient in certain
intermediate grain sizes (also called gap graded).
We can obtain a numerical measure of the gradation by defining the uniformity coefficient
Cu and the curvature coefficient Cz:
Uniformity coefficient:
D
Cu = 60 (2.28)
D10
Curvature coefficient:
2
D30
Cz = (2.29)
D60 × D10
Soils with Cu < 4 are said to be uniform, and soils with Cu > 4 (6 for sands) are well graded
as long as the grain-size distribution curve is smooth and symmetrical.
The curvature coefficient Cz is a measure of the symmetry and shape of the gradation curve.
For a well-graded soil, Cz will be around 1. For Cz much smaller or much greater than 1, the
soil is viewed as poorly graded.
The uniformity coefficient and curvature coefficient are used as part of the Unified and
Indian soil classification systems.
2.4 CONSISTENCY OF SOILS

2.4.1 Atterberg Limits
Consistency refers to the texture and firmness of a soil and is conventionally denoted as soft,
medium stiff, stiff, or hard. The consistency of a fine-grained soil is largely influenced by the
water content of the soil. A gradual decrease in water content of a fine-grained soil slurry
causes the soil to pass from the liquid state to a plastic state, from the plastic state to a semi-
solid state, and finally to the solid state. The water contents at these changes of state are dif-
ferent for different soils. The water contents that correspond to these changes of state are
called the Atterberg limits. These four consistency states are shown in Fig. 2.8. The water
contents corresponding to transition from one state to the next are known as the liquid limit
(wL), the plastic limit (wp), and the shrinkage limit (ws).
The liquid limit of a soil is the water content, expressed as a percentage of the weight of the
oven-dried soil, at the boundary between the liquid and plastic states of consistency of the soil
(IS: 2720 – Part 5, 1970). A specified test procedure has been given by Casagrande, which is
performed by placing a soil pat in a cup, with the pat grooved at the centre by a standard tool.
The cup is allowed to drop from a height of 10 mm. The water content of the soil pat when the
groove cut in it closes over 12 mm at 25 drops is referred to as the liquid limit of the soil. A plot of
water content versus number of blows (on a log scale) is called a flow curve (Fig. 2.9). Details of
the apparatus and procedure for this test and the one explained below are given in Chapter 10.
The cone penetrometer test is another procedure recommended by the Indian Standards
(IS: 2720 – Part 5, 1970) to find the liquid limit. Essentially, in this test, the penetration of a
standard cone (Fig. 2.10) into a saturated soil sample is measured for 30 seconds. If the pen-
etration is less than 20 mm, the wet soil is taken out and mixed thoroughly with water and
the test is repeated till the penetration is between 20 and 30 mm. The exact penetration value
M02_PURU1773_01_SE_C02.indd 40 2/12/2013 7:49:29 AM

Fig. 2.8 Consistency relationships
is noted down and the corresponding water content determined. The test is repeated for a
variety of water contents, and the water content corresponding to a penetration of 25 mm
is taken as the liquid limit of the soil. The test is quicker, and the results are accurate and
reproducible. This has several advantages over the mechanical method. Therefore, it has
been recognized as a standard method by Indian Standards.
The plastic limit of a soil is the water content, expressed as a percentage of the weight of
oven-dried soil, at the boundary between the plastic and semi-solid states of consistency of
the soil (IS: 2720 – Part 5, 1970). The plastic limit is determined by rolling a pat of soil into
Fig. 2.9 Flow curve
M02_PURU1773_01_SE_C02.indd 41 2/12/2013 7:49:30 AM

Fig. 2.10 Cone penetration test apparatus
a thread, and the water content at which the soil shows signs of crumbling at a diameter of
3 mm is the plastic limit. The detailed procedure is given in Chapter 10.
The plastic limit for different soils has a narrow range of numerical values. Sand has no
plastic stage, but very fine sand exhibits slight plasticity. The plastic limit is an important soil
property. Earth roads are easily usable at this water content. Excavation work and agricul-
tural cultivation can be carried out with the least effort with soils at the plastic limit.
Soil is said to be in the plastic range when it possesses water content in the range between
wL and wp. The range of the plastic state is given by the difference between wL and wp and is
defined as the plasticity index. That is,
I p = wL − w p (2.30)
The plasticity index represents the range of water content over which a soil is plastic. The
greater the plasticity index, higher will be the attraction between the particles of the soil
and greater the plasticity of the soil. Based on the plasticity index, the soils are classified by
Atterberg as follows:
Plasticity index (%) Plasticity
0 Non-plastic
<7 Low plastic
7–7 Medium plastic
>17 High plastic
The plasticity index is used in soil classification and in various correlations with other soil
properties as a basic soil characteristic.
The shrinkage limit is the maximum water content expressed as a percentage of oven-
dried weight at which any further reduction in water content will not cause a decrease in
volume of the soil mass, the soil mass being prepared initially from remoulded soil (IS: 2720 –
Part 6, 1972). Based on the above definition, the shrinkage limit is determined by completely
M02_PURU1773_01_SE_C02.indd 42 2/12/2013 7:49:30 AM

Fig. 2.11 Phase diagrams representing shrinkage limit conditions
drying out a lump of soil and measuring its final volume and mass. Thus, referring to the
phase diagram in Fig. 2.11, the shrinkage limit is given as
( M − M0 ) − (V − V0 ) ρw
ws = × 100 (2.31)
M0
where M is the initial wet mass of soil, M0 the final dry mass of soil, V the initial volume of
soil, and V0 the final volume of dry soil mass.
The shrinkage limit test can also be performed on undisturbed soil; in that case, the nota-
tion wsu is used.
The finer the particles of the soil, the greater is the amount of shrinkage. Soils that contain
montmorillonite clay mineral shrink more. Such soils shrink heterogeneously during sum-
mer, as a result of which cracks develop on the surface. Further, these soils imbibe more and
more water during the monsoon and swell. Soils that shrink and swell are categorized as
expansive soils. Indian black cotton soils belong to this group.
A detailed test procedure for determination of the shrinkage limit is given in Chapter 10.
The relationship between different limits of consistency and natural or in situ water con-
tent is given below (IS: 2720 – Part 5, 1970):
1. The liquidity index or water plasticity ratio (IL) is the ratio expressed as a percentage of
the natural water content (wn) of a soil minus its plastic limit to its plasticity index. That is,
wn − wp
IL = (2.32)
Ip
The in situ state of a soil is represented by the liquidity index: when IL < 0, the soil is
in the semi-solid state; when IL = 0, the soil is in the stiff state; when 0 > IL < 1, the soil
is in the plastic state, when IL = 1, the soil is in a very soft state; and when IL > 1, the soil is
in the liquid state.
2. The consistency index (Ic) of a soil is the ratio of the liquid limit minus the natural
water content to its plasticity index. That is,
wL − w n
Ic = (2.33)
Ip
M02_PURU1773_01_SE_C02.indd 43 2/12/2013 7:49:31 AM

Similar to the liquidity index, the consistency index also represents the in situ state of a
soil. When the consistency index is equal to 1.0, the water content is at the boundary between
the semi-solid and solid states. If the consistency index is negative, it represents the state
where the soil flows and is unsuitable for foundation purposes. Consistency index can have
a value greater than 1.0. For a fair load-bearing condition, the soil should have a value of
Ic = 0.5, which indicates the boundary between the soft and stiff plastic states.
3. The flow index (If) of a soil is the slope of the flow curve obtained from a liquid limit
test, expressed as the difference in water content at 10 blows (N1) and at 100 blows (N2):
w1 − w2
If = (2.34)
log10 ( N 2 N1 )
where w1 and w2 are the water contents corresponding to N1 and N2 drops, respectively.
Slopes of flow curves distinguish between the degree of cohesiveness and the shear
strength of various soils. Two soils with the same plasticity index but different liquid limits
will have different flow indices. The one with a steeper flow curve indicates soil of low shear
strength.
4. The toughness index (IT) of a soil is the ratio of the plasticity index to the flow index
Ip
IT = (2.35)
If
The shear strength of a fine-grained soil at a water content close to the plastic limit is a
measure of its toughness. The toughness of two fine-grained soils with the same plasticity
index is inversely proportional to the flow indices. For clay, the toughness index is generally
less than 3.
The concept of analysing various states in a soil based on water content is a sound one.
The limits have been fixed arbitrarily and cannot be accepted as fundamental properties.
Thus, not much significance should be attached to their accurate values.
2.4.2 Activity of Clays

As the particle size decreases, the surface area of the particle and the amount of water
attracted to the soil surface increase. Thus, the amount of water attracted will depend con-
siderably on the number of clay-size particles present in the soil. On the basis of this reason-
ing, Skempton (1953) proposed a relationship between the plasticity index and the percentage
of particle sizes finer than 2 μm and called the quantity, the activity of clay, A.
Plasticity index
A=
Percentage by weight of particlles finer than 2 μm
The activity of clay A gives a qualitative measure of the behaviour of the soil as active,
normal, or inactive. Skempton (1953) has classified the clay as inactive if A < 0.75, normal
if A is 0.75–1.25, and active if A > 1.25. Table 2.3 shows typical values of liquid limit, plastic
limit, and activity.
M02_PURU1773_01_SE_C02.indd 44 2/12/2013 7:49:32 AM

Table 2.3 Typical values of ωL, ωP, and A of some minerals
Mineral Liquid limit (ωL) Plastic limit (ωP) Activity (A)
Kaolinite 35–100 20–40 0.3–0.5

Halloysite 40–55 30–45 0.4–0.6
Illite 60–120 35–60 0.5–1.2
Moutmorillonite 100–900 50–100 1.5–7.0
Source: Das (2002).
WORKED EXAMPLES
Example 2.1 Adopting a routine laboratory procedure, the specific gravity of a river sand
was determined. The mass of dry sand was 198.6 g. The mass of the calibrated flask filled
with water was 1508.2 g. The masses of the flask, water, and sand were 1632.6 g. Determine
the particle specific gravity of the soil.
If the true specific gravity was 2.72 and an error was made in recording the mass of dry
sand, what is the correct mass of dry soil? The other two observations are correct.
Solution
Mass of dry soil = 198.6 g
Mass of an equal volume of water = (1508.2 – 1632.6) + 198.6 = 74.2 g
Mass of dry soil
Specific gravity of soil solids, G =
Mass of an equal volume of water
198.6
= = 2.68
74.2
Because of wrong recording of the mass of dry sand, both the numerator and denominator
are affected. Let Ms be the true mass of dry sand. Therefore,
Ms
2.72 =
(1508.2 − 1632.6) + Ms
Solving for Ms, the true mass of dry sand is 196.73 g.
Example 2.2 An attempt was made to determine the water content of a given moist soil of
known specific gravity, using a pycnometer. The usual laboratory procedure for specific
gravity determination of dry soil is used for the wet soil.
The following are the observations:
Mass of pycnometer (M1) = 545 g
Mass of pycnometer with moist soil (M2) = 790 g
Mass of pycnometer with soil and water (M3) = 1,540 g
M02_PURU1773_01_SE_C02.indd 45 2/12/2013 7:49:32 AM

Mass of pycnometer and water (M4) = 1,415 g

Specific gravity of soil grains = 2.67
Determine the water content of the soil from first principles.
Solution
Consider the two-phase diagrams shown in Fig. 2.12 representing the observations. Thus,
M3 − M4 = Ms − (mass of an equal volume of water)
⎛M ⎞ ⎛ ⎞
= Ms − ⎜⎜⎜ s ⎟⎟⎟ ρw ⎜⎜since V = Ms = V ⎟⎟
⎟⎟
⎜⎝ Gρw ⎟⎠ ⎜⎜⎝ s
ρs
w
⎠
⎛ G − 1⎞⎟
= Ms ⎜⎜⎜ ⎟
⎝ G ⎟⎠
⎛ G ⎞⎟
That is, Ms = ( M3 − M4 )⎜⎜ ⎟
⎝⎜ G − 1⎟⎠
Substituting the respective values,
2.67
Ms = (1540 − 1415) = 199.85 g
2.67 − 1
Therefore,
( M2 − M1 ) − Ms
w= × 100
Ms
Substituting,
(790 − 545) − 199.85
w= × 100 = 22.6%
199.85
Fig. 2.12
M02_PURU1773_01_SE_C02.indd 46 2/12/2013 7:49:34 AM

Example 2.3 A sample of soil, extracted in its natural state using a sampling tube of volume
0.001 m3, was found to have a mass of 1,730 g, the degree of saturation being 61.6%. The
oven-dried mass of soil was 1,440 g. Determine (i) the natural water content, (ii) the specific
gravity of soil solids, (iii) the void ratio, (iv) the bulk density, (v) the saturated density, and
(vi) the submerged density.
Solution
1730 − 1440
1. Natural water content w = × 100 = 20.14%
1440
1440
Dry density ρd = = 1.44 g cm 3
0.001×100 3
Also,
Gρw Gρw ⎛ ⎞
ρd = = ⎜⎜since e = wG ⎟⎟
1 + e 1 + (wG Sr ) ⎜⎝ Sr ⎟⎟⎠
G× 1 G
1.44 = =
1 + (20.14 61.6) G 1 + 0.3279
Rearranging,
2. G = 2.72
wG 20.14 × 2.72
3. e = = × 2.72 = 0.89
Sr 100 × (61.6 100)
⎛ 20.14 ⎞⎟
4. Bulk density ρt = ρd (1 + w) = 1.44 ⎜⎜⎜1 + ⎟ = 1.73 g cm
3
⎝ 100 ⎟⎠
⎛ e⎞
5. Saturated density ρsat = ρd ⎜⎜1 + ⎟⎟⎟ (Sr = 100%)
⎜⎝ G⎠
⎛ 0.89 ⎞⎟
= 1.44 ⎜⎜⎜1 + ⎟ = 1.91 g cm
3
⎝ 2.72 ⎟⎠
6. Submerged density ρ ′ = ρsat − ρw = 1.91 − 1.0 = 0.91 g cm 3
Example 2.4 A saturated specimen of undisturbed clay has a volume of 22.5 m3 and mass
of 35 g. After oven drying, the mass reduces to 20 g. Find its moisture content, specific gravity
of solids, void ratio, and dry density.
Solution
Mw 35 − 20
Moisture content w = = × 100 = 75%
Ms 20
M 20
Dry density ρd = s = = 0.898 g cm 3
V 22.5
M02_PURU1773_01_SE_C02.indd 47 2/12/2013 7:49:36 AM

Also, ρd = Gρw (1 + e) , and for a saturated soil, e = wG. Therefore,
Gρw
ρd =
Rearranging, 1 + wG
ρd 0.89
G= = = 2.697 (since rw = 1.0)
1 − ρd w 1 − 0.89 × 0.75
75
Therefore, the void ratio e = wG = × 2.697 = 2.02
100
Example 2.5 A wet soil sample weighs 3.46 N. After drying at 5°C, its weight is 2.84 N. The
bulk unit weight of the soil is 18.6 kN/m3. The specific gravity of the solid particles is 2.7.
Determine (i) the water content, (ii) the void ratio, (iii) the degree of saturation, and (iv) the
porosity.
Solution
3.46 − 2.84
1. Water content w = × 100 = 21.83%
2.84
γt 18.6
Dry unit weight γd = = = 15.27 kN m 3
1 + w 1 + (21.83 100)
Gγ w 2.7 × 9.81
2. Void ratio e = −1 = − 1 = 0.735
γd 15.27
wG 21.83 2.7
3. Degree of saturation Sr = = × × 100 = 80.19%
e 100 0.735
e 0.735
4. Porosity n = = × 100 = 42.36%
1 + e 1 + 0.75
Example 2.6 The bulk unit weight of a soil is 19.10 kN/m3, the water content is 12.5%, and
the specific gravity of soil solids is 2.67. Determine the dry unit weight, void ratio, porosity,
and degree of saturation.
Solution
γ 19.1
1. γd = = = 16.98 kN m 3
1 + w 1 + (12.5 100)
Gγ w
γd =
1+ e
Therefore,
Gγ w 2.67 × 9.81
2. e = −1 = − 1 = 0.54
γd 16.98
M02_PURU1773_01_SE_C02.indd 48 2/12/2013 7:49:38 AM

Therefore,
e 0.54
3. n = × 100 = × 100 = 35.07%
1+ e 1 + 0.54
wG 12.5 2.67
4. Sr = × 100 = × × 100 = 61.8%
e 100 0.54
Example 2.7 A soil sample has a porosity of 40%. The specific gravity of solids is 2.7. Cal-
culate the (i) void ratio, (ii) dry density, (iii) unit weight if the soil is 50% saturated, and
(iv) unit weight if the soil is completely saturated.
Solution
n 0.40
1. Void ratio e = = = 0.67
1 − n 1 − 0.40
Gρw 2.7 × 1
2. Dry density ρd = = = 1.62 g cm 3
1 + e 1 + 0.67
G + eSr 2.7 + 0.67 × 0.5
3. Wet unit weight γ t = γw = × 9.81
1+ e 1 + 0.67
= 17.82 kN m 3
G+e 2.7 + 0.67

4. Saturated unit weight γ sat = γw = × 9.81
1+ e 1 + 0.67
= 19.80 kN m 3
Example 2.8 How many cubic metres of fill can be constructed at a void ratio of 0.65 from
2,21,000 m3 of borrow material that has a void ratio of 1.25?
Solution
Let eb and ef be the void ratios of the borrow material and the fill, respectively. Also, let Vvb
and Vvf be the volume of voids in the borrow and the fill, respectively. The volume of soil
solids is the same in both the cases. From Fig. 2.13,
Fig. 2.13
M02_PURU1773_01_SE_C02.indd 49 2/12/2013 7:49:40 AM

Vvb V
eb = and ef = vf
Vs Vs
Therefore,
Vvb = e bVs and Vvf = ef Vs
Total volume of soil in the borrow is Vb = Vvb + Vs.
That is, Vb = e bVs + Vs = (1 + e b )Vs

Vb
Therefore, Vs =
1 + eb
Total volume of soil in the fill = Vf = (1 + ef )Vs.
Vb 1 + 0.65
Vf = ( 1 + e f ) = × 221000 = 162066.7 m3
1+e b 1 + 1.25
Example 2.9 For a stable packing of regular spheres at the minimum density, find the void
ratio and the dry unit weight. Unit weight of soil solids is 25 kN/m3.
Solution
Refer to Fig. 2.14.
Let D be the diameter of each sphere.
πD3
Volume of each sphere =
6
For the arrangement in Fig. 2.14, the density will be minimum.
Volume = 2D × 2D × D = 4D3
4D3 − 4 × πD3 / 6 1− π / 6 6 − π
Therefore, e = = = = 0.91
4 × πD3 / 6 π/ 6 π
Also,
Ms Vγ 4 × πD3 / 6 × γ s πγ s π × 25
γd = g= s s = = =
V V 4D 3 6 6
3
That is, γd = 13.09 kN/m
Fig. 2.14
M02_PURU1773_01_SE_C02.indd 50 2/12/2013 7:49:43 AM

Example 2.10 A fully saturated soil sample was extracted during an oil well drilling. The
wet mass of the sample was 3.15 kg, and the volume of the sampling tube was 0.001664 m3.
After analysis, the soil sample was found to contain 28.2% of the liquid as kerosene and had
a dry mass of 2.67 kg. The specific gravity of soil grains was 2.68. Determine the bulk density,
void ratio and water content of the sample.
Solution
3.15
Bulk density = = 1893 kg/m 3 = 1.89 Mg/m 3
0.001664
2.67
Volume of soil grains = = 0.000996 m 3
2.68 × 1000
Volume of voids = 0.001664 − 0.000996 = 0.000668 m 3
0.000668
e= = 0.67
0.000996
As the soil was fully saturated,
Volume of liquid = volume of voids = 0.000668 m 3
Volume of water = (1 − 0.282) × 0.000668 = 0.00048 m 3
Mass of water = 0.48 kg
0.48
Water content = × 100 = 17.89%
2.67
Example 2.11 A mass of soil is coated with a thin layer of paraffin wax. The paraffin wax
weighs 6.906 g, and the soil alone weighs 443 g. When the sample is immersed in water, it
displaces 346 ml of water. The specific gravity of the soil solids is 2.67, and that of wax is 0.89.
Find the void ratio and degree of saturation, if the water content is 17.2%.
Solution
Mp 6.906
Volume of paraffin wax Vp = = = 7.76 cm 3
Gpρw 0.89 × 1
Ms 443
Volume of soil solids Vs = = = 167.17 cm 3
Gs ρw 2.65 × 1
Volume of soil V = 346 − 7.76 = 338.24 cm 3
Volume of voids Vv = 338.24 − 167.17 = 171.07 cm 3

Vv 171.07
Void ratio e = = = 1.02
Vs 167.17
wG 17.2 2.65
Degree of saturation Sr = = × × 100
e 100 1.02
= 44.69%
M02_PURU1773_01_SE_C02.indd 51 2/12/2013 7:49:45 AM

Example 2.12 A compacted cylindrical specimen 50 mm in diameter and 100 mm long is to

be prepared from dry soil. If the specimen is required to have a water content of 15%, find the
percentage of air voids required in the preparation of the soil when the specific gravity is 2.69.
Solution
π
Volume of cylinder = Volume of soil = × 52 × 10 = 196.3 cm 3
4
e ⎛⎜ wG ⎞⎟ ⎛⎜ e − wG ⎞⎟
Air void ratio Av = ⎜1 − ⎟=⎜ ⎟
1 + e ⎜⎝ e ⎟⎠ ⎜⎝ 1 + e ⎟⎠
(1+ e) Av = e − wG
Rearranging,
wG + Av (15 100) × 2.69 + (20 100)
e= =
1 − Av 1 − (20 100)
0.15× 2.69 + 0.20 0.2035
= = = 0.75
0.880 0.80
Also,
Vv V − Vs V
e= = = −1
Vs Vs Vs
196.3
That is, 0.75 = −1
Vs
196.3
Vs = = 112.2 cm 3
1.75
Weight of soil Ms = ρ sVs = Gρ wVs = 2.69 × 1× 112.2
15
Weight of water Mw = w Ms = × 301.8 = 45.27 g
100
Example 2.13 A test of the density of the soil in place was performed by digging a small
hole in the soil, weighing the extracted soil, and measuring the volume of the hole. The soil
(moist) weighed 8.95 N; the volume of the hole was 426 cm3. After drying, the sample
weighed 7.78 N. Of the dried soil, 4 N was poured into a vessel in a very loose state. Its vol-
ume was subsequently determined to be 276 cm3. That same 4 N was then vibrated and
tamped to a volume of 212 cm3. The specific gravity of the solid particles is 2.7. Find the
relative density of the soil.
Solution
7.78
Natural density ρd = = 0.0183 N/cm 3
426
4
Loose density ρd min = = 0.0145 N/cm 3
276
4
Maximum density ρd max = = 0.01887 N/cm 3
212
M02_PURU1773_01_SE_C02.indd 52 2/12/2013 7:49:47 AM

From Eq. 2.23,

ρd max 0.0183 − 0.0145
ID = × × 100
ρd 0.01887 − 0.0145
0.01887 0.0038
= × × 100 = 92.29%
0.0183 0.00437
Example 2.14 In order to determine the in-place density of a highway sub-grade, a sand
bottle method was adopted. The mass of soil extracted from a hole at the surface was 4.87 kg.
The hole was then filled with sand from the sand bottle and found to have a mass of 3.86 kg.
While calibrating the sand bottle, to fill a container of volume 0.0048 m3, a mass of 6.82 kg of
sand was needed. In a moisture content determination, 28.26 g of the moist soil weighed
22.2 g after oven drying. If the specific gravity of the soil was 2.67, determine the bulk and
dry densities and the degree of saturation of the soil.
Solution
6.82
Density of sand in the sand bottle = = 1420.8 kg/m 3
0.0048
= 1.42 Mg/m 3
3.86
Volume of the hole = = 0.00272 m 3
1420.8
4.87
Bulk density = = 1790.4 kg/m 3 = 1.79 Mg/m 3
0.00272
28.26 − 22.2
Water content = × 100 = 27.3%
22.2
1790.4
Dry density = = 1406.4 kg/m 3 = 1.41 Mg/m 3
1 + (27.3 100)
Gρw 2.67 × 1000

e= −1 = − 1 = 0.899
ρd 1406.4
wG 27.3 2.67
Sr = × 100 = × × 100 = 81.08%
e 100 0.8899
Example 2.15 A relative density test conducted on a sandy soil yielded the following
results: maximum void ratio = 1.23, minimum void ratio = 0.48, relative density = 42%, and
G = 2.67.
Find the dry density of the soil in the present state. If a 3 m thickness of this stratum is
densified to a relative density of 62%, how much will the soil reduce in thickness? What will
be the new density in dry and saturated conditions?
Solution emax − e
ID = ×100
emax − emin
M02_PURU1773_01_SE_C02.indd 53 2/12/2013 7:49:49 AM

or
1.23 − e
0.42 = × 100
1.23 − 0.48
or
(1.23 − e ) = 0.42(1.23 − 0.48)
or
e = 0.915
Now,
Gρw 2.67 × 1
ρd = = = 1.37 g/cm 3
1 + e 1 + 0.915
and
Vv V − Vs
e= =
Vs Vs
Therefore,
3 −Vs
0.915 =
Vs
Therefore,
Vs = 1.57 m 3
For 62% relative density, the void ratio to which the soil has to be compacted is obtained from
1.23 − e
0.62 = × 100
1.23 − 0.48
or
e = 0.765
Therefore,
V − 1.57
0.765 =
1.57
or
V = 0.765 × 1.57 + 1.57 = 2.77
Therefore, the reduction in thickness is 0.23 m.
Example 2.16 From the results of a sieve analysis given below, plot a grain-size distribu-
tion curve and then determine (i) the effective size, (ii) the uniformity coefficient, and (iii) the
coefficient of gradation. Mass of soil taken for sieve analysis was 500 g.
IS sieve no. Mass of soil retained in each sieve (g)
480 3.8
240 32.2
120 52.8
60 38.7
30 122.5
15 15
8 26.4
M02_PURU1773_01_SE_C02.indd 54 2/12/2013 7:49:51 AM

Solution
Sieve no. Sieve opening Mass retained Percent Cumulative Percent

(mm) (g) retained percent retained finer
480 4.76 3.8 0.76 0.76 99.24

240 2.40 32.2 6.44 7.20 92.80
120 1.20 52.8 10.56 17.76 82.24
60 0.60 38.7 7.74 25.50 74.50
30 0.30 122.5 24.50 50.00 50.00
15 0.15 159.9 31.98 81.98 18.02
8 0.075 26.4 5.28 87.26 12.74
The grain-size distribution curve is plotted as given in Fig. 2.15.

Effective size D10 = 0.07 mm
D60 0.43
Uniformity coefficient Cu = = = 6.14
D10 0.07
D30 2 0.212
Coefficient of gradation Cz = = = 1.47
D60 × D10 0.43 × 0.07
Fig. 2.15
M02_PURU1773_01_SE_C02.indd 55 2/12/2013 7:49:51 AM

Example 2.17 In a sedimentation analysis 48 g of soil passing, 75 μm is dispersed in 1,000 ml

of water. In order to estimate the percentage of particles of size less than 0.003 mm, how long
after the commencement of sedimentation is the hydrometer reading to be taken? The centre
of the hydrometer is 165 mm below the surface of the water. The specific gravity of soil grains
is 2.72, and viscosity of water is 0.001 N-s/m2.
Solution
From Eq. 2.24,
D2 (γ s − γ w )
v=
18η
2
(0.003 / 1000) (2.72 − 1.0) 9.81
v= = 8.437 × 10−6 m/s
18 × (0.001 / 1000)
Also,
He
v=
t
That is,
165
8.437 =
1000t
Rearranging,
t = 19556.7 seconds = 5.4 hours
Example 2.18 In a pipette analysis, 25 g of soil was dispersed in water, and the suspension
was made to a volume of 1,000 ml. The viscosity of water is 0.0012 SI units. Thirty minutes
after the commencement of sedimentation, 20 ml of the suspension was taken at a depth of
100 mm. The sampled soil was dried and found to have a mass of 0.076 g and G of 2.71.
Compute (i) the largest size of particles remaining in suspension 30 minutes after the com-
mencement of sedimentation at a depth of 100 mm and (ii) the percentage of finer particles.
Solution
We know that
18ηw He
D=
(ρs − ρw ) t
Here, ρs = 2,710 kg/m3 and ρw = 1,000 kg/m3 (Fig. 2.16). Therefore,
18 × 0.0012 100
D= × 1000 mm
(2710 − 1000)9.81 1000 × 30 × 60
or
D = 0.00846 mm
3.8
Mass of soil material in suspension = × 100 = 15.2%
25
M02_PURU1773_01_SE_C02.indd 56 2/12/2013 7:49:52 AM

Fig. 2.16
Example 2.19 The liquid limit and plastic limit of a soil are 65% and 31%, respectively. The
natural water content is 25%. Find the liquidity index and activity number. Comment on the
consistency of the soil.
Solution
Plasticity index Ip = wL – wp = 65 – 31 = 34%
wn − wp 25 − 31
Liquidity index I L = = = − 0.176
Ip 34
Ip 34
Activity number A = = = 1.42
% Particle less than 2 μm 24
The consistency of the soil is very stiff as the liquidity index is negative. The soil is highly
plastic as the plasticity index is greater than 17%. The soil is active as the activity number is
greater than 1.25.
Example 2.20 The shrinkage limit of a clay is 22%, its natural moisture content 34.7%, and
its specific gravity 2.65. Calculate the percentage decrease to be expected in a unit volume of
clay if the moisture content is reduced by evaporation to 18.2%.
M02_PURU1773_01_SE_C02.indd 57 2/12/2013 7:49:53 AM

Solution
Volume of soil at natural saturated condition = Vs + volume of water
34.7
= Vs + Ms = Vs + 0.347 Ms
100
Volume of soil at shrinkage limit condition = Vs + 0.22 Ms

Although water content might have reduced to 18.2%, the volume cannot be less than that
at the shrinkage limit.
(V + 0.347 Ms ) − (Vs + 0.22 Ms )
Percentage volume reduction = s × 100
(Vs + 0.347 Ms )
0.127 Ms
= × 100
( Ms / qρw ) + 0.347 Ms
0.127
=
(1 / 2.65 × 1) + 0.347
= 17.5%
Example 2.20 A soil sample collected from the field was found to have a mass of 475 g and
its oven dry mass is 415.8 g. The soil was found to have a void ratio of 0.86 and the specific
gravity as 2.66. Determine the moist and dry densities. Moreover, find the mass of water, in
kilograms, to be added per cubic metre of soil in the field for saturation.
Solution
μω 475 − 415.8
ω= = ×100 = 12.46%
μλ 475
Gρω (1 + ω) 2.66 ×1, 000 (1 + 0.1246)

ρ= = = 1, 608.3 kg/m 3
1+ e 1 + 0.86
Gρω 2.66 ×1, 000
ρd = = = 1, 430.11 kg/m 3
1+ e 1.86
Mass of water to be added for saturation } = ρsat = ρ
(G + e) ρω (2.66 + 0.86)1, 000

ρsat = = = 1, 892.47 kg/m 3
1+ e 1.86
Therefore, the mass of water to be⎪⎫⎪

⎬ = 1, 892.47 − 1, 608.3
added per cubic metre ⎪⎪⎭
= 284.17 kg.
Example 2.21 A fully saturated clay sample has specific gravity of 1.98 at 25% water con-
tact. After oven-drying, the mass specific gravity reduces to 1.63. Find the specific gravity of
the clay and the shrinkage limit.
M02_PURU1773_01_SE_C02.indd 58 2/12/2013 7:49:54 AM

Solution
As the clay is fully saturated void ratio, e} = ωG = 0.25 G
⎛ G + e ⎞⎟
ρsat = ⎜⎜ ρ
⎜⎝ 1 + e ⎟⎟⎠ ω
We know
ρsat G + e
i.e., =
ρω 1+ e
G + 0.25G 1.25G
i.e., 1.98 = = .
1 + 0.25G 1 + 0.25G
By rearranging, we get
1.98 + 1.98 × 0.25G = 1.25G
1.98 + 0.495G = 1.25G
∴ G = 2.62
As the soil is allowed to dry gradually in an oven, the dry mass specific gravity is at the
shrinkage limit stage. Then the water contact at this stage is the shrinkage limit, ωs
ρω 1 ⎛⎜ 1 1 ⎞⎟
i.e., ωs = − =⎜ − ⎟ × 100
ρd G ⎜⎝ 1.63 2.62 ⎟⎠
= 23.2%.
Hence shrinkage limit = 23.2%.
POINTS TO REMEMBER
2.1 Soil deposits are particulate systems of three distinct phases, viz., soil solids, water,
and air. This is referred to as the three-phase system. Dry soil (absence of water phase)
and fully saturated soil (absence of air phase) constitute the two-phase systems.
2.2 Void ratio is an important parameter which governs the permeability, settlement, and
stability problems of soil. Values of void ratio may range from 0.50 to 1.50 in soils.
2.3 Water content and degree of saturation represent the amount of water present in a
soil. The behaviour of dry and saturated soils is easy to assess compared to partially
saturated soils. Control of compaction is governed by moisture content.
2.4 Specific gravity of soil solids has a narrow range of variation (2.65–3.00), and the pres-
ence of organic material reduces the specific gravity.
2.5 The density of a soil (dry, saturated, or submerged) is a function of void ratio and
moisture content and has a major role to play in all stability problems.
2.6 Grain-size distribution curves (obtained from sieve and sedimentation analyses)
reflect the range of particle sizes present. The flatter the curve, the larger is the range
of size of particles, and the steeper the curve, the smaller the range.
M02_PURU1773_01_SE_C02.indd 59 2/12/2013 7:49:56 AM

2.7 A numerical measure of the gradation of a soil is obtained by defining the uniformity
coefficient and curvature coefficient and is also used in soil classification.
2.8 Selective particle sizes are used in the classification (D60 and D10) design of filters
(D50, D85, and D15).
2.9 Consistency of a soil refers to the texture and firmness of a soil and is denoted as soft,
medium stiff, and stiff.
2.10 Consistency limits or Atterberg limits, viz., liquid limit, plastic limit, and shrinkage
limit, are the water contents at the changes of states from liquid to plastic, plastic to
semi-solid, and semi-solid to solid, respectively.
2.11 Liquidity index and consistency index represent the in situ firmness condition of a
soil.
QUESTIONS
Objective Questions
2.1 Choose the correct statement from the following:
(i) The porosity of a soil can be greater than 100%.
(ii) The water content of a soil cannot be greater than 100%.
(iii) The natural water content of a soil cannot exceed the liquid limit.
(iv) The consistency index of a soil can be negative.
2.2 Void ratios of a micaceous sand sample in the densest and the loosest conditions are
0.4 and 1.2, respectively. The relative density of the soil for the in-place void ratio of
0.6 will be
(a) 60% (b) 75% (c) 65% (d) 80%
2.3 Consistency, in general, is that property of a soil which is manifested by its resistance
to
(a) Impact (b) Rolling (c) Flow (d) None of the above
2.4 A clay is identified as a normal clay if the activity range is between
(a) 0.25 and 0.75 (b) 0.75 and 1.25 (c) 1.25 and 3.00 (d) 0.15 and 0.25
2.5 Swelling of clayey soil directly depends on the
(a) Percentage of clay fraction (b) Plasticity index of the soil
(c) Type of clay mineral (d) Liquid limit of the soil
2.6 Sand-bath method of determining water content is not suitable for
(a) Inorganic silts (b) Fine sands
(c) Soils with a high percentage of organic matter
(d) Soils with particle size ranging from 0.02 to 0.075 mm
2.7 For soils containing soluble salts, the specific gravity is determined using
(a) Salt water (b) De-aired water
(b) White spirit (d) Benzene
M02_PURU1773_01_SE_C02.indd 60 2/12/2013 7:49:56 AM

2.8 For a fine-grained soil with a plasticity index of 15 to 40%, the degree of plasticity is
referred to as
(a) Non-plastic (b) Moderately plastic
(c) Plastic (d) Highly plastic
2.9 Identify the incorrect statement. A semi-log plot is used for grain-size distribution so that
(a) A wide range of grain size can be accommodated
(b) Equal emphasis can be given to all grain sizes
(c) Comparison can be made between two or more soils
(d) An S-type curve can be obtained
2.10 Assertion A: Uniformity of a soil is reflected by the grain-size distribution curve.
Reason R: Uniformity coefficients indicate gradations of grain sizes in a soil sample.
Select the correct code.
(a) Both A and R are true, and R is the correct explanation of A.
(b) Both A and R are true, and R is not the correct explanation of A.
(c) A is true, but R is false.
(d) A is false, but R is true.
2.11 What is a unit phase diagram? Explain with examples.
2.12 Two soils of similar mineralogy have extreme percentage of clay content. How might
the plasticity indices of the soil vary?
2.13 Give reasons for determining the grain-size distribution of a soil mass. Why is the
study generally confined to coarse-grained soils?
2.14 Two clayey silty sands have identical particle sizes with 20% fines. When exposed to
air, one dries out easily while the other does not. Why? Explain.
2.15 It is said that consistency index of a fine-grained soil and density index of a coarse-
grained soil are synonymous. Explain.
EXERCISE PROBLEMS
2.1 The following data are obtained from a pycnometer test of a soil sample:
Mass of pycnometer full of water = 2770.6 g.
Mass of pycnometer with soil and water = 2948.8 g.
Mass of moist soil = 315.5 g.
Specific gravity of soil solids = 2.67.
Find the water content of the soil.
2.2 A fully saturated clay has a moisture content of 42.4% and specific mass gravity (or
bulk density) of 1.78 g/cm3. Determine from first principles the void ratio and specific
gravity of the soil grains.
2.3 Derive an expression for water content from first principles in terms of the unit weight of dry
soil, the unit weight of water, the degree of saturation, and the specific gravity of soil solids.
M02_PURU1773_01_SE_C02.indd 61 2/12/2013 7:49:56 AM

2.4 The volume of soil taken from a field is 450 cm3. The weight of soil mass is 760 g,
and the dry weight is 620 g. Taking G = 2.7, find the (i) water content, (ii) void ratio,
(iii) porosity, (iv) degree of saturation, and (v) mass specific gravity.
2.5 In a research project on synthetic soils, a soil with dry unit weight γd is mixed with
organic matter of unit weight γ0 to have varied organic content Oc. Organic content
is defined as the ratio of the dry organic matter to the total dry weight of the sample.
Derive an expression for the unit weight of the synthetic soil in terms of γd, γ0, and Oc.
2.6 An undisturbed sample was extracted using a sampling tube of volume 1,200 cm3.
The mass of the clay sample and tube was 5.00 kg, and the same sample after oven
drying was 4.31 kg. The mass of the empty tube was 2.12 kg. Determine the water con-
tent, wet density, and dry density of the sample. The specific gravity of the soil solids
was found to be 2.69. Find the void ratio and degree of saturation of the clay.
2.7 The porosity of a soil sample is 35%, and specific gravity of its particle is 2.70. Calcu-
late its void ratio, dry density, saturated density, and submerged density.
2.8 A clayey soil has moisture content of 15.8%. The specific gravity is 2.72, and the sat-
uration percentage is 70.8%. The soil is allowed to absorb water. After some time, the
saturation increased to 90.8%. Find the water content in the latter case.
2.9 A dry soil sample of volume 280 cm3 weighs 450 g. Determine the water content at
100% saturation without any change in volume. What will be the water content when
the volume is allowed to increase by 12% of the original dry volume?
2.10 A 1,000-cm3 container was filled with a sand first in its loosest possible state and then
in its densest possible state, and the weight of the sand was 1,520 g and 1,830 g, respec-
tively. The sand, in situ, had a void ratio of 0.64. If the specific gravity of the sand
particles is 2.65, determine the limiting void ratios and the relative density in situ.
2.11 A soil sample has 80% of particles (by weight) finer than 0.1 mm, 7.5% finer than 0.01 mm,
and 4% finer than 0.001 mm. Draw the grain-size curve and determine the percentage of
total weight in each of the various size ranges, the effective size, and the uniformity coef-
ficient of the soil.
2.12 Draw the grain-size distribution curve for the soil with the following data:
Aperture size (mm) 4.76 2.38 2.0 0.85

Percentage passing 100 97 92 87
Aperture size (mm) 0.425 0.25 0.15 0.075
Percentage passing 53 42 15 8
Find the uniformity and curvature coefficients.

2.13 In a hydrometer analysis of a fine-grained soil, the initial reading was found to be
1.05. The corrected hydrometer reading after 70 minutes was 1.03, which corresponds
to an effective depth of 11.5 cm. The suspension volume was 1,000 cm3. Calculate the
initial weight of the soil, the particle size corresponding to the 12 minute reading, and
the percentage of particles finer than this size. Take G = 2.68 and η = 0.1 poise.
2.14 For a particle of diameter 0.005 mm, how many hours are required to settle to a depth
of 3 m from the surface in a tank? The specific gravity of the particle is 2.70 and the
coefficient of viscosity is 0.001 M-s/m2.
M02_PURU1773_01_SE_C02.indd 62 2/12/2013 7:49:56 AM

2.15 In a liquid limit test, a soil sample showed water contents of 48%, 40%, 38.8%, and
37.1% against 12, 26, 28, and 31 blows, respectively. The plastic limit of the clay is
18.2%, and the natural water content is 34.5%. Find the liquid limit, plasticity index,
liquidity index, relative consistency, flow index, and toughness index of the soil.
2.16 The liquid limit of a soil is 86%, and its plastic limit is 34%. If the natural water content
is 48%, what is the state of consistency of the soil? What is the shrinkage limit of the
soil if the void ratio at the shrinkage limit state is 0.89? Take G=2.68.
2.17 A saturated specimen was immersed in mercury, and its displaced volume was 20.8
cm3. The weight of the sample was 0.312 N. After oven drying for 48 hours, the weight
reduced to 0.196 N, while the volume came down to 10.2 cm3. Find the shrinkage
limit, void ratio, specific gravity, and shrinkage ratio of the soil.
2.18 In a big project, the Atterberg limits and natural water contents of three soils are
determined as given below:
Soil wL (%) wp (%) wn (%)

1 126 42 165
2 63 32 42
3 86 36 78
Determine the consistency of the natural soil and the liquidity indices.
2.19 An undisturbed saturated specimen of clay has a volume of 18.9 cm3 and a mass
of 30.2 g. In oven drying, the mass reduces to 18.0 g. Assuming the volume of dry
specimen to be 9.9 cm3, determine the shrinkage limit, shrinkage ratio, and volumetric
shrinkage.
2.20 The Atterberg limits for a clay soil used for an earth dam are liquid limit 60%, plastic
limit 40%, and shrinkage limit 25%. If a specimen of the soil of volume 10 cm3 at the
liquid limit has a volume of 6.5 cm3 when dried, what would be the specific gravity of
the soil particles?
M02_PURU1773_01_SE_C02.indd 63 2/12/2013 7:49:56 AM

M02_PURU1773_01_SE_C02.indd 64 2/12/2013 7:49:56 AM
3
Identification and Classification
of Soils
CHAPTER HIGHLIGHTS
Field identification of soils – Engineering classification of soils: Purpose
of classification, Unified soil classification system, Indian soil classifica-
tion system, AASHTO soil classification system, Textural soil classification
system
3.1 INTRODUCTION
It is necessary to have a standard language for a careful description and classification of
a soil. In principle, soil description is different from soil classification.
A soil description should include the material characteristics (viz., primary characteristics:
particle size distribution and plasticity; secondary characteristics: colour, shape, texture,
and composition) and the in situ soil mass (viz., firmness or strength, bedding planes,
discontinuities, weathering, etc.). On the other hand, soil classification is the arrangement
of soils into various groups or sub-groups so as to express briefly the primary material
characteristics (viz., particle size distribution and plasticity) without detailed descriptions.
Generally, soils have various constituents in different proportions. The soil is denoted
by the major constituent, and the minor constituents are indicated by adjectives. Further,
the colour and density or stiffness and moisture conditions are added to fully describe
the field condition of a soil, e.g., brownish red loose to medium dense silty sand.
In this chapter, field identification tests for soils and different engineering classifica-
tion of soils are presented.
3.2 FIELD IDENTIFICATION OF SOILS

Soils can be broadly grouped as coarse-grained or non-cohesive and fine-grained or
cohesive soils.
M03_PURU1773_01_SE_C03.indd 65 2/12/2013 7:48:51 AM

3.2.1 Coarse-grained Soils

Coarse-grained soils are mineral fragments which are easily identified by the naked eye on
the basis of grain size. The major coarse-grained materials are gravel and sand. Based on
experience, one can identify the sand or gravel as fine, medium, or coarse. Further, the grains
may be rounded, sub-rounded, angular, or sub-angular. Sometimes other mineral grains
such as mica or shale may also be present, which can be identified with the help of a mag-
nifying glass. In addition to the major classification as sand or gravel, the lesser significant
percentage of material should be identified. For example, a gravel with a significant percent-
age of sand has to be categorized as sandy gravel.
3.2.2 Fine-grained Soils

Fine-grained soils are silts and clays. Depending on the significant absence or presence of
organic material, they are categorized as inorganic soils or organic soils, respectively.
Inorganic Soils. Field identification of these soils can be made by conducting the following
tests:
1. Dry strength test
2. Dilatancy test
3. Plasticity test
4. Dispersion test
Dry Strength Test. The strength of a soil in a dry state is an indication of the presence of
cohesion. A pat of soil about 6 mm thick is dried under natural conditions or in an oven. The
dry strength can be estimated by breaking and crushing between the fingers. Dry inorganic
clay shows high strength and can be broken only with a great effort. On the other hand,
inorganic silts have little or no dry strength and crumble easily between the fingers.
Dilatancy Test. A pat of soil is made with water so that it is soft and not sticky. The pat is
placed in the open palm in a horizontal position. Several times, the side of the hand is struck
against the other hand. The appearance of a shiny film of water on the surface of the pat
signifies silt. As clay is less permeable, no significant change on the surface of the pat can be
seen after shaking.
Plasticity Test. A small quantity of soil is rolled into a thread form on a flat surface or on
the palm. If the soil can be rolled into a long thread of about 3 mm diameter, it signifies that
it contains a large quantity of clay, but silts cannot be rolled into a long thread of 3 mm diam-
eter without severe cracking.
Dispersion Test. A small quantity of soil is put into a jar of water, allowing the particles to
settle. Coarse-grained particles settle initially, followed by fine-grained particles. In a 10 cm
depth of water, sand particles settle within 30 seconds, whereas silt particles may take 15 to
20 minutes, but clay particles remain in suspension for several hours or even days provided
flocculation does not take place.
Organic Soils. Organic soils contain a significant proportion of dispersed vegetable mat-
ter. The organic matter in soil is due to disintegrated plant roots and other vegetable matter,
such as muck or more fibrous materials. Organic soils have a distinctive odour and often are
dark brown, dark grey, or bluish grey in colour.
Organic silts are less plastic, containing silt-size particles and finer particles of organic
material and shell fragments.
M03_PURU1773_01_SE_C03.indd 66 2/12/2013 7:48:51 AM

Identification and Classification of Soils 67
Organic clay contains clay-size particles and finely divided organic material. Highly
organic soils such as peat consists predominantly of plant remains, usually dark brown or
black in colour and with a distinctive odour.
3.3 ENGINEERING CLASSIFICATION OF SOILS

3.3.1 Purpose of Classification Systems
The simple way of classifying a soil as non-cohesive or cohesive is inadequate as it does not
specify other properties, such as gradation, grain sizes involved, plasticity, activity, and other
relevant properties which help to identify the soil for a specific construction purpose. Thus, the
aim of a classification system is to establish a set of conditions which will allow useful com-
parisons to be made between different soils. Soils classified to have a preference for one set of
conditions may not be preferred for another set of conditions. Thus, a number of classification
systems are available to cater to a particular purpose. Hence, to be of general use, a system
must be simple, lucid, and directly involved with the engineering properties of the soil.
3.3.2 Unified Soil Classification System

Unified soil classification system is the most popular soil classification system among civil
engineers. As in many of the systems, the grain-size characteristic has been used as the basis
for grouping the soil particles into gravel, silt, or clay, i.e.,
Gravel > 4.75 to 80 mm
Sand > 0.075 to 4.75 mm
Silt > 0.002 to 0.075 mm
Clay < 0.002 mm
Further, Atterberg limits are used as an additional criterion for identifying the compress-
ibility or plasticity of fine-grained soils. This system was first developed by A. Casagrande
(1948) as the Airfield Classification system. After minor modifications, it was adopted by the
US Bureau of Reclamation and US Corps of Engineers and later (1969) accepted as a stan-
dard classification system by the American Society for Testing Materials (Table 3.1).
Unified soil classification system divides soils into two major groups, viz., coarse-grained
soils and fine-grained soils, and is defined by a set of two letters, a prefix and a suffix. Coarse-
grained soils are those for which more than 50% of the material has particle sizes greater
than 0.075 mm. They are basically divided into gravels (G) and sands (S) and are further
grouped according to gradation and the presence of silt and clay-size fraction.
They are (i) well graded (W), (ii) poorly graded (P), (iii) containing silt fines (M), and (iv) con-
taining clay fines (C). For example, the symbol SP refers to poorly graded sand with no fines.
Fine-grained soils are those for which more than 50% of material has particle sizes
< 0.0075 mm. They are divided into inorganic silts and very fine sand (M), inorganic clays
(C), and organic silts and clays (O). They are further classified based on the liquid limit of
the soil as low plasticity, L (wL < 50%) and high plasticity, H (wL > 50%). Highly organic soils
(peat) are visually identified. The fine-grained soils are presented in a chart form called the
plasticity chart (Fig. 3.1) based on their liquid limit and plasticity index. The “A” line (after
A. Casagrande) separates the inorganic clays from the silts and organic soils.
M03_PURU1773_01_SE_C03.indd 67 2/12/2013 7:48:51 AM

Table 3.1 Unified soil classification including identification and description
- -
- -
a
Boundary classifications: Soils possessing characteristics of two groups are designated by
combinations of group symbols,
M03_PURU1773_01_SE_C03.indd 68 2/12/2013 7:48:53 AM

-
-
-
for example, GW-GC, well-graded gravel–sand mixture with clay binder.
M03_PURU1773_01_SE_C03.indd 69 2/12/2013 7:48:54 AM

Fig. 3.1 Plasticity chart (Unified soil classification) (Source: A. Casagrande, 1948)
3.3.3 Indian Soil Classification System

The Indian soil classification (IS: 1498, 1970) is basically the same as that of the Unified soil
classification system but for a slight modification in the plasticity chart. In this system, the
fine-grained soils, viz., inorganic silts, inorganic clays, and organic silts and clays, are further
divided into three groups based on the liquid limit of the soil as low compressibility, L (wL<
35%), medium compressibility, I (35% < wL < 50%), and high compressibility, H (wL > 50%).
Figure 3.2 represents the plasticity chart as adopted by the Indian soil classification system.
Highly organic soils (e.g., peat) are classified as Pt. Table 3.2 gives the details of the Indian
soil classification system.
Fig. 3.2 Plasticity chart (Indian soil classification) (Source: IS: 1498, 1970)
M03_PURU1773_01_SE_C03.indd 70 2/12/2013 7:48:55 AM

Table 3.2 Indian soil classification system (including field identification and description) (IS: 1498, 1970)
Division Sub-division Group Hatching Mapping Typical names Field identification procedures Information required for
letter colour (excluding particles larger than describing soils
symbol 80 mm and using fractions of
estimated weights)
M03_PURU1773_01_SE_C03.indd 71
-
-
Identification and Classification of Soils
-
71
Table 3.2 Contd.
2/12/2013 7:48:56 AM
Table 3.2 Contd. 72
Division Sub-division Group Hatching Mapping Typical names Field identification procedures Information required for
letter colour (excluding particles larger than describing soils
symbol 80 mm and using fractions of
estimated weights)
Soil Mechanics and Foundation Engineering
2/12/2013 7:48:57 AM
3.3.4 AASHTO Soil Classification System

The American Association of State Highway and Transport Officials (AASHTO) system was
developed by the US Bureau of Public Roads (now referred to as the Federal Highway
Administration) primarily based on the Public Road Administration classification system,
1978. The system is based on both the particle size and the plasticity characteristics.
According to the system, the soils are classified into eight groups, viz., A-1 to A-7 with an
additional group A-8 for peat or muck. Several sub-groups are included in the system. The
details of the sub-groups are presented in Table 3.3. Soils within each group are evaluated
according to the group index (GI) obtained from the empirical formula (Eq. 3.1)
GI = 0.2 a + 0.005 ac + 0.01 bd (3.1)
where a is that part of the percentage of soil particles passing the 75 μm sieve greater than 35
and not exceeding 75, expressed as a positive whole number (1 to 40); b is that part of the
percentage of soil particles passing the 75 μm sieve greater than 15 and not exceeding 55,
expressed as a positive whole number (1 to 40); c is that part of the liquid limit of the soil
greater than 40 and not greater than 60, expressed as a positive whole number (1 to 20); and
d is that part of the plasticity index greater than 10 and not exceeding 30, expressed as a
positive whole number (1 to 20).
For using the above equation and Table 3.3, the grain-size distribution, liquid limit, and
plasiticity index values of the soil are to be determined.
If the specific index value for a soil falls below the minimum limit corresponding to a, b, c,
or d, the value of the respective term is taken as 0 and the term is dropped out while calcu-
lating the GI. Similarly, when the value of a, b, c, or d is more than the prescribed maximum
value, then the respective value of 20 or 40 is assigned.
The classification is carried out by proceeding from left to right in the Table 3.3, and the
first group which fits the test data is selected. The GI value shows if a soil is fit as a sub-grade
material or not.
A group index of 0 indicates a good sub-grade material, while a group index of 20 cor-
responds to a very poor sub-grade material.
3.3.5 Textural Soil Classification System

In this system, soil fractions as per the US Bureau of Soils and Chemistry System are used.
Accordingly, the following is the grain-size classification:
Gravel > 1.00 mm
Sand 1.00 to 0.05 mm
Silt 0.05 to 0.005 mm
Clay < 0.005 mm
A triangular chart has been developed by the Bureau using grain-size limits. In addition
to gravel, sand, silt, and clay, the system uses another term, loam. A loam is a mixture of sand,
silt, and clay particles in varying proportions. The term loam has originated from agricultur-
ists and is also adopted by highway engineers as they too deal with surface layers.
As a first step, the grain-size distribution of the soil is found and the percent soil fractions
are determined. With the known percentages of sand, silt, and clay, a point is located in the
triangular chart, as shown in Fig. 3.3. The specified term designated in the chart for the area
where the point falls is taken as the classification of the soil.
M03_PURU1773_01_SE_C03.indd 73 2/12/2013 7:48:58 AM

74
Table 3.3 AASHTO soil classification system
General Granular materials Silt–clay materials
classification (35% or less passing 0.075 mm) (>35% passing 0.075 mm)
Group classification A-1 A-2 A-7

A-1a A-1b A-3 A-2-4 A-2-5 A-2-6 A-2-7 A-7-5,
A-4 A-5 A-6 A-7-6
Sieve analysis
% Passing 50 max
No. 10 (2 mm) 30 max 50 max 51 max
No. 40 (0.425 mm) 15 max 25 max 10 max 35 max 35 max 35 min 35 min 36 min 36 min 36 min 36 min
No. 200 (0.075 mm)
Characteristics of
fraction passing
No. 40 (0.425 mm)
Liquid limit 40 max 41 max 40 max 41 max 40 max 41 min 40 max 41 min
Plasticity index 6 max NP 10 max 10 max 11 max 11 max 10 max 10 max 11 max 11 min
Group index 0 0 0 0 4 max 4 max 8 max 12 max 16 max 20 max
Usual types of Stone fragments, Fine Silty or clayey gravel and sand Silty soils Clayey soils
significant constitu- gravel, and sand sand
ent materials
General rating as Excellent to good Fair to
sub-grade poor
Note: A-8, peat or muck is by visual classification and is not shown in the table. NP, non-plastic.
2/12/2013 7:48:58 AM
Fig. 3.3 Textural soil classification system
This is a simple classification system widely used in the fields of agriculture and highway
engineering. This classification depends on the grain-size distribution and does not reveal
any other property of the soil.
WORKED EXAMPLES
Example 3.1 A soil has the following characteristics:

1. Percentage of soil passing 75 μm sieve = 55.
2. Percentage of coarse fraction passing 4.75 mm sieve = 60.
M03_PURU1773_01_SE_C03.indd 75 2/12/2013 7:48:58 AM

3. Liquid limit = 68%.

4. Plastic limit = 22%.
Classify the given soil according to Indian Standards.
Solution
As more than 50% of the material contains particles greater than 0.075 mm, it is a coarse-
grained soil with fines.
The fines show a plasticity index of (68 – 22)% = 46%, i.e., with highly compressible clay
as per the plasticity chart.
Since 60% of the whole material is gravel with compressible clay fines, the soil may be
classified as clayey gravel, and the symbol is GC.
Example 3.2 Grain-size analysis and consistency tests conducted on an inorganic soil
revealed the following results:
Size of particle (mm) Percentage passing
0.75 32
The liquid limit is 41% and plastic limit 33%. Classify the soil as per the AASHTO system.
Solution
Percentage of soil particles less than 0.075 mm = 32.
As per Eq. 3.1,
GI = 0.2a + 0.005ac + 0.01bd
a = 32 – 35 = –3 = 0
b = 32 – 15 = 17
c = 41 – 40 = 1
d = 8 – 10 = – 1 = 0
GI = 0.02 × 0 + 0.005 × 0 × 1 + 0.01 × 17 × 0
=0
From Table 3.3, on the basis of percentage of fine-grained soil, liquid limit, and plasticity
index values, the soil is classified as A–2–5 (0).
Example 3.3 The sieve analysis of a sample of a soil gave the following details. Classify the
soil as per the Textural soil classification system.
Sand = 36%
Silt = 42%
Clay = 22%
Solution
Using the above values, the triangular chart is entered, and the soil is fit to be classified as
clay loam.
M03_PURU1773_01_SE_C03.indd 76 2/12/2013 7:48:58 AM

POINTS TO REMEMBER
3.1 Coarse-grained soils are mineral fragments which are easily identified in the field by
the naked eye on the basis of grain size. The major coarse-grained materials are gravel
and sand.
3.2 Fine-grained soils are silts and clays which are classified as inorganic or organic soils
depending on the amount of organic material present.
3.3 Field identification tests for fine-grained inorganic soils are dry strength test, dilat-
ancy test, plasticity test, and dispersion test.
3.4 Field identification of fine-grained organic soils can be made by the presence of dis-
integrated plant roots and other vegetable matter, a distinctive odour, and often dark
brown, dark grey, or bluish grey.
3.5 The aim of a classification system is to establish a set of conditions which will allow
useful comparisons to be made between soils.
3.6 In the Unified soil classification system, the grain-size characteristics have been used as
the basis for grouping the soil particles into gravel, silt, or clay. Further, Atterberg limits
are used as an additional criterion for identifying the compressibility or plasticity of
3.7 The Indian soil classification system is basically the same as the Unified soil classifica-
tion system but for a slight modification in the plasticity chart.
3.8 The AASHTO soil classification system was developed by the US Bureau of Public
Roads. The system is based on both the particle size and the plasticity characteristics.
According to the system, the soils are classified into eight groups, viz., A-1 to A-7,
with an additional group A-8, for peat or muck.
3.9 Group index (GI) is obtained from the empirical formula GI = 0.2 a + 0.005 ac + 0.01 bd,
where a and b depend on a certain percentage of particles and c and d depend on
the liquid limit and plasticity index. GI is adopted in the AASHTO soil classification
system.
3.10 In the Textural soil classification system, soil fractions as per the US Bureau of Soils
and Chemistry are adopted. This system provides a triangular chart to classify the
soil as sand, silt, clay, or loam. This system is widely followed by agriculturists and
highway engineers.
QUESTIONS
Objective Questions
3.1 Choose the correct statements
1. A gravel with a significant percentage of clay has to be categorized as gravelly clay.
2. In a dilatancy test, the appearance of a shiny film of water on the surface of the soil
part signifies silt.
M03_PURU1773_01_SE_C03.indd 77 2/12/2013 7:48:58 AM

3. Plasticity chart used in the Unified soil classification system is the same as in the
Indian soil classification system.
4. The AASHTO soil classification system is based on particle size, shape, and rough-
ness only.
3.2 The strength of a soil in the dry state is an indication of a high amount of ____:
(a) Sand (b) Silt (c) Clay (d) Gravel
3.3 Silt particle size as per Unified soil classification system is
(a) 0.075 to 4.75 mm (b) 0.002 to 0.075 mm
(c) > 4.75 mm (d) <0.002 mm
3.4 As per the plasticity chart of Indian soil classification system, the soil is of medium
compressibility when the liquid limit is
(a) < 35% (b) 35% to 50 %
(c) >50% (d) 25% to 34 %
3.5 A loam is a mixture of
(a) Gravel and sand (b) Sand and silt
(c) Sand, silt, and clay (d) Sand and clay
3.6 In the AASHTO soil classification system, the group classification A-4 to A-7 signifies
the sub-grade as
(a) Excellent to good (b) Good to fair
(c) Fair to poor (d) Poor to very poor
3.7 Explain briefly the object of classifying soils for engineering purposes.
3.8 Discuss the physical properties and factors which are considered in any particular
system of soil classification.
3.9 Explain the tests to be conducted to identify the soils in the field.
3.10 List different systems of engineering classification of soils. Discuss the merits and
demerits of each system.
3.11 How is the plasticity chart useful for classifying fine-grained soils?
3.12 Explain the Indian soil classification system.
3.13 What are the advantages in using a triangular chart?
3.14 How is suitability of sub-grade soils assessed by the AASHTO soil classification
system?
EXERCISE PROBLEMS
3.1 The following data were obtained from a laboratory test on a soil:
Percentage of particles finer than 4.75 mm = 100
Percentage of particles finer than 75 μm = 96.9
Coefficient of uniformity = 1.40
M03_PURU1773_01_SE_C03.indd 78 2/12/2013 7:48:58 AM

Coefficient of curvature = 1.03

Plasticity index = 6.23
Classify the soil as per Unified soil classification system.
3.2 Liquid and plastic limits determination on a soil gave the following results:
(Casagrande’s apparatus was used for the liquid limit test).
Test no. Mass of cup (g) Mass of cup + Mass of cup + Number
wet soil (g) dry soil (g) of blows
Liquid limit test

1 23.68 40.86 34.68 13
2 22.93 42.82 35.78 20
3 26.27 38.02 34.27 47
Plastic limit test
1 25.34 32.17 31.01 —
2 24.83 30.48 29.51 —
Determine the plasticity index and classify the soil as per BIS plasticity chart.
3.3 The sieve analysis of a soil revealed that 58% of the particles are finer than 75 μm.
The liquid limit and plastic limit of the soil were 61% and 27%, respectively. Classify the
soil as per the AASHTO system.
3.4 The following results were obtained from a laboratory test on three soil samples. Clas-
sify the soil as per Indian soil classification system.
Sieve size Percentage passing

Soil A Soil B Soil C
4.75 mm 68 98 100
2.00 mm 55 96 100
1.00 mm 43 90 95
450 μm 30 90 84
200 μm 25 88 78
150 μm 16 86 75
75 μm 10 84 70
Liquid limit NP 22.4% 32.8
Plastic limit NP 15.2% 24.3
3.5 The following test results were obtained on a soil sample:

Percentage passing 4.75 mm IS sieve = 98.5%
Percentage passing 75 μm IS sieve = 42.0%
Coefficient of uniformity = 6.7
Coefficient of curvature = 1.2
Plasticity index = 2.2
Classify the soil by a suitable classification system.
M03_PURU1773_01_SE_C03.indd 79 2/12/2013 7:48:58 AM

M03_PURU1773_01_SE_C03.indd 80 2/12/2013 7:48:58 AM
4
Compaction of Soils
CHAPTER HIGHLIGHTS
Principles of compaction – Compactive effort – Laboratory compaction:
Standard Proctor test, Modified Proctor test, BIS Light Compaction Test, BIS
Heavy Compaction Test – Field compaction and equipment – Compaction
specification and control – Factors affecting compaction – Compaction of sand
4.1 INTRODUCTION
Compaction may be defined as the process of increasing the density of a soil using force
to pack the particles closer together, with a reduction in air voids without any significant
change in the volume of water in the soil.
The reduction in air voids is deliberately brought about by some mechanical means
during a construction process in the field or in the preparation of a sample in the labo-
ratory. The higher the compaction, the lower will be the compressibility of the soil and
higher the shear strength.
Typical examples are the construction of fills, embankments, and earth dams and
strengthening of sub-grades of highways and runways.
4.2 PRINCIPLES OF COMPACTION

Soil compaction is the process whereby soil particles are forced to pack more closely
together by reducing air voids. This is attained by applying some mechanical force
(static or dynamic loads) on the soil. The purpose of compaction is to produce a soil
having certain physical properties suitable for a given project. The state of compaction
of a soil is measured by the dry density and the associated moulding water content.
The increase in the dry density of the soil produced by compaction basically depends
on the water content of the soil and the applied energy. For each soil, there exists a moisture
M04_PURU1773_01_SE_C04.indd 81 2/12/2013 7:48:25 AM

content termed the optimum moisture content at which a maximum dry density is attained for a
given amount of compaction. Addition of water to dry soil results in adsorbed water around
particles. When the water content is low, the soil is stiff and difficult to compress, and this
results in a low dry density. As the adsorbed water films increase in thickness and act as a
lubricant and bring the particles more closely together, they increase the dry density by reduc-
ing the air content. After a certain point, the effect of lubrication stops and the adsorbed water
pushes the particles away, and any further increase in moisture decreases the dry density.
Thus the maximum dry density occurs at an optimum moisture content.
4.3 COMPACTIVE EFFORT

The concept of compactive effort is used in both field and laboratory compactions. For labo-
ratory conditions, compactive effort is defined as the application of a given amount of energy
per unit volume of compacted soil. For field conditions, compactive effort is defined as the
compaction obtained by allowing a piece of equipment to pass a given number of times on
a given thickness of lift. Compactive effort can be varied in the laboratory tests by changing
the weight of the compacting hammer, height of fall, number of blows per layer, and number
of layers. If a vibratory method is used, the compactive effort can be changed by changing
the frequency, amplitude, and time of vibration. In the field, the compactive effort can be
increased by increasing the number of passes of a roller. For all soils, both in the laboratory
and in field compaction, an increase in compactive effort results in an increase in the dry
density and decrease in the optimum moisture content.
4.4 LABORATORY COMPACTION

Laboratory compaction tests are designed to estimate the dry density of soils. Two such tests
were developed by Proctor (1933): (i) the Standard Proctor test (IS: 2720 – Part 7, 1974), which
causes adequate compaction for most applications, such as backfills, highway fills, and earth
dams, and (ii) the Modified Proctor test (IS: 2720 – Part 8, 1983), which is used for heavier-
load applications, such as airport and highway base courses. Procedures for conducting
laboratory tests are explained in Chapter 10.
The tests are performed by compacting a wet soil sample in a mould in a specified number
of layers. Each layer is compacted at a stipulated compactive effort. The compactive effort is
measured in terms of the energy per unit volume of compacted soil. The required compac-
tive effort is attained by controlling the weight of the hammer, height of drop, number of lay-
ers, and number of blows for each layer. After the final layer is compacted, the bulk density
of the soil and its moisture content are determined. Tests are repeated on fresh samples with
increasing moisture content. From the values of the bulk density and the moisture content
obtained, the dry density is calculated. Thus,
ρd = ρ/(1 + w)
A graph of dry density versus moisture content is plotted (Fig. 4.1) and the maximum dry
density and optimum moisture content are read from the graph.
It is not feasible to expel air completely by compaction and obtain the maximum dry den-
sity. The maximum theoretical value of the dry density is referred to as the zero air void–dry
density or the saturation dry density and can be computed from
M04_PURU1773_01_SE_C04.indd 82 2/12/2013 7:48:25 AM

Compaction of Soils 83
Fig. 4.1 Dry density–moisture content relationship
G ρw
ρd =
1+ wG
and the dry density for any particular degree of saturation can be computed from
G(1 − Av )ρw
ρd =
1 + wG
Theoretical curves for 0%, 5%, and 10% air voids are shown in Fig. 4.1.
4.4.1 Standard Proctor Test

In connection with the construction of earth dams, a standard compaction test was developed
by Proctor (1933).
The test apparatus consists of
1. A cylindrical mould of internal diameter 102 mm and an effective height of 177 mm, with
a volume of 0.945 litres
2. A detachable collar extension of the mould (used during compaction)
3. A detachable base plate
4. A 50 mm diameter metal rammer of weight 2.5 kg with a height of 300 mm moving in a
metallic sleeve
The soil is compacted in three equal layers with 25 blows on each layer, and the energy
transmitted to the soil, i.e., the compactive effort, is about 60.50 kg-m per 1,000 cm3 of soil.
About 3 kg of air-dried soil is used. An initial water content of 4% is added for coarse-
grained soils, and 10% for fine-grained soils. Other details of the test procedure are given in
Chapter 10.
This test is also adopted as the standard test by the American Association of State
Highway Officials (AASHO) and is usually called the AASHO test. The moisture content–
density relationship is obtained as discussed earlier.
M04_PURU1773_01_SE_C04.indd 83 2/12/2013 7:48:26 AM

4.4.2 Modified Proctor Test

The test apparatus consists of
1. A cylindrical mould of about 102 mm diameter with an effective height of 177 mm
2. A suitable detachable collar and a base
3. A rammer of weight 4.5 kg with a height of fall of 450 mm.
The mould is filled in five layers, each layer being compacted with 25 blows. The compac-
tive effort imparted is about 272.60 kg-m per 1,000 cm3, which is about 4.5 times that of the
Standard Proctor test.
The moisture content–dry density curve is obtained as discussed earlier. Because of the
higher compactive effort, the dry density will be high. This test procedure has emerged to
meet the heavier compaction requirements of airfield and airport pavements.
4.4.3 Indian Standard Compaction Tests

The Indian Standard equivalent of the Standard Proctor test is called the light compaction test
(IS: 2720 – Part 7, 1974), and the Indian Standard equivalent of the Modified Proctor test is
called the heavy compaction test (IS: 2720 – Part 8, 1974).
Light and heavy compaction details are given below:
Type of Weight of Height of No. of layers No. of blows on

Compaction rammer (kg) fall (mm) each layer
Light 2.6 310 3 25

Heavy 4.89 450 5 25
The other test procedure is the same, and the moisture content–dry density relationship
is obtained as discussed earlier.
It may be observed that the Indian compaction test is not significantly different from the
Proctor tests.
4.5 FIELD COMPACTION AND EQUIPMENT

Classification of soil layers using compacting equipment is referred to as field compaction. The
field compaction process involves one or more of the following (Lambe and Whitman,
1979): (i) selection of soil from borrow areas, (ii) transfer of soil from a borrow area to a
construction site, (iii) spreading of the soil to a suitable thickness, (iv) alteration of the
moisture content of the soil, (v) breaking of lumps and making the soil uniform, and (vi)
rolling the soil by adopting a specified procedure to attain the property required. Different
types of equipment are available commercially for following the above steps. The choice of
equipment will depend on the type of soil and economic considerations.
Compaction equipment consists of excavating and hauling equipment, rock separa-
tion equipment, spreading equipment such as bulldozers and graders, discs, harrows and
watering equipment, rollers, and special compacting equipment. Out of the above, the main
compaction equipment is the rollers, which are discussed below.
M04_PURU1773_01_SE_C04.indd 84 2/12/2013 7:48:26 AM

4.5.1 Smooth Wheel Rollers

Smooth wheel rollers consist of hollow steel drums. The mass of the main roller can be
increased using water or sand ballast. These are suitable for proof rolling of sub-grades and
for finishing the construction of fills with sandy or clayey soils. They can be successfully
used in places where a mixing or kneading action is not required. They provide excellent
coverage, and the contact pressure can be as high as 300 to 400 kN/m2. Smooth wheel rollers
can be either towed or self-propelled.
4.5.2 Pneumatic-Tyred Rollers

In these rollers, wheels are placed close together on two axles and placed such that the rear
set of wheels overlap the lines of the front set to ensure complete coverage of the soil surface.
Wide tyres with flat treads are provided so that the soil is not displaced laterally. The action
produced by these rollers is somewhat better than that of smooth wheel rollers in that they
produce a combination of pressure and kneading on the soil. Rubber-tyred rollers are effec-
tive for a wide range of soils from clean sand to silty clay.
4.5.3 Sheep’s Foot Rollers

These rollers consist basically of drums with numerous club-shaped tapered projections.
The mass of the drum can be varied by adding ballast. The area of each projection may be
4,000 to 6,500 mm2. The projections or feet penetrate into the layer during the rolling
operations. During compaction, the initial passes compact the lower portion of a lift.
In successive passes, compaction is obtained in the middle and the top sections of the
layer. For effective rolling, the lift thickness should be small and the contact pressure
under the projections very high, of the order of 1,500 to 7,500 kN/m2. These rollers are
most suitable for plastic and non-plastic fine-grained soils. Although not suitable for
clean granular soils, they may be used in such soils too if more than 20% fines are present.
Due to the excellent bonding caused by the kneading effect of the sheep’s foot, these roll-
ers are generally recommended for water-retaining earthworks.
4.5.4 Vibratory Rollers

Both smooth wheel and rubber-tyred rollers can be modified so that they impart an
impacting motion to the soil being compacted. A power-driven vibration mechanism is
provided to reasonably match the resonant frequency of the soil type and layer thick-
ness. These rollers are by far the predominant type used in the compaction of granular
soils.
4.5.5 Grid Rollers

These are intermediate between smooth wheel and sheep’s foot rollers, with their rotating
wheels, made of a network of steel bars, forming a grid with square holes. These rollers pro-
vide less kneading action but high contact pressures. These are most suitable for coarse-
grained soils.
M04_PURU1773_01_SE_C04.indd 85 2/12/2013 7:48:26 AM

4.5.6 Tampers and Rammers

Hand-held tampers or rammers operated by compressed air or gasoline power are com-
monly used to compact small areas to which access is difficult. They are also used for
compacting backfills in trenches.
4.5.7 Vibrating Plates

Manually controlled vibrating plates operate efficiently on clean granular soils. Satisfactory
compaction can also be achieved in other types of soils. Vibrating plates are also gang-
mounted on machines for use in less-restricted areas.
Sand or sand–gravel mixtures with silt (without clay) show good compaction characteris-
tics, and it is generally recommended to use vibratory drum rollers, vibratory rubber tyres,
or pneumatic tyre equipments for compaction of such soils. If some clay fraction is pres-
ent in sand or sand–gravel mixtures, vibratory sheep’s foot–type equipment can be used.
Fine-grained silt and clay show a varied compaction performance depending on the plas-
ticity of the material, and such soils can be compacted more effectively by using pneumatic
tyres, vibratory rubber tyres, or vibratory sheep’s foot–type equipment. Organic soils are
not recommended for structural earthfills.
Field compaction depends on several factors, such as the soil type, moisture content, lift
thickness, number of passes, and speed and type of compactor. The control of field compac-
tion is discussed in the next section.
Table 4.1 gives the compaction performance and recommended compaction equipment
(McCarthy, 1982).
Table 4.1 Compaction performance and recommended equipment
General soil description Unified soil Compaction Recommended compaction

classification characteristics equipment
Saud or sand–gravel SW, SP, GW, GP Good Vibratory drum roller, vibratory
mixtures (no silt or clay) rubber tyre, pneumatic tyre
Sand or sand–gravel SM, GM Good Vibratory drum roller, vibratory
with silt rubber tyre, pneumatic tyre
Sand or sand–gravel SC, GC Good to fair Pneumatic tyre, vibratory rubber tyre,
with clay vibratory sheep’s foot
{
Silt ML Good to poor Pneumatic tyre, vibratory rubber tyre,
vibratory sheep’s foot
MH Fair to poor Pneumatic tyre, vibratory rubber tyre,
vibratory sheep’s foot, sheep’s foot
Clay
{ CL
CH
Good to fair
Fair to poor
Pneumatic tyre, sheep’s foot, vibratory
Sheep’s foot and rubber tyre
Organic soil OL, OH, PT Not recommended for structural earthfills
Source: McCarthy (1982).
M04_PURU1773_01_SE_C04.indd 86 2/12/2013 7:48:26 AM

4.6 COMPACTION SPECIFICATION AND CONTROL

4.6.1 Specification
Two methods have been used (Foster, 1962) to specify the compaction requirements in the
various compancted layers. In one type called the performance-type specification, the com-
paction requirement is stated in terms of the physical properties of the compacted layer. A
typical specification of this type is the percentage of maximum dry density obtained in a
standard compaction test. Sometimes both the dry density and the moisture content are
specified. For example, a specification of 95% maximum dry density is frequently desig-
nated. Specification in terms of percentage of the maximum dry density in a standard test
is preferable to a stated value of the dry density because it ensures that a compactive effort
comparable to that of the standard laboratory test is applied in the field. This type of
specification is applicable for cohesive soils. Other physical properties that are used are
the percentage of air voids at a specific moisture content, void ratio, and relative density.
The last two are used primarily for non-cohesive materials. In performance-type specifica-
tion, the contractor is given a wide scope in the selection of equipment and lift thickness.
In the other type, generally called the work-type specification, the type of equipment, the
lift thickness, the moisture content, and the amount of work required to obtain the necessary
density are specified.
Performance-type specification has found widespread usage in highway and airfield pave-
ment work, whereas work-type specification has been used more for dam and levee work.
4.6.2 Field Control

It is the responsibility of the field engineer to check the density and water content during the
process of rolling each layer. The only way this can be accomplished is by taking soil samples
for moisture and density determination. Density readings are taken, usually a designated num-
ber, for every lift or for a specified volume of fill placed. It is common to take such readings for
each layer for every 500 to 1,000 m2, depending on the importance of the site. During the initial
stages of the work itself, there should be enough technicians and equipment to conduct the test.
The most important aspect of construction control is the speed with which the moisture
and density are determined when the soil is under rolling for rectification. Conventional
methods of measuring the moisture content are slow. Rapid methods of determining the
moisture content involve the use of a frying pan, a hot plate, a forced-draft, radiant-heat
oven, a Proctor needle, and a nuclear moisture gauge. The latter two methods are important
and are described below.
A Proctor needle (Fig. 4.2) consists of a spring-loaded plunger with a calibrated stem and
a needle. The calibration is made in kg/cm2 so that the penetration resistance can be read
directly. The needle is provided with various bearing points to measure a wide range of pen-
etration resistance. A laboratory “penetration resistance curve” of the field soil is needed to
use the Proctor needle. A routine laboratory compaction test is performed on the soil. Before
determining the wet density for each moisture content, the penetrometer (with a known bear-
ing area) is pushed into the wet soil with a uniform push, up to a depth of 7.5 cm, and the
penetration resistance versus moisture content along with dry density is plotted (Fig. 4.3). To
M04_PURU1773_01_SE_C04.indd 87 2/12/2013 7:48:26 AM

Fig. 4.2 Proctor needle
determine the moisture content in the field, the wet soil from the rolling yard is compacted
into a mould under conditions similar to laboratory conditions. For this compacted soil, using
the Proctor needle, the penetration resistance is read off. Then, from the laboratory curve, the
moisture content corresponding to this penetration resistance is obtained. This method is suf-
ficiently fast and accurate for fine-grained soils.
The nuclear moisture gauge is a modern instrument which is rapid and gives precise
results. When using this, a source of fast neutrons is placed in the soil, and the neutrons
move randomly and collide with atoms in the soil and rebound as slow neutrons. A coun-
ter is provided to record the counts of these slow neutrons. The quantity of hydrogen
atoms in a soil is due to the presence of water, and hence such a count may be used to indi-
cate the amount of water at the location of the source of fast neutrons. A radium–beryllium
mixture is commonly used as the source of fast neutrons.
The wet density of the compacted soil in the field is determined using a core-cutter
method or sand replacement method.
The Bureau of Indian Standards (IS: 10379, 1982) recommends three methods for
non-gravelly soils and one method for soils containing gravels and rockfills. As per the first
method for non-gravelly soils, the in-place density is determined using conventional meth-
ods such as the sand replacement method, core-cutter method, or rubber balloon method
(IS: 2720 – Part 34, 1972). The moisture content is obtained using any of the rapid methods of
Fig. 4.3 Penetration resistance and dry denstiy/moisture content curves
M04_PURU1773_01_SE_C04.indd 88 2/12/2013 7:48:26 AM

water content determination such as the calcium carbide method, torsion balance method,
sand bath method, or alcohol method (IS: 2720 – Part 2, 1973). Further, it is recommended
that control tests be performed after removing the top 5 cm layer of earth. In the second
method, Hilf’s method for compaction control may be adopted (IS: 2720 – Part 38, 1977). This
method suggests a relationship between the field moisture content, dry density, and labora-
tory optimum conditions determined without measurement of the water content. The third
method is intended for certain weathered soils. For such soils, a test embankment under
identical field conditions is used to determine the field moisture content and dry density.
In soils containing gravels and rockfills, the total density of the soil increases and the moisture
content decreases with increasing gravel per cent up to 60% to 75%, beyond which the density
again decreases. For soils with 30% gravel, the use of conventional light or heavy compaction
methods (IS: 2720 – Parts 7 and 8, 1974, 1983) on the soil fraction passing a 40 mm IS sieve is rec-
ommended. The field density may preferably be determined using the ring and water replace-
ment method (IS: 2720 – Part 33, 1971) or alternatively using the sand replacement method.
4.7 FACTORS AFFECTING COMPACTION

The degree of compaction of a soil is measured in terms of the dry density, which is the mass
of soil solids per unit volume of the soil. The degree of compaction contributes to the shear
strength, permeability, compressibility, and sustainability for repeated loads.
The major factors which affect compaction are (i) the moisture content, (ii) the compactive
effect, (iii) the type of soil, and (iv) the method of compaction.
4.7.1 Effect of Moisture Content

As explained earlier, at lower levels of moisture content, the soil particles offer more resis-
tance to compaction and the soil behaves like a stiff material. Increasing the moisture con-
tent helps the particles to move closer because of the lubrication effect. On further
increasing the moisture content beyond a certain limit, the water starts to replace the soil
particles. Thus, the dry density increases up to a limiting moisture content (optimum
moisture content), beyond which an increase in the moisture content decreases the dry
density. The effect of the formation of a structure with increasing moisture content is
another meaning given for the increase in the dry density and the subsequent decrease
beyond a certain limit (Lambe, 1958).
4.7.2 Effect of Compactive Effort

The maximum dry density and the optimum moisture content are both affected by a change in
the compactive effort. An increase in the compactive effort increases the maximum dry density
but decreases the optimum moisture content. However, the air void ratio at the peak density
remains approximately the same (reflected by the constant air content line passing through the
peak points). Further, it can be seen (Fig. 4.4) that there is only a marginal increase in the density
with an increase in the compactive effort. It reflects the fact that only a very small improvement
in the dry density results from the use of heavier equipment. Heavy equipment is generally
preferred for economic reasons as it can produce the required compactive effort more cheaply.
M04_PURU1773_01_SE_C04.indd 89 2/12/2013 7:48:26 AM

H
(rd)max
High compactive
L
(rd)max effort
E
B
Compacted density
D
C
A
Low
(OMC)H (OMC)L compactive
effort
Moulding water content
Fig. 4.4 Effect of compactive effort (Source: Lambe, 1962)
4.7.3 Effect of Type of Soil

Well-graded coarse-grained soils with smooth rounded particles show a high dry density,
whereas uniform sands have a low maximum dry density (Fig. 4.5). Clayey soils have lower
dry densities and higher optimum moisture contents than do sands. The effect of increasing
the compactive effort is also more in clayey soils.
Figure 4.6 shows typical curves for different soils at the same compactive effort.
4.7.4 Effect of Method of Compaction

It is ideal to develop a laboratory test which could produce a reasonable moisture–density
curve so as to assess the maximum dry density and optimum moisture content. As the pro-
cesses of imparting energy to the soil are different in the field and laboratory, there may be dif-
ferent degrees of compaction depending on the method of compaction. Field compaction is
essentially a rolling or kneading type of compaction, whereas the laboratory compaction is
Fig. 4.5 Dry density versus moisture content for two grades of sands
M04_PURU1773_01_SE_C04.indd 90 2/12/2013 7:48:27 AM

Fig. 4.6 Typical curves for different soils at the same compactive effort
of the dynamic-impact type. Dry unit weight–moisture curves (Turnbull, 1950) for the
same soil subjected to different methods of compaction are given in Fig. 4.7. It can be seen
that Curves 3 to 6 approximately give the same maximum dry density although the meth-
ods of compaction are different. Further, it may be reasoned out that the Standard AASHO
test is the best fit to simulate the field moisture–unit weight relationship.
19
Laboratory
static compaction
1 (13.8 MN/m2)
S
18
=
10
0%
Dry unit weight (kN/m3)
2
17
Modified AASHO
16 Standard AASHO
3 Field compaction (sheeps-foot)

6
5
15 Field compaction (rubber–tired)
4
Laboratory
static compaction
(1.38 MN/m2)
14
10 15 20 25
Moisture content (%)
Fig. 4.7 Dry unit weight–moisture content curves for different methods of compaction
(Source: Lambe and Whitman, 1979)
M04_PURU1773_01_SE_C04.indd 91 2/12/2013 7:48:27 AM

4.8 EFFECT OF COMPACTION ON SOIL STRUCTURE

For a given compactive effort, the soil structure varies with the moulding moisture content (Fig.
4.4). At a low moisture content, the repulsive forces between the particles are weaker than the
attractive forces, and hence the soil structure is flocculated for compaction on the dry side (Point
A). Increasing the moisture content increases the repulsive forces, permitting the particles to ori-
ent in a more orderly array (Lambe, 1962). The degree of orientation becomes such that the dry
density is maximum at the optimum moisture content (Point B). Beyond this point, the degree of
particle orientation further improves, leading to a particle parallelism, resulting in a dispersive
soil structure (Point C). Thus, the compacted soil tends to be more flocculated on the dry side of
the optimum moisture content than on the wet side. Further, an increase in the compactive effort
tends to disperse the soil for a given moisture content (Fig. 4.4 – Points A and E and C and D).
The soil structure discussed above shows a similarity between compacted soils (dry side
and wet side of optimum) and undisturbed and remoulded soils. Undisturbed soils and
compacted soils (at dry of optimum) both show a flocculated structure, whereas remoulded
soils and compacted soils (at wet of optimum) both show a dispersed structure.
The effects of compaction, dry of optimum (Point A) and wet of optimum (Point C),
on several engineering properties are listed in Table 4.2 (Lambe, 1958).
Table 4.2 Comparison of properties of cohesive soil on dry and wet sides of OMC
Property Comparison
Structure
(i) Particle arrangement Dry side more randomly oriented (flocculated)
(ii) Water deficiency Dry side more deficient, takes more water, swells more;
low pore water pressure
(iii) Permanence Dry side susceptible to change
Permeability
(i) Quantity Dry side more permeable
(ii) Permanence Dry side permeability may decrease
Compressibility
(i) Quantity Wet side more compressible in low-pressure range
Dry side compressible only in high-pressure range
(ii) Rate Dry side rapidly compressible
Strength
As moulded
(i) Undrained Dry side very high
(ii) Drained Dry side somewhat high
At saturation
(i) Undrained Dry side somewhat higher if swelling prevented
Wet side can be higher if swelling is permitted
(ii) Drained Dry side almost the same or slightly higher
Pore pressure at failure Wet side higher
Stress–strain modulus Dry side much greater
Sensitivity Dry side more sensitive
Source: Lambe (1958).
M04_PURU1773_01_SE_C04.indd 92 2/12/2013 7:48:27 AM

Fig. 4.8 Moisture–density relationship for sand
4.9 COMPACTION BEHAVIOUR OF SAND

The moisture–density curve (inverted V shaped) discussed so far is applicable to soils pos-
sessing some value of plasticity. In non-plastic soils like sands, the moisture–density curve is
different.
The moisture–density relationship, as obtained from a laboratory test, on a cohesionless
sand is shown in Fig. 4.8. The curve shows that a thin film of water around the grains keeps
the particles away due to surface tension and causes a decrease in density in the initial
stages. This phenomenon is referred to as bulking of sand, which occurs at a moisture content
of 4% to 5%.
With increase in moisture content, the menisci are broken and the particles come closer,
leading to an increase in the dry density. This density increases till the saturation is 100%,
beyond which the density decreases due to occupation by water of the positions of grains.
An increase in the compactive effort in sand causes no significant change compared to
cohensive soils. The above moisture–dry density relationship is of less practical importance.
Usually, cohesionless soils are densified by vibration in the dry or fully saturated condi-
tion or by simply flooding the area to be compacted. The degree of compaction is measured
by the relative density or void ratio.
4.10 CALIFORNIA BEARING RATIO TEST

The California Bearing Ratio (CBR) test for the design of flexible pavements was developed
by the California Division of Highways. The basic procedure of this test was adopted by the
Corps of Engineers of the US Army. Certain modifications were made in the test procedure,
and now the modified method is adopted by the Corps of Engineers and regarded as
the standard method of determining the CBR. The Bureau of Indian Standards (IS: 2720 –
Part 16, 1987) has also adopted the modified procedure.
The Corps of Engineers have developed design curves using CBR values for determining
the required thickness of flexible pavements for airport runways and taxiways.
The detailed laboratory test procedure is explained in Chapter 10 for a remoulded soil.
However, this penetration test can also be performed on undisturbed samples. A field CBR
test is also available (IS: 2720 – Part 31, 1990) for finding the CBR of existing sub-grades.
M04_PURU1773_01_SE_C04.indd 93 2/12/2013 7:48:28 AM

WORKED EXAMPLES
Example 4.1 A laboratory compaction test conducted on a sample of soil gave the following
results:
Bulk density (g/cm3) Moisture content (%)
2.06 12.85
2.13 14.28
2.15 15.65
2.16 16.86
2.14 17.89
1. Find the maximum dry density and optimum moisture content

2. Plot the zero-air void line and 5% air void line. The specific gravity of soil is 2.72.
Solution
The dry density ρd = ρ/(1 + w),

Dry density values are calculated from the above formula as
2.06 2.13 2.15
= 1.82 g / cm3 ; = 1.85 g / cm3 ; = 1.86 g / cm3 ;
1 + 0.129 1 + 0.143 1 + 0.157
2.16 2.14
= 1.84 g / cm3 ; = 1.81 g / cm3
1 + 0.169 1 + 0.179
The moisture–dry density curve is plotted as given in Fig. 4.9. The maximum dry density
and optimum moisture content (OMC) are read as
Max. ρd =1.87 g/cm3
OMC =14.9%
The dry density in terms of air voids is given as
G ρw (1 −V a )
ρd =
1+mG
Using this expression, the ρd values for zero-air void (i.e., Va = 0) and 5% air void
(i.e., Va = 0.05) are obtained as
w (%) 14 15 16 17 18
Va = 0 1.98 1.94 1.90 1.87 1.83
Va = 5% 1.88 1.84 1.80 1.78 1.74
These lines are plotted as shown in Fig. 4.9.
Example 4.2 As per the compaction specification, a highway fill has to be compacted to
95% of Standard Proctor Compaction test density. A borrow area available near the project
site has a dry density of 1.65 g/cm3 at 100% compaction and a natural void ratio of 0.61.
The specific gravity of the soil solids is 2.65. Compute the volume of borrow material needed
M04_PURU1773_01_SE_C04.indd 94 2/12/2013 7:48:28 AM

Zero air
voids line
5% air
1.90 voids line
Dry density, rd (g/cc)
1.865
1.80
14.9
12 14 16 18
Moisture content (%)
Fig. 4.9 Compaction curve and zero and 5% air void lines.
to construct a highway fill of heigh 5 m and length 1 km with side slopes of 1:1.5. The top
width of the fill is 8 m.
Solution
Top width of highway fill =8m

Base width of highway fill = 2 (1.5 × 5) + 8 = 23 m
Volume of highway fill = ½(8 + 23) × 5 × 1 × 1000
= 77,500 m3
Dry density of soil at borrow = 1.65 g/cm3
Void ratio at borrow eb = 0.61
Dry density of highway fill = 0.95 × 1.65 = 1.57 g/cm3
2.65
Void ratio at fill ef = − 1 = 0.69
1.57
(1 + e b )
Volume of soil from borrow Vb = Vf
(1 + ef )
⎛ 1 + 0.61 ⎞⎟
= 77 , 500 ⎜⎜ = 73 , 832 m 3
⎜⎝ 1 + 0.69 ⎟⎟⎠
Example 4.3 The undisturbed soil at a given borrow pit is found to have a water content
of 16.8%, a void ratio of 0.62 and G of 2.70. The soil from the borrow pit is to be used to
construct a rolled fill having a finished volume of 4,800 m3. The soil is excavated and
dumped in trucks. In the construction process, the trucks dump their loads on the fill, and
the material is spread and broken up after which water is added until the water content
M04_PURU1773_01_SE_C04.indd 95 2/12/2013 7:48:30 AM

is 18.2%. The soil and water are thoroughly mixed and compacted until the wet density is
1.85 Mg/m3.
1. How many truck loads of soils were transferred if each truck load is 15 m3? Assume the
void ratios of the excavated soil and the soil loaded in the truck are the same.
2. If the fill should become saturated at some time subsequent to construction and does not
change volume appreciably, what will the saturation moisture content be?
3. What will the saturation moisture content be if the soil swells to increase its original
volume by 15.8%?
Solution
1.85
Dry density of fill ρd = = 1.57 g /cm 3
1 + 0.182
2.70
Void ratio in fill ef = − 1 = 0.72
1.57
⎛ 1 + e b ⎞⎟
Volume of soil taken from borrow Vb = ⎜⎜ ⎟
⎜⎝ 1 + e ⎟⎠⎟Vf
f
1 + 0.62
= × 48000 = 45, 209 m 3
1 + 0.72
45209
1. Number of truck loads required = = 3 , 014
15
eS 0.72×100
2. Saturation moisture content, w = r = = 26.67%
G 2.70
3. Increased volume = 48000 (1 + 0.158) = 55,584 m3
Vf
Void ratio in the swelled condition = (1 + e b ) − 1
Vb
55584
= (1 + 0.62 ) − 1 = 0.99
45209
eSr 0.99×100
4. Saturation moisture content, w = = = 36.73%
G 2.70
Example 4.4 A sub grade soil of G = 2.67 and dry density 1.53 Mg/m3 is available. With this
soil an aggregate of the same specific gravity in a proportion of 75% of aggregate to 25% soil
is mixed. The mixture is then compacted to a dry density of 1.84 Mg/m3. At 100% saturation,
the aggregate has a moisture content of 3%. What is the saturation moisture content for the
soil in the compacted mixture?
Solution
Consider 1 m3 of soil mixture.

Of this 25% is soil, i.e., 0.25 × 1.53 = 0.383 Mg
The balance is aggregate = 1.84 − 0.383 = 1.457 Mg
M04_PURU1773_01_SE_C04.indd 96 2/12/2013 7:48:32 AM

0.383
Vs of soil grains = = 0.143 m 3
2.67 ×1
1.457
Vs of aggregate = = 0.546 m 3
2.67 ×1
Soil grains fill the voids formed by the aggregate, and the balance volume of the voids is
filled by soil grains = 1 − 0.546 − 0.143 = 0.311 m3.
0.111
Void ratio e = = 0.45
0.546 + 0.143
0.45×1
At saturation of the compacted mixture, w = ×100 = 16.91%
2.67
Of the water, 3% is for aggregate saturation.
Therefore, the total water content at 100% saturation of mixture = 19.91%
Example 4.5 Some soil has been dumped loosely from a scraper. It has a unit weight of
16 kN/m3, a water content of 10.5%, and a specific gravity of solids of 2.68. Find the void
ratio, porosity, density, and unit weight of the soil in the loose condition.
To make the compaction process more workable, an optimum water content of 15% is
necessary. How much of water should be added in litres per cubic metre of soil to raise the
water content to the optimum?
The soil is compacted with the optimum water content until it is 95% saturated. Find the
new void ratio, porosity, dry density, and dry unit weight.
Solution
1. In loose condition
16
Dry unit weight γd = = 14.40 kN / m3
1 + 0.105
Gγ w 2.68 × 9.81
Void ratio e = −1 = − 1 = 0.826
γd 14.40
γd 14.40
Dry density ρd = = = 1.468 g / cm3
g 9.81
e 0.826 ×100
Porosity n = ×100 = = 45.2%
1+ e 1 + 0.826
2. Water to be added
Volume of soil in fill Vf = (1 + ef) Vs
Hence in an unit volume of fill,
Vf 1
Vs = = = 0.548 m 3
1 + ef 1 + 0.826
Water to be added per cubic metre of fill is given as
M04_PURU1773_01_SE_C04.indd 97 2/12/2013 7:48:33 AM

15 − 10.5 Mw M Mw
= = w =
100 Ms ρsVs 2.68 ×1000 × 5.48
3.5
That is, Mw = × 2.68 ×1000 × 0.548 = 51.4 kg.
100
Therefore, water to be added per cubic metre of fill = 51.4 litres.
Example 4.6 In the construction of a levee, the compaction specification required was 95%
of Proctor maximum dry density at a field moisture content within 2% of the optimum mois-
ture content.
The maximum dry density and optimum moisture content obtained in the laboratory
from the Proctor test were 1.94 Mg/m3 and 13.5%, respectively. A field supervisor conducted
sand-cone tests at two locations and obtained the results presented below. The sand in the
sand bottle was found to have a density of 1.87 Mg/m3. Check whether the specification was
satisfied.
Location Mass of soil removed (g) Mass of sand

no. used (g)
Wet Dry
1 43.86 38.46 39.51

2 37.38 32.21 32.39
Solution
95% of ρd max = 0.95 × 1.94 = 1.843 g/cm3

2% of OMC = 13.5 ± 2% = 15.5% to 11.5%
Location 1
3951
Volume of pit = = 2112.8 cm 3
1.87
4386
Wet density = = 2.08 g / cm3
2112.8
4386 − 3846
Water content = ×100 = 14.04%
3846
2.08 3
Dry density ρd = 1 + 0.1404 = 1.82 g / cm
Therefore, the moisture content requirements are satisfied, but the density requirement is
not satisfied.
Location 2
3239
Volume of pit = = 1732.1 cm 3
1.87
3738
Wet density = = 2.16 g / cm3
1732.1
M04_PURU1773_01_SE_C04.indd 98 2/12/2013 7:48:35 AM

3738 − 3221
Water content = ×100 = 16.05%
3221
2.16
Dry density ρd = = 1.86 g / cm 3
1 + 0.1605
Therefore, the density requirement is satisfied, but the moisture content requirement is
not satisfied.
Example 4.7 An airfield sub-grade is compacted with a thickness of 350 mm. The rammer
used for compaction has a foot area 0.06 m2 and imparts an energy of 50 kg m. Find the num-
ber of passes required to develop a compactive energy of 30,000 kg fm/m3.
Solution
50
Compactive energy imparted by rammer per cubic metre of the soil =
0.06 × 350 ×10−3
= 2380.95 kg fm/m3
30000
No. of passes required = = 12.6 , say 13
2380.95
POINTS TO REMEMBER
4.1 Soil compaction is the process of increasing the density of the soil by applying some
mechanical energy and thereby reducing air voids.
4.2 The compactive effort is the energy input to a soil for compacting it. The compactive
effort can be varied in the laboratory and in the field. Increasing the compactive effort
increases the dry density and decreases the optimum moisture content.
4.3 For most soils, the Standard Proctor test (BIS Light Compaction Test) is applicable.
In situations where a heavy load is anticipated, the BIS Heavy Compaction Test is
adopted. The compactive effort used in the Modified Proctor test is 4.5 times that of
the Standard Proctor test.
4.4 Field compaction equipment consists of excavating and handling equipment, rock
separation equipment, spreading equipment, discs, harrows, watering equipment,
rollers, vibrators, and other special compacting equipment.
4.5 Smooth wheel rollers are suitable for rolling of sub grades and for finishing construc-
tion fills of sandy or clayey soils. Rubber-tyred rollers are effective for a wide range of
soils from clean sand to silty clay. Sheep’s foot rollers are effective for a wide range of
soils from clean sand to silty clay.
4.6 Field control of compaction is the process of checking the density and moisture con-
tent during compaction by rollers or other compacting equipment. The most impor-
tant aspect of construction control is the speed with which the moisture and density
are determined and rectified if needed.
4.7 Factors affecting the compaction of a soil are the moisture content, compactive effort,
and method of compaction.
M04_PURU1773_01_SE_C04.indd 99 2/12/2013 7:48:35 AM

4.8 For compaction on the dry side of optimum, the soil structure is flocculated and dis-
persed in the wet side of optimum.
4.9 The permeability, swelling, undrained strength, and sensitivity are high on the soil
compacted dry of optimum.
4.10 The moisture-density curves for non-plastic soils, like sand, are not the same as those for
plastic soils. The curve for a non-plastic soil shows a trough at a low moisture content
and a peak at a high moisture content. Increasing the compactive effort in sand causes
no significant change in the maximum dry density and the optimum moisture content.
QUESTIONS
Objective Questions
4.1 Choose the correct statements from the following:
1. The compactive effort in the laboratory can be varied only by varying the height of
fall.
2. It is not feasible to expel air completely by compaction.
3. In field compaction, the effect of compaction is more with a small lift thickness.
4. The conventional method of measuring the moisture content is used in field control
of moisture content.
4.2 In the light compaction test, the number of blows used per layer is
(a) 15 (b) 25 (c) 30 (d) 35
4.3 In the nuclear moisture gauge a source of ______ is used.
(a) Slow protons (b) Slow neutrons
(c) Fast protons (d) Fast neutrons
4.4 An increase in the compactive effort in laboratory compaction causes the OMC to
(a) Remain the same (b) Decrease
(c) Increase by 5% (d) Decrease by 5%
4.5 The soil structure at the dry side of optimum is
(a) Partially flocculated (b) Fully flocculated
(c) Fully dispersed (d) None of the above
4.6 Assertion A: The process of compaction is accompanied by the expulsion of air.
Reason R: The degree of compaction of a soil is charactenzed by its dry density.
Select the correct code:
(a) Both A and R are correct, and R is the correct explanation of A.
(b) Both A and R are correct, and R is not the correct explanation of A.
4.7 A zero-air void density can
(a) Be obtained with a high compactive effort
(b) Be obtained with a low compactive effort
(c) Be obtained using static compaction
(d) Not be obtained in practice
M04_PURU1773_01_SE_C04.indd 100 2/12/2013 7:48:35 AM

4.8 The use of sheep’s foot rollers to compact cohesion-less soils in

(a) Very effective (b) Moderately effective
(c) Effective (d) Ineffective
4.9 The relative compaction corresponding to zero relative density is
(a) 80% (b) 70% (c) 60% (d) 50%
4.10 The swelling is greater and shrinkage is less for clay compacted
(a) at OMC (b) at OMC ± 2% (c) on the dry side of optimum
4.11 Distinguish between the Standard Proctor and Modified Proctor tests.
4.12 How do you differentiate between the compactive effort imparted in the laboratory
and that in the field.
4.13 How is the required compactive effort for a particular soil to attain a desired dry
density assessed in the field?
4.14 What effect does increased compaction have on the properties of a granular soil?
4.15 What is the ratio of the compactive energy of the IS heavy compaction test and the IS
light compaction test?
4.16 With illustrative compaction curves, discuss various factors which influence the com-
paction of a cohesive soil of high compressibility.
4.17 Explain the methods of finding the placement density of a compacted fill. Name the
method which is suitable for all types of soils.
4.18 How can the Standard Proctor test be modified to suit the compacting machinery
used at the site for compacting a cohesive soil?
EXERCISE PROBLEMS
4.1 The following data refer to a compaction test as per Indian Standards using light com-
paction:
Water content (%) 8.5 12.2 13.75 15.5 18.2 20.2

Mass of wet 18.0 19.4 20.0 20.5 20.3 19.8
sample (N)
Plot the compaction curve and obtain the maximum dry unit weight and optimum
moisture content. Also plot the 80% saturation line. Take G = 2.7 and the volume of
the mould = 1,000 cm3.
4.2 A laboratory compaction test conducted on a 900 ml volume of mould yielded the
following results:
Mass of dry soil (g) 160.1 154.80 155.75 165.35 182.74

Moisture content (%) 9.2 11.56 13.45 15.20 17.34
Plot the moisture–dry density curve and find the maximum dry density and optimum
moisture content. Find the void ratio at OMC and at 5% of OMC. Take G = 2.65.
M04_PURU1773_01_SE_C04.indd 101 2/12/2013 7:48:35 AM

4.3 In an embankment filling, the field density of the dry soil is 19.2 kN/m3 and the
maximum dry density (Proctor’s density) of the soil is 20.0 kN/m3. Calculate the
percentage compaction.
4.4 The bulk unit weight and the moisture content of a borrow area are 16.85 kN/m3 and
12.8%, respectively. It is intended to construct an embankment of 5 m height and, 10 m
top width with 1:1.5 side slopes and 2 km length with a finished dry unit weight of
19.50 kN/m3. Specific gravity of soil = 2.67.
1. Determine the quantity of soil required from the borrow pit for construction of 1 m
of the embankment.
2. If the construction is to be made with a moisture content of 15.2%, estimate the
amount of water to be added.
4.5 The details of two borrow areas identified for the construction of an embankment are
given below:
Borrow area Bulk density (g/cc) Moisture content (%) Specific gravity
A 1.65 14.5 2.65

B 1.30 15.6 2.67
The borrow areas are approximately at the same distance. The embankment is of
length 1 km, top width 8 m, height 6 m, and side slopes 1:1.6.
Which of the above two borrow areas would you recommend? Reason out your
choice.
4.6 A sand replacement test was conducted on a compacted field soil. The following are
the observations made:
Bulk density of sand used for the test = 1.5 g/cm3
Mass of soil excavated from the pit = 980 g
Mass of sand filling the pit = 720 g
Moisture content of compacted fill = 14.8%
Specific gravity of soil solids = 2.68%
Compute the wet density, dry density, void ratio, and degree of saturation of the
compacted fill.
4.7 A core cutter of 10 cm diameter and 18 cm height is used in an in-place density
determination of a compacted fill. The following are the other observations made:
Mass of empty core cutter = 2,330 g
Mass of soil + core cutter = 5,020 g
Mass of wet soil sample for
water content determination = 54.8 g
Mass of oven-dried sample = 50.2 g
Specific gravity of soil solids = 2.68
Compute the field dry density, void ratio, and degree of saturation.
M04_PURU1773_01_SE_C04.indd 102 2/12/2013 7:48:36 AM

5
Permeability and Capillarity
CHAPTER HIGHLIGHTS
Water flow – Darcy’s law – Validity of Darcy’s law – Laboratory and field
permeability tests – Permeability of stratified soils – Values of permeability –
Factors affecting permeability – Capillary phenomenon in soils – Shrinkage
and swelling of soils
5.1 INTRODUCTION
The amount, distribution, and movement of water in soil have an important bearing on
the properties and behaviour of soil. The engineer should know the principles of fluid
flow, as groundwater conditions are frequently encountered on construction projects.
Water pressure is always measured relative to atmospheric pressure, and water table is
the level at which the pressure is atmospheric. Soil mass is divided into two zones with
respect to the water table: (i) below the water table (a saturated zone with 100% degree
of saturation) and (ii) just above the water table (called the capillary zone with degree
of saturation ≤100%). Below the water table, the pore water may be static or seeping
through the soil under hydraulic potential. This chapter and the next have been devoted
to give an accurate and complete knowledge of the water condition in the soil.
5.2 WATER FLOW

Soil is a particulate material and has pores that provide a passage for water. Such pas-
sages vary in size and are tortuous and interconnected. A sufficiently large number of
such paths of flow are grouped to act together, and the average rate of flow is consid-
ered to represent a property of the soil. This property is termed permeability of the soil
and may be defined as the capacity of a soil to permit water to pass through its intercon-
nected void spaces. As in any other porous medium, water transmission takes place
between two points due to the difference in pressure heads or total heads (h).
M05_PURU1773_01_SE_C05.indd 103 2/12/2013 7:47:30 AM

As per Bernoulli’s equation, the total head consists of three components, viz., position or
elevation head (z), pressure head due to water pressure, uw (hw = uw /γw), and velocity head
(hv = v2/2g, where v is the velocity). Seepage velocities in soils are normally small, and hence
the velocity head is ignored. Thus, the total head causing the flow of water in soil is
uw
h= +z (5.1)
γw
The movement of water through a pipe or pore may take on either of the two char-
acteristic states of motion, viz., laminar or turbulent. Laminar flow indicates that each
water particle follows a definite path and never crosses the path of another particle. This
is an orderly and steady flow with no mixing. Turbulent flow indicates a random path of
irregular and twisted movement. This is a disorderly and unsteady flow with more mix-
ing. The velocity of flow depends directly on the nature of motion. Because of small pores
in most soils, the flow of water is steady and laminar except in a few cases, such as flow in
very coarse-grained soils and high velocities causing internal soil erosion (Taylor, 1948).
In general, for flow in the laminar range, energy losses are proportional to the first power
of velocity.
5.3 DARCY’S LAW

Considering one-dimensional flow in a saturated medium obeying laminar flow, Darcy (1856)
demonstrated experimentally (a schematic set-up of Darcy’s sand filtration experiment is
shown in Fig. 5.1) that the flow velocity is proportional to the hydraulic gradient (i), i.e.,
v∝i (5.2)
or
v = ki (5.3)
where v is the flow velocity (mm/s or m/s), k the coefficient of permeability (mm/s or m/s),
i the hydraulic gradient = h/L, h the difference in pressure heads = (h1 – h2), where h1 and
Fig. 5.1 Schematic set-up of Darcy’s experiment
M05_PURU1773_01_SE_C05.indd 104 2/12/2013 7:47:31 AM

Permeability and Capillarity 105
h2 are the pressure heads (m), L = length of specimen (mm), and the rate of flow, q (m3/s) is
given as
q = k iA (5.4)
where A is the cross–sectional area.
The velocity v is the overall velocity, also called discharge velocity, Darcian velocity, or
superficial velocity. This velocity is different from the velocity inside the soil pores, which is
known as the seepage velocity, vs. As the flow is continuous, q must be the same throughout
the system. Thus,
q = Av = Av vs
where Av is the cross-sectional area of the voids.
⎛A⎞ ⎛k ⎞
v s = ⎜⎜⎜ ⎟⎟⎟v = = k i / n = ⎜⎜ ⎟⎟⎟ i = k p i
v
A
⎝ v⎠⎟ n ⎜⎝ n ⎠
where kp is the coefficient of percolation, or

kp = k/n (5.5a)
and
⎛ 1 + e ⎞⎟
v = v s n = ⎜⎜
⎜⎝ e ⎟⎟⎠ s
v (5.5b)
This shows that the seepage velocity is greater than the superficial velocity, since Av< A
and vs > v, to keep the flow constant. In soil flow problems, it is more convenient to use the
total cross-sectional area of flow rather than the area of voids.
Microscopically, the flow follows a tortuous path, but macroscopically it may be
presumed to be orderly and in a straight line.
5.4 RANGE OF VALIDITY OF DARCIAN FLOW

Darcy’s type of flow is stable in character as long as the four basic conditions are always
satisfied, viz., (i) the steady-state laminar flow, (ii) 100% saturation (no compressible air pres-
ent), (iii) flow fulfilling continuity conditions, and (iv) no volume changes (compression or
swelling) during or as a result of flow. The validity of Darcy’s flow may be analysed with
respect to particle size, velocity, and hydraulic gradient.
Similar to flow through pipes, for flow through soils, Reynold’s number Rn may be
expressed as
v D γw
Rn = (5.6)
ηw g
where D is the average diameter of the soil particle.
It has been accepted that Darcy’s law is valid and flow will be laminar as long as Reynold’s
number is equal to one. For Rn = 1, the corresponding value of D = 0.5 mm, which is in the
coarse-sand range. This appears to be the upper limit of particle size beyond which the flow
M05_PURU1773_01_SE_C05.indd 105 2/12/2013 7:47:32 AM

Table 5.1 Realm of validity for Darcy’s flow of water in granular soils
D10 (mm) 76.2 25.4 9.52 2.0 0.59 0.25 0.074 0.02
Gravel Sand Silt
Coarse Medium Fine Coarse Medium Fine Coarse Fine
Realm of Practically always Darcy’s laminar flow only for Always laminar
flow of turbulent flow i less than about 0.2 to 0.3 flow for the
water for the loose state and range of i
0.3 to 0.5 for the dense state found in nature
Source: Burmister (1954).
may be turbulent. It has been shown by several authors that the flow is laminar in fine-
grained soils for the range of gradients found in nature.
Darcian linear relationship between velocity and gradient deviates faster in very fine
sands than in coarser sands with respect to gradient. In coarse materials, the pores are wider,
and therefore, the turbulence may begin at lower values of gradient than in fine sands. Bur-
mister (1954), based on experiments in granular soils, fixed a certain range for gradients,
which is presented in Table 5.1.
In dense clays and heavy loams, in which the water is of molecularly bound nature, seep-
age starts only when the gradient exceeds a certain value, i0 , called the initial or threshold
gradient. This gradient represents the gradient required to overcome the maximum binding
energy of mobile pore water. For dense clays, i0 may attain values from 20 to 30.
However, for a wide range of soils (silts through medium sands) for which the range of
gradients usually met with in nature, Darcy’s law stands valid. This is also true for clays
under steady state of flow.
5.5 LABORATORY PERMEABILITY TESTS

Basically, there are two laboratory experiments for the determination of the coefficient of
permeability, viz., the constant head and falling or variable head permeameters.
5.5.1 Constant Head Permeameter

This test is preferred for coarse-grained soils, such as gravels and sands, for permeabilities
>10−4 m/s. A schematic diagram of the apparatus is shown in Fig. 5.2 (details of the appara-
tus and procedure are given in Chapter 10).
The soil specimen is placed at an appropriate density in the permeameter. A steady ver-
tical flow of water under a constant total head is maintained. After saturation, a certain
quantity of water passing through the soil for a given time is collected and q is calculated.
Thus, from Darcy’s law,
h
q = Ak
L
M05_PURU1773_01_SE_C05.indd 106 2/12/2013 7:47:32 AM

Fig. 5.2 Constant head permeameter
or
qL
k= (5.7)
Ah
5.5.2 Falling Head Permeameter

For fine-grained soils, such as silt and clay (with k between 10−4 and 10−7 m/s), this is gener-
ally used. The experimental set-up is shown schematically in Fig. 5.3. After saturation, the
standpipe (with area of cross-section a) is filled with water, and time t1 corresponding to h1
is noted down. Water is allowed to fall to h2, and time t2 is noted. The coefficient of perme-
ability, k, is calculated from Eq. 5.7.
At any intermediate time t, let the water level be h and its rate of change be –dh/dt.
At time t, the head difference is h, hence i = h/L.
dh h
−a = Ak
dt L
h2 t2
dh Ak
−a ∫
L ∫t
= dt
h1
h
1
or
h1 t2
dh Ak
a∫
L ∫t
= dt
h2
h
1
M05_PURU1773_01_SE_C05.indd 107 2/12/2013 7:47:33 AM

Fig. 5.3 Falling head permeameter
or
aL h
k= log e 1
A(t2 − t1 ) h2
Let (t2 – t1) be t, then
aL h
k = 2.303 log10 1 (5.8)
At h2
Tests should be repeated using different values of h1 and h2. The permeability of fine-
grained soils can also be found from the consolidation test (Chapter 8).
5.6 FIELD PERMEABILITY TESTS

Data from field permeability tests are needed in the design of various civil engineering
works, such as cut-off wall design of earth dams, to ascertain the pumping capacity for
dewatering excavations and to obtain aquifer constants. The in situ tests, although expen-
sive, take into account the effects of stress release, the direction of flow, and boundary condi-
tions. However, field measurements are not sometimes precise because of the uncertainty of
soil and water conditions at the location (Lambe and Whitman, 1979).
Under field conditions, the rate of flow of water is measured by a quantity, coefficient of
transmissivity (T). It is defined as the rate of flow of water through a vertical strip of aquifer
of unit width under a unit hydraulic gradient (IS: 5529 – Part 1, 1985). This coefficient depicts
the ability of an aquifer to transmit water. This is related to the coefficient of permeability as
T = k ht (5.9)
where ht is the aquifer thickness.
M05_PURU1773_01_SE_C05.indd 108 2/12/2013 7:47:34 AM

Generally, the tests are carried out in boreholes where sub-surface explorations are carried
out. These tests can be done effectively up to a depth of 30 m and give the most reliable
results. The tests may be either pumping in or pumping out type. Pumping in test can be
conducted irrespective of the position of the water table in a stratum, while pumping out
test is suited for tests below the water table. The water table (or phreatic surface) is the level
to which undergroundwater will rise in a soil and will be at atmospheric pressure.
The pumping in test is suitable for low permeability and thin strata where adequate yield
may not be available for pump out test. By this test, permeability of the soil at the bottom of
the borehole is obtained. Thus, this is recommended for permeability determination of strat-
ified deposits and, hence, to check the effectiveness of grouting in such deposits. This test
is economical since it does not require an elaborate test arrangement as in pump out tests.
The types of pumping in tests as recommended in IS: 5529 – Part 1 (1985) are constant head
method, falling head method, and slug method. The constant head method is recommended
in highly permeable strata. The falling head method is more suitable for tests below the
water table. Further, this method is applicable for strata with low permeability and where
the soil below the casing pipe can stand without collapsing. The slug method is conducted
in artesian aquifers with small to moderate permeabilities. The reader may refer to IS: 5529 –
Part 1 (1985) for details of these methods.
The pumping out test is a more general and accurate method for permeability determination
below the water table. This method is most suitable for all groundwater problems. There are
basically two conditions of flow; accordingly, the pump out tests may be grouped as uncon-
fined flow (or gravity well) test or confined flow (or artesian well) test. In IS: 5529 – Part 1 (1985),
three methods are given, viz., unsteady-state, steady-state, and Bailer methods. Of these meth-
ods, the steady-state method, also known as Theim’s steady state or equilibrium method, is the
most accurate. The steady-state method for two flow conditions is explained below.
5.6.1 Unconfined Flow Pumping Out Test

Figure 5.4 represents a permeable layer underlain by an impermeable stratum and the
arrangement of wells. The soil is assumed to be homogeneous and coarse grained. The initial
Fig. 5.4 Pumping test from unconfined aquifer
M05_PURU1773_01_SE_C05.indd 109 2/12/2013 7:47:34 AM

water table position is considered horizontal, and the hydraulic gradient is assumed to be
constant at any given radius.
One large-diameter perforated casing is sunk up to the impervious stratum or to a consid-
erable depth, and this well is used as the main or test well. Two additional small-diameter
perforated casings are sunk at some distance from the test well. Water is pumped from the
main well at a constant rate. The draw-down of the water table takes place, and the steady-
state water table in each of the nearby observation wells is recorded.
The steady state is established when the water level in the main well and the observation
wells becomes constant.
Assume that the water is flowing into the well in a horizontal and radial direction. Con-
sider an elementary cylinder of soil having radius r, thickness dr, and height h. Let the water
level fall in the observation wells at the rate of dh. At the steady state, the rate of discharge,
q, due to pumping is
q = k iA
where
dh
i≈ (this is referred to as Dupit’s assumption)
dr
and
A = 2πrh
Therefore,
dh
q=k 2πrh
dr
Rearranging and integrating
r2 h2
dr 2πk
∫ r
=
q ∫
h dh
r1 h 1
Therefore,
2.303 q log10 (r2 / r1 )
k= (5.10)
π ( h22 − h12 )
Another form of the expression is obtained by substituting h1= H0, h2 = H, r1 = r0 (radius

of the main well), r2 = R (radius of influence), and H = depth of the original groundwater
table from the impervious stratum. Thus,
2.303 q log10 (R / r0 )
k= (5.11)
π ( H 2 − H02 )
Dupit’s assumption of i = dh/dr is reasonably accurate except at points close to the well.
Equations 5.10 and 5.11 have been developed for full penetration of the well, and in
case of partial penetration (Fig. 5.5), the coefficient of permeability is given as (Mansur and
Kaufman, 1962)
2.303 q log10 (R / r0 )
k= (5.12)
⎡ ⎛ 10 r0 ⎞⎟ 1.85 ⎤
[π ( H − S)2 − H02 ] ⎢1 + ⎜⎜⎜0.30 + ⎟⎟ sin ⎥
⎢⎣ ⎝ H ⎠ H ⎥⎦
M05_PURU1773_01_SE_C05.indd 110 2/12/2013 7:47:35 AM

Fig. 5.5 Partial penetration of well in unconfined aquifer
5.6.2 Confined Flow Pumping Out Test

An artesian well penetrating the full depth of the aquifer is shown in Fig. 5.6. Herein, a permeable
layer is sandwiched between two impermeable layers. Because of the artesian effect, the piezo-
metric surface locates itself above the upper surface of the aquifer. In the steady-state condition,
consider an elementary cylinder of radius r, thickness dr, and height h. The rate of flow is given as
dh
q = kiA = k 2π rHc
dr
where Hc is the depth of the confined aquifer. Thus,
r2 h
dr
2
2π k Hc dh
∫ r
=∫
q
r1 h 1
Fig. 5.6 Pumping test from confined aquifer
M05_PURU1773_01_SE_C05.indd 111 2/12/2013 7:47:36 AM

or
2.303 q log10 (r2 / r1 )
k= (5.13)
2π Hc ( h2 − h1 )
Substituting h2 = H0, r1 = r0, h2 = H, and r2 = R,

2.303 q log10 (R / r0 )
k= (5.14)
2π Hc ( H − H0 )
Equations 5.13 and 5.14 are valid as long as H0 > Hc.
5.7 PERMEABILITY OF STRATIFIED SOILS

In nature, soils are usually deposited in successive layers, and the permeabilities of the
layers may differ. It is not justifiable to find the numerical average of the coefficient of per-
meabilities of different layers. The stratifications can be considered as horizontal and
continuous, and the effective or overall coefficient of permeability for flow in horizontal and
vertical directions can be estimated.
5.7.1 Horizontal Flow

Consider the soil profile, shown in Fig. 5.7, consisting of two layers with isotropic permea-
bilities k1 and k2 in the first and second layers, respectively.
Assume the total head along the line AB to be constant. Similarly, the total head along CD
may also be taken as constant, but the value will be less than that along AB. Thus, the overall
gradient (i) and the individual gradients in each layer (i1 and i2) are equal, that is,
i = i 1 = i2 (5.15)
Let q1 and q2 be the rates of flow through unit thickness of the stratum. Let q be the total
rate of flow and kH be the effective coefficient of permeability i in the horizontal direction.
Thus,
q = q1 + q2
or
kH i A = k1 i1A1 + k2i2A2
Fig. 5.7 Flow parallel to stratification
M05_PURU1773_01_SE_C05.indd 112 2/12/2013 7:47:37 AM

or
kH A = k1 A1 + k2A2 (Since i = i1 = i2)
or
kH (Ht × 1) = k1 (H1 × 1) + k2 (H2 × 1)
or
k1 H1 + k 2 H 2
kH = (5.16)
Ht
or
kH Ht = k1H1+k2H2 (5.17)
If there are m layers, then
m
∑ kj Hj
j =1
kH = m
(5.18)
∑ Hj
j =1
or
m m
kH ∑ H j = ∑ k j H j (5.19)
j =1 j =1
5.7.2 Vertical Flow

In this condition, the flow takes place in the direction perpendicular to the stratification
(Fig. 5.8). To satisfy the continuity condition,
q = q1 = q2
Let h1 and h2 be the loss in heads in the first and second layers, respectively. Let kV be the
effective coefficient of permeability in the vertical direction. Now,
h1 h h
i1 = , i 2 = 2 , and i =
H1 H2 Hi
kv i A = k1 i1 A = k2 i2A
Fig. 5.8 Flow perpendicular to stratification
M05_PURU1773_01_SE_C05.indd 113 2/12/2013 7:47:38 AM

or
h
kV = k1i1
Ht
or
⎛ h + h2 ⎞⎟
k V ⎜⎜⎜ 1 ⎟ = k1i1
⎜⎝ H t ⎟⎟⎠
or
⎛i H + i H ⎞
k V ⎜⎜⎜ 1 1 2 2 ⎟⎟⎟ = k1i1
⎜⎝ Ht ⎟⎠
or
H t k1 i1
kV =
i1 H1 + i2 H 2
Dividing by k1i1,
Ht
kV = (Since k1i1 = k 2 i2 )
i1 H1 / k1i1 + i2 H 2 / k 2 i2
or
Ht
kV = (5.20)
H1 / k1 + H 2 / k 2
or
Ht H1 H2
= + (5.21)
kV k1 k2
For m layers,
m
∑ Hj
j =1
kV = m H
(5.22)
∑k
j
j =1 j
or
m
∑ Hj m Hj
j =1
=∑ (5.23)
kV j =1 kj
The above equations are valid only when one-dimensional flow takes place in the hori-
zontal or vertical direction.
M05_PURU1773_01_SE_C05.indd 114 2/12/2013 7:47:39 AM

5.8 VALUES OF PERMEABILITY

Table 5.2 represents a classification of soil based on permeability values (Terzaghi and
Peck, 1967).
Typical permeability values for different soils are as follows:
Uniform sand to fine sand 0.4 × 10−3 to 10−7 m/s

Silty sand to fine silt 10−7 to 0.5 ×10−7 m/s
Clay 0.5 ×10−7 to 0.01×10−7 m/s
Colloidal clay ≤10−9 m/s
Table 5.2 Permeability values
Degree of permeability k (m/s)
High > 10−3

Medium 10−3 to 10−5
Low 10−5 to 10−7
Very low 10−7 to 10−9
Practically impermeable < 10−9
Source: Terzaghi and Peck (1967).
5.9 FACTORS AFFECTING PERMEABILITY

The coefficient of permeability of a soil depends basically on the characteristics of both the
soil medium and the pore fluid. Lambe and Whitman (1979) have grouped particle size, void
ratio, composition, fabric, and degree of saturation as major soil characteristics, and viscos-
ity, unit weight, and polarity as major pore fluid characteristics. For a civil engineer dealing
with soils, the permeant is water, whose variation in property may be presumed to be very
less. Thus, soil characteristics may have to be given more importance.
5.9.1 Soil Characteristics

Based on Poiseuille’s law for flow through a bundle of capillary tubes, Taylor (1948) has
given a theoretical expression for flow through soil medium as
γw e3
k = Ds2 Cs (5.24)
ηw 1 + e
where Ds is the effective particle diameter and Cs the composite shape factor.
Particle Size. Equation 5.24 considers only particle size and void ratio among the soil
characteristics. This shows that permeability may be empirically related to the square of
some representative particle diameter. Such estimations may be true only in coarse-grained
soils, like silts and sands. Hazen (1911) proposed an expression for k for filter sands as
2
k = C D10 (mm / s) (5.25)
M05_PURU1773_01_SE_C05.indd 115 2/12/2013 7:47:40 AM

Table 5.3 Values for Hazen’s coefficient
⎛ 1 ⎞⎟
Sand (one or more applies) Cs ⎜⎜⎜ ⎟
⎝ mm ⎟⎠
(i) Very fine, well graded or with appreciable fines (≤75-μm size) 4–8
(ii) Medium, coarse, very poorly graded, clean, coarse but well graded 8–12
(iii) Very coarse, very poorly graded, gravelly, clean 12–15
where D10 is the effective size (mm) and C the experimental coefficient dependent on the
nature of the soil. Values of Hazen’s coefficient are given in Table 5.3.
Several correlations have been reported in literature utilizing some characteristic grain size.
Void Ratio. Different linear relationships have been attempted, relating void ratio and
permeability, viz., k ∝ e3(1 + e), k ∝ e2, and log k ∝ e. These relationships have been found to
indicate a straight-line relationship in non-cohesive soils but not in fine-grained soils. But e
versus log k is always close to a straight line for nearly all soils (Lambe and Whitman, 1979;
Taylor, 1948).
Composition. The effect of soil composition is more predominant in clayey soils than in
silts and sands. Depending on the type of clay mineral and the exchangeable cation present
in the clay, the permeability varies from 10−6 to 10−10 m/s, and accordingly the variation of
void ratio is from 16 to 1. The effect of exchangeable ion on permeability is less for low ion
exchange capacity of a soil. An increase in the thickness of the diffuse double layer (effect by
cation exchange capacity and cation valency) decreases its permeability, as the pore path is
reduced by the thickness of water held onto the soil particles.
Fabric. The permeability of a soil deposit is greatly influenced by the in-place micro- and
macrostructure. Clays are very significantly affected by the fabric component of a structure.
At similar void ratios in a clay, the permeability has been shown to be many times greater in
a flocculated state than in a dispersed state (Lambe, 1955). The above discussion is confined
to microstructural changes. The significance of macrostructure is extremely important, e.g.,
the effect of stratification. Variation in the permeabilities of layered soils contributes more
to the effective coefficient of permeability parallel to stratification than to the coefficient of
permeability perpendicular to stratification.
5.9.2 Pore Fluid Characteristics

Pore Fluid. Water may be considered as the pore fluid generally met with, and accordingly,
the discussion may be confined to water as the pore fluid. Equation 5.24 indicates that per-
meability is directly proportional to unit weight γw and inversely proportional to viscosity
ηw. Of these two parameters, viscosity is considerably affected by change in temperature.
A viscosity decrease caused by increase in temperature results in high permeability. A cor-
rection for temperature has to be effected, and k at 27°C has to be reported as per Indian
Standards (IS: 2720 – Part 17, 1986), i.e.,
η
k 27 ° C = kT wT (5.26)
ηw 27 ° C
M05_PURU1773_01_SE_C05.indd 116 2/12/2013 7:47:40 AM

The variables of water, i.e., γw and ηw , may be eliminated by defining another permeabil-
ity term as specific or absolute permeability, i.e.,
k ηw
K= (5.27)
γw
and K will have units of length square. Michaels and Lin (1954) conducted permeability
studies on kaolinite with permeants of different polarity and observed marked variation in
permeability. Thus, apart from viscosity and unit weight of the permeant, a factor represent-
ing polarity should be included in Eq. 5.24 (Lambe and Whitman, 1979).
Degree of Saturation. The degree of saturation has an important bearing on permeability.
In general, the higher the degree of saturation, the higher is the permeability. As the degree
of saturation increases, there is an increase in the flow channels for water and, hence, high
permeability.
5.10 SURFACE TENSION

If the water in soil pores is interconnected and subjected only to gravity, the soil mass above
the water table would be dry. But in nature, the soil pores are saturated, fully above the
water table and partially at large distances from it. The water in the soil voids located above
the water table is referred to as soil moisture. The phenomenon of water rise in the pores of
soils above the water table against the gravitational pull is called the capillary rise.
The principle of capillary rise of water is basically related to the rise of water in a glass
capillary tube. Water, like any other liquid, behaves as though the surface is tightly stretched
due to the intermolecular attraction of forces. This phenomenon is termed surface tension.
When a clean capillary tube is brought in contact with a source of water, the water rises up
in the tube and remains there. The rise is attributed to the combination of surface tension
with the attraction between the glass and water molecules. The shape of air–water is concave
in the downward, and the curved liquid surface is termed the meniscus. The rise of water in
the tube reaches a maximum height (hc) when the equilbrium condition is reached (Fig. 5.9).
At this stage, the downward pull of gravity on the capillary column of water is balanced by
the surface of water due to surface tension (Ts) effects.
Fig. 5.9 Rise of water in a capillary tube
M05_PURU1773_01_SE_C05.indd 117 2/12/2013 7:47:41 AM

10 20 30 40
Fig. 5.10 Surface tension of water as a function of temperature
Upward force due to surface tension acting around the periphery = (2πr) × Ts cos α
Downward pull due to gravity on the column of water = (πr2) × hc × γw
where r is the radius of the capillary tube (m), Ts the surface tension of water (N/m), α the
contact angle (pore water makes zero contact angle with glass), and hc the capillary rise (m).
2πrTs cos α = π r2hcγw
2Ts 4Ts
hc = = (5.28)
r γw d γw
where d is the diameter of the capillary tube.
The surface tension Ts is temperature dependent (Fig. 5.10). The height of the capillary
rise is not affected by the variations in the shape and size of tubes at levels below the menis-
cus. The capillary pore water pressure (negative) is given as
uc = hc rw (5.29)
This is a measure of the suction exerted on the pore water by the soil.
5.11 CAPILLARY PHENOMENON IN SOILS

The capillary rise in soil is similar to the capillary rise in a capillary tube. But in soils, the pores
are irregularly shaped and interconnected in more directions than only in the vertical. Because
of these limitations, a satisfactory analysis is impossible. However, the capillary tube concept
does serve as a sound basis for understanding the capillary phenomenon in soils.
The main difficulty with the use of Eq. 5.28 is the proper assessment of the diameter of
pores. Keeping the basic idea that fine-grained particles make smaller voids and coarse-
grained particles larger voids, attempts have been made to relate the grain size and void
ratio to the capillary head. One such equation has been given by Hazen as
C
hcr = (5.30)
e D10
where C = 0.01 × 10−3 to 0.05 × 10−3 (m2) and hcr = maximum capillary rise (m).
M05_PURU1773_01_SE_C05.indd 118 2/12/2013 7:47:41 AM

Fig. 5.11 Capillary zones
This expression shows that decrease in the effective grain size causes a decrease in the
void ratio and an increase in the capillary rise in soils.
Equation 5.28 indicates that the effect of molecular attraction will be near the water table,
and hence, irrespective of the void space, all the pores will be filled. At a distance from the water
table, only smaller voids would be filled with water. Thus, the capillary zone may be divided
into three zones of arbitrary boundaries. The zone just above the water table is called the zone of
capillary saturation (with almost 100% saturation). Above the zone of capillary saturation is the
zone of partial saturation wherein only small pores are filled with water and the large ones with
air; evidently, here the degree of saturation is less than 100%. The third zone near the ground
surface contains water surrounding the particles at contact points, but there is no continuity.
This zone is referred to as the zone of contact water (Fig. 5.11). Water may also reach this zone
from the ground surface by percolation and may be held in suspension by the capillary forces.
Capillarity of a dry soil is its capacity to draw up water to elevations above the phreatic line
and also to retain the water above the phreatic line in a draining soil. The height of water that
a soil can support is generally called the capillary head. So far, only the first aspect is considered.
Figure 5.12a represents a column of cohesionless soil. The maximum capillary rise, hcr,
and the minimum capillary head, hcn (for the maximum degree of saturation) are identified
and represented in Fig. 5.12b.
Fig. 5.12 Capillary heads in soil
M05_PURU1773_01_SE_C05.indd 119 2/12/2013 7:47:42 AM

Table 5.4 Capillary heads
Soil Particle size, D10 Void ratio Capillary head (mm)

(mm)
hcr hcs
Coarse gravel 0.82 0.27 54 60

Sandy gravel 0.20 0.45 284 200
Fine gravel 0.30 0.29 195 200
Silty gravel 0.06 0.45 1,060 680
Coarse sand 0.11 0.27 820 600
Medium sand 0.02 0.48–0.66 2,396 1,200
Fine sand 0.03 0.36 1,655 1,120
Silt 0.006 0.95–0.93 3,592 1,800
Source: Lane and Washburn (1946).
Let us consider a situation where the same soil has been saturated up to a height h above
the water table and allowed to drain. Then, the moisture will be as shown in Fig. 5.12c. Point
“a” represents the highest elevation up to which there exists a continuous water path above
the phreatic line. This distance is referred to as the maximum capillary head, hcm. Point “b”
shows the point up to which the soil is fully saturated, and this height is called the saturation
capillary head, hcs.
Thus, any capillary head associated with drainage has a maximum value of hcm, and that
with capillary rise has a maximum value of hcr. There is more possiblity of bridging effect of
surface water on large voids during draining than the pulling effect during rising. Accord-
ingly, it is reasonable to expect that hcs > hcn and hcm > hcr. Many capillary heads might exist
between the two extremes hcm and hcn. Table 5.4 shows the range of capillary heads for cohe-
sive soils (Lane and Washburn, 1946).
In certain practical problems, the time necessary for the attainment of maximum capillary
rise is more. The term indicating the rate of capillary rise is called the capillary conductivity
or capillary permeability. Factors which influence capillary conductivity are pore size, water
content, and temperature of the soil. The rate of capillary conductivity is low in fine-grained
soils and high in coarse-grained soils.
5.12 SHRINKAGE AND SWELLING OF SOILS

Volume changes occur in soil deposits due to changes in water content and in the effective
stresses (discussed in Chapter 7) produced by neutral stresses. When a saturated soil is
allowed to dry, a meniscus develops in each void at the soil surface. The formation of such a
meniscus brings in tension in the soil water, leading to a compression in the soil structure,
which is termed shrinkage. In a partially saturated soil, the force causing shrinkage arises
from the curved air–water interfaces. The compression caused by shrinkage is as effective as
that produced by external load. Pressures as high as 500 kN/m2 can be produced in fine-
grained soils due to shrinkage.
M05_PURU1773_01_SE_C05.indd 120 2/12/2013 7:47:42 AM

Table 5.5 Volume change potential
Volume change Plasticity index (%) Shrinkage

potential limit (%)
Arid area Humid area
Low 0–15 0–30 >12

Moderate 15–30 30–50 10–12
High >30 >50 <10
Source: Holtz and Gibbs (1956).
The degree of shrinking depends on the initial water content, the type and amount of clay
content, and the mode and environment of geological deposition. The presence of sand and
silt-size particles in a clay deposit reduces the total shrinkage. Shrinkage occurs horizon-
tally as well as vertically, causing vertical shrinkage cracks. In highly compressible clays, the
cracks may be as high as 0.5 m wide and 5 m deep.
Some soils (expansive soils like the black cotton soils of India) not only shrink on drying
but also swell when they come in contact with water. Swelling is caused mainly by repulsive
forces which separate the clay particles causing volume increase. The mechanism causing
swelling is more complex, and the factors which contribute to it are
1. the elastic rebound of soil grains,
2. the clay mineral’s affinity for water,
3. the cation exchange capacity and electrical repulsive forces, and
4. the expansion of entrapped air.
Free swell of a soil is defined as the increase in the volume of a soil, without any external
constraints, on submergence in water (IS: 2720, 1977). In general, the free swell ceases when
the moisture reaches the plastic limit, i.e., at a water plasticity ratio of about 0.25.
Clay soils with high montmorillonite clay mineral show an almost reversible swelling
and shrinking on rewetting and redrying. Clay soils with illite or kaolinite show a large ini-
tial volume decrease on drying, with only a limited swelling on rewetting. Holtz and Gibbs
(1956) have given an indication of the potential volume change based on shrinkage limit and
plasticity index (Table 5.5).
Volume change is a serious problem in shrinkage-susceptible soils present in arid or semi-
arid areas.
WORKED EXAMPLES
Example 5.1 A sand sample of 25 cm length was subjected to a constant head permeability
in a permeameter having an area of 30 m2. A discharge of 100 cm3 was obtained in a period
of 1 minute under a head of 39 cm. Mass of dry sand in the sample was 1,350 g. The specific
gravity of the sand particles was 2.67. Determine (i) the coefficient of permeability, (ii) the
superficial velocity, and (iii) the seepage velocity.
M05_PURU1773_01_SE_C05.indd 121 2/12/2013 7:47:43 AM

Solution
From Eq. 5.7,
qL ⎛ 100 ⎞⎟ 25
(i) k= = ⎜⎜ ⎟ = 0.0356 cm / s.
Ah ⎜⎝ 60 ⎟⎠ 30 × 39
Dry density ρd = G ρw/(1 + e). That is,
Gρw
e= −1
ρd
2.67 ×1
= −1
1350 /( 30 × 25 )
= 1.483 – 1
= 0.483
e 0.483 0.483
Porosity n = = = = 0.326.
1 + e 1 + 0.483 1.483
q 100
(ii) Superficial velocity vs = = = 0.056 cm / s.
A 60 × 30
(iii) Seepage velocity v = vsn = 0.056 × 0.326
= 0.0183 cm/s.
Example 5.2 A soil sample has a length of 3.5 m and a cross-sectional area of 2 m2. If water
flows through such a soil sample and the fluid energy lost is 1,650 N-m for every cubic metre
flow of water, estimate Darcy’s velocity and permeability. The time of flow for 1 m3 of water
is 26 hours. Find also the seepage velocity if the void ratio of the sample is 0.58.
Solution
Q 1
q= = = 1.068 ×10−5 m 3 / s.
t 26 × 60 × 60
q 1.068
Darcy’s velocity v = = ×10−5 = 5.34 ×10−6 m / s.
A 2
Total energy loss
Loss of head h =
Energy loss per unit length
Total energy loss
=
Volume × Unit weight of water
1650
= = 0.168 m
1× 9810
Now,
h 0.168
i= = = 0.048
L 3.5
M05_PURU1773_01_SE_C05.indd 122 2/12/2013 7:47:45 AM

v 5.34
k= = ×10−6 = 1.11×10−4 m / s
i 0.048
v (1 + e ) (1 + 0.58)
Seepage velocity vs = = 5.34 ×10−6
e 0.58
= 1.46 × 10−5 m/s.
Example 5.3 In a falling head permeability test, the time taken for the head to fall from h1
to h2 is t. If the test is repeated with the same initial head, what would be the final head in a
time interval of t/2?
Solution
Let h’ be the head after the lapse of time t/2.
Now, the time taken for the head to fall from h1 to h′ and from h′ to h2 is the same. Thus,
for a given soil sample,
⎛ 2.303 aL ⎞⎟ h1
k1 = ⎜⎜ ⎟ log
⎜⎝ At / 2 ⎟⎠ 10 h ′
Again,
⎛ 2.303 aL ⎞⎟ h′
k 2 = ⎜⎜ ⎟⎟ log10
⎜⎝ At / 2 ⎠ h2
As the test is performed on the same sample on the same permeameter, k1 = k2. Therefore,
h1 h′
log10 = log10
h′ h2
That is,
(h′)2 = h1h2
h ′ = h1 / h2
Example 5.4 In a falling head permeability test, the head causing flow was initially 500 mm,
and it drops 20 mm in 5 minutes. How much time is required for the head to fall to 250 mm?
Solution
From Eq. 5.8, the coefficient of permeability
aL h
k = 2.303 log10 1
At h2
Therefore,
2.303 aL h
t= log10 1
kA h2
M05_PURU1773_01_SE_C05.indd 123 2/12/2013 7:47:46 AM

Therefore, the values in the bracket are constant for a given permeameter and the soil at a
particular void ratio.
Hence,
⎡ h ⎤ ⎡ 500 ⎤
t1 = (c) ⎢ log10 1 ⎥ = (c) ⎢ log10 ⎥
⎢ h ⎥ ⎢
⎣ 20 ⎥⎦
⎣ 2 ⎦1
⎡ h ⎤ ⎡ 500 ⎤
t2 = (c) ⎢ log10 1 ⎥ = (c) ⎢ log10 ⎥
⎢ ⎥
h2 ⎦ 2 ⎢⎣ 250 ⎥⎦
⎣
t1 ⎡ 500 ⎤ ⎡ 500 ⎤
= ⎢ log10 ⎥ ⎢log10 ⎥
t2 ⎣⎢ ⎥ ⎢
20 ⎦ ⎣ 250 ⎥⎦
= 5 [log102]/[log1025]
0.301
= 5×
0.699
= 2.15 minutes
Example 5.5 A falling head permeameter contains a soil sample 8 cm high and 60 cm3 in
cross-sectional area. The permeability of the sample is expected to be 1 × 10−4 cm/s. If it is
desired that the head in the standpipe should fall from 30 to 10 cm in 40 minutes, determine
the size of the standpipe which should be used.
Solution
From Eq. 5.8,
aL h
k = 2.303 log10 1
At h2
Rearranging,
k At
a=
2.303 × L log10 h1 / h2
1×10−4 × 60 × 40 × 60
a=
2.303 × 8 × log10 30 / 10
1×10−4 × 60 × 40 × 60
=
2.303 × 8 × 0.477
= 1.64 cm2
Example 5.6 A permeameter of 80 mm diameter with a sample length of 300 mm has been
used for constant head and falling head tests. While conducting a constant head test, the loss
of head was 1,150 mm for a length of 250 mm and the rate of flow was 2,700 mm3/s. Find the
coefficient of permeability (in mm/s).
M05_PURU1773_01_SE_C05.indd 124 2/12/2013 7:47:47 AM

If a falling head test were then performed on the same sample at the same void ratio, find
the time taken for the head to fall from 900 to 450 mm. The diameter of the standpipe in the
falling head test was 25 mm.
Solution
Constant head test:
h 1150
Gradient i = = = 4.6
L 250
⎛ 80 2 ⎞⎟
Area of sample A = π ⎜⎜⎜ ⎟ = 5026.6 mm 2
⎜⎝ 4 ⎟⎟⎠
Rearranging Eq. 5.4,
q 2700
k= = = 0.117 m / s.
iA 4.6 × 5026.6
Falling head test:
⎛ 252 ⎞⎟
Area of standpipe a = π ⎜⎜⎜ ⎟ = 490.9 mm 2
⎜⎝ 4 ⎟⎟⎠
2.303 aL h
t= log10 1
kA h2
2.303 × 490.9× 300 900
t= log10
5026.6 × 0.117 450
= 173.6 seconds
Example 5.7 Prove that for stratified deposits of soils, the average permeability in the
horizontal direction is greater than the average permeability in the vertical direction.
Solution
Consider two layers of thicknesses H1 and H2 with coefficients of permeability k1 and k2 in
the first and second layers, respectively.
From, Eqs. 5.16 and 5.20,
k H + k2 H 2
kH = 1 1
Ht
and
Ht
kV =
( H1 / k1 ) + ( H 2 / k 2 )
Therefore,
kH (k1 H1 + k 2 H 2 )/ H t
=
kV H t /[( H1 / k1 ) + ( H 2 / k 2 )]
M05_PURU1773_01_SE_C05.indd 125 2/12/2013 7:47:49 AM

H12 + H 22 + [(k12 + k 22 )/ k1 k 2 ]H1 H 2

=
H t2
H12 + H 22 + [(k12 + k 22 )/ k1 k 2 ]H1 H 2

=
H12 + H 22 + 2 H1 H 2
Since k1 and k2 are always positive,
(k1 – k2)2 ≥ 0
or
k12 + k 22 ≥ 2 k1 k 2
or
k12 + k 22
≥2
k1 k 2
Only when k1 = k2 the ratio is equal to 2, otherwise it is greater than 2. Here, k1 ≠ k2.
Therefore, in kH/kV, the numerator is greater than the denominator.
Therefore, kH > kV
Hence, the average coefficient of permeability in the horizontal direction is greater than the
average coefficient of permeability in the vertical direction.
Example 5.8 A horizontal stratified soil deposit consists of three layers, each uniform in
itself. The permeabilities of the layers are 4 × 10−4, 25 × 10−4 and 7.5 × 10−4 mm/s; their
thicknesses are 6, 3, and 12 m, respectively. Find the effective average permeability of the
deposits in the horizontal and vertical directions.
Solution
From Eq. 5.19,
m
∑ kjHj
j =1
kH = m
∑ Hj
j =1
( 4 × 6 + 25× 3 + 7.5×12)1000 ×10−4

kH =
(6 + 3 + 12)1000
189×10−4
= = 9×10−4 mm / s
21
M05_PURU1773_01_SE_C05.indd 126 2/12/2013 7:47:50 AM

From Eq. 5.20,

m
∑ kjHj
j =1
kv = m
∑ Hj /kj
j =1
(6 + 3 + 12)1000 ×10−4 21
kV = =
[(6 / 4) + (3 / 25) + (12 / 7.5)]×1000 3.22
kV = 6.5 × 10−4 mm/s
Example 5.9 In a falling head permeameter, the sample was 18 cm long with cross-
sectional area of 22 cm2. Calculate the time required for the drop of head from 25 to 10 cm if
the cross-sectional area of the standpipe was 2 cm2. The sample of soil was heterogeneous,
with a coefficient of permeability of 3 × 10−4 cm/s for the first 6 cm, 4 × 10−4 cm/s for the
second 6 cm, and 6 × 10−4 cm/s for the last 6 cm of thickness. Assume the flow taking place
perpendicular to the bedding planes.
Solution
From Eq. 5.20, the effective vertical coefficient of permeability
m
∑ kjHj
j =1
kV = m
∑ Hj /kj
j =1
(6 + 6 + 6) 18
kV = ×10−4 = ×10−4
[(6 / 3) + (6 / 4) + (6 / 6)] 4.5
kV = 4 × 10−4 cm/s
2.303 a L h
t= log10 1
kA h2
2.303 × 2×18 1 25
t= −4
× log10
4 ×10 × 22 60 10
829080
t= × 0.398 = 62.5 minutes
5280
Example 5.10 In a falling head permeability test on a soil of length l1, the head of water in the
stand pipe takes 5 seconds to fall from 900 to 135 mm above the tail water level. When another
M05_PURU1773_01_SE_C05.indd 127 2/12/2013 7:47:51 AM

soil of length 60 mm is placed on the first soil, the time taken for the head to fall between the
same limits is 150 seconds. The permeameter has a cross-sectional area of 4,560 mm2 and a
standpipe area of 130 mm2. Calculate the permeability of the second soil.
Solution
Refer to Fig. 5.13 for a two-layer system.
From Eq. 5.21, we have
l l l
= 1 + 2
k V k1 k 2
From the permeability test on Sample 1 only, we have

2.303 al1 h
k1 = log10 1
At1 h2
or
l1 At1 4560 × 50
= =
k1 2.303 a log10 ( h1 / h2 ) 2.303 ×130 × log10 (900 / 135)
or
l1
= 924.31 seconds
k1
For permeability tests on both the soils,

2.303 al h
kV = log10 1
At2 h2
or
l At2 4560 ×150
= =
k V 2.303 a log10 ( h1 / h2 ) 2.303 ×130 × log10 (900 / 135)
= 2772.93 seconds
Fig. 5.13
M05_PURU1773_01_SE_C05.indd 128 2/12/2013 7:47:52 AM

Therefore,
l2 l l
= − 1 = 2772.93 − 924.31
k 2 k V k1
or
l2
= 1848.62 seconds
k2
Therefore,
l2 60
k2 = = = 0.0325 mm / s
1848.62 1848.62
Example 5.11 A sand deposit of 12 m thickness overlies a clay layer. The water table is 3 m
below the ground surface. In a field permeability pump out test, the water is pumped out at a
rate of 540 litres/min when steady-state conditions are reached. Two observation wells are
located at 18 m and 36 m from the centre of the test well. The depths of the draw-down curve
are 1.8 m and 1.5 m, respectively, for these two wells. Determine the coefficient of permeability.
Solution
This is an unconfined aquifer, hence k is given by Eq. 5.10. That is,
2.303 q log10 (r2 / r1 )
k=
π( h22 − h12 )
Here, q = 540 litres/min = 540 × 1000 cm3/min

r1 = 18 m and r2 = 36 m
h1 = 9 – 1.8 = 7.2 m = 720 cm
h2 = 9 – 1.5 = 7.5 m = 750 cm
Therefore,
2.303 × 540 ×1000(36 / 18)
k=
π[(750)2 − (720)2 ]
2.303 × 540 ×1000 × 2

k= = 17.96 cm / min
π× 44100
Example 5.12 A pumping test is conducted in an unconfined aquifer with a partially

penetrated well. The following are the details:
Diameter of well = 1 m
Height of water level from the impermeable layer before pumping = 30 m
Depth of water in the well = 8 m
Depth of the bottom of the well from the impervious layer = 10 m
Radius of influence = 50 m
Discharge at steady state of pumping = 0.45 m3/min
Find the coefficient of permeability of the soil.
M05_PURU1773_01_SE_C05.indd 129 2/12/2013 7:47:53 AM

Solution
Here, R = 50 m, r = 0.5 m, s = 10 m, H0 = 8 m
q = 0.45 m3/min
From Eq. (5.12),
2.303 q log10 (R / r0 )
k=
{
π[( H − S)2 − H02 ] 1 + [0.3 + (10 rc / H )]sin
1.8S
H }
Substituting the above values,
2.303 × 0.45× log10 (50 / 0.50)

k=
π[(30 − 10) − 8 2 ]{1 + [0.3 + (10 × 0.5 / 30)]sin (8 ×10 / 30)}
2
Reducing,
k = 1.95 × 10−3 m/min = 0.0326 mm/s
Example 5.13 A glass capillary tube is 0.2 mm in diameter. What is the theoretical maxi-
mum height of capillary rise for a tube of this size? The surface tension is 0.0735 N/m.
Estimate the pressure in the capillary water just under the meniscus.
Solution
From Eq. 5.28,
4Ts 4 × 0.0735
hc = =
dγ w (0.20 / 1000)× 9.81×10 3
hc = 0.15 m
From Eq. 5.29
uc = hcγw = 0.15 × 9.81 = 1.47 kN/m2
Example 5.14 A silt deposit has a series of clay seams of about 5 mm thick at an average
vertical spacing of about 1.65 m. Permeability of silt deposit is about 120 times as that of clay
seam. Find the ratio of equivalent horizontal permeability to vertical permeability.
Solution
Naturally, alternate layers are found in silt and clay; therefore it is enough to find the ratio
only for the two adjacent layers.
Height of silt layer, H1 = 1.65 m.
Height of clay seam, H2 = 0.005 m.
Permeability of silt, k1 = 120 times k2, where k2 is permeability of clay.
Hence,
1.65k1 + 0.005 (k 2 )
Horizontal flow, kh =
1.655
= 119.64 k2
M05_PURU1773_01_SE_C05.indd 130 2/12/2013 7:47:54 AM

1.655
Vertical flow, kv = ⎜ 1.65 ⎞⎟ ⎛⎜ 0.005 ⎞⎟
⎛
⎜⎜ ⎟⎟ + ⎜ ⎟⎟
⎝ k1 ⎟⎠ ⎜⎝ k 2 ⎟⎠
1.655
=
(0.0825 + 0.005)
k2
= 19.94 k2
k h 119.64 k 2
= = 6.0.
kv 19.94 k 2
POINTS TO REMEMBER
5.1 Permeability of a soil is its capacity to permit water to pass through its inter-connected
void spaces.
5.2 Potential or total head causing flow of water through soil consists of position or eleva-
tion head, pressure head due to water pressure, and velocity head. As seepage veloci-
ties are small in soil, the velocity head is ignored.
5.3 Darcy’s law (velocity proportional to gradient) is valid as long as the flow is laminar,
the soil is fully saturated, no volume change occurs during flow, and the continuity
condition is present. Further, Darcy’s law is valid for Reynold’s number equal to unity.
5.4 Constant head permeameter is preferred for coarse-grained soils, whereas falling
head permeameter is suitable for fine-grained soils.
5.5 In a layered soil, the average coefficient of permeability in the horizontal direction is
greater than the average coefficient of permeability in the vertical direction.
5.6 Pumping out test is suitable for tests below the water table, whereas pumping in test
can be conducted irrespective of the position of the water table for finding field k.
5.7 Factors affecting permeability are soil characteristics, viz., particle size, void ratio,
composition, and fabric, and pore fluid characteristics, viz., pore fluid and degree of
saturation.
5.8 Water table is the level of water in a soil stratum at which the pressure is atmospheric.
Soil below the water table is called saturated zone (sr = 100%), while soil above is
called capillary zone (sr ≤ 100%)
5.9 In capillaries, the water surface is tightly stretched due to intermolecular attraction of
forces, which is referred to as surface tension.
5.10 Capillary rise is the phenomenon of water rise in the pores of soils above the water
table against the gravitational pull.
5.11 Shrinkage of soil depends on the initial water content, the type and amount of clay
content, and the mode and environment of geological deposition. The presence of
sand and silt-size particles in clays reduces the shrinkage.
5.12 Clayey soils with montmorillonite clay mineral show reversible swelling and shrink-
ing, whereas kaolinite and illite show less swelling than shrinking.
M05_PURU1773_01_SE_C05.indd 131 2/12/2013 7:47:55 AM

QUESTIONS
Objective Questions
5.1 State whether the following are true or false:
(i) Coefficient of permeability is greater for coarse-grained soils than for fine-
grained soils.
(ii) The constant head permeability test is the most reliable and accurate for clayey
soils.
(iii) Moisture rises above the groundwater table as a result of capillary tension.
(iv) In sandy soils, the seepage velocity is equal to Darcy’s flow velocity.
(v) The capillary pressure in a soil may be more than 5 m head of water.
5.2 Coefficient of permeability of a fine-grained soil increases with
(a) Increase in temperature of the pore fluid
(b) Increase in viscosity of the pore fluid
(c) Increase in density of soil
(d) None of the above
5.3 Select the correct range of permeability (m/s) of a soil whose degree of permeability
is low:
(a) 10−8 to 10−9 (b) 10−3 to 10−5 (c) 10−5 to 10−7 (d) 10−1 to 10−3
5.4 Artesian conditions are said to exist when the piezometric surface lies
(a) Below ground level
(b) Between ground level and the aquifer
(c) Above ground level
(d) Below groundwater level
5.5 The pressure on a phreatic surface is
(a) Less than atmospheric pressure
(b) Equal to atmospheric pressure
(c) Greater than atmospheric pressure
(d) Not related to atmospheric pressure
5.6 Compacted well-graded, gravelly sands with little or no fines will be
(a) Impervious (b) Semi-pervious to pervious
(c) Semi-pervious (d) Pervious
5.7 Identify the wrong factor. The following three major characteristics influence perme-
ability of clays:
(a) Fabric (b) Composition
(c) Degree of saturation (d) Particle shape
5.8 In a sedimentary soil deposit, the permeability
(a) Is uniform in all directions
(b) Is greater in the horizontal direction than in the vertical direction
(c) Is lesser in the horizontal direction than in the vertical direction
(d) Is double in the vertical direction of that in the horizontal direction
M05_PURU1773_01_SE_C05.indd 132 2/12/2013 7:47:55 AM

5.9 Which of the following pairs are correctly matched?

(1) Aquifer — Source for obtaining surface water
(2) Unconfined aquifer — Porous soil constitutes the surface stratum
(3) Confined aquifer — Porous soil bounded above and below aquitards
(4) Artesian aquifer — Confined aquifer with high pressure
Select the correct answer using the codes given below:
(a) 1, 2, and 3 (b) 2, 3, and 4 (c) 3, 4, and 1 (d) 4, 1, and 2
5.10 The mechanical effect of cohesion in sand due to contact moisture depends on the
(a) Shape of sand grains (b) Surface roughness of sand grains
(c) Relative density of sand (d) Gradation of sand
5.11 Two fluids with extreme viscosities are to be passed through a porous material. Reason
out the condition for which the coefficient of permeability will be the greatest.
5.12 In fine-grained soils, what effects does the presence of adsorbed water have on the
coefficient of permeability?
5.13 In what condition is the capillary system if the angle of wetting is zero?
5.14 Explain the ways by which the capillary water and the effect of capillarity can be
removed from soil.
5.15 Capillary rise is greater for fine-grained soils than for coarse-grained soils. Substanti-
ate this statement.
5.16 What are the applications of the capillary tube theory to soil engineering?
EXERCISE PROBLEMS
5.1 A constant head permeability test was conducted on a sand sample of 250 mm length
and 2,000 mm2 area. The head loss was 500 mm, and the discharge was found to be
260 ml in 130 seconds. Determine the coefficient of permeability of the sand sample.
Find the superficial and seepage velocities if the dry unit weight and specific gravity
of the samples were 17.98 kN/m3 and 2.62, respectively.
5.2 Three soil samples, x, y, z, with coefficients of permeability 1 × 10−1, 2 × 10−2 and
5 × 10−3 m/s are placed in a tube of cross-section 100 mm × 100 mm, as shown in
Fig. 5.14. Water is supplied through the apparatus such that the head difference is
maintained at 300 mm. Find the rate of supply in litres per hour.
5.3 The changes caused by a rise in temperature in viscosity and unit weight of a pore
fluid are 82.5% and 97.8%, respectively. Compute the percent change in the coefficient
of permeability assuming other factors to remain constant.
5.4 A variable head permeability test is conducted on a 100 mm long specimen. The
diameter of the standpipe is 1/10th that of the specimen. The test took 900 sec-
onds to fall from a height of 300 to 100 mm. Determine the permeability of the
specimen.
M05_PURU1773_01_SE_C05.indd 133 2/12/2013 7:47:55 AM

Fig. 5.14
5.5 A field pumping test was performed for a horizontal stratum of sandy soil 4 m thick,
sandwiched between two impermeable strata. After the steady-state flow equilib-
rium, the rate of flow was 90 litres/hour. The elevation of water level in a borehole
3 m away from the test well was 2.1 m, and in a borehole 6 m away, it was 2.7 m above
the top of the lower impermeable stratum. Estimate the coefficient of permeability of
the soil.
5.6 In a falling head permeameter test, the initial head at t = 0 was 600 mm. The head
dropped 30 mm in a time of 40 seconds. Find the time required to run the test to a final
head of 200 mm.
5.7 A sand sample at a void ratio of 0.52 has a permeability of 0.4 × 10−3 m/s. Assuming
a reasonable relationship between void ratio and permeability for this soil, estimate
the permeability at a void ratio of 0.65.
5.8 During a falling head permeability test, the sample on close investigation was found
to be in two layers 60 and 40 mm thick. The routine falling head test on this sample
yielded the following results: diameter of standpipe, 4 mm; sample diameter, 80 mm;
length of sample 100 mm; initial head, 1,100 mm; final head, 420 mm, and time for
fall in head, 6 minutes. After the test, independent tests were made on each soil; the
permeabilities were found to be 5 × 10−4 and 17 × 10−4 mm/s, respectively. Check
the average permeability through the sample in the laboratory test with the estimated
value considering the layer effect. Also, estimate the average permeability in a direc-
tion at right angles to sampling. Comment on the results.
5.9 A graded filter has to be constructed with four soils of different layer thicknesses.
The layer thicknesses are 350, 250, 200, and 100 mm and are to be placed at different
compacted densities such that the permeabilities are 2 × 10−2, 3 × 10−1, 6.8 × 10−1, and
1.5 mm/s, respectively. Calculate the average coefficients of permeability in directions
parallel and orthogonal to the layers.
M05_PURU1773_01_SE_C05.indd 134 2/12/2013 7:47:55 AM

5.10 Three layers of soil represent the soil profile beneath a reservoir. The depth of water
in the reservoir is 15 m and the area of spread is 4,500 m2. The permeability and thick-
ness of each layer are given below:
Layer Thickness (m) k (mm/s)
1 2.5 3.6 × 10−5

2 1.5 2.8 × 10−5
3 2.8 1.8 × 10−5
A sand layer lies below this profile. The sand has horizontal drainage. Assuming ver-
tical flow through the layers, compute the water loss in a period of 30 days from the
reservoir.
5.11 The rise in a capillary tube is 520 mm above the free water surface. Determine the
surface tension if the radius of the tube is 0.03 mm.
5.12 A sand sample has a porosity of 32.4%, and Hazen’s effective grain size is 0.056 mm.
Estimate the capillary rise in the soil sample.
M05_PURU1773_01_SE_C05.indd 135 2/12/2013 7:47:56 AM

M05_PURU1773_01_SE_C05.indd 136 2/12/2013 7:47:56 AM
6
Seepage
CHAPTER HIGHLIGHTS
Seepage forces – General flow equation – Significance of Laplace’s equation –
Properties and applications of flow nets – Construction of flow nets: boundary
conditions, construction methods, flow nets for sheet piles and dams – Aniso-
tropic soil conditions – Non-homogeneous soil conditions – Piping –Design of
filters
6.1 INTRODUCTION
Groundwater is frequently encountered in construction projects. The movement of
water through soil is referred to as seepage, and such movement leads to several groups
of problems in civil engineering. Seepage of water has a bearing on three major types of
problems, viz., (i) loss of stored water through an earth dam or foundation, (ii) instabil-
ity of slopes and foundations of hydraulic structures due to the force exerted by the
percolating water, and (iii) settlement of structures founded on or above compressible
layers due to explusion of water from the voids caused by load applications.
Theoretical solutions based on simple assumptions in problems related to stability
and settlement have been successful. But hydraulic problems do not lead to simple solu-
tions because of adverse field conditions. This chapter discusses some of the techniques
used for analysing seepage flow.
6.2 SEEPAGE FORCES

When water flows through soil, the water head is dissipated in viscous friction. During
energy dissipation, a drag force is exerted on the soil particles in the direction of flow.
M06_PURU1773_01_SE_C06.indd 137 2/12/2013 8:33:14 AM

Saturated soil
L
(a) Upward flow of water

Buoyancy (B)
W
Seepage force (SF)
A
Uw Soil particle Weight (Ws)
(b) Free body of soil (c) Equilibrium of forces
Fig. 6.1 Flow showing seepage and other forces
Consider the column of soil shown in Fig. 6.1a. If the height h of the water surface in the
reservoir is raised, the water pressure at the bottom of the soil sample is increased and the
drag force on the soil particle becomes greater. The drag force and the buoyant weight of
the particles are in balance at a critical height h = hc, and an increase in height will cause
the soil particles to be washed out of the container. At this critical condition, the force acting
on the bottom of the soil sample will just equal the weight of the soil and water mass in the
container.
Now, consider the upward and the downward forces at the bottom of the soil mass
(Fig. 6.1b).
⎛ G + 1⎞⎟
Downward force, W = ⎜⎜ γ AL
⎜⎝ 1 + e ⎟⎟⎠ w
Upward force due to water pressure U = (b + L) γw A.

Assuming no friction on the sides of the container, and considering the critical condition
(i.e., h = hc),
G+e
( hc + L)γ w A = γ w AL
1+ e
that is,
hc G − 1
ic = = (6.1)
L 1+ e
where ic is the critical hydraulic gradient.
This condition also occurs when individual soil particles are freely suspended in flowing
water. The equilibrium of forces is shown in Fig. 6.1c for such a condition.
M06_PURU1773_01_SE_C06.indd 138 2/12/2013 8:33:14 AM

Seepage 139
G γw
Ws = AL = weight of soil particles acting downwards
1+ e
B = γwVs = force due to buoyancy acting upwards
SF = seepage force on particles
For equilibrium,
Ws = SF + B
G γ w AL G γ w AL
or SF = Ws − B = − γ wVs = − γ w (1 − n)AL
1+ e 1+ e
⎛G γ γ ⎞
or SF = ⎜⎜ w − w ⎟⎟⎟ AL
⎜⎝ 1 + e 1 + e ⎠
or SF = ic γ wV (6.2)
The seepage force expressed per unit volume is referred to as the seepage pressure. If h is
less than hc, the seepage force is proportionately less than icγwV.
This critical condition described above is responsible for the phenomenon of boil in soils,
usually referred to as quicksand. The quicksand condition is likely to occur at hydraulic gra-
dients of about 1.0 in non-cohesive soils. Contrary to common belief, quicksand is not a type
of sand but a phenomenon caused due to the flow condition. In cohesive soils, the cohesive
strength of soil must be overcome before soil particles are washed out of the soil mass.
6.3 GENERAL FLOW EQUATION

In formulating the general flow equation for soils, the following assumptions are made:
1. The soil medium is saturated, incompressible, homogeneous, and isotropic with respect
to permeability.
2. The flow is laminar and follows Darcy’s law.
3. Water is incompressible.
Consider an element of soil of dimensions dx, dy, and dz with velocities vx, vy, and vz in
the x, y, and z directions, respectively. This represents a generalized flow condition in three
dimensions for a homogeneous isotropic medium (Fig. 6.2a).
There are many seepage problems in which the flow is essentially two-dimensional, for
example, seepage under long sheet pile walls, dams, and water-retaining structures and
through embankments and earth dams. Thus, ignoring the flow in the y direction (i.e., vy = 0),
a two-dimensional flow condition may be considered (Fig. 6.2b).
Let the hydraulic gradients be ix and iz in the x and z directions and the permeability be k.
The quantity of water entering the element in unit time is vx dy dx + vz dx dy and that leav-
ing is (vx + ∂vx / ∂x dx ) dy dz + (vz + ∂vz / ∂z dz)dx dy.
As the element is undergoing no volume change and the water is incompressible, the
quantities of water entering and leaving should be equal, and thus
⎛ ∂v ⎞ ⎛ ∂v ⎞
vx dy dz + vz dx dy = ⎜⎜vx + z dx⎟⎟⎟ dy dz + ⎜⎜vz + z dz⎟⎟⎟ dx dy
⎜⎝ ∂x ⎠ ⎜
⎝ ∂z ⎠
M06_PURU1773_01_SE_C06.indd 139 2/12/2013 8:33:16 AM

Vz + ∂Vz dz Vy + ∂Vy dy
∂z ∂y
Vx + ∂Vx dx
z
Vz + ∂Vz dz
∂x ∂z
dz Vx + ∂Vx dx
Vx Vx ∂x
dz
dy
Vy dx dx
x
Vz
Vz
(a) Flow in three dimensions (b) Flow in two dimensions
Fig. 6.2 Generalized flow condition
or
⎛ ∂vx ∂vz ⎞⎟
⎜⎜ + ⎟ dx dy dz = 0
∂z ⎟⎠
⎜⎝ ∂x (6.3)
Since dx dy dz ≠ 0, Eq. 6.3 becomes
∂vx ∂vz
+ =0 (6.4)
∂x ∂z
Equation 6.4 is referred to as the equation of continuity in two dimensions.* Now, based
on Darcy’s law,
∂h ⎪⎫
vx = kix = k ⎪⎪
∂x ⎪
⎬ (6.5)
∂h ⎪
vz = kiz = k ⎪⎪
∂z ⎪⎪⎭
The partial derivatives in Eq. 6.5 suggest a potential function of the form φ(x, z), such that
∂φ ∂φ
vx = and vz = (6.6)
∂x ∂z
Substituting Eq. 6.6 in Eq. 6.4, we obtain
∂ 2φ ∂ 2φ
+ =0 (6.7)
∂x 2 ∂ z 2
Equation 6.7 is the Laplace equation which presents the two-dimensional steady flow
of an incompressible fluid through an incompressible isotropic porous medium. In simple
terms, it represents the balancing of gradient changes in the x and z directions when the
volume is constant.
* For an element which experiences volume change, the continuity equation becomes [(∂vx / ∂x ) +
(∂vz / ∂z )]dx dy dz = (dv / dt ), the volume change per unit time (refer to Chapter 8).
M06_PURU1773_01_SE_C06.indd 140 2/12/2013 8:33:17 AM

Seepage 141
The existence of a potential function requires an irrotational flow such that

∂vz ∂vx
− + =0 (6.8)
∂x ∂z
If a flow function (ψ) is defined such that
∂ψ ∂ψ
vx = and vz =
∂z ∂x
then
∂2ψ ∂2ψ
+ =0 (6.9)
∂x 2 ∂z 2
Equation 6.9 also satisfies the Laplace condition.
Since φ = φ(x, z) and vx = ∂φ / ∂x and vz = ∂φ / ∂z
∂φ ∂φ
dφ = dx + dz
∂x ∂z
= vx dx + vz dz
For φ = constant, dφ = 0, then

dz v
=− x (6.10)
dx vz
Similarly, since ψ = (x, z) and vx = (∂ψ / ∂z) and vz = −(∂ψ / ∂x )
∂ψ ∂ψ
dψ = dx + dz
∂x ∂z
= −vz dx + vx dz
For ψ = constant, dψ = 0, then

dz vz
= (6.11)
dx vx
Thus, the curves of constant φ are normal to curves of constant ψ since the product of their
gradients is –1. The form of the curve depends on the boundary conditions of the problem.
6.4 SIGNIFICANCE OF LAPLACE EQUATION

The solution of two Laplace equations for the potential and flow functions takes the form of
two families of orthogonal curves.
One set of curves (constant-ψ lines) represent the trajectories of seepage and are termed
flow lines (Fig. 6.3). The space between two adjacent flow lines may be imagined to be a flow
channel with an impervious boundary such that water does not cross the flow lines.
The other set of curves (constant-φ lines with φ = kh) represent lines of equal head and are
termed equipotential lines. The head loss caused by water crossing two adjacent equipotential
lines is termed the potential drop.
M06_PURU1773_01_SE_C06.indd 141 2/12/2013 8:33:19 AM

Δh1
Δh2
Flow
lines
I
h1 b Flo
h2 wc
han
h3 nel
Flo
wc
han
hn nel
Equipotential lines
Fig. 6.3 Flow net definitions
The entire pattern of flow lines and equipotential lines is referred to as a flow net. Thus,
a flow net is a graphical representation of the head and direction of seepage at every point.
Seepage losses and their related flow pattern, the uplift pressure, and pore pressures are
frequently estimated using flow nets.
6.5 PROPERTIES AND APPLICATIONS OF FLOW NETS

6.5.1 Properties
Let us consider the example of the flow net shown in Fig. 6.4. Each flow line starting at the
upstream boundary with head h1, dissipates the head in viscous friction and attains head h2
when terminating at the downstream boundary. All such lines are shown by continuous
lines. Along each flow line there must be a point where the total head may have a specific
value. A line connecting all such points of equal head represents an equipotential line, and
such lines are shown by broken lines.
An infinite number of flow lines and equipotential lines could be drawn for any given
condition. But an important consideration to be borne in mind is that the geometric figures
formed by the equipotential and flow lines should approach a square shape.
h1
C h2
A B D
Q1
R1 Q R
I1
P1 I
P S
S1 Q2 R2
I2
J
P2 S2
Impervious
Fig. 6.4 Flow under dam
M06_PURU1773_01_SE_C06.indd 142 2/12/2013 8:33:19 AM

Seepage 143
Choose P1Q1R1S1 and P2Q2R2S2, the figures formed between two pairs of flow and equipotential
lines in two different channels, and P′Q′R′S′, an auxiliary figure bounded by the same pair of
flow lines of the first figure and by the same pair of potential lines of the second figure.
Flow through any one channel may be given as
Δh h
Δq = ki ( h ×1) = kih = k h = kΔh
l l
Considering the three figures, the discharge equation for each case is given as
h1 ⎫⎪
Δq 1 = k Δh 1 (Fig. P1 Q 1 R 1 S 1 ) ⎪⎪
l1 ⎪⎪
⎪⎪
h2 ⎪
Δq 2 = k Δh 2 (Fig. P2 Q 2 R 2 S 2 )⎬ (6.12)
l2 ⎪⎪
⎪⎪
h’ ⎪
Δq ’ = k Δh ’ (Fig. P ’Q ’R ’S ’) ⎪⎪
l’ ⎪⎪⎭
where Δh1, Δh2 and Δh′ are the potential drops considering two successive equipotential
lines.
If all the figures in the flow net are drawn as squares, then
h1 = l1 , h2 = l2 , and h ’ = l ’
The auxiliary square and the first square have the same flow boundaries; thus Δq′ = Δq1,
and they have the same equipotential boundaries as the second square, and thus Δh′ = Δh2.
Hence,
Δq1 = Δq2 and Δh1 = Δh2 (6.13)
Thus, it is shown that when all the figures are squares, there must be the same quantity
of flow in each channel and the same potential drop in crossing each figure. To have these
conditions, it is just sufficient if the ratio h/l is maintained the same, but drawing square
figures is far more convenient than drawing rectangular figures.
From the above discussion for a flow net with square figures, the properties of flow nets
can be summarized as
1. Flow lines and equipotential lines intersect or meet orthogonally.
2. The quantity of water flowing through each channel is the same.
3. The potential drop between any two successive equipotential lines is the same.
4. The velocity of flow is more (because of high gradients) in figures of small dimensions so
that the discharge remains the same.
5. Flow lines and equipotential lines are smooth continuous curves, being either elliptical or
parabolic in shape.
6.5.2 Applications
Seepage Quantity. We have shown that in a flow net with square figures the flow through
one channel is
Δq = kΔh (6.14)
Let H be the total head loss, i.e., H = h1 – h2.

Let Nd be the number of potential drops
M06_PURU1773_01_SE_C06.indd 143 2/12/2013 8:33:20 AM

Therefore,
H
Δh = (6.15)
Nd
Therefore, H
Δq = k (6.16)
Nd
Let Nf be the number of flow channels.
The total discharge through the complete flow net per unit length is given as
H
q = Nf Δq = k Nf
Nd
Nf
that is, q=kH (6.17)
Nd
The ratio Nf/Nd is independent of k and H and is characteristic of the flow net. This is
called the shape factor of the flow net.
Seepage Pressure. Let nd be the number of potential drops (each of vale Δh) lost by a
water particle before reaching point J, the point where the seepage pressure is needed. Let hl
be the net potential at point J, that is,
hl = H − nd Δh (6.18)
Hence, the seepage pressure ps = hl γw
or
ps = ( H − nd Δh)γ w (6.19)
This pressure acts in the direction of the flow.
Uplift Pressure. The uplift pressure uw (also called hydrostatic pressure) at any point
within the soil mass is given by
uw = hw γ w (6.20)
where hw is the piezometric head = h1 − z , where z is the position head of the point.
The downstream water level is usually considered as the datum, and all points above the
datum are considered as positive.
Exit Gradient. The maximum hydraulic gradient at the downstream end of the flow lines
is termed the exit gradient. This is given as
Δh
ie = (6.21)
l
where l is the length of the smallest square in the last field.
6.6 CONSTRUCTION OF FLOW NET

6.6.1 Boundary Conditions
Prior to the construction of a flow net, it is essential to study the hydraulic boundary
conditions associated with that particular problem and ascertain the characteristics of flow
lines and equipotential lines.
Two practical problems associated with their boundary conditions are shown in Fig. 6.5.
The boundaries are as detailed below:
M06_PURU1773_01_SE_C06.indd 144 2/12/2013 8:33:22 AM

Seepage 145
A B D E
Q
R
C S
F G P T
(a) Sheet pile wall (b) Earth dam
Fig. 6.5 Boundary conditions
1. A–B, D–E, P–Q, and S–T are permeable boundaries. These surfaces have a constant head
and hence are equipotential lines.
2. F–G, B–C–D, and P–T are impermeable boundaries. Further, there is no flow across these
boundaries; i.e., ∂ht/∂z=0 and ψ is constant. Thus, these are flow lines.
3. R–S is a seepage surface, and along this surface the pore pressure is zero, and if Δφ is
constant the equipontential lines meet the seepage surface at constant vertical intervals.
4. Q–R is the piezometric surface or a free surface, and the pore pressure is zero and since ψ
is constant it is a flow line (Whitlow, 1983).
6.6.2 Construction Methods

A solution of the Laplace equation for the boundary condition of the given seepage problem
may yield data for plotting the flow net; that is, for such a solution the functions φ(x, z) and
ψ(x, z) have to be determined for the relevant boundary conditions.
The construction of a flow net can be done by adopting any one of the following methods.
Graphical Method. The commonest procedure for obtaining flow nets is a graphical, trial
and error sketching method for seepage problems with well-defined boundary conditions.
For this method, both practice and a natural aptitude are needed. Reasonable good flow nets
can be obtained by practice by adhering to the correct boundary conditions and by the use
of square figures. A square figure may be defined as the one in which the median lengths
of the flow lines and equipotential lines are equal and have right angle corner intersections.
The following procedure may be adopted to obtain a reasonably good flow net:
1. Make a scale drawing showing the structure, soil mass, the pervious boundaries (through
which water enters and leaves the soil), and the impervious boundaries (which confine
the flow).
2. Keeping the properties of a flow net in mind, sketch two or more flow lines entering
and leaving at right angles to the pervious boundaries and approximately parallel to the
impervious boundaries.
3. Then draw equipotential lines at right angles to the flow lines such that the median
lengths of the flow lines and equipotential lines are equal. This cannot be achieved in the
first trial as the positions of the flow lines were approximate ones.
4. Readjust the flow lines and equipotential lines such that the condition stipulated in Step 3
is attained.
5. Introduce more flow lines and equipotential lines.
M06_PURU1773_01_SE_C06.indd 145 2/12/2013 8:33:22 AM

The following hints suggested by A. Casagrande (1937) are valuable for a beginner in
flow net construction.
1. Use every opportunity to study the appearance of well-constructed flow nets; when the
picture is sufficiently absorbed in your mind, try to draw the same flow net without look-
ing at the available solution; repeat this until you are able to sketch this flow net in a
satisfactory manner.
2. Four or five flow channels are usually sufficient for the first attempt; the use of too many
flow channels may distract the attention from essential features.
3. Always watch the appearance of the entire flow net. Do not try to adjust details before the
entire flow net is approximately correct.
4. The beginner usually makes the mistake of drawing overly sharp transitions between
the straight and curved sections of flow lines or equipotential lines. Keep in mind that
all transitions are smooth, of elliptical or parabolic shape. The size of the squares in each
channel will change gradually.
Typical flow nets for flows below sheet pile walls and dams are shown in Figs. 6.6 and 6.7.
Electrical Analogy Method. Laplace equation not only governs a steady-state flow of
ground-water but is also encountered in a steady flow of electric current through a conduc-
tor and the flow of heat through a plate. The correspondence between water and current
flows is reflected from the following comparison:
Steady-State Seepage Electric Current
Total head ht Voltage V

Coefficient of permeability k Coefficient of electric conductivity kE
Discharge velocity v Current I
Darcy’s law v ∝ hydraulic gradient Ohm’s law I ∝ voltage gradient
∇2ht = 0 ∇2V = 0
Equipotential lines = ht = constant Equipotential lines = V = constant
Impervious boundary (∂ht/∂x) = 0 Insulated boundary (∂V/∂x) = 0
Thus, the flow domain of a porous medium has to be transferred into an electrical con-
ductor field with similar configuration and boundary conditions (Fig. 6.8). The methodology
is to obtain the locus of the lines of equal voltage drop which is in correspondence to the
location of equipotential lines for the given flow domain.
A typical electrical analogy set-up is shown in Fig. 6.9. The flow domain is simulated with
different conducting materials such as various metal sheets, heavy paper coated with graph-
ite, dilute copper sulphate solution, salt water, and gelatins. The inflow face is at a potential
V1 and the outflow surface at a potential V2. Points of equal voltage drop are found using the
probe. Alternatively, the voltages at different pre-defined nodal points may be found. Then
the contours of equal voltage may be sketched by hand after transferring the voltage values
of nodal points on a separate sheet of paper with pre-drawn boundaries.
It should be noted that the electrical analogue simply provides only the equipotential
lines. To get the complete flow net, flow lines are drawn orthogonal to the equipotential lines
conforming to the boundary conditions. However, direct determination of flow lines is pos-
sible by interchanging the locations of metal bars and insulators.
M06_PURU1773_01_SE_C06.indd 146 2/12/2013 8:33:22 AM

Seepage 147
Sheet pile
(a)
h1
h2
Flow line
Equipotential line
Impervious
(b)
Sheet pile
Impervious
Sheet piles
(c)
Impervious
Fig. 6.6 Flow net for (a) sheet pile with level ground surface, (b) sheet pile with varied ground
surface, and (c) double sheet piles
Other Methods. Flow tanks and viscous flow models have also been in use for construct-
ing flow nets. In the flow tank model, the scale model of the prototype is used with sand
as the porous medium. In the viscous flow model, a viscous fluid like glycerine is used for
the medium. In both the cases, coloured dyes are injected at the upstream boundary which
traces the path of flow lines. The equipotential lines are estimated later.
M06_PURU1773_01_SE_C06.indd 147 2/12/2013 8:33:22 AM

(a)
Sheet pile
cut-off
Impervious
(b) Sheet pile

cut-off
Impervious
(c)
Sheet pile cut-off
Sheet pile cut-off
Impervious
Fig. 6.7 Flow net for flow under dam with (a) upstream sheet pile cut-off, (b) downstream sheet
pile cut-off, and (c) double sheet pile cut-off
Pervious boundary Conduction

boundary
V1 V2
h1 h2
Porous medium Conduction medium
Impervious boundary Insulated boundary
Fig. 6.8 Seepage and current flow analogy
M06_PURU1773_01_SE_C06.indd 148 2/12/2013 8:33:23 AM

Seepage 149
Switch Battery
Rheostat
Resistor A
Inflow face Outflow face

B D
V
Probe
Fig. 6.9 Typical electrical analogy set-up
The solution to the Laplace’s equation may also be obtained with the help of computers
using analytical methods such as the finite difference method, finite element methods, and
complex variable methods.
Construction of flow nets for flow through earth dams is discussed in Chapter 20.
6.7 ANISOTROPIC SOIL CONDITIONS

Most natural soil deposits depict anisotropy in permeability, with a higher permeability
coefficient in the horizontal direction (x direction) than in the vertical direction (z direction),
i.e., with kx > kz. Thus, Eq. 6.5 can be re-written as
∂h ⎫⎪
vx = k x ix = k x ⎪⎪
∂x ⎪
⎬ (6.22)
∂h ⎪⎪
v z = k z iz = k z ⎪⎪
∂z ⎪⎭
and substituting Eq. 6.22 in the continuity equation, we have

∂2 h ∂2 h
kx + k z =0
∂x 2 ∂z 2 (6.23)
M06_PURU1773_01_SE_C06.indd 149 2/12/2013 8:33:24 AM

or
kx ∂ 2 h ∂ 2 h
+ =0
k z ∂x 2 ∂z 2
Let a2 = kx/kz; then

∂2 h ∂2 h
a2 + =0
∂x 2 ∂z 2
or
∂2 h ∂2 h
+ =0
⎛ x ⎞2 ∂z 2
∂ ⎜⎜ ⎟⎟⎟
⎝a⎠
or
∂2 h ∂2 h
+ =0
∂xl2 ∂z 2
(6.24)
where
x k
xl = =x z
a kx
Equation 6.24 satifies Laplace’s condition for an isotropic soil in an xl–z plane. Trans-
form an anisotropic flow region into a fictitious isotropic flow region by transferring
all x dimensions as xl (Fig. 6.10b). Now construct an artificial flow net by the usual
method which will satisfy all the requirements of seepage. Redraw this flow net on the
true scale by multiplying each x dimension (measured from some arbitrary baseline,
such as the centre line of the dam) by a, while keeping the z dimension the same. This
flow net on the true scale may consist of parallelograms and rectangles but not squares
(Fig. 6.10a).
Figure 6.11 shows one flow field in natural and transformed scales. The quantity of flow
ΔqN and ΔqT through these sections may be expressed as
Δh b
ΔqN = k x
b kx / kz
and
Δh
ΔqT = ke b
b (6.25)
where ke is the effective coefficient of permeability.
But
ΔqT = ΔqN
M06_PURU1773_01_SE_C06.indd 150 2/12/2013 8:33:25 AM

Seepage 151
h1
h2
A P B
x
(a) Flow net for actual section
h1
h2
A′ P′ B′
xt
(b) Flow net for transformed section
Fig. 6.10 Transformed flow net for anisotropic soil
z z
b k x /k z b
Flow Flow
b b
xl
x
(a) Natural scale (b) Transformed scale
Fig. 6.11 Flow fields
Therefore,
Δh
ke Δh = k x
kx / kz
that is
ke = k x k z
(6.26)
Thus,
Nf
q = ke H
Nd
(6.27)
M06_PURU1773_01_SE_C06.indd 151 2/12/2013 8:33:26 AM

6.8 NON-HOMOGENEOUS SOIL CONDITIONS

Consider a flow across a soil interface. If there is a change in soil conditions, the flow lines
are deflected at the interface of the soil with varying permeabilities k1 and k2. If the flow takes
place into a less permeable soil (i.e., k1 > k2, e.g., at the upstream casing and clay core junc-
tion), the flow lines are refracted towards the normal at the interface, and away from the
normal when k1 < k2 (e.g., flow from clay core to downstream casing). Figure 6.12 shows the
interface condition to two soils with permeabilities k1 and k2 (k1 > k2).
Let the potential drop from point P to Q and from R to S be Δh; then
⎛ Δh ⎞ ⎛ Δh ⎞
Δq = k1 ⎜⎜⎜ ⎟⎟⎟ PQ = k 2 ⎜⎜ ⎟⎟⎟ RS
⎝ PR ⎠ ⎜⎝ QS ⎟⎠
But
PR QS
tan α1 = and tan α2 =
PQ RS
Hence,
k1 k2
=
tan α1 tan α2
or
k1 tan α1
= (6.28)
k 2 tan α2
When k2 ≥ 10 k1, the second soil offers no resistance and hence may be treated as an open
drain and no deflection correction is needed.
No
rm
al
90° Interface of soils 1 and 2

α1 (k1 > k2)
R
Soil 1 P
Δq α1 α2
k1 90°
α2 Δq
Q S
Soil 2
k2
PQ and RS – Equipotential lines

Flow lines
Fig. 6.12 Interface condition of two soils
M06_PURU1773_01_SE_C06.indd 152 2/12/2013 8:33:27 AM

Seepage 153
6.9 PIPING
Because of local instability caused by a high hydraulic gradient at the exit face of a percolating
soil mass, soil grains are dislodged and eroded. Such erosions gradually cause a pipe-shaped
discharge channel. The width of the channel and the hydraulic gradient will increase with
time and lead to a failure of the structure constructed on or with the soil. Such a mode of
failure is called failure by piping.
Failures by piping may be due to scour or sub-surface erosion starting downstream and
propagating inwards, causing an ultimate failure. Such a piping failure is called failure by
sub-surface erosion, and no theoretical approach is possible. Piping failure is also initiated
when the upward seepage pressure at the toe becomes greater than the effective weight of
the soil (i.e., due to a quick condition). Such a piping failure is referred to as failure by heave.
The mechanics of failure by piping due to heave is discussed below.
It has been found that the failure due to piping takes place within a distance of D/2 from
the sheet pile, where D is the depth of the sheet pile (Fig. 6.13).
Consider prism of soil ABB′A′ with width A′B′ = ½AA′ at the exit end of the structure
shown in Fig. 6.13. The effective vertical pressure at the time of failure on any horizontal sec-
tion through the prism is approximately equal to zero. Thus piping occurs when the seepage
force on the base of the prism (U) becomes equal to the effective weight of the overlying sand
(W). Let the hydraulic potential at A and B be hA and hB, Then,
U = ½ D γ w ( hA + hB )
A′
B′
A B
Impervious
(a) Flow net with location of piping
D/2
A′ B′
Sheet pile cut-off D

W
A B
U
Equipotential lines
(b) Equilibrium of forces
Fig. 6.13 Effect of piping
M06_PURU1773_01_SE_C06.indd 153 2/12/2013 8:33:28 AM

and
W = ½ D2 γ ’
Let
hA + hB
ha =
2
Therefore, factor of safety with respect to piping,
W Dγ ′
Fp = =
U ha γ w
and
′
( hA + hB )/ 2 − ( hA + hB′ )/ 2
iav =
D (6.29)
where h′A and h′B are the hydraulic potentials at A′ and B′.
Then,
Fp = ic / iw
(6.30)
where Fp = factor of safety against failure by piping, and ic = (G – 1)/(1 + e) and is normally
greater that 3 or 4. In cohesive soils, because of cohesion this method gives conservative values.
If the factor of safety against failure by piping is small, this may be increased by providing
inverted filters. If the weight of the filter is Wf over the prism, then the increased factor of safety
Fp′ = (W + Wf )/U
(6.31)
The filter material should satisfy the condition explained in the next section.
6.10 DESIGN OF FILTERS

Filter or drain materials used for preventing piping should satisfy two requirements apart
from adding weight as follows:
1. The gradation of filter material should be capable of forming small-size pores such that
the migration of adjacent particles through the pores is prevented.
2. The gradation of filter material should be such that it allows a rapid drainage without
developing large seepage forces.
The above requirements are satisfied on adopting a suitable grain-size distribution for the
filter material, based on the material to be protected. If the following filter criterion is met,
piping will be adequately controlled (Bertram, 1940):
D15 (filter)
< 4 to 5
D85 (protected soil)
(6.32)
This criterion emphasizes that the D15 size of the filter soil should not be more than four
or five times the D85 size of the protected soil. The second criterion is
M06_PURU1773_01_SE_C06.indd 154 2/12/2013 8:33:29 AM

Seepage 155
D15 (filter)
> 4 to 5
(6.33)
The second requirement is that the D15 size of the filter soil should not be more than four
or five times the D15 size of the protected soil.
The US Corps of Engineers have recommended that
D50 (filter)
≤ 25
(6.34)
Based on this criterion the D50 size of the filter should be less than or equal to 25 times the
D50 size of the protected soil.
Generally, the filter is not of one material but of different materials placed in layers. Each
of these layers satisfies the requirements with respect to the preceding layer. Further, as
a rough guideline, the grain-size distribution curves of the fine- and coarse-grained soils
should be roughly parallel.
WORKED EXAMPLES
Example 6.1 At the toe of a dam, the foundation soil has a void ratio of 0.72. The specific
gravity of the soil solids is 2.65. To ensure safety against piping, the upward gradient must
not exceed 30% of the critical gradient at which quicksand conditions occur. Estimate the
maximum permissible upward gradient.
Solution
G − 1 2.65 − 1
Critical gradient ic = = = 0.959
1 + e 1 + 0.72
The permissible upward gradient is 30% of the critical gradient.
30
Maximum permissible upward gradient = 0.959× = 0.288
100
Example 6.2 A concrete gravity dam, 150 m long and 90 m wide, lies on a permeable soil
with a coefficient of permeability of 30 × 10–3 mm/s. The head of water is maintained at
30 mm upstream and zero at the tail-end. The soil is underlain by an impervious stratum.
The depth from the base of the dam to the impervious stratum is 40 m. A flow net con-
structed for this condition yielded 7 flow channels and 16 equipotential drops. What is the
seepage loss per day under the dam, considering a two-dimensional flow. Estimate also the
approximate seepage loss under the dam using Darcy’s law directly.
Solution
For a two-dimensional flow,
Nf
q=kH
Nd
M06_PURU1773_01_SE_C06.indd 155 2/12/2013 8:33:30 AM

or
3 ×10−3 7
q= × 30 × × 60 × 60 × 24 = 3.40 m 3 / day / m
1000 16
For the entire length of the dam, q = 3.402 × 150 = 510.3 m3/day.
Using Darcy’s law (one-dimensional flow) directly,
q= kiA
3 ×10−3 30
= × ×( 40 ×150)× 60 × 60 × 24
1000 90
= 518.4 m 3 / day
Example. 6.3 For the dam of Fig. 6.14, draw the flow net and determine the following:
1. the quantity of flow,
2. the seepage pressure in the middle of square B,
3. the uplift pressure at point B, and
4. the exit gradient at point A.
The coefficient of permeability is 4.0 × 10–2 mm/s.
Solution
The flow net is drawn as in Fig. 6.14. Number of flow channels, Nf = 5
2m
Scale
10 m
20 m
Datum
Sheet pile cut-off
II I 16
III
1 15
IV
14
13
2 12 15 m
10 11
9
3 7 8
V 5 6
4
B
Impervious
Fig. 6.14
M06_PURU1773_01_SE_C06.indd 156 2/12/2013 8:33:31 AM

Seepage 157
Number of potential drops, Nd = 16

Head loss, H = 10 m
Potential drop
H 10
Δh = = = 0.625
Nd 16
1. The quantity of flow

Nf
q=kH
Nd
4.0 ×10−2 5
q= ×10 × = 1.25×10−4 m 3 / s / m length
1000 16
2. The potential head at B is

ht = H − nd Δh
= 10 − 6.5× 0.625 = 5.94 m
Seepage pressure ps = ht γ w = 5.94 × 9.81 = 58.27 kN / m 2
3. The uplift pressure head hw = ht – z. Consider the downstream water level as the datum
hw = 5.94 + 11.6 = 17.54 m
2
and uplift pressure head uw = 17.54 × 9.81 = 172.07 kN / m
Δ h 0.625
4. Exit gradient ie = = = 1.04
l 0.60
Example 6.4 A masonry dam 50 m long and overlying an impermeable soil is founded on
a soil with anisotropy in permeability. The upstream water level is 9.6 m, and the tail water
level is 0.6 m above the ground level. The vertical permeability of the soil is 1.39 m/day and
the horizontal permeability is six times the vertical permeability. The flow net drawn on a
transformed section yields five flow channels and eight equipotential lines. Determine the
seepage flow per day.
Solution
The horizontal permeability = 6 ×139 = 8.34 m/day
Effective coefficient of permeability ke = k x k z
or
ke = 1.39× 8.34 = 3.4 m/day
Nf
Seepage flow per day = ke × H × ×(length of dam)
Nd
5
= 3.4 × 9.0 × × 50 = 9556.25 m 3/ day
8
M06_PURU1773_01_SE_C06.indd 157 2/12/2013 8:33:33 AM

Example 6.5 A section through a dam is shown in Fig. 6.15. Plot the distribution of the
uplift pressure on the base of the dam.
11.5 m
10.2 m Datum
7
1 2 3 4 5 6
0
7
50 6
15 m kN/m2 5
100 3 4
1 2
150
0 3 6 9
Scale
Fig. 6.15
Solution
The flow net is drawn as shown in Fig. 6.15.
11.5
Potential drop Δh = = 1.438 m
8
For point 1, the potential head ht = H − nd Δh
or
ht = 11.5 − 1×1.438 = 10.06 m
The datum head z = – 0.9 m

Therefore, uplift pressure head hw = ht − z = 10.06 + 0.9 = 10.96 m
and the uplift pressure at point 1 = 10.96 × 9.81 = 107.5 kN / m 2
Similarly, the uplift pressures at other points are calculated and tabulated as follows. The
uplift pressure diagram is shown in Fig. 6.15.
Point nd ht (m) z (m) hw (m) u (kN/m2)

1 1 10.06 –0.9 10.96 107.5
2 2 8.62 –1.8 10.42 102.2
3 3 7.18 –0.9 8.08 79.27
4 4 5.75 –0.9 6.65 65.24
5 5 4.31 –1.8 6.11 59.94
6 6 2.87 –1.8 4.67 45.81
7 7 1.43 –0.8 2.23 21.88
8 8 0 0 0 0
M06_PURU1773_01_SE_C06.indd 158 2/12/2013 8:33:34 AM

Seepage 159
Example 6.6 Find the factor of safety against piping for the sheet pile wall shown in
Fig. 6.16. The saturated density of the sand is 20.1 kN/m3.
Sheet pile
10 m
A′ B′
12.5 m
25 m
A B
0 5 10
Scale
Fig. 6.16
Solution
The flow net is drawn as shown in Fig. 6.16.
Potential drop
10
Δh = = 1.25 m
8
Pressure head at A
hA = 10 − 4 ×1.25 = 5.0 m
Pressure head at B
hB = 10 − 5.9×1.25 = 2.63 m
5 + 2.63
ha = = 3.82 m
Therefore, 2
Now, depth of penetration D =12.5 m.

Therefore, factor of safety with respect to piping
Dγ ′
Fp =
ha γ w
12.5 (20.1 − 9.81)
= = 3.43
3.82× 9.81
Example 6.7 A filter is required to be provided at the downstream side of a weir. A sieve
analysis conducted on the soil to be protected is as follows.
Sieve no. 1.2 (mm) 600 μm 300 μm 150 μm 75 μm

Per cent finer 96 88 83 23 2
Suggest a suitable grain-size distribution range for the filter.
M06_PURU1773_01_SE_C06.indd 159 2/12/2013 8:33:35 AM

Solution
The grain-size distribution curve of the soil is plotted as shown in Fig. 6.17. From the plot the
following grain sizes are taken.
(D15 )s = 0.12 mm Therefore (D15 )f > 4 × 0.12, i.e., 0.48 mm
(D50 )s = 0.21 mm Therefore (D15 )f < 4 × 0.21, i.e., 5.25 mm
(D85 )s = 0.40 mm Therefore (D15 )f < 5× 0.4 , i.e., 0.48 mm
The grain-size distribution range for the filter is shown in Fig. 6.17.
Grain-size distribution
100 range for filter
Grain-size distribution of
Percentage finer
80
protected soil
60
40
20
0
100 10 1.0 0.1 0.01
Particle size (mm)
Fig. 6.17
POINTS TO REMEMBER
6.1 Quicksand is not a type of sand but a phenomenon caused due to a flow condition.
Quicksand condition is likely to occur at hydraulic gradients of about 1.0.
6.2 The general flow equation for soils is based on the assumptions that the soil medium
is saturated, incompressible, and homogeneous; has isotropic permeability; the flow
is laminar; and the fluid is incompressible.
6.3 Solution of the Laplace equations yields two sets of curves: flow lines which represent the
trajectories of seepage and equipotential lines which represent the lines of equal head.
M06_PURU1773_01_SE_C06.indd 160 2/12/2013 8:33:35 AM

Seepage 161
6.4 The entire pattern of flow lines and equipotential lines is referred to as the flow net.
Thus, a flow net is a graphical representation of the head and direction of seepage at
every point.
6.5 The properties of a flow net are as follows: (i) the flow lines and equipotential lines meet
orthogonally, (ii) the quantity of flow through each channel is the same, and (iii) the
head loss (potential drop) between any two successive equipotential lines is the same.
6.6 The uplift pressure uw, also called the hydrostatic pressure, at any point within a soil
mass is the pressure caused by the piezometric head at that point (i.e., the total head
minus the position head).
6.7 Piping is caused by a high hydraulic gradient at the exit face of the percolating soil
mass. Failures by piping may be due to scour or sub-surface erosion starting down-
stream and propogating inwards, causing an ultimate failure.
6.8 Filter or drain materials are used for preventing piping. Apart from providing weight,
the filters should satisfy two grain-size requirements.
QUESTIONS
Objective Questions
6.1 State whether the following statements are true or false:
(1) In practically all seepage problems, velocity heads are disregarded.
(2) In a flow through a porous medium, lines connecting points of equal total energy
head are termed equipotential lines.
(3) The uplift pressure at any point within a soil mass is independent of the position
of the point.
(4) The flow of water through a soil specimen in a laboratory constant head perme-
ability test is under two-dimensional flow conditions.
(5) The seepage loss through an anisotropic soil medium is less than in an isotropic
medium.
6.2 For a flow under a concrete dam founded on a homogeneous isotropic porous
medium, will the flow net alter (answer yes or no)
(a) If the horizontal permeability is altered?
(b) If the difference in head is changed?
(c) If the shape factor of the net is increased?
(d) If the width of the dam is reduced?
6.3 Piping in soils occur when
(a) The effective pressure becomes zero
(b) A sudden change of permeability takes place
(c) The soil is fissured and cracked
(d) The soil is highly porous
M06_PURU1773_01_SE_C06.indd 161 2/12/2013 8:33:35 AM

6.4 The seepage taking place beneath a long masonry dam founded on pervious soils is
often considered as a _________ flow.
(a) Three-dimensional
(b) One-dimensional
(c) Two-dimensional
(d) Radial
6.5 Seepage flow in a porous medium is determined by the absolute value of
(a) Nf
(b) Nd
(c) Nf / Nd
(d) Nd / Nf
6.6 The velocity potential defined in the Laplace equation is a
(a) Scalar function of space (b) Vector function of space
(c) Scalar function of space and time (d) Vector function of space and time
6.7 The quantity of seepage depends on
(1) The coefficient of permeability (2) The length of the flow path
(3) The differential head across the flow path (4) The number of flow paths
Of these statements,
(a) 1, 2, and 3 are correct (c) 2, 3, and 4 are correct
(b) 3, 4, and 1 are correct (d) All are correct
6.8 Identify the incorrect flow net property
(a) Flow lines and equipotential lines intersect orthogonally.
(b) The quantity of water flowing through each channel is the same.
(c) The potential drop between any two successive equipotential lines is different.
(d) Flow lines and equipotential lines are smooth curves.
6.9 In order to prevent piping, the exit gradient should be
(a) Equal to the critical gradient (c) Greater than the critical gradient
(b) Much less than the critical gradient (d) Not a function of the critical gradient
6.10 Which of the following pairs are correctly matched?
(1) Piping A progressive failure
(2) Piping ratio A filter criterion
(3) Graded filter Material provided to prevent seepage
(4) Quicksand condition When the pore pressure equals the total pressure
(a) 1, 2, and 3 (b) 2, 4, and 1
(c) 2 and 4 (d) 3 and 1
6.11 What do you understand by the mechanism of piping? Explain the methods that are
adopted to increase the factor of safety against piping.
6.12 What constitutes a flow net? State any four methods of obtaining flow net in any given
case.
M06_PURU1773_01_SE_C06.indd 162 2/12/2013 8:33:35 AM

Seepage 163
6.13 State reasons for the quantity of seepage between two successive flow lines being
equal.
6.14 Give reasons for limiting the size of particles used in constructing drainage filters.
6.15 What soil conditions in the foundations are vulnerable to the problem of piping
danger?
6.16 Explain how weighted filters are useful in seepage problems for improving the
stability.
6.17 What methods do you suggest to reduce the exit gradient in the case of flows under
concrete dams?
6.18 Distinguish between seepage pressure and uplift pressure. Which one should be
considered in the design of a masonry weir? Why?
6.19 Discuss the effects of anisotropy and non-homogeneity of a soil on the seepage loss.
EXERCISE PROBLEMS
6.1 A sandy soil collected from an excavation showed void ratios of 0.48 and 0.97 in its
densest and loosest states, respectively. The range of critical hydraulic gradients at
which quicksand conditions might occur is needed to decide the depth of excavation.
Take G = 2.65 and estimate the range.
6.2 Explain the phenomenon of quicksand. What hydraulic head is required to create a
quicksand condition in a non-cohesive soil sample of length = 6 m, void ratio = 0.65,
G =2.65?
6.3 In a vertically upward flow of groundwater (an artesian condition) the hydraulic gra-
dient in a sand mass is 0.95. Check whether a condition for quicksand or erosion could
develop.
6.4 From the flow net shown in Fig. 6.18 find
1. the flow rate through the soil,
2. the water pressure in the middle of square X,
3. the seepage force per unit volume at X, and
4. the factor of safety against piping.
17 m
28 m
Xe
Impervious
10 m
Scale
Fig. 6.18
M06_PURU1773_01_SE_C06.indd 163 2/12/2013 8:33:36 AM

9m
15 m 0.5 m
1.5 m
4.5 m
Sheet pile cut-off
12 m
Impervious
Fig. 6.19
6.5 It is proposed to construct a dam 100 m long on a permeable soil of permeability

0.00152 mm/s. The cross section of the dam is shown in Fig. 6.19. Estimate the quan-
tity of water that will be lost per day by seepage. Also calculate the percentage reduc-
tion in the rate of flow if a 6 m upstream impervious apron is provided.
6.6 It is proposed to design a dam of 150 m length with a pool level of 40 m (Fig. 6.20).
Three designs, all with impervious cut-off walls to decrease seepage, are being consid-
ered. Calculate the seepage per day for each design. Use flow net solutions. The value
of k = 2.5 × 10–8 m/s.
Design Depth of cut-off wall (m)

1 20
2 30
3 45
40 m 70 m
d Sheet pile cut-off

70 m
Impervious shale
Fig. 6.20
6.7 A masonry weir is constructed on a permeable stratum of 6 m thickness and underlain

by an impervious rock (Fig. 6.21). The coefficient of permeability of the soil is 0.54 ×
10–4 m/s. Plot graphically a flow net for the permeable foundation of the weir and
estimate the seepage loss per metre length of the weir. Also compute the hydrostatic
uplift pressure at 1, 1.5, 2, and 4 m from the upstream edge of the floor.
M06_PURU1773_01_SE_C06.indd 164 2/12/2013 8:33:36 AM

Seepage 165
8m d/s Bed level

3m 0.2 m
2m 2m
Sheet pile cut-off
6m
Bed rock
Fig. 6.21
6.8 The cross section of a dam is shown in Fig. 6.22. Make a flow net and determine the
quantity of seepage under the weir. Also plot the distribution of the uplift pressure on
the base of the dam. The coefficient of permeability of the soil is 2.8 × 10–5 m/s.
6.9 A single-row vertical sheet piling penetrates 6 m into a soil of 15 m thickness overly-
ing an impermeable rock. The coefficients of permeability of the soil in the vertical
and horizontal directions are 2 ×10–2 and 4 ×10–2 mm/s, respectively. The depth of
water on one side of the piling is 9 m and on the other side 2 m. Draw a neat sketch of
a flow net and estimate the quantity of seepage in m3/day/m run of piling. Re-plot
the flow net from the transformed section to the natural section.
18 m
12 m
0.6 m
7m 7m
Sheet pile cut-off
14 m
Bed rock
Fig. 6.22
M06_PURU1773_01_SE_C06.indd 165 2/12/2013 8:33:37 AM

6.10 For the dam section shown in Fig. 6.23, construct a flow net if the coefficients of per-
meability in the horizontal and vertical directions are 3.8 ×10–3 and 32.4 ×10–3 mm/s,
respectively. Compute the seepage loss per linear metre of the dam. Compare this
value with the seepage loss beneath the same dam if the soil is assumed to have an
isotropic permeability of 11.6×10–3 mm/s.
10.5 m
27 m
9m
15 m Sheet pile cut-off
Bed rock
Fig. 6.23
6.11 In a tidal estuary, during low tide, the depth of water in front of a sheet pile wall is
5 m and the water table behind the wall lags 3 m behind the tidal level (Fig. 6.24). Plot
the net distribution of water pressure on the piling.
Sheet pile
4m
Water table
3m
4.5 m
6m
15 m
Impervious
Fig. 6.24
M06_PURU1773_01_SE_C06.indd 166 2/12/2013 8:33:38 AM

7
Stress and Stress Distribution in
Soil
CHAPTER HIGHLIGHTS
Stresses at a point – Mohr’s circle – Stress paths – Stress concepts: total stress,
neutral stress, effective stress – Geostatic stresses – Different positions of water
table – Stresses due to surface loads: elastic half space, Boussinesq’s theory,
pressure distribution diagrams, Westergaard equation, types of surface loads,
Newmark’s influence chart, approximate solutions – Contact pressure
7.1 INTRODUCTION
Internal stress develops in a soil mass by the weight of the overburden and due to
external loadings caused by the construction of structures. It is impossible to keep track
of forces acting at different points of a soil mass because of its heterogenic nature, and
thus stress developed over a zone is used. A stress-induced soil is associated with defor-
mation (may be settlement or heave). Depending on the method of application of a load
and the mode of distribution of stresses, the stress developed might strengthen the soil
by expelling pore water pressure or induce a soil mass failure by actuating the stresses.
This chapter and Chapters 8 and 9 discuss the predication of stress and the associated
volume change and strength of soil.
7.2 STRESSES AT A POINT

In most branches of engineering, materials are regarded as a continuum and stresses
and strains are evaluated considering an infinitesimal element having the same proper-
ties as the whole mass. This is generally so in geotechnical engineering when dealing
with soil and rock. The stress at a point within a soil mass has to be viewed as a large
point with representative materials of the whole mass.
M07_PURU1773_01_SE_C07.indd 167 2/12/2013 8:02:44 AM

sz
tzx
tzy txz
tyz sx
txy
tyx
sy x
(a)
y
s3
sl
sl s3
(b) Major and minor principal planes and stresses
sn
1 Unit
s1 cos q cos q tn B
s3 sin q cos q
A q
s1 cos q sin q s1 cos q s3 sin q sin q
s3 sin q sin q
C
(c) Cut element with forces
Fig. 7.1 State of stresses
Consider an incremental element and the stresses acting on the planes to represent
the stress conditions at a point as shown in Fig. 7.1a. Here, σx, σy, and σz are the normal
stresses and τxy, τyz, and τzx are the shear stresses. To satisfy the rotational movement
equilibrium condition, the shear stress acting on orthogonal planes should be zero (i.e.,
τ xy = τ yx , τ yz = τ zy and τ xz = τ zx ).
M07_PURU1773_01_SE_C07.indd 168 2/12/2013 8:02:44 AM

Stress and Stress Distribution in Soil 169
Now consider all the planes passing through the point and locate the planes on which
there are no shear stresses. Such planes also mutually maintain orthogonality and are repre-
sented in Fig. 7.1b.
These normal stresses are called principal stresses and the planes, principal planes. These
three principal stresses are termed the major principal stress, σ1 (the largest stress), the inter-
mediate principal stress, σ2, and minor principal stress, σ3 (the smallest stress). The respective
strains in these three directions may be taken as ε1, ε2 , and ε3. In many practical geotechnical
problems in soils, the principal stresses act vertically and horizontally, e.g., the stress condi-
tion below a horizontal ground surface.
It is convenient to assume one of the principal stresses or principal strains to be zero and
convert the problem to one in a two-dimensional state. These two-dimensional states are
called (i) plane stress (ignoring strain, ε2) or (ii) plane strain (ignoring stress, σ2). Many geo-
technical problems are plane strain problems. Further, compression is considered as positive
and tension as negative, and a shear stress causing a counter-clockwise torque about the
centre of a free body is considered positive.
Now consider the cut element ABC as shown in Fig. 7.1c. Let the plane AB be 1 unit in
length; then AC = cos θ and CB = sin θ. The normal and shear stresses on plane AB can be
determined by resolving the forces parallel and normal to plane AB as
σn = σ1 cos 2 θ + σ3 sin 2 θ
where σn is the normal stress acting on plane AB, that is,

σ1 + σ3 σ1 − σ3
σn = + cos 2θ (7.1)
2 2
and
τ n = (σ1 − σ3 ) sin θ cos θ (7.2)
where τn is the shear stress acting on plane AB.

This shows that the normal and shear stresses on any plane orthogonal to the intermedi-
ate principal plane may be determined from Eqs. 7.1 and 7.2. It may be observed that these
two equations have not included any material properties, but are based merely on the prin-
ciples of mechanics.
From Eq. 7.2 it may be seen that the maximum shear stress value is (σ1 – σ3)/2 for 45°, and
on this plane the normal stress is always (σ1 + σ3)/2.
7.3 MOHR’S CIRCLE

The normal and shear stresses given by Eqs. 7.1 and 7.2 for different values of θ may be pre-
sented graphically on a coordinate system and the locus of these points represents a circle
(Fig. 7.2). This graphical representation of the state of stress in a lucid form is known as a
Mohr circle after Mohr (1882). The graphical representation is a convenient aid in solving
problems.
The circle has a radius of (σ1 − σ3)/2, with the centre on the x- axis with coordinates
[(σ1 + σ3)/2, 0]. Any point on the circle represents σn and τn on some plane. Point D represents
M07_PURU1773_01_SE_C07.indd 169 2/12/2013 8:02:45 AM

Op q tn
F
O
E X
s3
s1 + s3
2 s
n
s1
Fig. 7.2 Mohr’s stress circle
the state of stress on the plane inclined at θ with the major principal plane. Points E and F
represent the major and minor principal stresses, respectively.
The Mohr diagram is an excellent visualization of the orientations of various planes. If,
through the coordinates of σn and τn on the Mohr circle, a line is drawn parallel to the plane
on which these stresses act, this line intersects the Mohr circle at a unique point. If parallels
are drawn from E(σ1, 0) and F(σ3, 0) to the respective planes, these planes pass through the
same unique point. This point is referred to as the origin of planes or pole, Op. Thus, any line
drawn from the pole, parallel to a plane (on which the stresses are needed), intersects the
circle at a point, the coordinates of which represent the normal and shear stresses acting on
that plane.
7.4 STRESS PATHS

Progressive changes in the state of a particular load application can be represented by a
series of Mohr circles. For example, Fig. 7.3a represents successive states as σ1 is increased
with σ3 constant. Such a diagram with several complete stress circles can appear cluttered.
Stress s1 + s3
E q q = Constant
path E 2
D s1 = Constant
D
Shear stress
C s3 = Constant
B C 1 1
A B 1 1
A s s1 = s3 s
Principal
stress
(a) (b) (c)
Fig. 7.3 Stress paths
M07_PURU1773_01_SE_C07.indd 170 2/12/2013 8:02:46 AM

It is convenient to plot only the point of maximum shear stress, and if needed the complete
circle can be reconstructed using such a point. Thus, the locus of points (Fig. 7.3b) on the
Mohr diagram whose coordinates represent the maximum shear stress and the associated
principal stress for the entire stress history is defined as a stress path (Lambe, 1967). For given
principal stresses σ1 and σ3, the coordinates of a point on the stress path are
σ1 + σ3 σ1 − σ3
p= and q =
2 2
Such a plot is referred to as a p–q diagram.

The stress path for σ3 = constant and σ1 increasing is a 45° line (Fig. 7.3b). Figure 7.3 shows
stress paths for different variations of σ1 and σ3. A stress path diagram may be constructed
for total or effective stress conditions.
7.5 EFFECTIVE STRESS CONCEPT

In a natural soil stratum or in man-made earth structures, three conditions based on
moisture content may be visualized, viz., dry, saturated, or partially saturated. All earth
structures or structural foundations may experience one or all of these conditions during
their life span. Thus, the stress conditions present during these stages have to be com-
pletely understood.
7.5.1 Dry Soil

In a dry soil system any stress has to be visualized as the force in the mineral skeleton per
unit area of the soil. Accordingly, a normal stress can be defined as the sum of the normal
components of the forces (ΣN) over a plane divided by the area of the plane (A). Let us con-
sider a dry soil medium of unit width (Fig. 7.4a) and unit length (normal to the plane of the
paper) as the surface over which the normal stress is to be computed. The soil at the section
would have attained equilibrium due to the overburden pressure, σ1, and this overburden
pressure would have changed the mechanical properties of the soil. Thus, this pressure may
be termed the effective pressure or effective stress, σ′.
Ground surface
B B′
1
s = s¢
z Grains
contact
A Soil
A+ grains
(a) (b)
Fig. 7.4 Vertical stress in dry soil
M07_PURU1773_01_SE_C07.indd 171 2/12/2013 8:02:46 AM

Weight of prism of soil = γd (z × 1 × 1)

Area along section AA′ = 1
Therefore, the total stress is
γd z
σ= (7.3)
1
Thus the effective stress σ ′ = σ. (7.4)
Figure 7.4b shows a close-up view of the particles on plane AA′.
7.5.2 Saturated Soil

Consider a saturated soil condition as shown in Fig. 7.5. This condition is similar to the pre-
vious case, but the voids are completely filled with water (Fig. 7.5a). As before, the overbur-
den pressure, σ, is given as
σ = γ sat z (7.5)
This total normal stress acting on section AA′ has two components, one of which acts on
the pore water and the other on the soil skeleton. The component on the water acts equally
in all directions and does not cause any change in the mechanical properties of the soil and
is known as the neutral stress or pore water pressure, uw.
The remaining part
σ ′ = σ − uw (7.6)
is that component of the total stress which rests entirely on the soil skeleton of the soil. Thus,
only this component of the total stress will cause a change in the mechanical properties and,
hence, is known as the effective stress.
This classical equation was put forth by Terzaghi (1925, 1943).
σ ′ = γ sat z − uw
σ ′ = γ sat z − zγ w
= z(γ sat − γ w )
σ ′ = zγ ′ (7.7)
Ground
Water surface
table
B B′
1 s = s¢+uw
z
uw s¢
Water
A
A′
Grains
contact
(a) (b)
Fig. 7.5 Vertical stress in saturated soil
M07_PURU1773_01_SE_C07.indd 172 2/12/2013 8:02:47 AM

Let A be the total area (Fig. 7.5 b), Aw the area of water in contact with the total area minus
the mineral contact area), and Ac the mineral contact area. Then
Aw + Ac = A
Aw Ac
+ =1
A A
aw + ac = 1 (7.8)
where aw = Aw /A is the ratio of area of water contact to the total area and ac = Ac/A the ratio
of area of mineral contact to the total area.
Actually, the pore water pressure acts only on aw, rather than on the complete area.
Hence,
σ ′ = σ − aw uw
or
σ ′ = σ − (1 − ac ) uw
It has been widely accepted that ac is negligible (Lambe and Whitman, 1979; Reosenqvist, 1959).
Hence
ac = 0 and σ ′ = σ − uw
The above relationship is generally valid. This expression indirectly assumes that no other
stresses except the external applied stress, σ, and the pore water pressure, uw, exist within
the system. Further, the attractive and repulsive forces between particles are not accounted
for in this expression.
7.5.3 Partially Saturated Soil

Consider the situation shown in Fig. 7.6a, which represents a partially saturated soil. Again,
the total stress
σ=γz (7.9)
Ground
surface
B B′
1
s =s ′+u*
z
Air uw s′
Water
A′ Grains
A contact
u* = ua aa + uw uw
Water
table (a) (b)
Fig. 7.6 Vertical stress in partially saturated soil
M07_PURU1773_01_SE_C07.indd 173 2/12/2013 8:02:48 AM

Here the contact areas may be considered as (Fig. 7.6b)
ac + aa + aw = 1 (7.10)
where aa is the ratio of area of air contact to the total area.

Again, considering ac = 0 and aa + aw = 1, σ ′ can be defined as
σ′ = σ − u * (7.11)
where u* = ua aa + uw aw and ua is the pressure in the gas and vapour phase; that is,
u* = ua + aw (uw − ua ) (7.12)
σ ′ = σ − [ua − aw (ua − uw )] (7.13a)
Bishop (1959) based on his intuition replaced aw in the above expression by an empirical
parameter χ, and thus
σ ′ = σ − [ua − χ(ua − uw )] (7.13b)
The parameter χ has to be determined experimentally (Bishop et al., 1960). It is believed

that the parameter depends on the degree of saturation, i.e., it has a value of 1.0 for saturated
soil and 0 for dry soil. χ may have different values at a given degree of saturation in relation
to the shear strength and volume change. Further, in certain situations where the stress his-
tory plays a more important role than does the degree of saturation, the factor χ has been
found to take negative values and values greater than unity. Thus, the factor χ is purely an
empirical one and may to some extent depend on the degree of saturation.
There was another school of thought which related the total external pressure to the inter-
nal stress in the particulate soil system (Lambe, 1960).
σ ′ = σ ac + ua aa + uw aw + R′ − A′ (7.13c)
That is,
σ ′ = σ ac + u * +R′ − A′ (7.13d)
where σ is the mineral–mineral contact stress, R′ the total inter-particle repulsion divided
by total inter-particle area, and A′ the total inter-particle attraction divided by total inter-
particle area.
From Eq. 7.13d the conventional effective stress, σ′, can be written as
σ ′ = σ − u* = σ ac + R′ − A′ (7.13e)
The above expression shows an increase in effective stress with an increase in the repul-
sive forces and a decrease in the attractive forces. This is contrary to the general physical
behaviour in a particulate soil system.
Having found the anomaly in the above expression, Sridharan (1968) rewrote Lambe’s
equation in the following manner.
c = σ ac = σ − uw − ua − R′ + A′ (7.13f)
where c is the effective contact stress, uw the effective pore water pressure, and ua the
effective pore air pressure.
M07_PURU1773_01_SE_C07.indd 174 2/12/2013 8:02:51 AM

This can be represented as a combination of two effective stresses as
c = σ ′ + σ ′′ (7.13g)
where
σ ′ = conventional effective stress
= σ − u* = σ − u a − u w
= σ − uw (for saturated soil)
σ″ = intrinsic effective stress
= A ′ − R′
The effective contact stress, c , has been defined as the modified effective stress. Equation 7.13f
agrees with the general behaviour of a soil system showing an increase in the effective stress
with an increase in the attractive forces and a decrease in the repulsive forces. The applica-
tion of this expression to predicting the volume change and shear strength behaviour of clays
has been hypothesized by Sridharan and Venkatappa Rao (1973, 1979). In a soil system with
low A′ – R′ forces, such as in granular soils, the expression for c tends to σ′, the conventional
effective stress.
7.6 GEOSTATIC STRESSES

Stresses within a soil mass are caused by the self-weight of the soil and the external applied
load. The stress patterns due to these effects are complicated. The magnitude of the subsur-
face stresses is affected by the presence of groundwater. Stresses induced by the overburden
pressure are called geostatic stresses. This situation gives rise to simple stress calculations
when the ground surface is horizontal, and there is no marked variation of the soil proper-
ties in the horizontal direction.
Hence, the vertical stress caused by the soil at a point below the surface is equal to the
weight of the soil lying directly above the point. Considering the unit weight, γ, to be con-
stant with depth, the vertical stress, σv due to overburden at a depth z from the ground
surface is given as (Fig. 7.7.a)
σv = γ z (7.14)
G.S G.S
z1
Layer 1
g1z1
z2
z z Layer 2
g1z1 + g2z2
z3
Layer 3
Yz g1z1 + g2z2+ g3z3
(a) Uniform soil (b) Layered soil
Fig. 7.7 Vertical stress distribution in uniform and layered soils
M07_PURU1773_01_SE_C07.indd 175 2/12/2013 8:02:51 AM

As the ground surface is horizontal, there are no shear stresses upon the horizontal or
vertical planes. The vertical stress increases with depth.* If the soil stratum is layered with
different total unit weights, then the vertical stress at a depth z will be equal to the total
weight of the individual soil layers (Fig. 7.7b); that is,
σ v = γ1 z1 + γ 2 z2 + γ 3 z3 = Σγ z (7.15)
The situation will be different depending on the groundwater position. In general, five
situations may be recognized for the static water condition, and they are explained below.
7.6.1 Case 1 – Soil Entirely Dry

Consider a level ground with the water table at a lower depth (Fig. 7.8a). The total stress, σv,
at any depth z is given as
σ v = γd z (7.16)
uw = 0 (7.17)
σ′v = σ v (7.18)
where σ′v is the effective vertical stress due to overburden. The total, neutral, and effective
stress variations up to a depth z are shown in Fig. 7.8a.
7.6.2 Case 2 – Moist Soil

This is a situation of partially saturated soil wherein it is difficult to predict the neutral
pressure distribution. Thus, this condition is treated as in Case 1 but with γ instead of γd.
Hence,
σv = γ z (7.19)
uw = 0 (7.20)
σ v′ = σ v (7.21)
The stress distributions up to a depth z are shown in Fig. 7.8b.
7.6.3 Case 3 – Completely Submerged Soil with

Water Table at Ground Surface
In this situation, the total stress is governed by the saturated unit weight of the soil. Thus,
σ v = γ sat z (7.22)
*But the unit weight is not constant. It generally increases with depth. Under such conditions, σv is
z
given as σ v = ∫ γ dz .
0
M07_PURU1773_01_SE_C07.indd 176 2/12/2013 8:02:53 AM

(a)
H
gdH gdH
z Dr y soil
Ggw
gd =
1+e
gdz O gdz
sv uw s¢
(b) G.S.
H
gH gH
z
Moist soil
(G + eSr) gw
g =
1+e
gz O gz
sv uw s v′
Water
table
(c) G.S.
Submerged H
gsat H gw H g ′H
Soil
z
G+e g
gsat = w
1+e
G–e g
g¢= w
1+e
gsat z gw z g ′z
G.S. gwz gwz

(d)
H
Saturated soil gsat H
(by capillary action) gw z + g ′ H
z gw (z − H)
Water G+e g
table gsat = w
1+e g ′z
gsat z gsat z
sv uw s v′
Fig. 7.8 Vertical stress distribution in (a) dry soil, (b) moist soil, (c) submerged soil, and (d) saturated
soil
M07_PURU1773_01_SE_C07.indd 177 2/12/2013 8:02:54 AM

G.S.
(e)
Ggw Dr y
gd = H gdH
1+e soil gwHc
Saturated soil gd H +gw Hc

z
(by capillary action)
Water Hc
table G+e g gsatHc
gsat = w
1+e
gdH +gsatHc gdH+gsatHc
sv uw s v¢
Fig. 7.8 (e) Vertical stress distribution in partially saturated soil
uw = γ w z (7.23)
σ v′ = σ v − uw = (γ sat − γ w )z
σ v′ = γ ′ z (7.24)
The stress distributions are shown in Fig. 7.8c.
7.6.4 Case 4 – Completely Saturated by Capacity Action Above

Water Table But No Flow
Consider the stress conditions up to the water table from the ground surface. The neutral
pressure is zero at the water table level. Since the distribution of pressure in continuous col-
umns of water is hydrostatic and the pressure at the water table level is zero, the pressure in
the water above the water table will be less than atmospheric or negative. Hence, the stress
condition at the ground surface is
σv = 0 (7.25)
uw = −γ w z (7.26)
σ ′ v = σ v− u w = γ w z (7.27)
The stress condition at the water table level is
σ v = γ sat z (7.28)
uw = 0 (7.29)
σ ′ v = σ v− u w = γ sat z (7.30)
The condition shows that the soil at the ground surface is under stress (Fig. 7.8d). This
situation also explains why damp sand, as on a beach, is hard and dry sand loose on the
surface.
M07_PURU1773_01_SE_C07.indd 178 2/12/2013 8:02:55 AM

7.6.5 Case 5 – Same as Condition 4 Except that the

Height of Capillary Rise Is Less than z
Consider a situation in which the capillary rise is less than z and equals (z – H) = H c and soil
above this level is dry up to the ground surface. The stresses at the ground surface are zero.
The stresses at a depth H from the ground surface are given as
σ v = γd H (7.31)
uw = −γ w Hc (7.32)
σ ′ v = σ v− u w = γ d H − (−γ w Hc ) (7.33)
σ ′ v = γ d H + γ w Hc
That is, the stresses at a depth z from the ground surface are given as
σ v = γd H + γ sat Hc (7.34)
uw = 0 (7.35)
σ ′ v = σ v− u w = γ d H + γ sat Hc (7.36)
The stress distributions for the conditions explained above are given in Fig. 7.8e.
For calculation of the stresses below the water table in Cases 4 and 5, Case 3 is combined
with Case 1.
In the design of structures, such as retaining walls, sheeting, and pile foundations, the
geostatic stresses acting in the horizontal direction are needed. The horizontal stress, σh, is
a function of the vertical stress at the point under consideration. The ratio of the horizontal
or lateral stress to the vertical stress is represented by a coefficient K, termed the coefficient of
lateral pressure, that is,
σh
K= (7.37)
σv
σh = Kσ v (7.38)
Depending on the stress history of the soil medium, K has a wide range of values.
A detailed discussion of the parameter K is given in Chapter 11.
7.7 STRESSES DUE TO SURFACE LOADS

7.7.1 Elastic Half-Space
A soil medium with a horizontal ground surface extending laterally to infinite length and
downwards from the horizontal is called a semi-infinite medium or semi-infinite half-space. If
such a medium is assumed to be homogeneous, isotropic, and elastic, then it is called an
elastic half-space. The theoretical treatment for determining the stresses in such a medium
involves the theory of elasticity.
M07_PURU1773_01_SE_C07.indd 179 2/12/2013 8:02:57 AM

7.7.2 Elastic Properties of Soil

The relationship between the deformation or strain with stress is important in understand-
ing the behaviour of any material. For some building materials, Hooke’s law provides an
useful approximation between the stress and strain. But this law does not necessarily hold
good for soils, as their behaviour in general is non-linear and not perfectly elastic. It is
observed that the entire stress–strain graph is a curve, unlike in steel, where the initial por-
tion is predominantly a straight line. The slope of the initial portion of the curve is defined
as the stress–strain modulus, E.
The determination of the stress–strain modulus is a much more critical problem. The most
common method of computing the stress–strain modulus is to use the initial tangent value
or the slope of the stress–strain curve at the origin (Fig. 7.9). This is referred to as the initial
tangent modulus (Ei). The stress–strain modulus is also taken as the initial secant modulus
(Es), which is obtained using the origin and the secant line intercept at a stress level of 1/3
to 1/2 of the ultimate or failure stress. In another procedure, a cyclic loading test is done
and the tangent modulus after the fifth or sixth cycle is taken. The tangent modulus for the
re-loading curves is called the re-load modulus (Er).
The initial tangent modulus is quite often used to represent the stress–strain modulus of a
soil. The two main reasons for this choice are that the soil is elastic only near the origin and
the region near the origin is nearly the same for different test plots. However, it has been
recommended by many researchers that the re-load modulus is a better choice. The re-load
modulus is generally higher than the initial tangent modulus of the first cycle due to the
effect of strain hardening.
The other two important elastic properties of soil are the shear modulus, G, and Poisson’s
ratio, ν. The shear modulus is used in soil dynamic problems to estimate amplitudes of vibra-
tions. The stress–strain modulus and Poisson’s ratio are used in the evaluation of stresses
and settlements.
Typical values of the stress–strain modulus and Poisson’s ratio are given in Tables 7.1 and
7.2, respectively (Bowles, 1982). Table 7.2 shows that ν has a very narrow range of variation.
An accurate prediction of ν is neither possible nor necessary. Fortunately, the value of ν usu-
ally has a relatively small effect on engineering predictions.
Ei Er
1 1
Deviator stress
Ei Initial tangent modulus

Es Initial secant modulus
Es
Er Re-load modulus
1
Axial strain
Fig. 7.9 Definitions of modulus
M07_PURU1773_01_SE_C07.indd 180 2/12/2013 8:02:57 AM

Table 7.1 Typical range of values for stress–strain modulus, Es for soils
Soil Es (kN/m2 ×103)
Clay
Very soft 2–15
Soft 5–25
Medium 15–50
Hard 50–100
Sandy 25–250
Glacial till
Loose 10–153
Dense 144–720
Very dense 478–1,440
Loess sand
Silty 7–21
Loose 10–24
Dense 48–81
Sand and gravel
Loose 48–144
Dense 96–192
Shale 144–14,400
Silt 2–20
Table 7.2 Typical range of values for Poisson’s ratio
Soil ν
Clay, saturated 0.4–0.5

Clay, unsaturated 0.1–0.3
Sandy clay 0.2–0.3
Silt 0.3–0.35
Sand 0.15–0.40
Loess 0.10–0.30
Soil does not completely fulfil the basic assumptions of homogeneity and isotropy made
in the elastic half-space concept. However, the civil engineer has to apply the results of this
theory with judgement.
7.7.3 Boussinesq’s Theory

One of the methods for computing stresses based on the theory of elasticity was given by
Boussinesq (1885). Boussinesq assumed a weightless, elastic half-space, and given compo-
nents of stresses caused by a vertical-point-surface load (Fig. 7.10). The stress components
M07_PURU1773_01_SE_C07.indd 181 2/12/2013 8:02:57 AM

q
z
R
sz
tzx
trz
r
sr
sq
Fig. 7.10 Stresses at a point due to point load
due to the surface load, viz., the vertical stress, σz, radial stress, σr, circumferential stress, σθ,
and shear stress, τrz, are given as (using polar coordinates, r, θ, and z),
3Q ⎡ 1 ⎤ 5/ 2
σz = ⎢ ⎥ (7.39)
2πz 2 ⎢ 1 + ( r / z )2 ⎥
⎣ ⎦
Q ⎡⎢ 3 r 2 z 1 − 2ν ⎤
⎥
σr = − (7.40)
2π ⎢⎣ (r 2 / z 2 )5/ 2 r 2 + z 2 + z(r 2 + z 2 )1/ 2 ⎥⎦
⎡ ⎤
(1 − 2ν) ⎢⎢ 2 2 3 / 2 − ⎥
Q z 1
σθ = ⎥ (7.41)
2π ⎢ (r / z ) 2 2 2 2 ⎥
r + z + z (r + z ) ⎦
⎣
and
3Q ⎡⎢ rz 2 ⎤
⎥
τ rz = (7.42)
2π ⎢⎣ 1 + (r / z)5/ 2 ⎥⎦
Equation 7.39 is most frequently used in practice. This equation represents a high stress
beneath the point of load application (z = 0) and a decrease in stress with increase in depth.
Further, the stress decreases with increasing distance from the point of load application.
It should also be observed that it does not depend on the elastic or other properties of the
soil, i.e., it is independent of the material content of the medium (i.e., clay or sand). This
equation can be written in terms of an influence factor, called Boussinesq’s vertical stress
coefficient, NB , when
3Q ⎡⎢ 1 ⎤ 5/ 2
NB = ⎥ (7.43)
2π ⎢⎣ 1 + (r 2 / z 2 ) ⎥⎦
Then
Q
σz = 2 N B (7.44)
z
M07_PURU1773_01_SE_C07.indd 182 2/12/2013 8:02:59 AM

0.5
0.4
NB
NB 0.3
or
Nw Nw
0.2
sz = N Q2
Z
0.1
0 0.5 1.0 1.5 2.0 2.5

r/z
Fig. 7.11 Boussinesq and Westergaard coefficients for a concentrated load
Equation 7.44 shows that the vertical stress is (i) directly proportional to the load,
(ii) inversely proportional to the depth squared, and (iii) proportional to some function of
the ratio r/z.
The solid line in Fig. 7.11 shows a variation of NB with the ratio r/z. Table 7.3 presents
Boussinesq’s vertical stress coefficients.
Table 7.3 Boussinesq coefficients
r/z NB r/z NB r/z NB r/z NB r/z NB r/z NB
0.00 0.4775 0.34 0.2733 84 0.0844 1.35 0.0251 1.85 0.0085 2.35 0.0034
0.01 0.4773 0.35 0.2679 85 0.0823 1.36 0.0245 1.86 0.0084 2.36 0.0033
0.03 0.4764 0.37 0.2571 87 0.0783 1.38 0.0234 1.88 0.0081 2.38 0.0032
0.04 0.4756 0.38 0.2518 88 0.0764 1.39 0.0229 1.89 0.0079 2.39 0.0032
0.05 0.4756 0.39 0.2466 89 0.0744 1.40 0.0224 1.90 0.0078 2.40 0.0031
0.06 0.4723 0.40 0.2414 90 0.0727 1.41 0.0219 1.91 0.0076 2.41 0.0031
0.07 0.4717 0.41 0.2363 91 0.0709 1.42 0.0214 1.92 0.0075 2.42 0.0030
0.08 0.4699 0.42 0.2313 92 0.0691 1.43 0.0209 1.93 0.0073 2.43 0.0030
0.09 0.4679 0.43 0.2263 93 0.0674 1.44 0.0204 1.94 0.0072 2.44 0.0029
0.10 0.4657 0.44 0.2214 94 0.0658 1.45 0.0200 1.95 0.0070 2.45 0.0029
0.11 0.4633 0.45 0.2165 95 0.0641 1.46 0.0195 1.96 0.0069 2.46 0.0028
0.12 0.4607 0.46 0.2117 96 0.0626 1.47 0.0191 1.97 0.0068 2.47 0.0028
0.13 0.4579 0.47 0.2070 97 0.0610 1.48 0.0187 1.98 0.0066 2.48 0.0027
0.14 0.4548 0.48 0.2040 98 0.0595 1.49 0.0183 1.99 0.0065 2.49 0.0027
0.15 0.4516 0.49 0.1978 99 0.0581 1.50 0.0171 2.00 0.0064 2.50 0.0026
0.16 0.4482 0.50 0.1934 1.00 0.0567 1.51 0.0171 2.01 0.0063 2.51 0.0026
0.17 0.446 0.51 0.1889 1.01 0.0553 1.52 0.0171 2.02 0.0062 2.52 0.0025
0.18 0.4409 0.52 0.1846 1.02 0.0539 1.53 0.0167 2.03 0.0060 2.53 0.0025
Table 7.3 Contd.
M07_PURU1773_01_SE_C07.indd 183 2/12/2013 8:02:59 AM

Table 7.3 Contd.
r/z NB r/z NB r/z NB r/z NB r/z NB r/z NB
0.19 0.4370 0.53 0.1804 1.03 0.0526 1.54 0.0163 2.04 0.0059 2.54 0.0025
0.20 0.4329 0.54 0.1762 1.04 0.0513 1.55 0.0160 2.05 0.0058 2.55 0.0024
0.21 0.4286 0.55 0.1721 1.05 0.0501 1.56 0.0157 2.06 0.0057 … …
0.22 0.4242 0.56 0.1681 1.06 0.0489 1.57 0.0153 2.07 0.0056 2.57 0.0023
0.23 0.4197 0.57 0.1641 1.07 0.0477 1.58 0.0150 2.08 0.0055 … …
0.24 0.4151 0.58 0.1603 1.08 0.0466 1.59 0.0147 2.09 0.0054 2.59 0.0023
0.25 0.4103 0.59 0.1565 1.09 0.0454 1.60 0.0144 2.10 0.0053 … …
0.26 0.4054 0.60 0.1527 1.10 0.443 1.61 0.0141 2.11 0.0052 2.61 0.0022
0.27 0.4004 0.61 0.1491 1.12 0.0433 1.62 0.0138 2.12 0.0051 … …
0.28 0.3954 0.62 0.1455 1.13 0.0422 1.63 0.0135 2.13 0.0050 2.63 0.0021
0.29 0.3902 0.63 0.1420 1.14 0.0412 1.64 0.0132 2.14 0.0049 … …
0.30 0.3849 0.64 0.1386 1.15 0.0402 1.65 0.0129 2.15 0.0048 2.65 0.0021
0.31 0.3796 0.65 0.1353 1.16 0.0393 1.66 0.0126 2.16 0.0047 … …
0.32 0.3742 0.66 0.1320 1.17 0.0384 1.67 0.0124 2.17 0.0047 2.67 0.0019
0.33 0.3687 0.67 0.1288 1.18 0.0374 1.68 0.0121 2.18 0.0046 … …
0.34 0.3632 0.68 0.1257 1.19 0.0365 1.69 0.0119 2.19 0.0045 2.69 0.0017
0.35 0.3577 0.69 0.1226 1.20 0.0357 1.70 0.0116 2.20 0.0044 … …
0.36 0.3521 0.70 0.1196 1.21 0.0348 1.71 0.0114 2.21 0.0043 2.71 0.0015
0.37 0.3465 0.71 0.1166 1.22 0.0340 1.72 0.0112 2.22 0.0043 … …
0.38 0.3408 0.72 0.1138 1.23 0.0332 1.73 0.0109 2.23 0.0042 2.73 0.0013
0.39 0.3351 0.73 0.1110 1.24 0.0324 1.74 0.0107 2.24 0.0041 … …
0.40 0.3294 0.74 0.1083 1.25 0.0317 1.75 0.0105 2.25 0.0040 2.75 0.0011
0.41 0.3238 0.75 0.1057 1.26 0.0309 1.76 0.0103 2.26 0.0040 … …
0.42 0.3181 0.76 0.1031 1.27 0.0302 1.77 0.0101 2.27 0.0039 2.77 0.0009
0.43 0.3124 0.77 0.1005 1.28 0.0295 1.78 0.0099 2.28 0.0038 … …
0.44 0.3068 0.78 0.0981 1.29 0.0288 1.79 0.0097 2.29 0.0038 2.79 0.0007
0.45 0.3011 0.79 0.0956 1.30 0.0282 1.80 0.0095 2.30 0.0037 … …
0.46 0.2955 0.80 0.0933 1.31 0.0275 1.81 0.0093 2.31 0.0036 2.81 0.0005
0.47 0.2899 0.81 0.0910 1.32 0.0269 1.82 0.0091 2.32 0.0036 … …
0.48 0.2843 0.82 0.0887 1.33 0.0263 1.83 0.0089 2.33 0.0035 2.83 0.0003
0.49 0.2788 0.83 0.0865 1.34 0.0257 1.84 0.0087 2.34 0.0034 … …
2.85 0.0001
6.15 0.0001
7.7.4 Pressure Distribution Diagrams

Boussinesq’s vertical stress equation may be used to draw three types of pressure distribu-
tion diagrams (Fig. 7.12). They are
1. the stress isobar,
2. the vertical stress distribution on a horizontal plane at a depth of z below the ground
surface, and
3. the vertical stress distribution on a vertical plane at a distance of r from the load point.
M07_PURU1773_01_SE_C07.indd 184 2/12/2013 8:03:00 AM

Q
r
G.S.
Vertical stress
Vertical stress z1 distribution
distribution on
with depth
a horizontal
plane z2
Stress
isobar
Fig. 7.12 Stresses at a point due to point load
The stress isobar is a stress contour connecting all points of equal stress below the ground
surface. There are many isobars for a given load system. The stress isobar is also referred to
as bulb of pressure or pressure bulb. The soil mass bounded within a pressure bulb furnishes
the support power of a footing.
The vertical stress distribution on a horizontal plane at a depth z1 from the ground
surface is obtained by varying r. The magnitude of the vertical stress along the load-line
decreases with an increase in depth and this is reflected in the distribution diagram at a
depth z2 (Fig. 7.12).
The vertical stress distribution on a vertical plane at a distance r from the load points is
obtained by varying z. The diagram represents a maximum value at a depth nearer to the
ground surface, which decreases with depth. The magnitude of the maximum value will
decrease with increasing distance from the load point (Fig. 7.12).
7.7.5 Westergaard Equation

Some fine-grained soils are interspersed with thin lenses of coarse-grained material that par-
tially prevent lateral deformation of the soil. Such a situation represents the non-homogeneous
condition. Westergaard (1938) suggested a solution to such a material by considering an elastic
medium in which the lateral strain was assumed to be zero. As some of the soils, e.g., sedimen-
tary soils, are of this type, Westergaard’s solution may be taken as a better approximation for
such soils than compared with that proposed by Boussinesq for homogeneous soils. Wester-
gaard’s expression for the vertical stress is given as
Q (1 − 2ν)
σz = 2
1/ 2π 3/2
(7.45)
z (2 − 2ν)/ ⎡⎢(1 − 2ν)/(2 − 2ν) + (r / z)2 ⎤⎥
⎣ ⎦
Westergaard further considered obtaining a maximum by letting ν = 0; hence, Eq. 7.45
reduces to
Q 1/ π
σz = (7.46)
z ⎡1 + 2(r / z)2 ⎤ 3 / 2
2
⎣⎢ ⎦⎥
M07_PURU1773_01_SE_C07.indd 185 2/12/2013 8:03:00 AM

This can be rewritten in the form

Q
σz = Nw (7.47)
z2
where
1/ π
Nw = 3/2
⎡ 2⎤
⎢⎣1 + (r / z) ⎥⎦
This expression resembles Eq. 7.44 of Boussinesq and a comparison of Nw is made with NB
in Fig. 7.11. The stresses given by Westergaard’s solution range down to two-thirds of those
of Boussinesq’s solution.
7.7.6 Types of Surface Loads

In practice, the loads are applied over finite areas and never as point loads. Boussinesq’s
point load solution may be conveniently integrated to obtain stresses due to surface loads
distributed over a particular area.
Line Loads of Infinite Length. Stresses at a point A due to a line load q per unit length on
the surface are given as (Fig. 7.13)
2q z3
σz = (7.48)
π ( x + z 2 )2
2
and
2q x z2
τ rz = (7.49)
π ( x 2 + z 2 )2
The lateral pressure on earth-retaining structures caused by a line load (e.g., a railway) on
the surface of the backfill may be computed using Eq. 7.48.
q
∞
G.S.
q
∞
z
sx sz
A
x
Fig. 7.13 Uniformly distributed infinite linear load
M07_PURU1773_01_SE_C07.indd 186 2/12/2013 8:03:01 AM

B
q
G.S.
a b sz
sx A
Fig. 7.14 Uniformly distributed infinite strip load
Strip Area Carrying Uniform Pressure. A strip of width B and infinite length, loaded
with a uniform pressure, is shown in Fig. 7.14 (similar to the pressure of a wall footing). The
stresses at point A are given as
q
σz = [α + sin α cos(α + 2β )] (7.50)
π
and
q
τ rz = [sin α sin(α + 2β )] (7.51)
π
A plot of contours of equal vertical stresses is shown in Fig. 7.15 for different stress ratios.
As explained earlier, this enables one to fix the depth of the stress influence. The distribution
of stress beneath an uniform strip load is important in estimating settlements.
Strip Area of Triangular Shape. A triangular strip area carrying a linearly increasing
pressure over width B and of infinite length is shown in Fig. 7.16a. The vertical stress at a
point A due to such a surface load is
q⎛x 1 ⎞
σ z = ⎜⎜⎜ α − sin 2β ⎟⎟⎟ (7.52)
π ⎝B 2 ⎠
B q
Width
0.95q
B/2
0.90q
0.0
0.80q 1q
0.70q
B/2
0.
0.60q
05
q
0.50q
B/2
0.40q 0.10q
0.30q 0.20q
Fig. 7.15 Bulbs of vertical pressure under uniform strip load
M07_PURU1773_01_SE_C07.indd 187 2/12/2013 8:03:03 AM

x
B
B
q q
G.S. G.S.
R2
R2
R1 z
R1 z
b
a a
sz sz
(a) (b)
sx A
Fig. 7.16 Uniformly distributed infinite triangular load
The vertical stress beneath the vertical face (Fig. 7.16b) is obtained by making β = 0 and
x = B; thus,
q
σz = α (7.53)
π
The shear stress is given as
q⎛ z ⎞
τ xz = ⎜⎜⎜1 + cos 2β − 2 α⎟⎟⎟ (7.54)
π⎝ B ⎠
For a symmertrically distributed triangular load (Fig. 7.17a), the stresses are
2q ⎡⎢ B R R ⎤
σz = ⎢ (α1 + α2 ) + x(α1 − α2 ) − 2 z log e 1 2 2 ⎥⎥ (7.55)
πB ⎢⎣ 2 R0 ⎥⎦
B/2 B/2
b/2 b/2
B/2 B/2
q1
q = q1 + q2
Q q2
G.S. G.S.
R′1 R′0
R ′1 R ′2
a′1
R0
R1 R2
z R1 a a′2 R2
z
a1 a2 1 a2
A A
(a) (b)
Fig. 7.17 Symmetrically distributed triangular load
M07_PURU1773_01_SE_C07.indd 188 2/12/2013 8:03:04 AM

B /2 B /2
q
0.9q
0.05q
0.8q
0.7q
B /2
0.6q
0.5q 0.1q
0.4q
0.2q
0.3q
Fig. 7.18 Bulbs of vertical pressure under symmetrical triangular load
and
2qz
τ rz = (α1 − α2 ) (7.56)
πB
Equation 7.55 is of use in the estimation of settlement of embankments. An embankment

section may be considered as the difference between two triangles of equal angles but of
unequal base width (Fig. 7.17b). The stress beneath such an embankment may be obtained
by subtracting the stresses due to the small triangle from those due to the large triangle.
The pressure bulbs under a triangular strip load are shown in Fig. 7.18.
Strip Area Loaded with Embankment Loading. An increase in vertical stress in a soil
mass due to embankment type loading may be handled using the method of superposition.
One half of the embankment is shown in Fig. 7.19a as a half-sectional elevation. The
vertical stress at A due to this loading is equal to the stress caused at A by the large triangle
(Fig. 7.19b) minus the stress caused at A by the small triangle (Fig. 7.19c).
Applying Eq. 7.53 to Fig. 7.19b, the vertical stress
q + (b / a)q
(σ z )1 = (α1 + α2 ) (7.57)
π
a b b q
a
q q b q
a
a b b
a2 z
a2 z z
a1 a1 + a2
A
(a) A
(b) A
(c)
Fig. 7.19 Vertical stress due to embankment loading
M07_PURU1773_01_SE_C07.indd 189 2/12/2013 8:03:05 AM

Similarly, for the triangle in Fig. 7.19c the vertical stress

bq
(σ z )2 = α2 (7.58)
aπ
Therefore, the stress due to embankment loading
⎡ q + (b / a)q bq ⎤
σ z = (σ z )1 − (σ z )2 = ⎢ (α1 + α2 ) − α2 ⎥ (7.59)
⎢⎣ π aπ ⎥⎦
Therefore,
q ⎡⎛ a + b ⎞⎟ b ⎤
σz = ⎢⎜⎜ ⎟⎟ (α1 + α2 ) − α2 ⎥ (7.60)
⎢
π⎣ a⎜
⎝ ⎠ a ⎥⎦
or sz = NEq, where
1 ⎡⎛⎜ a + b ⎞⎟ b ⎤
NE = ⎢⎜ ⎟ (α + α ) − α2 ⎥ (7.61)
π ⎢⎣⎜⎝ a ⎟⎠ a ⎥⎦
1 2
1 ⎛a b⎞
= f ⎜⎜⎜ , ⎟⎟⎟ (7.62)
π ⎝z z⎠
The values of the influence factor for various a/z and b/z are given in Fig. 7.20
(Osterberg, 1957).
0.50 3.0
2.0
1.6
0.45 1.4
1.2
1.0
0.40 0.9
0.8
0.7
0.35
0.6
0.30 0.5
NE
0.4
0.25
0.3
0.20
0.2
0.15
0.10 0.1
b/z = 0
0.05
0
0.001 0.01 0.1 10
a/z
Fig. 7.20 Influence chart of embankment loading (Source: Osterberg, 1957)
M07_PURU1773_01_SE_C07.indd 190 2/12/2013 8:03:07 AM

2a
q
G.S.
sz
Fig. 7.21 Vertical stress at centre of circular loaded area
Circular Area Carrying Uniform Pressure. Two cases of stresses due to a uniform pres-
sure on a circular area are available, viz., (i) stresses under the centre of the circular area and
(ii) stresses at any point on the soil.
The vertical stress at a depth z under the centre of a circular area of diameter 2 a is (Fig. 7.21)
⎡ ⎧⎪ ⎫⎪⎪3 /2 ⎤⎥
σz = q ⎢⎢ 1 − ⎪⎨
1 (7.63)
2⎬ ⎥
⎢⎣ ⎪⎪⎩ (1 + a /z ) ⎪⎪⎭ ⎥⎦
= q Ncc (7.64)
where
⎡ ⎧⎪ ⎫⎪3 / 2 ⎤⎥
⎢ ⎪ 1 ⎪⎬
Ncc = ⎢1 − ⎨ ⎥ (7.65)
⎢ ⎪⎪⎩ (1 + a / z)2 ⎪⎪⎭ ⎥
⎣ ⎦
The value of the influence factor Ncc is given in Fig. 7.22. Equation 7.63 is valid only for
the stress along the centre line. Foster and Ahlvin (1954) have given a chart for finding σz at
any point lying under as well as outside the loaded area (Fig. 7.23).
1.0
0.8
0.6
Ncc
0.4
0.2
0 2 4 6 8
2a
z
Fig. 7.22 Influence chart for vertical stress under centre of circular area
M07_PURU1773_01_SE_C07.indd 191 2/12/2013 8:03:08 AM

q dA = db dl
a db
l dl
z
sz
Fig. 7.23 Vertical stress at any point due to uniformly loaded circular area
The expression for σz is of the form

⎛ 3 qa 3 ⎞⎟2π a
l dβ dl
σ z = ⎜⎜⎜ ⎟⎟
⎜⎝ 2π ⎟⎠ ∫ ∫ (l2 + z2 + x 2 − 2xl cos β )5/2 (7.66)
0 0
or
σ z = q N CA (m, n) (7.67)
where NCA is a shape function of dimensionless variables,

z x
m= , n=
a a
This chart (Fig. 7.24) is based on the assumption that Poission’s ratio ν = 0.5. This is
applicable to points under the centre as well as at all points away from the centre. The pressure
bulbs for a uniform circular load are given in Fig. 7.25.
0 1.25
2.5 2.0 1.5 1.0 0.0
1 3.0 0.25
4.0 0.5
2 5.0 0.75
Depth ratio m = z/a
3 6.0
7.0
4 n = x/a
8.0
5
9.0
6 10.0
7
8
9
10
0.1 1.0 10 100
Nca , %
Fig. 7.24 Influence chart for vertical stress at any point due to uniformly loaded circular area
(Source: Foster and Ahlvin, 1954)
M07_PURU1773_01_SE_C07.indd 192 2/12/2013 8:03:09 AM

2a q = Maximum applied
Diameter pressure
0.90q 0.80q
a 0.70q
0.60q
0.50q
a
0.30q 0.40q
0.20q
a
0.15q
0.05q
0.10q
Fig. 7.25 Bulbs of vertical pressure under uniform circular load
Rectangular Area Carrying Uniform Pressure. The vertical stress beneath the corner of a
uniformly loaded rectangular area can be expressed as (Fig. 7.26)
q ⎡⎢ 2mn(m2 + n2 + 1)1/ 2 m2 + n2 + 2 2 2
−1 2mn( m + n + 1)
1/ 2 ⎤
⎥
σz = . + tan (7.68)
4π ⎢⎣ m2 + n2 + m2 n2 + 1 m2 + n2 + 1 m2 + n2 + 1 − m2 n2 ⎥⎦
Here the width, B, and length, L, of the rectangle are given as mz and nz, where z is the
depth under consideration. Equation 7.68 can be written as
σ z = qNR (7.69)
where
1 ⎢⎡ 2mn(m2 + n2 + 1)1/ 2 m2 + n2 + 2 2 2
−1 2mn( m + n + 1)
1/ 2 ⎤
⎥
NR = . + tan (7.70)
4π ⎢⎣ m + n + m n + 1 m + n + 1
2 2 2 2 2 2
m + n + 1 − m n2 ⎥⎦
2 2 2
Figure 7.27 shows the variation of NR with m and n (Fadum, 1948). The factors m and n
in the chart are interchangeable. This chart can be adopted for any area based on rectangles
under any point within or outside the area to be obtained by the method of superposition.
mz
nz q
sz
Fig. 7.26 Vertical stress under corner of uniformly loaded rectangular area
M07_PURU1773_01_SE_C07.indd 193 2/12/2013 8:03:10 AM

0.28
mz n
0.26 nz q
2.0
0.24 1.4
z
0.22 sz
1.0
0.20
sz = qNR 0.8
0.18
0.6
0.16
0.5
Na 0.14
0.4
0.12
0.10 0.3
0.08
0.2
0.06
0.04 0.1
0.02
0
0.00
0.1 1 10
m
Fig. 7.27 Influence chart for vertical stress under corner of uniformly loaded rectangular area
(Source: Fadum, 1948)
Newmark’s Influence Chart. The preceding sections outline the stress distributions due
to loaded areas of regular geometry and cannot be applied without error to irregularly
shaped areas. Newmark (1942) devised a graphical procedure based on the expression for
the vertical stress under the centre of a loaded circular area (Fig. 7.21).
Equation 7.63 for a circular loaded area can be rewritten in the form
⎡ ⎫⎪3 / 2 ⎤⎥
σ z ⎢ ⎧⎪⎪ 1 ⎪⎬
= ⎢1 − ⎨ ⎥
q ⎢ ⎪⎪⎩ 1 + ( a / z)2 ⎪⎪⎭ ⎥
⎣ ⎦
For different values σz/q, the values of the ratio are calculated. The a/z values for σz /q
varying from 0 to 1.0 are given in Table 7.4.
A suitable depth scale is chosen and all the radii for nine circles are calculated and drawn
(Fig. 7.28); e.g., the first circle, with σz /q = 0.10, will have a radius a = 0.27z. If the depth z
(in metres) is represented by the length AB (in mm), then the radius a (in mm) = 0.27AB.
Table 7.4 σz/q versus a/z
σz/q a/z σz/q a/z
0 0 0.60 0.92
0.10 0.27 0.70 1.11
0.20 0.40 0.80 1.39
0.30 0.52 0.90 1.91
0.40 0.64 1.00 ∞
0.50 0.77
M07_PURU1773_01_SE_C07.indd 194 2/12/2013 8:03:11 AM

11
a = 1.
z a = 1.39
z
Depth unit = z
A B
Scale
Influence value = 0.005
Fig. 7.28 Newmark influence chart for vertical stress at any depth z = AB
(Source: Newmark, 1942)
Now, divide each circular ring into 20 convenient parts (i.e., 200 influence units). The stress
transferred by one annular ring is 0.10, which is divided into 20 parts. Thus, the influence
value for one block (irrespective of size) is
0.10
NN = = 0.005 (7.71)
20
The stress at a depth z for a specifc point is
σ z = q× N N ×(number of influence blocks) (7.72)
To use this chart, the loaded surface is drawn to a scale such that the distance AB equals
the depth of the point in question. The point beneath the loaded area for which the vertical
stress is sought is then located over the centre of the chart. The plotted area covers a number
of influence blocks, and the number of influence units are counted. Thus, the vertical stress
is found from Eq. 7.72.
Figure 7.29 shows Newmark’s influence chart for the vertical stress based on Wester-
gaard’s theory.
Approximate Soultion of Vertical Stress. Approximate estimates of vertical stress at a
depth z due to a uniformly loaded circular or rectangular area can be obtained by the 60°
distribution (Fig. 7.30) or a 2:1 distribution (about 63°) assumption (Fig. 7.30b). This method
M07_PURU1773_01_SE_C07.indd 195 2/12/2013 8:03:11 AM

5
a = 2.2
z
A B
Scale
Influence value = 0.005
n = 0.0
Fig. 7.29 Influence chart for vertical stress based on Westergaard theory
(Source: Bowles, 1982)
B B
q q
30°
z 2
1
60°
sz sz
B + 1.5z B +z
(a) 30° distribution (b) 2:1 distribution
Fig. 7.30 Approximate solutions to vertical stress
M07_PURU1773_01_SE_C07.indd 196 2/12/2013 8:03:12 AM

predicts values nearer to those obtained from elastic solutions when z/B is in the range
1.5 < z/B < 5. For smaller depths this approach yields lower values (i.e., z/B < 1.5) and
higher values at greater depth (i.e., z/B > 5.0). However, this method may be used for the
determination of σz for a preliminary analysis.
For a retangular area of dimensions B×L, the vertical stress is given below. For a 30° slope
(or distribution),
⎡ BL ⎤
σz = q ⎢ ⎥ (7.73)
⎢ (B +1.5z)(L +1.5z) ⎥
⎣ ⎦
For a 2:1 slope,
⎡ BL ⎤
σz = q ⎢ ⎥ (7.74)
⎢ (B + z)(L + z) ⎥
⎣ ⎦
7.7.7 Contact Pressure

The pressure transmitted from the base of a foundation to the soil is termed the contact pres-
sure. This depends on the rigidity of the foundation structure and the nature of the soil.
The presence of a thick compressible layer, like soft clay, beneath a flexible foundation
presents a bowl-shaped settlement profile with more settlement at the centre and almost
zero at the edge. But the pressure distribution is uniform. This is the conventional distribu-
tion pattern used in the calculation of stressed settlements (Fig. 7.31a).
An extremely rigid footing on the same clay will settle an uniform amount across its
breadth. Thus, the compressible cohesive soil under a rigid footing has to transmit a higher
contact pressure near the edges than at the centre so as to maintain a uniform settlement. The
contact pressure distribution is shown in Fig. 7.31b.
q q
qmax = q
qmax Uniform
qmin settlement
Dish-shaped
qmax > q > qmin
settlement profile
(a) Flexible footing on cohesive soil (b) Rigid footing on cohesive soil
q
q
qmax = q qmax
qmax = q Uniform
Settlement
settlement
profile
(c) Flexible footing on (d) Rigid footing on cohesionless soil
cohesionless soil
Fig. 7.31 Effect of rigidity of footing on contact pressure
M07_PURU1773_01_SE_C07.indd 197 2/12/2013 8:03:13 AM

For a flexible foundation resting on a non-cohesive soil, the distribution of contact pressure
is uniform, but the edges of the foundation experience a large settlement. Because of the lack
of confining pressure at the edges, the foundation settles more (Fig. 7.31c).
The settlement of a rigid footing on a sand layer is uniform and the contact pressure
increases from zero at the edge to a maximum at the centre (Fig. 7.31d).
In actual practice, no foundation is perfectly flexible or infinitely rigid, and hence the
actual distribution of the contact pressure is somewhere between the extreme values. Suffi-
cient accuracy in the calculation of stresses and displacements can be obtained by assuming
a uniform distribution of the contact pressure.
7.7.8 Validity of Elastic Theory Application

The stresses obtained from the application of elastic theory can be accepted only when
favourable field results are available. Only a limited number of field measurements are
reported in the literature (e.g., Taylor, 1945; Turnbull et al., 1961). A great number of such
observations are needed.
Based on a few excellent comparisons, it has been reported that a good agreement was
found in the case of vertical stresses. For a lack of better knowledge, the civil engineer is
compelled to adopt elastic theory for the computation of stresses. In any case, an error of
±25% between the field observations and elastic theory may be expected.
WORKED EXAMPLES
Example 7.1 Three soil samples are tested with the state of stresses shown in Fig. 7.32.
s1 = 600 kN/m2
2
k N/m
s2 = 100 kN/m2 00
=1
s2
s3 = 100
kN/m2
00
s1 = 600 =6 2
kN/m2 s 1 N/m
k
30°
(a) (b) (c)
Fig. 7.32
M07_PURU1773_01_SE_C07.indd 198 2/12/2013 8:03:13 AM

1. Draw the Mohr’s circle for each case.

2. Locate the pole for each case.
3. Locate the state of stress acting on a plane at 30° with the major principal plane for each case.
Solution
Mohr’s circles are drawn as shown in Fig. 7.33. Also, the pole is marked in each case.
For each case, the state of stress acting on a plane at 30° with the major principal plane is
marked in Fig. 7.33 and given below as
1. σn = 500 kN/m2
τn = 220 kN/m2
2. σn = 500 kN/m2
τn = 220 kN/m2
3. σn = 500 kN/m2
τn = 220 kN/m2
Minor pr. pl.
Minor pr. pl.
(sn, tn) (sn, tn)
tn = 220 30°
tn = 220
Op 30°
O O Op
s2
sn = 500
100
s1 = 600 sn = 500
Major pr. pl. sl = 600

s3
= 100 (a) (b)
Major pr. pl.
Minor pr. pl.
Op
Major pr. pl.
30°
30°
O
s2
100 tn = 200
Note: Stresses are given

(sn, tn) in kN/m2
sn = 500
s1 = 600
(c)
Fig. 7.33
M07_PURU1773_01_SE_C07.indd 199 2/12/2013 8:03:14 AM

Example 7.2 In the process of an excavation for a wall footing, the water table level was
lowered from a depth of 1.2 m to a depth of 4.5 m in a clayey sand deposit. Considering that the
soil above the water table remains saturated at a water content of 28%, compute the following:
1. The effective stress at a depth of 4 m after the lowering of the water table. Take G = 2.68.
2. The increase in effective stress at a depth of 5 m.
Solution
G+e G + wG
γ sat = γw = γw
1+ e 1 + wG
2.638 + 0.28 × 2.68
= × 9.807 = 19.2 kN / m 2
1 + 0.28 × 2.68
Before lowering of water table

1. Stress condition at the surface
Because of capillary saturation, the water pressure will be negative, Thus,
σ=0
uw = −1.2× 9.807 = −11.77 kN / m 2
σ ′ = σ − uw = 0 + 11.77 = 11.77 kN / m 2
2. Stress condition at a depth of 1.2 m
σ = 1.2×19.2 = 23.04 kN / m 2
uw = 0
σ ′ = σ − uw = 23.04 − 0 = 23.04 kN / m 2
3. Stress condition at a depth of 5 m
σ = 5×19.2 = 96 kN / m 2
uw = (5 − 1.2)× 9.807 = 37.27 kN / m 2
σ ′ = 96 − 37.27 = 58.73 kN / m 2
The stress distributions are shown in Fig. 7.34a.
After lowering of water table
1. Stress condition at the surface
The lowering of the water table induces tension in the water between the levels 1.2 and
4.5 m from the surface (Fig. 5.34b). Thus,
σ=0
uw = −4.5× 9.807 = −44.13 kN / m 2
σ ′ = σ − uw = 44.13 kN / m 2
σ = 4 ×19.2 = 76.8 kN / m 2
uw = −( 4.5 − 4.0)× 9.807 = −4.9 kN / m 2
σ ′ = σ − uw = 76.8 + 4.9 = 81.7 kN / m 2
M07_PURU1773_01_SE_C07.indd 200 2/12/2013 8:03:15 AM

–11.77 kN / m2 11.77 kN / m2
1.2 m
23.04 kN / m2 23.04 kN / m2
4m
5m
96 kN / m2 37.27 kN / m2 58.73 kN / m2
s uw s¢
(a) Before lowering of water table
–44.13 kN / m2 44.13 kN / m2
1.2 m
4m
4.5 m
5.0 m
4.9 kN / m2
76.8 kN / m2 81.7 kN / m2
96 kN / m2 4.9 kN / m2 91.1 kN / m2
s uw s¢
(b) After lowering of water table
Fig. 7.34

σ = 5×19.2 = 96 kN / m 2
uw = (5 − 4.5)× 9.807 = 4.9 kN / m 2
σ ′ = σ − uw = 96 − 4.9 = 91.1 kN / m 2
Therefore, change in effective stress at 5 m depth = 91.1 – 58.73 = 32.37 kN/m2
Example 7.3 An overhead water tank is supported at a depth of 3 m by four isolated square
footing of sides 2 m each placed in a square pattern with a centre-to-centre spacing of 8 m
(Fig. 7.35). Compute the vertical stress at the foundation level (i) at the centre of the four
footings and (ii) at the centre of one footing. Adopt Boussinesq’s point load approximation.
The load on each footing is 700 kN.
M07_PURU1773_01_SE_C07.indd 201 2/12/2013 8:03:16 AM

+ +
1m
8m .3
11
+ +
8m
Fig 7.35
Solution
Boussinesq vertical stress
3 Q ⎡ 1 ⎤ 5/ 2
σz = ⎢ ⎥
2π z 2 ⎢ 1 + ( r / z )2 ⎥
⎣ ⎦
⎡ ⎪⎫⎪ ⎤⎥
5/ 2
⎢ 3 700 ⎪⎧⎪ 1
The stress at the centre = 4 ⎢ × 2 ⎨ ⎬ ⎥
⎢ 2π 3 ⎪⎪⎩ 1 + (5.655 / 3)2 ⎪⎪⎭ ⎥
⎣ ⎦
⎡ 700 ⎤
= 4⎢ × 0.0226⎥ = 3.36 kN / m 2
⎢⎣ 6π ⎥⎦
The stress at the centre of any corner footing

5/ 2
⎡ ⎫⎪5/ 2 ⎧⎪ ⎫⎪5/ 2 ⎧⎪ ⎫⎪⎤⎥
3 Q ⎢ ⎧⎪⎪ 1 ⎪⎬ + ⎪⎨ 1 ⎪⎬ + ⎪⎨ 1 ⎪⎬
= ⎢ 2×⎨
2π z 2 ⎢ ⎪⎪⎩ 1 + (8 / 3)2 ⎪⎪⎭ ⎪⎪⎩ 1 + (0 / 3)2 ⎪⎪⎭ ⎪⎪⎩ 1 + (11.31/ 3)2 ⎪⎪⎭⎥⎥
⎣ ⎦
3 700
= × [0.0107 + 1 + 0.001108]
2π 9
= 37.58 kN / m 2
Example 7.4 Two railway wagon lines in a harbour yard are located at 6 m centre-to-centre.
The average loads per metre run in the lines are 100 and 80 kN/m. Find the vertical stress
induced by this loading at a depth of 2 m beneath each load and halfway between them. If a
100 kN crane is installed exactly midway between the lines, what additional stress is caused
below the crane at the same depth.
Solution
Consider the railway wagon load as a line load of infinite extent. The vertical stress is given as
M07_PURU1773_01_SE_C07.indd 202 2/12/2013 8:03:16 AM

2qz 3
σz =
π( x 2 + z 2 )2
The stress below the 100 kN/m load
2×100 ⎡⎢ 23 ⎤ 2× 80 ⎡
⎥+ ⎢ 23 ⎤
⎥
= ⎢ 2 2⎥ ⎢ 2 2⎥
π ⎣ (0 + 2 ) ⎦
2
π ⎣ (6 + 2 ) ⎦
2
= 31.83 + 0.226 = 32.09 kN / m 2
The stress below the 80 kN/m load
2× 80 ⎡⎢ 23 ⎤ 2×100 ⎡
⎥+ ⎢ 23 ⎤
⎥
= ⎢ 2 2⎥ ⎢ 2 2⎥
π ⎣ (0 + 2 ) ⎦
2
π ⎣ (6 + 2 ) ⎦
2
= 25.78 kN / m 2
The stress midway between the two loadings
2×100 ⎡⎢ 23 ⎤ 2× 80 ⎡
⎥+ ⎢ 23 ⎤
⎥
= ⎢ 2 2⎥ ⎢ 2 2⎥
π ⎣ (3 + 2 ) ⎦
2
π ⎣ (3 + 2 ) ⎦
2
= 5.42 kN / m 2
The additional stress below the crane, considering the crane load as a vertical concentrated
load, is given as
3Q ⎡ 1 ⎤ 5/ 2
σz = ⎢ ⎥
2 π z2 ⎢ 1 + ( r / z )2 ⎥
⎣ ⎦
3 ×100 ⎢ ⎡ 1 ⎤ 5/ 2
= ⎥ = 11.94 kN / m 2
2 ⎢ 2⎥
2 × π × 2 ⎣ 1 + ( 0 / 2) ⎦
Example 7.5 An embankment is to be constructed with the following dimensions:

Top width = 8 m; height = 4 m; side slopes = 1:1½.
The unit weight of the soil is 21 kN/m3. Compute the vertical pressure at a depth of 6 m
below the ground surface at the following locations:
1. On the central longitudinal plane of the embankment
2. Below the toes of the embankment
If a surcharge load of 50 kN/m2 is acting on the road surface, what is the increase in stress
at the same central point. Assume the surcharge load is distributed vertically downwards.
Solution
Refer to Fig. 7.36.
M07_PURU1773_01_SE_C07.indd 203 2/12/2013 8:03:17 AM

8m
Q¢ Q R
4m
P S
U T
6m 6m
6m
Fig. 7.36
The stress at the centre of the embankment

a 6 b 4
= = 1, = = 0.67
z 6 z 6
q = γ z = 21× 4 = 84 kN / m 2
(σ z )A = 2× q NE = 2× 84 × 0.425 = 71.4 kN / m 2
The stress below the toe = stress due to PQ’QRS – stress due to PQ’Q.
For the stress due to PQ’QRS,
a 6 b 14
= = 1, = = 2.33
z 6 z 6
For the stress due to PQ’Q,
a 6 b
= = 1, =0
z 6 z
The respective NE values from Fig. 7.20 are 0.49 and 0.24.
Hence, (σz)B = 84 (0.49 – 0.24) = 21 kN/m2.
The additional stress due to the surcharge load can be obtained using Fig. 7.20, assuming a
very low value of a/z, say a/z = 0.1. Here, b/z = 8/10 = 0.80. For a/z = 0.1 and b/z = 0.80,
the value of NE = 0.37. Therefore, the additional stress at A = 50 × 0.37 = 18.5 kN/m2.
Example 7.6 Calculate the stress in a soil mass below the centre of a uniformly loaded cir-
cular area of radius 1.5 m with a pressure of 60 kN/m2 and thus obtain the exact depth at
which the stress reduces to 10% of the applied stress.
Solution
The vertical stress at a depth z under the centre of a circular area of diameter 2a is given as
⎡ ⎪⎧ ⎪⎫⎪ ⎤⎥
3/2
⎢ ⎪ 1
σ z = q ⎢1 − ⎨ ⎬ ⎥
⎢ ⎪⎪⎩ 1 + ( a / z)2 ⎪⎪⎭ ⎥
⎣ ⎦
M07_PURU1773_01_SE_C07.indd 204 2/12/2013 8:03:18 AM

Here σ z = 0.10 q. Therefore,

⎡ ⎪⎧ ⎪⎫⎪ ⎤⎥
3/2
⎢ ⎪ 1
0.10 q = q ⎢1 − ⎨ ⎬ ⎥
⎢ ⎪⎪⎩ 1 + (1.5 / z)2 ⎪⎪⎭ ⎥
⎣ ⎦
or
1
= 0.90
{1 + (1.5 / z)2 }
3/2
or z2 = 2.25/0.075 = 30, or z = 5.48 m. Therefore, the depth at which the stress is 10% of the
applied stress is 5.48 m.
Example 7.7 A total load of 900 kN is uniformly distributed over a rectangular footing of
size 2 m × 3 m. Find the vertical stress at a depth of 2.5 m below the footing at point C, under
one corner, and D, under the centre. If another footing of size 1 m × 3 m with a total load of
450 kN is constructed adjoining the previous footing, what is the additional vertical stress at
the point C at the same depth due to the construction of the second footing.
Solution
Refer to Fig. 7.37.
For the first footing,
900
q= = 150 kN / m 2
2× 3
For the stress under corner C,
3 2
m= = 1.2 and n = = 0.8
2.5 2.5
3m 1m
2m
D
3m
Fig. 7.37
M07_PURU1773_01_SE_C07.indd 205 2/12/2013 8:03:20 AM

From Fig. 7.27, NR = 0.168. Therefore,
(σ z )C = qNR = 150 × 0.168 = 25.2 kN / m 2
For the stress under centre D,
1.5 1
m= = 0.6 and n = = 0.4
2.5 2.5
Again, from Fig. 7.27, NR = 0.08. Therefore,
(σ z )D = 4(qNR ) = 4 ×150 × 0.08 = 45 kN / m 3
For the second footing,

450
q= = 150 kN / m 2
1× 3
For the additional stress under corner C,
3 1
m= = 1.2 and n = = 0.4
2.5 2.5
From Fig. 7.27, NR = 0.105. The additional stress at corner C, due to the construction of the
second footing = 150×0.105 = 15.8 kN/m2.
Example 7.8 A foundation is constructed to take a stress of 150 kN/m2 and is flush with
another existing foundation (Fig. 7.38) taking a load of 100 kN/m2. Find the vertical stress at
a depth of 2 m below the point D. Use the Newmark’s chart given in Fig. 7.39 with an influ-
ence value of 0.002.
Solution
Considering the depth scale AB = 2 m, draw the loaded area to this scale with D at the centre
of the chart. This is drawn and shown in Fig. 7.39.
The number of stress blocks occupied by areas ABCD and EFGH are separately counted
and given as
No. of blocks in ABCD = N1 = 72
No. of blocks in EFGH = N2 = 50
Stress under point D = q1 × N N × n1 + q2 × N N × n2
= N N (q1 n1 + q2 n2 )
6 kN / m 2
= 0.002 (150 ×72 + 100 × 50) = 31.6
M07_PURU1773_01_SE_C07.indd 206 2/12/2013 8:03:21 AM

1m 1m
E F
1m
A B
150 100
2m kN/m2 kN/m2
D C
1m
H G
Fig. 7.38
A
D
H
E
C B
F
G
A Z B
Influence value per field

= 0.002
Fig. 7.39
M07_PURU1773_01_SE_C07.indd 207 2/12/2013 8:03:21 AM

Example 7.9 A shallow pond of 1 m depth has a 4.5 m thick layer of silty fine sand below
the bottom of the pond with a total unit weight of 17.8 km/m3. This is underlain by a layer
of medium sand. It has been found that the sand layer is subjected to an artesian pressure
throughout the year as given below.
January to March 72 km/m2
April to June 68 km/m2

July to September 102 km/m2
October to December 85 km/m2
During which period the bottom of the pond is unsafe to use?

Solution
As there is an artesian pressure there is a possibility of quick sand formation at the bottom
of the pond. The resisting pressure is the overburden pressure above the sand layer.
P
1m
4.5 m Silty fine sand

g t = 17.8 kN / m3
A
Medium sand
A point A at the interface is considered.
Overburden pressure at ⎫⎪⎪

⎬ = γ ω hω + γt h1
A⎪⎪⎭
= 9.81 × 1 + 17.8 × 4.5 = 89.91 km/m2
Out of the four periods only during July to September the artesian pressure (102 km/m2) is
more than the overburden pressure (89.91 km/m2). Hence, July to September is an unsafe
period as the effective stress reduces to zero.
POINTS TO REMEMBER
7.1 Principal planes are those where only normal stresses act and no shear stresses exist.
Such a normal stress is called a principal stress. Three principal stresses act on three
mutually orthogonal planes, viz., the major principal stress (the largest stress), the
intermediate principal stress, and the minor principal stress (the smallest stress).
M07_PURU1773_01_SE_C07.indd 208 2/12/2013 8:03:22 AM

7.2 The graphical representation of the state of the normal and shear stresses acting on
different planes in a lucid form is known as the Mohr’s circle.
7.3 The locus of points on the Mohr diagram whose coordinates represent the maximum
shear stress and the associated principal stress for the entire stress history is defined
as a stress path.
7.4 In a saturated soil, the total stress (σ) has two stress components. The stress compo-
nent on the water, which does not cause any change in the mechanical properties of
the soil, is known as the neutral stress or pore water pressure (uw). The other compo-
nent of total stress which rests entirely on the soil skeleton of the soil is responsible for
changes in the properties of the soil.
7.5 The elastic half-space is an idealized soil medium which is homogeneous, isotropic,
and elastic. The behaviour of the medium is governed by the stress–strain modulus,
E, and Poisson’s ratio, ν.
7.6 Boussinesq’s theory assumes a weightless, elastic half-space, and gives components
of stresses caused by a vertical point surface load.
7.7 The stress isobar or pressure bulb is a stress contour connecting all points of equal
stress below the ground surface. The soil bounded within a pressure bulb furnishes
the support power of a footing.
7.8 Westergaard assumed an elastic half-space medium in which there is no lateral strain
and the medium suits the condition of a sedimentary soil.
7.9 Newmark’s chart is a graphical procedure for determining the vertical stress due to a
surface load of any shape. The chart is based on the expression for the vertical stress
under the centre of a loaded circular area.
7.10 The contact pressure is the pressure transmitted from the base of a foundation of the
soil and depends on the rigidity of the foundation structure and the nature of the soil.
QUESTIONS
Objective Questions
7.1 State which of the following statements are true or false:
1. In several situations the effective stress will be greater than the total applied stress.
2. The effective stress in a soil mass is always the actual grain-to-grain contact stress.
3. Application of Boussinesq’s vertical stress overestimates the settlement.
4. The effective stress of a soil is not affected by the type of pore fluid.
5. Westergaard’s expression for the vertical stress considers the weight of the soil
medium.
7.2 The contact pressure distribution under a rigid footing on saturated clay and dense
sand is _______ and ________ , respectively.
(a) Uniform
(b) Concave parabolic
(c) Convex parabolic
M07_PURU1773_01_SE_C07.indd 209 2/12/2013 8:03:22 AM

7.3 For a vertical concentrated load acting on the surface of a semi-infinite elastic soil
mass, the vertical normal stress at depth z is
(a) Directly proportional to z (b) Inversely proportional to z
(c) Directly proportional to z2 (d) Inversely proportional to z2
7.4 The approximate depth at which the effective vertical pressure is equal to 100 kN/m2
in a typical deposit of submerged soil is
(a) 5 m (b) 10 m
(c) 20 m (d) 100 m
7.5 If the entire semi-infinite mass is loaded with a load intensity of q at the surface, the
vertical stress at any depth is
(a) 0.2q (b) q
(c) Zero (d) Infinity
7.6 A rise in the groundwater table up to the capillary zone results in
(a) A decrease in the degree of saturation (b) An increase in the effective stress
(c) A decrease in the effective stress (d) No change in the pore water pressure
7.7 Assertion A: The effective stress is that part of the load which is transmitted by the
particles divided by the gross area.
Reason R: The effective stress is not a stress that can be measured directly but a com-
puted value.
7.8 The diameter of the Mohr’s circle for plane stress conditions is the
(a) Octahedral stress (b) Shear stress
(c) Deviator stress (d) Principal stress ratio
7.9 Which of the following pairs is correctly matched?
1. Westergaard’s theory — For soils with a thin layer of coarse materials only
2. Newmark’s chart — Graphical procedure based on circular loaded area
3. Boussinesq’s theory — For vertical stress only
Select the correct answer using the codes given below.
Codes:
(a) 1 and 2 are correct (b) 1 and 3 are correct
(c) 2 and 3 are correct (d) 2 alone is correct
7.10 As the depth of the stress isobar increases, the intensity of stress
(a) Increases (b) Decreases
(c) Remains constant (d) Initially decreases and then increases
7.11 Discuss the essential differences between Boussinesq’s and Westergaard’s theories.
For which condition do both these theories yield approximately the same value of
vertical stress?
M07_PURU1773_01_SE_C07.indd 210 2/12/2013 8:03:22 AM

7.12 How far is it justifiable to adopt Boussinesq’s theory for predicting the vertical stress
in sand deposits?
7.13 What is the basic principle involved in the development of Newmark’s chart?
7.14 Explain why the effective stress evaluation in a partially saturated clay is very complex.
7.15 Discuss the limitations of Lambe’s effective stress expression.
EXERCISE PROBLEMS
2
7.1 A sample is subjected to the following principal stresses: σ1 = 700 N / m and σ3
= −200 N / m 2 . Draw the Mohr’s circle and locate the origin of planes. Find σn and
τn on a plane making an angle 50° with the major principal plane. Also find the maxi-
mum shear stress.
7.2 The normal stresses acting on two mutually perpendicular planes are 150 and 60 kN/m2
and the shear stress on each plane is 110 kN/m2. Draw Mohr’s circle and find
1. The principal stresses and the planes
2. The possibility of tension occurring on any plane for this stress condition
3. The shear and normal stresses acting on a plane making an angle of 70° with the
major principal stress
7.3 An impervious, saturated clay layer of 12.5 m thickness lies over a sand aquifer. Piezom-
eters inserted into the sand layer show an artesian pressure condition with the piezo-
metric surface 3.5 m above the surface of the clay. Determine the effective stress at the
top of the sand layer. The parameters of the clay layer are e = 1.26 and G = 2.72. How
deep an excavation can be made in the clay layer without the danger of a bottom heave?
7.4 The soil conditions shown in Fig. 7.40 are revealed during a boring operation.
Represent by diagrams the variation with depth of the total vertical overburden pres-
sure, the pore water pressure, and the effective overburden pressure. Assume the top
1 m is dry. Compute the total and effective overburden pressure at the bottom of the
clay layer immediately after the lowering of the water table, if the water level in the
gravel layer is suddently lowered by 2.1 m.
Before lowering After lowering
1m
2.1 m
5m g = 19.6 kN/m3
Gravel
gd = 15.7 kN/m3
8m Clay g = 18.6 kN/m3
Rock
Fig. 7.40
M07_PURU1773_01_SE_C07.indd 211 2/12/2013 8:03:22 AM

7.5 In a fine sand deposit, the water table is located at 4 m from the ground surface. Due to cap-
illary action only 1 m depth of sand above the water table is saturated and the remaining
is dry. The dry and saturated unit weights of sand are, respectively, 19.2 and 21.6 kN/m3.
Estimate the effective vertical stress in the sand at a depth of 10 m below the surface.
7.6 A boring log indicates the presence of 5 m of silty sand from the ground surface followed
by 3 m of coarse sand, which in turn rests on a deep deposit of gravel. The groundwater
table is located at the top of the sand layer. The soil characteristics are as given below:
Specific gravity Void ratio

Silty sand 2.67 0.90
Sand 2.65 0.60
Compute the total, neutral, and effective stresses and draw the stress diagrams from the
ground surface to the top of the gravel layer. Assume there is no capillary rise of water.
7.7 A point load of 1,200 kN acts on the surface of a deep clay layer. Compute the vertical
stresses in horizontal layers spaced at 1 m increments of depth up to 5 m and take a
radial distance up to 4 m on either side of the point load with 1 m increments. From
these results plot
1. the vertical pressure bulb for 20 kN/m2
2. the distribution of stress directly beneath the load,
3. the distribution of stress on horizontal place at 3 m depth.
Adopt Boussinesq’s theory.
7.8 Two concentrated loads Q1 = 900 kN and Q2 = 1,200 kN are spaced 4 m apart. Draw
the 5 kN/m isobar of the system. Adopt Boussinesq’s theory for a single, vertical con-
centrated load.
7.9 A 200 kN load is transferred through a steel stanchion. Compute the vertical stresses
beneath the stanchion at depths of 1, 3, and 9 m. Estimate the depth at which the load
is 25% of the applied load. Adopt Westergaard’s theory (ν = 0).
7.10 Three load-bearing walls meet at a point X, as shown in Fig. 7.41. Compute the verti-
cal stress at a point 3 m under X. The loads transferred by the walls may be approxi-
mated as line loads. (Hint: Take half of the stress caused by an infinite line load.)
120 kN/m
90 kN/m
X
C
150 kN/m
B
Fig. 7.41
M07_PURU1773_01_SE_C07.indd 212 2/12/2013 8:03:23 AM

7.11 A uniformly distributed load of infinite extent in both lateral directions, when applied
at the surface of a natural soil formation, produces an increase of 75 kN/m 2 in the ver-
tical stress at a depth of 3 m. Find the stress increment at a depth of 5 m.
7.12 Show that the vertical stress, σ’z, at depth z, for a long strip area with a triangular dis-
tributed load (Fig. 7.42) is
a b
a
b
Fig. 7.42
q ⎡a+b+r r ⎤
σz = ⎢ β − α⎥
π ⎢⎣ b a ⎥⎦
7.13 A very long embankment is to be built with a top width of 10 m and side slopes of
1:1½. The height of the embankment is 10 m. Compute the vertical stresses at a depth
of 5 m from the base at the following points: (i) below the toe, (ii) below the central
line, and (iii) below a point midway on the slope. Assume γ = 21 kN/m3.
7.14 A circular area is loaded with a uniform load intensity of 100 kN/m2 at the ground
surface. Calculate the vertical pressure at point A, so situated on the vertical line
through the center of the loaded area that the area subtends an angle of 90° at it.
7.15 A circular ring footing for an overhead water tank carries a load of 1,000 kN whose
outer diameter is 3 m and inner diameter is 1.5 m. Assume the surface pressure to
be uniform over the area. Determine the vertical stress at depths of 2 m and at radial
distances of 2 and 4 m from the center. Use Newmark’s chart with an influence value
of 0.005.
7.16 A square footing of 3 m×3 m carries a uniformly distributed load of 200 kN/m2.
Find the vertical stress at 3 m below the footing and under a point (i) 1.2 m away
from the corner and in line with the side, (ii) 1.2 m inside the corner and in line with
the side.
7.17 A wheel load of 1,200 kN is applied at the surface of a road. What will be the total load
on the crest of a culvert situated at 2.5 m below the surface? Assume that an area of
2 m×3 m at the crest level is transferring the load. Use Boussinesq’s stress coefficients
for a uniformly loaded rectangular area.
M07_PURU1773_01_SE_C07.indd 213 2/12/2013 8:03:23 AM

7.18 The ground surface of a building is lowered 2 m below the existing surface. A 1.2 m
square footing carrying a load of 200 kN/m2 is then constructed at the level of the new
surface. Estimate the net increase in stress in the soil mass 1.2 m below the center of
the foundation. Take the unit weight of soil to be 18 kN/m3.
7.19 Construct a Newmark’s chart for vertical stresses based on Boussinesq’s theory with
an influence value of 0.0025. Using this chart, determine the vertical stress induced
at a depth of 8 m below the circumference of a uniformly loaded circular area of 6 m
diameter, with an intensity of 120 kN/m2.
7.20 A continuous strip footing of 3 m width carries a uniformly distributed load of 110
kN/m2. Plot the vertical stress distribution on a plane situated at 2 m from the surface.
Compare the vertical stress distribution with that of the 60° approximation.
M07_PURU1773_01_SE_C07.indd 214 2/12/2013 8:03:23 AM

8
Consolidation and Consolidation
Settlement
CHAPTER HIGHLIGHTS
Rheological models of soil – Compressibility of soils – One-dimensional con-
solidation – Consolidation test – Compressibility characteristics – Types of
clay deposits – Prediction of pre-consolidation pressure – Terzaghi’s theory
of one-dimensional consolidation, time factor, coefficient of consolidation,
fitting methods – Secondary compression – Consolidation settlement and its
rates – Acceleration of consolidation – Compressibility of sands
8.1 INTRODUCTION
When a soil layer is subjected to compressive stress due to construction activities, or
otherwise, it undergoes compression. The compression may be caused by rearrange-
ment of particles, seepage of water, crushing of particles, and elastic distortions. The
compression may be progressive and cumulative, dependent on the type, magnitude,
and duration of load and on the properties of the materials. Although the stresses
induced may not cause a failure, the civil engineer is concerned with them as the
magnitude of the compression may be detrimental for some special structures or block
the normal function of conventional structures. Compressibility is one of the three fun-
damental principles of geotechnical engineering to be understood by a civil engineer.
Although the mass is heterogeneous and does not have simple predictable charac-
teristics, the engineer is compelled to provide a safe and economical design. Thus, the
stress change – compression behaviour – may be dealt with by idealizing the soil mate-
rial as elastic for certain conditions and treating the soil as a mathematical model.
Settlement of a structure has to be analysed for three reasons: appearance of the
structure, utility of the structure, and damage to the structure. The aesthetic view of
a structure may be spoiled due to the presence of cracks or tilt of the structure caused
by settlement. Settlement caused to a structure may damage some of the utilities like
M08_PURU1773_01_SE_C08.indd 215 2/12/2013 8:01:30 AM

cranes, drains, pumps, electrical lines, etc. Further settlement can cause a structure to fail
structurally and collapse.
Settlement is the combination of time-independent (e.g., immediate compression) and
time-dependent compression (called consolidation). The engineer is interested in assessing
the magnitude and rate of compression, as well as the total and relative or differential settle-
ment of a structure.
This chapter deals with the process of consolidation and the methods of evaluating
consolidation settlement for different field conditions.
8.2 RHEOLOGICAL MODELS OF SOILS

The study of the behaviour of a material in a fluid state is referred to as rheology. Soil is a
particulate system in which the soil skeleton undergoes a sort of statistical flow caused by
particle rolling, sliding, and slipping, resulting in a void ratio reduction, and such a behav-
iour can be considered as a problem in rheology.
Three basic rheological models (Fig. 8.1a–c), have been identified to represent soil behav-
iour. They are the spring element (spring constant, ks) or Hookean model, which represents
the elasticity of the soil; the dashpot element or Newtonian element (constant C), which
relates to the permeability of the soil; and the yield stress model, which depicts the perma-
nent reduction in void ratio of the soil. The basic spring element is combined with one or
both of the other elements.
The combined Hookean and yield stress model is used as a simple rheological model to
represent immediate soil settlement (Fig. 8.1d). The yield or slip here occurs at a particular
stress level f, and in soil, this may stop after a certain strain level. If pore water is involved, the
Newtonian model may be combined with other elements. The combination of the Hookean
and the Newtonian models in parallel is referred to as the kelvin model (Fig. 8.1e). This can
also be represented as in Fig. 8.1f (piston–spring combination), where the dashpot effect is
produced by the pore water pressure and the subsequent drainage through the valve. This
piston–spring combination is used to represent the compressibility and consolidation of soils.
f
ks
C
(a) Spring model (b) Dashpot model (c) Yield stress model
(uw)1 Piezometer
σ Datum
Valve
σ σ
Piston
ks
c ks
(uw)1 ks
f
(d) Spring and yield (e) Kelvin model (f) Piston–spring

stress model combination
Fig. 8.1 Rheological soil models
M08_PURU1773_01_SE_C08.indd 216 2/12/2013 8:01:30 AM

Consolidation and Consolidation Settlement 217
8.3 COMPRESSIBILITY OF SOILS

A soil mass when subjected to a change in the stress system results in a change in volume of
the mass. Consider the condition that a change in stress causes a volume decrease. The
reduction of volume causes compression or settlement of the boundaries of the mass.
All soils, dry or partially saturated, undergo elastic distortion almost immediately after
the load application. Similarly, in a saturated soil under no drainage condition, shear stresses
induce elastic shear strain, which also occur more or less simultaneously with the applica-
tion of load (Leonards, 1962). Compression caused due to these processes is termed initial or
immediate compression.
The time-dependent volume decrease may be attributed to (i) a compression of soil mat-
ter, (ii) a compression of water and air within the voids, and (iii) escape of water and air
from the voids. Under the stresses normally encountered in civil engineering problems, the
soil solids and pore water (relatively incompressible fluid) undergo a negligible percentage
of the total compression. Thus, all the compression is mainly due to the reduction in void
volume within the soil. In a saturated system, the reduction in void volume is due to escape
of water. But in partially saturated soils, because of the compressible nature of air, there
may be an appreciable reduction in volume, even though there may be no seepage of water.
The time-dependent compression in a partially saturated soil is beyond the scope of current
knowledge. The time-dependent compression in a saturated soil is explained below.
When a pressure is applied to a saturated soil–water system, the applied pressure is
immediately transferred as an excess pressure in pore water. The resulting hydraulic gradi-
ent initiates a flow of water and the soil mass begins to compress, and the portion of applied
stress is transferred to the soil skeleton. This causes a reduction of void volume and dissipa-
tion of excess pore water pressure because of seepage of water from the voids. This process
of gradual load transfer from pore water to soil skeleton and the corresponding gradual
compression is called consolidation. That part of consolidation which is completely controlled
by the resistance to flow of water under the induced hydraulic gradient is called primary
consolidation. The other part, called secondary consolidation (creep), is due to the plastic defor-
mation of the soil at zero excess pore water pressure. The primary consolidation is normally
more than the secondary consolidation. The primary consolidation is generally referred to
as consolidation settlement, and the same is followed in this book.
The immediate compression (including for undrained condition) is computed by assum-
ing the soil mass as an elastic medium, whereas the time-dependent compression is com-
puted assuming one-dimensional consolidation (explained in the subsequent paragraphs).
8.4 ONE-DIMENSIONAL CONSOLIDATION

Consider a fine-grained soil layer (say clay) of thickness H, sandwiched between two permeable
sand layers and below the water table (Fig. 8.2a). If a pressure intensity, Δσ, is applied on the
ground surface, the immediate increase of the pore water pressure will be Δuw, which will be
equal to the applied total pressure Δσ. Thus, immediately after application of pressure (t ≈ 0),
(Δuw )0 = Δσ (8.1)
and
Δσ ′ = 0 (8.2)
M08_PURU1773_01_SE_C08.indd 217 2/12/2013 8:01:30 AM

Initial Final
(Δuw)0 = Δs
Δs
(Δuw)t = 0
Permeable layer (sand)
H Compressible layer (clay)

t=0 t=∞
Permeable layer (sand)
(a) One-dimensional compression-field condition
Final
Initial
s,u (Δuw)t = 0
(Δuw)0 = Δs ∞
Δs ′ = 0 Δσ ′ = Δ σ
(Δs ′)t1
Δs
(Δuw)t1
t=0 t = t1 t=∞ Time
(b) Stress–time curve
Fig. 8.2 Process of one-dimensional consolidation
Once the consolidation process starts by gradual squeezing of water from the soil pores,
the excess pore water pressure decreases, and the effective stress also increases [(Δσ ′)t1] by
the same amount such that the total stress always remains equal to Δσ at t = t1. That is,
Δσ = (Δuw )t1 + (Δσ ′)t1 (8.3)
where
(Δuw )t1 = Δσ − (Δσ ′)t1 (8.4)
This fact is represented in Fig. 8.2b. At t = ∞, the excess pore water pressure at all depths
of the clay layer will be dissipated completely such that
Δσ = (Δσ′ )t∞ = Δσ′ (8.5)
where
Δu w = (Δu w )t∞ = 0
This is the stage at which consolidation is said to be completed. In fact, only the primary
consolidation is over, and secondary consolidation may be on.
Terzaghi (1925, 1943) postulated a rigorous mathematical solution to the process of
consolidation, with a piston-spring rheological model (Fig. 8.1f) to explain the load transfer
M08_PURU1773_01_SE_C08.indd 218 2/12/2013 8:01:32 AM

Δσ ′
ΔH ΔV Δe
Water e0
Water e1
H V
H1 V1
Soil Soil
solids 1 1
solids
(a) Before loading (b) After loading
Fig. 8.3 Interpretation of compressibility
technique. He considered the one-dimensional vertical consolidation, as this is the general

field situation corresponding to the state of the soil during deposition and compression under
its own weight of overburden. The lateral extent of the stratum is very large in comparison
with the thickness of the layer, and it is justified to assume the lateral strains to be small.
The one-dimensional consolidation is based on the following considerations: (i) all
displacements are vertical so that there are no lateral strains, (ii) all the flow of water from the
soil layer is in a vertical direction only, and (iii) the change in void ratio is a direct function
of the vertical component of effective stress.
As the consolidation is one-dimensional, the change in volume, ΔV, per unit of original
volume, V, may be taken equal to the change in height, ΔH, per unit of original height, H
(Fig. 8.3), i.e.,
ΔH ΔV (8.6)
=
H V
It is convenient to represent V and ΔV in terms of void ratio as
ΔH Δe
=
H 1 + e0
where Δe is the change in void ratio and e0 is the original void ratio. Rearranging,
Δe
ΔH = H (8.7)
1 + e0
This relationship is very general in nature and independent of the degree of saturation of
soil and the mechanism causing volume change.
8.5 CONSOLIDATION TEST

The consolidation or oedometer test is used to determine the compressibility characteristics
of a saturated undisturbed or remoulded soil. A trimmed sample is fitted in a cylindrical
container, and a seating pressure of about 12 kPa is applied (Fig. 8.4). Water moves from the
saturated soil both in upward and downward directions towards the porous stones.
When equilibrium is attained under the seating load, an additional increment of load is
applied and allowed to consolidate. In the standard test (explained in detail in Chapter 10), the
pressure is doubled every time until the maximum anticipated pressure in the field is attained.
M08_PURU1773_01_SE_C08.indd 219 2/12/2013 8:01:33 AM

Load Compression
Confining dial
ring
Water
Soil specimen Porous stone
Fig. 8.4 Consolidometer
Each pressure is normally maintained for a period of 24 hours, compression readings being
observed at suitable intervals during this period. The effective stress in the specimen is equal
to the applied pressure at the end of the load increment period. The expansion of the specimen
due to the successive decrease in applied pressure may be measured.
The above test procedure is referred to as the conventional procedure. Based on a detailed
study of this procedure, Leonards and Ramiah (1959) have reported that the void ratio and
effective stress relationship was not significantly affected by
1. moderate variations in room temperature,
2. variations in specimen size (the diameter to height ratio of about 2.75 or more), and
3. variations in the duration of load increment (provided primary consolidation is complete
and secondary compression is not important).
The void ratio at the end of each increment is obtained from the difference in dial gauge
readings and dry weight of the specimen at the end of the test. The method of calculation is
as follows (Fig. 8.5):
Mass of sample measured at the end of test = Ms
Thickness at the end of any increment period = H1
Area of specimen =A
Ms
Equivalent thickness of soil solids, Hs =
AGρw
Void ratio e1 corresponding to pressure p1 is calculated as follows:
H1 − H s H
e1 = = 1 −1 (8.8)
Hs Hs
In the same way, void ratios at the end of each increment period are calculated.
ΔH
Water
H H1
Soil
solids Hs
Fig. 8.5 Phase diagram for specimen
M08_PURU1773_01_SE_C08.indd 220 2/12/2013 8:01:34 AM

8.6 COMPRESSIBILITY CHARACTERISTICS

8.6.1 Pressure–Void Ratio Curves
Figure 8.6 represents the results of a typical laboratory one-dimensional pressure–void ratio
curve. In Fig. 8.6a, the pressure is taken on an arithmetic scale, whereas in Fig. 8.6b, it is
taken on a logarithmic scale. The shape of the e–p curve is dependent on the consolidation
history of the soil.
A flat and somewhat straight curve up to a certain pressure (AB, Fig. 8.6b) is followed
by a steep and fairly straight line (CD) with a smooth transition from the first limb. If the
pressure is released, the soil rebounds and does not reach back to the previous void ratio,
depicting permanent deformation. The recompression curve is somewhat parallel to the first
and gradually blends into the straight line (CDE) (Fig. 8.6b). The point of transition, B, in the
first loading curve corresponds to a state of pressure known as pre-consolidation pressure.
More discussion on the e–p curve is presented in the next section.
8.6.2 Compression Index

The compression index (Cc) is the slope of the linear portion of the pressure–void ratio curve
on a semi-log plot, with pressure on the log scale (IS: 8009 – Part 1, 1976). This is a dimen-
sionless parameter. For any two points on the linear portion of the plot,
e0 − e1 Δe
Cc = = (8.9)
log( p1 / p0 ) log( p1 / p0 )
where p0 and p1 are pressures corresponding to e0 and e1.
The slope of the expansion or decompression part of the e–log p plot (approximated to a
straight line) is referred to as the expansion index, Ce.
Virgin curve
1.3 1.3
Virgin
A B curve
1.2 1.2
Recompression
Void ratio
Void ratio
1.1 1.1
curve C
1.0 1.0
0.9 Rebound D
0.9
Rebound curve
curve E
0.8 0.8
0 400 800 1,200 10 100 1,000
Vertical effective Vertical effective
stress, kN/m2 stress, kN/m2
(a) Arithmetic plot (b) Logarithmic plot
(e –p curve) (e–log p curve)
Fig. 8.6 Pressure–void ratio curves
M08_PURU1773_01_SE_C08.indd 221 2/12/2013 8:01:34 AM

8.6.3 Coefficient of Compressibility

The coefficient of compressibility, av (m2/N), is the secant slope, for a given pressure incre-
ment, of the effective pressure–void ratio curve (IS: 8009 – Part 1, 1976); that is,
Δe
av = (8.10)
Δp
8.6.4 Coefficient of Volume Compressibility

The coefficient of volume compressibility, mv (m2/N), is the compression of a soil layer per
unit of original thickness due to a given unit increase in pressure (IS: 8009 – Part 1, 1976). If
for an increase in effective pressure from p0 to p1 the void ratio decreases from e0 to e1, then
1 ⎛⎜ e0 − e1 ⎞⎟ 1 ⎛⎜ Δe ⎞⎟
mv = ⎜⎜ ⎟⎟ = ⎜ ⎟⎟ (8.11)
1 + e0 ⎜⎝ p1 − p0 ⎟⎠ 1 + e0 ⎜⎜⎝ Δp ⎟⎠
or
1 ⎛⎜ H − H1 ⎞⎟ 1 ⎛⎜ ΔH ⎞⎟
mv = ⎜ ⎟⎟ = ⎜ ⎟⎟ (8.12a)
H ⎜⎜⎝ p1 − p0 ⎟⎠ H ⎜⎜⎝ Δp ⎟⎠
The coefficient of volume compressibility is numerically related to the coefficient of
compressibility as
av
mv = (8.12b)
1 + e0
The value of mv for a particular soil is not constant but depends on the stress range
considered.
8.6.5 Degree of Consolidation

The degree of consolidation (or per cent consolidation), Uz, is the ratio, expressed as a per-
centage of the amount of consolidation at a given time, within a soil mass to the total amount
of consolidation obtainable under a given stress condition (IS: 8009 – Part 1, 1976). This is
expressed as
e 0 − et
Uz = (8.13)
e0 − ef
where ef is the void ratio at the end of consolidation and et the void ratio during consolida-
tion at time t.
For an assumed linear e–p curve, the stress in question is as shown in Fig. 8.7; then, Uz can
be expressed in terms of p as
p − p0
Uz = (8.14)
p1 − p0
Let the stress be increased from p0 to p1 and p be the pressure at any time. Also, let (uw)0
be the pore water pressure before the increase in total stress, (uw)i the increase in pore water
M08_PURU1773_01_SE_C08.indd 222 2/12/2013 8:01:35 AM

e0
e1
uw
(uw) i
p0 p p1
Fig. 8.7 Assumed linear e–p curve
pressure above (uw)0 due to increase in pressure, and uw the pore water pressure at any time
in excess of (uw)0 then,
p1 = p0 + (uw )i = p + uw
(uw )i − uw u
Uz = = 1− w (8.15)
(uw )t (uw )t
8.7 TYPES OF CLAY DEPOSITS

In the natural process of deposition, fine-grained soils, like silt and clay, undergo the process
of consolidation under their own weight of overburden pressure. A state of equilibrium is
reached after a lapse of several years, and the compression ceases. This process continues,
season after season, and sometimes erosion or removal of overburden takes place, and
sometimes the process of consolidation may be continuously taking place due to frequent
deposition. So it is evident that clay soil deposits exist in the field under different conditions,
and their stress history should be known.
8.7.1 Normally Consolidated Clay

If the present effective overburden pressure in the deposit is the maximum pressure to which
the deposit has ever been consolidated at any time in the past, such a deposit is called a
normally consolidated clay deposit. There is no reliable procedure available to predict the
in situ effective stress–void ratio relationship.
A field e–p relationship has to be obtained only from a carefully obtained undisturbed
soil sample. Whatever the care with which the sampling operation is performed, there is
bound to be some disturbance due to stress removal. Accordingly, the shape of the e–log
p (or e–p) curve is strongly influenced depending on the degree of disturbance. Increase
in the degree of disturbance flattens the curve considerably, but the straight line por-
tion (in the e–log p curve) converges at a low void ratio (Fig. 8.8). Schmertmann (1955)
assumed that the straight line portion of the laboratory and in situ curves coincide at
0.42 e0. The in situ void ratio may be taken equal to the initial void ratio at the start of the
test without appreciable loss of accuracy. Further, for normally consolidated clays the
M08_PURU1773_01_SE_C08.indd 223 2/12/2013 8:01:36 AM

p0
e0
Virgin curve
Undisturbed
Disturbed sample
e sample
Completely
re-moulded
sample
log p
Fig. 8.8 Effects of sample disturbance (Source: Leonards, 1962)
pre-consolidation pressure (pc) is equal to the effective overburden pressure (p0). This is
not the case in the deposits which are subjected to thixotropic, secondary compression,
or cementation effects (Leonards, 1962). Thus, the in situ (or unsampled) compression
curve, called virgin curve, may be taken as line CD in Fig. 8.6b. Values of mv or Cc have
to be obtained from the virgin curve and used in the computation of the settlement.
Terzaghi and Peck (1967), based on Skempton’s earlier concept, proposed an empirical
equation for Cc, as Cc = 0.009(wL−10%). Indian Standard (IS: 8009 – Part 1, 1976) suggests
Cc = 0.30(e0 − 0.27).
8.7.2 Over-consolidated Clay

A clay soil deposit that has been fully consolidated under a pressure pc in the past, larger
than the present overburden pressure p0, is called an over-consolidated (pre-consolidated or
pre-compressed) clay deposit. The ratio (pc−p0)/p0 is called the over-consolidation ratio
(OCR). Over-consolidation of clay may be caused by any or a combination of the following
loads (Leonards, 1962):
1. Pressures due to overburden which have been removed (e.g., due to erosion or due to
removal of an old structure)
2. Glacial ice sheets which have since disappeared
3. Sustained seepage forces
4. Tectonic forces caused due to movements in the earth’s crust
5. Fluctuation of the water table
The in situ e–p relationship is radically changed by over-consolidation. Factors other than
pressure which may affect the in situ e–p relationship are weathering, deposition of cementa-
tion materials, and leaching of ions from the pore water. The in situ e–log p curve is obtained
following the procedure given below (Leonards, 1962) (as shown in Fig. 8.9):
1. Consolidation tests are performed on undisturbed samples and loading; unloading
and reloading are continued till a reasonable straight line portion of the e–log p curve is
obtained.
M08_PURU1773_01_SE_C08.indd 224 2/12/2013 8:01:36 AM

p0
Range of pc
e0 G
Most probable value of pc
G′ Beginning of straight
line portion of curve
Parallel
0.42e0 F
p0 log p
Fig. 8.9 In situ e–log p curve (Source: Leonards, 1962)
2. The pre-consolidation pressure, pc, is estimated (discussed in the next section).

3. A line from the point G (e0, p0) is drawn parallel to the mean slope of the rebound curve.
4. Select a point G′ on this line corresponding to pc. From this point, a line is drawn to
meet the point (F) of intersection of the laboratory curve at 0.42 e0. (Schmertmann, 1955,
suggested a range of 0.35 e0 to 0.45 e0.)
5. The dot–dashed lines GG′ and G′F are then used to obtain Cc for calculating the settlement.
8.7.3 Under-consolidated Clay

Rapid natural deposition or deposits under recent fillings may not be fully consolidated
under the present overburden pressure; such clay deposits are called under-consolidated
clays (Fig. 8.10). In such cases, pc < p0, and structures constructed on this deposit will cause
additional compression. No specific procedure is available to get the in situ e–p curve. How-
ever, for all practical purposes, this may be treated as normally consolidated for the purpose
of calculating the settlement.
Net pressure
increment
p p0
e0
Effect of
overburden Total
reduction
Effect of in void ratio
structure
e
log p
Fig. 8.10 Under-consolidated clay deposit
M08_PURU1773_01_SE_C08.indd 225 2/12/2013 8:01:37 AM

8.8 PREDICTION OF PRE-CONSOLIDATION PRESSURE

The earliest and most widely used procedure was suggested by A. Casagrande (1936). The
point B corresponding to maximum curvature (or minimum radius) is chosen on the first
laboratory loading curve (Fig. 8.11). A tangent BT is drawn to the curve at B, and a horizontal
BL is also drawn. The angle α (LBT) is bisected, and the straight line portion of the curve is
projected backwards to intersect the bisector at P. The pressure corresponding to this point
is the pre-consolidation pressure, pc.
Burmister (1951) and Schmertmann (1955) have also suggested procedures to determine
the pre-consolidation pressure.
If pc > p0, the soil may be taken as pre-consolidated; if pc = p0, the soil is normally consoli-
dated; and if pc < p0, the soil is probably under-consolidated (if pressure difference is large)
or nearing normal consolidation condition.
8.9 RATE OF CONSOLIDATION

8.9.1 Terzaghi’s Theory of One-Dimensional Consolidation
A consolidation theory in three dimensions involving stress and strain condition would be
highly complicated because of the heterogeneous nature of the soil. The solutions to three-
dimensional consolidations have to be attended through numerical methods. But the one-
dimensional consolidation theory, as proposed by Terzaghi (1925, 1943), simplifies the
problem smoothly and, at the same time, satisfies the relevant factors connected with
settlement.
In analysing the rate of one-dimensional consolidation, Terzaghi (1925, 1943) made the
following assumptions:
1. The soil mass is homogeneous.
2. The void spaces are completely filled with water.
3. The soil solids and water are incompressible.
4. Darcy’s law is valid.
5. The seepage flow and deformation are in one-dimensional direction.
6. Strains are small.
7. The permeability is constant over the range of effective stresses.
P
L
B a /2
a /2
e T
pc log p
Fig. 8.11 Determination of pc
M08_PURU1773_01_SE_C08.indd 226 2/12/2013 8:01:37 AM

8. There is an unique pressure–void ratio relationship, i.e., the coefficient of compressibility

is constant.
9. The time lag in consolidation is entirely due to the low permeability of the soil.
Consider an element of soil of constant area dx×dy and thickness dz within a clay layer
of 2 d depth free to drain in the z direction only (Fig. 8.12). Let an increment of vertical stress
Δσ be instantaneously applied and maintained constant.
The volume of water flowing into the element in unit time is
qin = vz dx dy (8.16a)
and the volume of water flowing out of the element
⎛ ∂v ⎞
qout = ⎜⎜vz + z dz⎟⎟⎟ dx dy (8.16b)
⎜⎝ ∂z ⎠
As the soil is fully saturated and solid particles and water are incompressible, the law of
conservation of matter requires that qout − qin = change in volume of the element per unit
time. That is,
∂V
Δq = (8.17)
∂t
or
∂V ∂vz
= dx dy dz (8.18)
∂t ∂z
and
V = Vs (1 + e0 ) = dx dy dz (8.19)
or
∂V ∂e
= Vs (8.20)
∂t ∂t
Sand
z
dx dy
dz
2d
Clay layer
Sand
Fig. 8.12 Element within clay layer
M08_PURU1773_01_SE_C08.indd 227 2/12/2013 8:01:38 AM

Substituting for
dx dy dz
Vs =
1 + e0
∂V dx dy dz ∂e
= (8.21)
∂t 1 + e0 ∂t
or
dx dy dz ∂e ∂vt
= dx dy dz
1 + e0 ∂t ∂t
∂e ∂v
= (1 + e0 ) z (8.22)
∂t ∂z
The hydraulic gradient i = ∂h/∂z
or
1 ∂uw ⎛ ⎞
i= ⎜⎜∵ , h = uw ⎟⎟
γ w ∂z ⎜⎜⎝ ⎟
γ w ⎟⎠
or
k ∂uw
vz = ki =
γ w ∂z
or
∂vz k ∂ 2 uw
=
∂z γ w ∂z 2
∂e (1 + e0 ) k ∂ 2 uw
=
∂t γw ∂z 2
Since ∂e = av∂uw
∂e ∂u
= av w
∂t ∂t
or
∂uw ⎛ k ⎞ ∂ 2 uw
av = (1 + e0 )⎜⎜⎜ ⎟⎟⎟
∂t ⎜⎝ γ w ⎟⎠ ∂z 2
or
∂uw ⎛⎜ 1 + e0 ⎞⎟⎛⎜ k ⎞⎟ ∂ 2 uw
= ⎜⎜ ⎟⎜ ⎟
∂t ⎜⎝ av ⎟⎟⎠⎜⎜⎝ γ w ⎟⎟⎠ ∂z 2
M08_PURU1773_01_SE_C08.indd 228 2/12/2013 8:01:40 AM

or
∂uw ∂ 2 uw
= cv (8.23)
∂t ∂z 2
where
k ⎛⎜ 1 + e0 ⎞⎟
cv = ⎜ ⎟⎟ (8.24)
γ w ⎜⎜⎝ av ⎟⎠
Equation 8.23 is the Terzaghi’s one-dimensional consolidation equation, where cv is called
the coefficient of consolidation (m2/year).
As consolidation proceeds, k, e, and av all decrease with time, but the ratio cv remains
approximately constant. The main limitations of Terzaghi’s theory (apart from its one-
dimensional nature) is the non-linearity of the void ratio–pressure relationship.
8.9.2 Time Factor

The solution of Eq. 8.23 for a constant initial pore pressure [(uw)i = (uw)0] satisfies the follow-
ing conditions:
at z = 0 uw = 0
at z = 2d uw = 0
at t = 0 uw = (uw )0
where
∞
2(u )
uw = ∑ [sin ( Mz
w 0
/ d)]
exp(−M 2 Tv ) (8.25)
m= 0
where M = (π/2)(2m+1), m = any integer (0, 1, 2, 3), and d = length of the longest drainage path.
Now, time factor
cv t
Tv = (8.26)
d2
This is a dimensionless factor containing the physical constants of a soil stratum influenc-
ing its time rate of consolidation.
But
⎪⎧ u ⎪⎫
U z = ⎪⎨1 − w ⎪⎬×100 (8.27)
⎪⎪⎩ (uw )i ⎪⎪⎭
Combining Eqs. 8.25 and 8.26, a basic Uz versus Tv relationship can be developed; thus,
∞
2 ⎛⎜ Mz ⎞⎟
Uz = 1− ∑ ⎜⎜⎝sin
2
⎟⎟ exp(−M Tv ) (8.28)
m= 0 M d ⎠
A numerical relationship between Uz and Tv may be obtained by substituting values for

m from 0 to ∞ (Fig. 8.13).
M08_PURU1773_01_SE_C08.indd 229 2/12/2013 8:01:41 AM

20
40
Uz
60
80
100
0 0.2 0.4 0.6 0.8 1.0 1.2
Time factor, Tv
Fig. 8.13 Average percent consolidation versus time factor
The above numerical relationship is not valid for other types of distributions (e.g.,
rectangular, sinusoidal, and trapezoidal) except constant initial excess hydrostatic pressure.
However, since the variation is very small, Fig. 8.13 may be used with reasonable accuracy
(Taylor, 1948).
Two empirical equations instead of Eq. 8.28 are in use, viz.,
π
Tv = (U z / 100)2 when U z ≤ 60% (8.29)
4
and
⎛ U ⎞
Tv = −0.933 log10 ⎜⎜1 − z ⎟⎟⎟ − 0.0851 when U z > 60% (8.30)
⎜⎝ 100 ⎠
8.9.3 Determination of Coefficient of Consolidation

A comparison of laboratory compression versus time and the theoretical Uz versus Tv has
shown similar shapes. This observation resulted in two types of transformation plots, one
using the square root of time (Taylor, 1948) and the other the logarithm of time (Casagrande,
1936b), which are used to determine the coefficient of consolidation and are called fitting
methods.
The Square Root of Time Fitting Method. The theoretical curve Uz versus Tv is a straight
line up to 60% consolidation (Fig. 8.14a), and the abscissa of curve at 90% consolidation is
1.15 times the abscissa of an extension of the straight line. The characteristics of the theoreti-
cal curve have been used by Taylor (1948) to determine a point of 90% consolidation on the
laboratory time curve.
A plot of compression dial reading versus time is made (Fig. 8.14b). In the early portion
of the laboratory curve, a straight line is drawn through the observed points. A second line
is drawn coinciding with the first line at t = 0, such that the abscissa for the new line is 1.15
M08_PURU1773_01_SE_C08.indd 230 2/12/2013 8:01:42 AM

0
4
AB=0.9 π = 0.7976
0 Corrected
AC = 0.849 = 0.9209 zero point
1 Corrected factor
0.9209 20
= = 1.15
0.7976
4 40
U z% π U z%
Experimental
60 curve
A B C
90%
80
A a B C
90
100 1.15a
Root time t min t 90
Tv
(a) Theoretical curve (b) Experimental curve
Fig. 8.14 Square root of time method
times the abscissa of the previous line at a given dial reading. The observed zero reading
and the points of coincidence of the two straight lines will not be the same, but the latter will
usually lie below. This point is called the corrected zero point. The point of intersection of
the second straight line and the laboratory curve corresponds to 90% consolidation, and the
time is designated as t90. Therefore,
cv t90
(Tv )90 = (8.31)
d2
Finding (Tv)90 and deciding the drainage path, d, which is equal to half the thickness of
the layer for double drainage and full thickness for single drainage, we have
0.848 d 2
cv = (8.32)
t90
The Logarithm of Time Fitting Method. The intersection of the tangent and the asymp-
tote to the theoretical consolidation curve is at 100% consolidation (Fig. 8.15a). As the shapes
of the theoretical curve (Uz versus log Tv) and the laboratory compression dial versus log t
curve resemble one another, Casagrande (1936b) suggested that the 100% consolidation
point be obtained by drawing tangents to the straight line portions of the primary consoli-
dation curve.
The early portion of the curve approximates a parabola. The corrected zero point may be
located by taking the difference in ordinates (z0) between two points corresponding to time
t and 4 t on the early portion of the curve and laying off this value (z0) above the curve at
point t (Fig. 8.15b).
As the dial reading corresponding to zero and 100% primary consolidation is known, the
time for 50% consolidation (t50) can be obtained; thus,
cv t50
(Tv )50 = (8.33)
d2
M08_PURU1773_01_SE_C08.indd 231 2/12/2013 8:01:43 AM

0
10
20
Tangents
30
40
Uz% 50
60
70
80 Asymptote
90
100
0.001 0.01 0.1 1 2
Log T
(a) Theoretical curve
Corrected zero point

0
z0
z0
Uz%
100
t 4t
Log time (log t )
(b) Experimental curve
Fig. 8.15 Logarithm of time method
or
0.196 d 2
cv = (8.34)
t50
In general, both fitting methods show good agreement.
Hyperbola Method. The following procedure is adopted for the determination of Cv:
1. From laboratory consolidation test, the time (t) and the specimen deformation (ΔH) are
obtained.
2. A graph of t/ΔH against t is drawn in (Fig. 8.16).
3. The straight line portion bc is projected back to point d. The intercept D is determined.
4. The slope m of the line bc is determined.
5. Then coefficient of consolidation, Cv is calculated as
⎛ mH 2 ⎞⎟
Cv = 0.3 ⎜⎜⎜ ⎟.
⎜⎝ D ⎟⎟⎠
M08_PURU1773_01_SE_C08.indd 232 2/12/2013 8:01:44 AM

t
ΔH
c
m
b 1
d
D
a Time, t
Fig. 8.16 Hyperbola method of determination of Cv
As the unit of D is time/length and the unit of m is (time/length)/time = 1/length, the unit
of Cv is (length)2/time.
The hyperbola method is fairly simple to use, and it gives good results for U = 60%–90%.
8.10 SECONDARY COMPRESSION

A soil mass is said (theoretically) to be fully consolidated under a given pressure when the
excess pore water pressure is zero, depicting the termination of primary consolidation. Actu-
ally, consolidation does not cease but continues slowly even after the excess pore water pres-
sure dissipation, and this continued time-dependent compression is referred to as the
secondary compression. In natural soil deposits where the ground surface is inclined, creeps
and, perhaps, volume changes due to dissipation of non-measurable excess pore water pres-
sure will be taking place (Fig. 8.17). Compression occuring during creep due to readjustment
of the soil skeleton is so low that it may continue for a very long time. Practical evidence
shows that creep ceases or becomes so small that it is not measurable.
Evaluation of secondary compression is difficult. However, by maintaining a constant
pressure on a clay long enough past the point of primary consolidation, a relationship
between secondary compression and time may be obtained, provided temperature control
and equipment corrosion are taken care of. The magnitude of secondary compression is
Compression
Start of
End of secondary compression
primary compression
Cα 10 years
Δt
tlog = tα
Log time
Fig. 8.17 Secondary compression and definition of its rate
M08_PURU1773_01_SE_C08.indd 233 2/12/2013 8:01:44 AM

Table 8.1 Typical values of Cα
Types of soil Cα
1 Normally consolidated clays 0.005–0.02

2 Very plastic soils, organic soils ≥0.03
3 Pre-compressed clays with OCR > 2 <0.001
Source: Ladd (1967).
referred to as a coefficient of secondary compression and expressed as the slope Cα of the

final portion of the time compression curve (Fig. 8.17).
That is,
ΔH / H
Cα =
log(tp + Δt / tp )
(8.35)
where tp is the time when the primary consolidation is complete and Δt the time increment
producing ΔH.
Therefore, the secondary compression or creep in situ for the time increment Δt for a
stratum of thickness H is
tp + Δt
ΔH = HCα log
tp
(8.36)
Secondary compression may be insignificant for inorganic soils but may be predominant
in highly organic soils. Based on different relationships between Cα and Cc, Bowles (1984)
suggested for inorganic soils a value of Cα/Cc ≤ 0.05 and for organic soils a range of 0.07 to
0.10. Typical values for the coefficient of secondary compression for some soils are given in
Table 8.1 (Ladd, 1967).
8.11 CONSOLIDATION SETTLEMENT AND ITS RATES

The discussion in the previous section enabled us to identify two general types of settlement,
viz., immediate settlement and time-dependent settlement (Fig. 8.18). The time-dependent
settlement may be divided into settlement due to consolidation and settlement due to sec-
ondary compression or creep. Hence, the total settlement (S) of any soil mass under an
applied stress is given as
S = Si + Sc + Ss (8.37)
where Si is the immediate settlement, Sc the settlement due to consolidation, and Ss the settle-
ment due to secondary compression.
Depending on the type of soil, each component gets its importance. For coarse-grained
soils Si predominates, in fine-grained soils Sc predominates, and in organic soils Ss has a
vital role to play. Consolidation theory is usually used for the consolidation settlement in
saturated fine-grained soils.
M08_PURU1773_01_SE_C08.indd 234 2/12/2013 8:01:45 AM

Immediate
settlement
Total
settlement
S = Si + Sc + Ss
Time-dependent
Settlement
settlement
Primary
compression = Sc
Theoretical
Secondary compression
= Ss
Period of Period of
Primary compression Secondary Actual
compression
ti ≈ C t100
Time
Fig. 8.18 Components of total settlement
In this section, only the consolidation settlement is discussed, and immediate settlement
component is considered in Chapter 15.
8.11.1 Consolidation Settlement

Consolidation settlement is calculated knowing the following:
1. Thickness of compressible layer, H
2. Initial void ratio, e0
3. Field e–log p curve or Cc, mv, or av corresponding to the pressure range
Considering Eq. 8.7 and letting ΔH = Sc, we get the generalized equation for consolida-
tion settlement; that is,
Δe
Sc = H (8.38)
(1 + e0 )
Compression index from the field e–log p curve is given in Eq. 8.9 as
Δe
Cc =
log ( p1 / p0 )
p1
Δe = Cc log
p0
M08_PURU1773_01_SE_C08.indd 235 2/12/2013 8:01:46 AM

Substituting in Eq.8.38, we get
Sc = H C log p1 (8.39)
1 + e0 c p0
For over-consolidated clays, Cc has to be determined from the line GG’ or G’F (Fig. 8.9)
depending on the present pressure.
From the laboratory e–log p curve, plot the field e–p curve taking corresponding e and p
values. From such a curve, find av corresponding to the pressure range; then,
Δe
av =
Δp
or
Δe = av Δp
Substituting in Eq. 8.38, we get

av
Sc = HΔp (8.40)
(1 + e0 )
But
av
mv =
1 + e0′
Thus,
Sc = mv HΔp (8.41)
Equations 8.38 to 8.41 can be used for the computation of settlement.

Equation 8.41 can be used only if the thickness of the compressible layer is less and the
assumption of constant mv and Δp with depth is valid. Settlement computation for thick
layers is discussed below.
As the lateral strain is zero, the decrease in volume per unit volume is equal to the decrease
in thickness per unit thickness. This is nothing but the settlement per unit depth, that is,
ΔV = e0 − e1
V 1 + e0
or
dSc e0 − e1
=
dz 1 + e0
or
⎛ e − e1 ⎞⎟⎛ p1 − p0 ⎞⎟ a
dSc = ⎜⎜ 0 ⎜ dz = v Δp dz
⎜⎝ p1 − p0 ⎟⎟⎠⎜⎜⎝ 1 + e0 ⎟⎟⎠ 1 + e0
M08_PURU1773_01_SE_C08.indd 236 2/12/2013 8:01:47 AM

p0 Δp mv mv Δp
H
Sc = Area
(a) (b) (c)
Fig. 8.19 Graphical procedure for consolidation settlement
or
dSc = mv Δp dz (8.42)
H
Sc = ∫ mv Δp dz (8.43)
0
If mv and Δp are constants, Eq. 8.43 reduces to Eq. 8.41.

The variations of initial effective vertical stress (p0) and effective vertical stress increment
(Δp) over the depth of the layer are shown in Fig. 8.19. The variation of mv is shown in
Fig. 8.19b. Figure 8.19c represents the variation of non-dimensional factor mvΔp with depth.
The area under this curve is the settlement of the layer.
8.11.2 Construction Period Correction

In practice, real loads are added to the soil over a period of time, perhaps after many months
or years, and not instantaneously. Due to excavation, there is a reduction in the net load,
resulting in swelling of the clay. Settlement begins only when the applied load exceeds the
weight of the excavated soil. The corrected settlement may be obtained adopting Terzaghi’s
approximate method (also adopted by IS: 8009 – Part 1, 1976). This is based on the following
assumptions: (i) the actual settlement at the end of the construction time is the same as that
resulting from the total load acting for half of the loading time and (ii) the load–time rela-
tionship is linear. Thus, the settlement at the end of the loading period is equal to the settle-
ment on the instantaneous load curve corresponding to one-half of the total loading period.
The instantaneous settlement–time curve is plotted from point O (Fig. 8.20). As per the
assumption, the amount of consolidation settlement at time tc is the same as that obtained
from the instantaneous curve at 1/2tc. A point P is located on the instantaneous curve at
time tc/2, and a point Q is established by drawing a parallel to the time axis from P to meet
the ordinate at tc. From points P and Q onwards, the abscissa of the corrected curve will be
1/2tc greater than for the instantaneous curve. To get points during the construction period,
choose a point R at time 1/2t, and point S is found by projecting horizontally onto the ordi-
nate at t with the line OS. That is, the settlement at t (i.e., ordinate BT), is equal to (OB/OA)
AS, i.e., t/tc As. This process is repeated until additional points are obtained for construction
of the corrected curve.
M08_PURU1773_01_SE_C08.indd 237 2/12/2013 8:01:48 AM

Effect of
Actual construction period
0 Load
Time
Load
q
Time
t/2 tc/2 t tc Time

0
B A
T
S Corrected curve
Settlement
R Q
P tc /2
Instantaneous
curve
Fig. 8.20 Correction for construction period
8.11.3 Settlement Due to Secondary Compression

Considering Eq. 8.35 and letting ΔH = Ss, the settlement due to secondary compression is
tp + Δt
Ss = HCα log (8.44)
tp
8.11.4 Predicting the Rate of Settlement

The settlement Sc is computed from Eqs. 8.38 to 8.41. An approximate value of cv is selected
corresponding to the average pressure during consolidation. The time rate of settlement is
calculated using the equation given below:
Tv (d)2
t= (8.45)
cv
where Tv is calculated corresponding to the desired degree of consolidation and d is the
drainage path in the field.
Now,
εav = U z εf (8.46)
Sc
εf = (8.47)
H
M08_PURU1773_01_SE_C08.indd 238 2/12/2013 8:01:49 AM

where εav is the average strain in the clay layer, Sc the total calculated consolidation
settlement, H the height of the field clay layer, and
St
εav = (8.48)
H
where St is the settlement of field layer at time t.
The degree of consolidation for the full clay layer is
St
Uz = (8.49)
Sc
or
St = U z Sc (8.50)
Settlement of clay layer, St, and the time for consolidation are computed for Uz varying
from 10% to 100%. The time values are calculated from Eq. 8.45 and the settlement values
from Eq. 8.50.
For very thick clay layers, to determine cv, several tests on undisturbed samples have to
be made. If variation in cv for each layer is not wide, an arithmetical average may be taken.
8.12 ACCELERATION OF CONSOLIDATION BY SAND DRAINS

The normally slow process of consolidation of fine-grained soils may be accelerated by
shortening the length of the drainage path of the pore water escaping from the consolidating
soil. This may be accomplished by installing a system of vertical sand drains as shown in
Fig. 8.21a. A sand drain essentially consists of a vertical borehole put down through the
saturated fine-grained soil and extending to a relatively firm bottom. By this provision, the
excess pore water pressures dissipate by both vertical and radial flow. However, the result-
ing settlement will still be in the vertical direction.
The basic theory of sand drains was presented by Rendulic (1935) and Barron (1948).
A review of the theories of sand drains was made by Richart (1957). In the analysis of theo-
ries, two conditions prevail:
1. Free-strain condition: Application of a flexible surcharge load will cause an uneven settle-
ment at the surface, and this condition is referred to as free-strain condition.
2. Equal-strain condition: Application of a rigid surcharge load will cause an equal settlement
at the surface, and this condition is referred to as equal-strain condition.
Considering an element of soil A with vertical and radial flows (Fig. 8.21b) and devel-
oping the governing differential equations similar to one-dimensional theory (for detailed
derivation, refer Leonards, 1962), we have
∂ 2 uw ⎡ ∂2u 1 ∂uw ⎤⎥ ∂uw

cv + ch ⎢⎢ w
+ = (8.51)
∂z 2 ⎢⎣ ∂r
2 r ∂r ⎥⎥ ∂t
⎦
M08_PURU1773_01_SE_C08.indd 239 2/12/2013 8:01:50 AM

q Horizontal sand
blanket
rc
Vertical sand drains
Clay Ar rw
z Impermeable base
(a) Sand drains arrangement
Vz + ΔVz
Vr
Vr + ΔVr Control line of
Δz sand drain
θ
Δr r rw
Vz
(b) Element A
Fig. 8.21 Three-dimensional consolidation
where
k v (1 + e0 )
cv =
γ w av
is the coefficient of vertical consolidation (as previously defined, Eq. 8.24) and
k h (1 + e0 )
ch = (8.52)
γ w av
is the coefficient of radial and horizontal consolidation, where kv and kh are referred to as
vertical and horizontal permeabilities, respectively.
Equation 8.51 is the governing differential equation for three-dimensional consolidation
and may be considered to consist of two parts:
One-dimensional flow:
∂ 2 uw ∂uw
cv =
∂z 2 ∂t
Radial flow:
∂ 2 uw ∂u ∂u
ch +1 w = w (8.53)
∂r 2 r ∂r ∂t
A solution to Eq. 8.51 has been considered by Carilo (1942) as a combination of solutions
of Eqs. 8.23 and 8.53, and accordingly we have
M08_PURU1773_01_SE_C08.indd 240 2/12/2013 8:01:51 AM

Square arrangement (re = 0.564)

0 S
0.1 S
Average degree of radial
0.2 n = 7 10 15 2
consolidation Ur
5 0 3
0.3 0 4
0 6 Hexagonal arrangement
0 1
0.4 00
(re = 0.525)
Sand drains
0.5
S S
0.6
0.7
0.8
0.9
1.0
0.01 0.02 0.04 0.05 0.1 0.2 0.3 0.4 0.5 1 2 3 4 5 6 8 10
Time factor, Tr
Fig. 8.22 Average degree of consolidation for radial flow
(1 − U ) = (1 − U z )(1 − Ur ) (8.54)
where U is the degree of consolidation for three-dimensional flow, Uz the degree of consoli-
dation for one-dimensional flow,and Ur the degree of consolidation for radial flow
The time factor Tr for radial flow is
Ch t
Tr = (8.55)
(2re )2
where re is the radius of influence (Fig. 8.21a). The time factor Tv for one-dimensional con-
solidation is as given by Eq. 8.25.
The solution to Eq. 8.23 is already discussed, and the solution for Eq. 8.53 was given by
Rendulic (1935) for free-strain condition and by Barron (1948) for equal-strain condition.
It has been reported by Richart (1957) for values of n(= re/rw) greater than 5 that both the
solutions give closer values. Hence, for all practical purposes, Fig. 8.22 (Barron, 1948) may
be used regardless of the imposed boundary strain condition.
During drilling operation, a zone of soil surrounding the drain, called smear zone, is
formed due to re-moulding of the clay, which causes a reduction in horizontal permeability.
This factor has to be considered while selecting the diameter of the drain.
8.13 COMPRESSIBILITY OF SANDS

Sandy soils deform less under static loads than under dynamic loads. A static load does
cause a reduction in the void space in the sand mass. In general, the settlement in non-
cohesive soils occurs immediately after the application of load. This settlement is attributed
to the volume changes caused by lateral yielding or shear strain that occurs in the soil. The
settlement may be calculated considering the sand layer as an elastic medium (discussed in
Chapter 15).
In uniform non-cohesive soil deposits, the soil rigidity (or modulus of elasticity) increases
with depth because of increase in confining pressure, which is a function of overburden
pressure. Further, if the loaded area is large, the settlement is governed by the contact
M08_PURU1773_01_SE_C08.indd 241 2/12/2013 8:01:51 AM

pressure, which is a function of the modulus of sub-grade reaction at different points. In

such situations, compression versus pressure curves may be used, the curve being obtained
by field testing or by the use of a simple compression test in the laboratory in which the lat-
eral deformations are prevented.
However, sufficient judgement must be exercised in using the labortory test results to
predict field settlement of sand layers. The most common method to predict the settlement is
by adopting semi-empirical rules relating settlement to the results of field penetration tests.
WORKED EXAMPLES
Example 8.1 The dial readings recorded during a consolidation test at a certain load are as
follows:
Time Dial reading Time Dial reading

(minutes) (mm×10−5) (minutes) (mm×10−5)
0.0 240 15 622
0.10 318 30 738
0.25 340 60 842
0.50 360 140 930
1.00 385 240 975
2.00 415 1,200 1,070
4.00 464 1,620 1,090
8.00 530
The initial thickness of the sample is 25 mm. Compute the coefficient of consolidation by
square root of time and logarithm of time-fitting methods.
Solution
Dial reading Time Dial reading Time
(mm×10−5) (mm×10−5)
( min ) ( min )
240 0 622 3.873
318 0.316 738 5.477
340 0.500 842 7.746
360 0.707 930 11.832
385 1.000 975 15.492
415 1.414 1,070 34.641
464 2.000 1,090 40.249
530 2.828
M08_PURU1773_01_SE_C08.indd 242 2/12/2013 8:01:52 AM

200
300
400
t90 = 7.2 min
t90 = 51.84 min.
500
Dial reading, mm × 10–5
600
700
800
900
1,000
a
1,100 1.15a
1,200
0 8 16 24 32 40 48 56 64
Root time, min
Fig. 8.23
Dial reading versus Time is plotted as shown in Fig. 8.23. The Time for 90% consoli-
dation is read from the graph as
t90 = 7.2 min
t90 = 51.84 min
(1090 − 240)×10−5
Average thickness of the sample = 25 − = 24.96 min
2
Length of drainage path d = 24.96 = 12.48 mm
2
(Tv )90 d 2 0.848 ×12.48 2

cv = = = 2.55 mm 2/min
t90 51.84
M08_PURU1773_01_SE_C08.indd 243 2/12/2013 8:01:53 AM

Dial reading
200
Zero point = 310
Z0
Z0 100% consolidation = 930
400
50% consolidation = 620
Dial reading, mm × 10−3
t0 = 15 min
600
800
1,000
1,200
1,400
0.1 1 10 100 1,000 10,000
Log time, minutes
Fig. 8.24
Dial reading Vs log t is plotted as shown in Fig. 8.24. The time for 50% consolidation is
read from the graph as
t50 = 15 min
(Tv )50 d 2 0.196 ×12.48 2

cv = = = 2.04 mm 2/min
t50 15
Example 8.2 A large highway interchange fill was built over a 15 m thick soil structure. The
initial void ratio was 0.78. After 70% settlement, the void ratio was 0.69. Estimate the total
anticipated settlement.
Solution
From Eq. 8.13,
e 0 − et
Uz =
e0 − ef
M08_PURU1773_01_SE_C08.indd 244 2/12/2013 8:01:54 AM

That is,
0.78 − 0.69
0.70 =
0.78 − ef
Therefore,
ef = 0.65
From Eq. 8.7,
ΔH = H Δe
1 + e0
Therefore,
(0.78 − 0.65)
ΔH = 15× = 1.096 m
1 + 0.78
Example 8.3 A layer of clay of 8 m thickness is located between two sand layers. The coef-
ficient of consolidation of the clay is 5×10−2 m2/s. The fill material was placed over a very
large area above the top sand layer. After 3 years, what percentage of the additional pressure
due to the fill material will be carried by the soil grains at the middle of the clay layer?
Solution
From Eq. 8.26,
cv t
Tv =
d2
That is,
Tv = 5×10−2 × 3 × 365× 24 × 60 × 60 = 0.296
1000 ×1000 (8 / 2)2
From Eq. 8.29,

Tv = π (U z / 100)2
4
That is,
2
⎛U ⎞
U v = π ⎜⎜ z ⎟⎟⎟
4 ⎜⎝ 100 ⎠
That is,
Uz
= 2× 0.296 = 0.434
100 π
Therefore,
U z = 43.4%
Additional pressure due to fill material carried by the soil grains is 43.4%.
M08_PURU1773_01_SE_C08.indd 245 2/12/2013 8:01:56 AM

Example 8.4 A laboratory specimen of clay 20 mm thick, drained at the top and bottom,
has taken 360 seconds to reach 50% consolidation when the pressure was increased from
75 to 150 kN/m2. The initial void ratio was 0.80, and the final void ratio due to the increment
of load was 0.45. Determine the coefficient of permeability of the clay.
Solution
For 50% consolidation, Tv = 0.196. From Eq. 8.26,
Tv d 2
cv =
t
That is,
( )
2
cv = 0.196 × 20 = 5.44 ×10−8 m 2/s
360 100 × 2
From Eq. 8.24,

⎛ 1 + e0 ⎞⎟
cv = k ⎜⎜ ⎟
γ w ⎜⎝ av ⎟⎠
That is,
cv av γ w 5.44 ×10−8 × 4.67 ×10−3 × 9.81
k= =
1 + e0 1 + 0.80
Therefore,
k = 1.39×10−9 m / s
Example 8.5 A soil has a compression index, Cc, of 0.28. At a stress of 120 kN/m2, the void
ratio was 1.02. Calculate (i) the void ratio if the stress on the soil is increased to 180 kN/m2
and (ii) the total settlement of the stratum of 6 m thickness.
Solution
e0 − e1
Cc =
log10 ( p1 / p0 )
or
1.02 − e1
0.28 =
log10 (180 / 120)
or
e1 = 1.02 − 0.28 log10 180 = 0.971
120
M08_PURU1773_01_SE_C08.indd 246 2/12/2013 8:01:57 AM

Now,
Sc = H C log p1
1 + e0 c 10
p2
or
Sc = 5×1000 × 0.28 × log10 180 = 122 mm

1 + 1.02 120
Example 8.6 A soil profile at a certain location is as shown in Fig. 8.25. A rigid circular foun-
dation of 5 m diameter rests on the sand. The contact pressure at the underside of the foun-
dation is 250 kN/m2. The average coefficient of compressibility of the clay is 0.606×10−3 m2/kN
for the stress range encountered. Estimate the ultimate settlement of the foundation assuming
a 30° load distribution.
Solution
Consider that there is no capillary rise of water in sand. Thus,
Sand:
Gγ w 2.65× 9.807
γd = = = 16.9 kN / m 3
1+ e 1 + 0.54
( )
γ sat = G + e γ w = 2.65 + 0.54 9.807 = 20.3 kN / m 3
1+ e 1 + 0.54
Clay:
( )
γ sat = 2.7 + 1.2 9.807 = 17.4 kN / m 3
1 + 1.2
The initial effective stress at the centre line of the clay layer
5m
250 kN/m2
1m
5m 30° e = 0.54 30°

G = 2.65 Sand
e = 1.20
G = 2.70 Clay
9m
Sand
Fig. 8.25
M08_PURU1773_01_SE_C08.indd 247 2/12/2013 8:01:59 AM

p0 = (1×16.9 + 4 × 20.3 + 4.5×17.4 − 8.5× 9.807 ) = 93.0 kN / m 2

The final effective stress at the centre line of the clay layer
π× 52 4
p1 = 93 + 250 × ×
4 π(5× 2× 4.75)2
= 122.7 kN / m 2
Now,
av 0.606 ×10−3
mv = = = 0.275×10−3 m 2/ kN
1 + e0 1 + 1.2
Sc = mv HΔp
= 0.275×10−3 × 9×1000(122.7 − 93.0)
= 73.5 mm
Example 8.7 In an oedometer test, a clay specimen initially 25 mm thick attains 90%
consolidation in 10 minutes. In the field, the clay stratum from which the specimen was
obtained has a thickness of 6 m and is sandwiched between two sand layers. A structure
constructed on this clay experienced an ultimate settlement of 200 mm. Estimate the
settlement at the end of 100 days after construction.
Solution
Tv d 2
cv =
t
For the oedometer test for 90% consolidation, Tv is 0.848. Therefore,
2
0.848 ⎛⎜ 25 ⎞⎟
cv = ⎜ ⎟ = 0.221 mm 2/ s
10 × 60 ⎜⎝ 2 ⎟⎠
For the field condition,

Tv (6 ×1000 / 2)2
0.221 =
100 × 24 × 60 × 60
0.221×100 × 24 × 60 × 60
Tv = = 0.212
3000 2
For Uz ≤ 60%, the maximum Tv value is 0.283. Therefore, the expression for Tv to be used is
⎛ π ⎞⎛ U ⎞2
Tv = ⎜⎜⎜ ⎟⎟⎟⎜⎜ z ⎟⎟⎟
⎝ 4 ⎠⎜⎝ 100 ⎠
M08_PURU1773_01_SE_C08.indd 248 2/12/2013 8:02:00 AM

⎛ π ⎞⎛ U ⎞2
0.212 = ⎜⎜⎜ ⎟⎟⎟⎜⎜⎜ z ⎟⎟⎟
⎝ 4 ⎠⎝ 100 ⎠
0.212× 4 ×1002
Uz = = 51.96%
π
We know Uz is also given as

St
Uz = ×100
Sc
U z Sc 51.96 × 200
St = = = 103.9 mm
100 100
Example 8.8 A 2 m thick layer of compressible soil with an average void ratio of 1.02 is
subjected to a total overburden pressure of 150 kN/m2. Laboratory consolidation tests show
that the equation for the e–log p curve is
p
e = e0 − 0.125 log10
p0
A large structure at the ground surface increases the pressure in the compressible layer
by 70 kN/m2.
1. Find the change in volume per unit volume of the compressible layer.
2. If the soil is laterally confined so that all the volume change results in change in height,
calculate the settlement that would result due to the increased load.
Solution
After an increased pressure of 70 kN/m2,
150 + 70
e = 1.02 − 0.125 log10 = 0.999
150
1. From Eq. 8.11,

Δe / Δp
mv =
1 + e0
1 (1.02 − 0.999)
mv = = 1.49×10−4 m 2/ kN
1 + 1.02 70
2. From Eq. 8.41, Sc = mvHΔp. Therefore,

Sc = 1.49×10−4 × 2×100 ×70 = 2.09 cm
M08_PURU1773_01_SE_C08.indd 249 2/12/2013 8:02:02 AM

Example 8.9 A normally consolidated clay layer of 10 m thickness has a unit weight of
20 kN/m3 and specific gravity of 2.72. The liquid limit of the clay is 58%. A structure
constructed on the clay increases the overburden pressure by 10%. Estimate the ultimate
consolidation settlement. There is no secondary compression.
Solution
For normally consolidated clays,
Cc = 0.009(wL − 10%)
= 0.009(58 − 10) = 0.432
Considering the unit weight as saturated unit weight, e0 is calculated from

G + e0
γ sat = γw
1 + e0
That is,
2.72 + e0
20 = 9.81
1 + e0
Solving, e0 = 0.656. From Eq. 8.39, the consolidation settlement
H p
Sc = Cc log 1
1 + e0 p0
10 p + 0.1p0
= × 0.432 × log 0 × 100
1 + 0.656 p0
That is,
Sc = 10.8 cm
Example 8.10 The overburden pressure on a normally consolidated 5 m thick clay struc-
ture is 250 kN/m2. A laboratory consolidation test followed by a suitable correction gave
two points on the field curve:
e1 = 1.12 p1 = 150 kN / m 2
e2 = 1.02 p2 = 450 kN / m 2
How much settlement does the clay layer experience due to an additional load intensity
of 150 kN/m2?
Solution
Overburden pressure = 250 kN/m2
e1 = 1.12 p1 = 150 kN / m 2
e2 = 1.02 p2 = 450 kN / m 2
M08_PURU1773_01_SE_C08.indd 250 2/12/2013 8:02:03 AM

From Eq. 8.9,

e0 − e1 1.12 − 1.02
Cc = = = 0.209
log( p1 / p0 ) log( 450 / 150)
From Eq. 8.39,

0.209 450
Sc = × 5× log ×100
1 + 1.12 150
= 12.6 cm
Example 8.11 A homogeneous clay layer, 9 m thick, is expected to have an ultimate

settlement of 308 mm. After a time span of 2 years, the average settlement was measured
to be 108 mm. How much longer will it take for the average settlement to attain 220 mm?
Solution
Degree of consolidation Uz = (St/Sc) × 100%
108
Uz = ×100 = 35.07%
308
For Uz≤ 60%, Tv is obtained from

2
⎛ π ⎞⎛ U ⎞
Tv = ⎜⎜⎜ ⎟⎟⎟⎜⎜ z ⎟⎟⎟
⎝ 4 ⎠⎜⎝ 100 ⎠
That is,
⎛ π ⎞⎛ 35.07 ⎞⎟2
(Tv )1 = ⎜⎜⎜ ⎟⎟⎟⎜⎜⎜ ⎟ = 0.0966
⎝ 4 ⎠⎝ 100 ⎟⎠
Now, cv = (Tvd2/t) from the first condition

(Tv )1 d 2
cv =
t
Again,
220
Uz = ×100 = 71.4%
308
Now, Tv is obtained from

⎛ U ⎞
Tv = −0.933 log10 ⎜⎜1 − z ⎟⎟⎟ − 0.0851
⎜⎝ 100 ⎠
That is,
⎛ 71.4 ⎞⎟
(Tv )2 = −0.933 log10 ⎜⎜⎜1 − ⎟ − 0.0851 = 0.422
⎝ 100 ⎟⎠
M08_PURU1773_01_SE_C08.indd 251 2/12/2013 8:02:05 AM

(Tv )2 d 2 (Tv )2 d 2 t1 (Tv )2

t= = = t1
t (Tv )1 d 2 (Tv )1
0.422
= × 2 = 8.733 years
0.0966
The additional time for the average settlement to attain 220 mm is 8.73 years.
Example 8.12 In a consolidation test, the pressure on a sample was increased from 150 to
300 kN/m2. The void ratio after 100% consolidation under 150 kN/m2 was 0.945, and that
under 300 kN/m2 was 0.812. The coefficient of permeability of the soil was 25×10−6 mm/s,
and the initial height of the sample was 20 mm. Determine (i) the coefficient of compress-
ibility, (ii) the coefficient of volume compressibility, (iii) the coefficient of consolidation, and
(iv) the time taken in days for 90% consolidation of the layer of this clay, 0.5 mm thick in the
field and sandwiched between an impervious layer beneath and the pervious layer on top.
Solution
Δe = 0.945 − 0.812 = 0.133
Δp = 300 − 150 = 150 kN / m 2
Δe 0.133
av = = = 8.87 ×10−4 m 2/ kN
Δp 150
av 8.87 ×10−4
mv = = = 4.56 ×10−4 m 2/ kN
1 + e0 1 + 0.945
k 25×10−6 1
cv = = × = 5.59×10−6 m 2/ s
γ w 1000 × 9.807 4.56 ×10−4
d2
t= Tv
cv
52 × 0.848
t= = 43.9 days
5.59×10−6 × 60 × 60 × 24
Example 8.13 A clay deposit of 5 m thick is covered with coarse sand at top and bottom
and has a permeability of 0.025 m/yr. It is additionally loaded by a pressure of 65 kpa. Labo-
ratory tests on the field specimen indicate that the consolidation in the field will be only 50%
in a six-month period. Find the settlement of the clay deposit in one year.
M08_PURU1773_01_SE_C08.indd 252 2/12/2013 8:02:06 AM

Solution
2
Coefficient of consolidation Cv = Tr d
t
2
= 0.197 × 2.5 = 2.463 w2/y
0.5
Coefficient of volume⎫⎪⎪ k 0.025

⎬ mv = =
compressibility ⎪⎪⎭ Cv γ ω 2.463 × 9.8
= 1.036 × 10−3 m2/km.

Settlement, Sf = mrH0Δσ’
= 1.036 × 10−3 × 5.0 × 65
= 0.337 m.
As the consolidation will be 50% in six months, in one year the consolidation will be com-
plete and the settlement will be 0.337 m.
Example 8.14 Two clay specimens, P and Q, whose initial void ratios at equilibrium under
a pressure of 1 km/m2 are 0.56 and 0.65, respectively, decrease to 0.51 and 0.63, respectively,
when the pressure is increased to 1.5 km/m2. The thickness of specimens initially was 35 and
25 mm for P and Q, respectively. The time taken by specimen P to reach 50% consolidation
is 30% of that required by specimen Q for reaching the same degree of consolidation of per-
meability of P and Q.
Solution
Δe 1
mv = .
1 + e0 Δσ ′
Hence for specimen P:
0.56 − 0.51 1 0.05

( m v )P = × = .
1 + 0.56 0.5 1.56 × 0.5
For specimen Q:
0.65 − 0.63 1 0.02

( m v )Q = × = .
1 + 0.65 0.5 1.65 × 0.5
( m v )P 0.05 1.65 × 0.5

Hence, = = = 2.64.
( m v )Q 1.56 × 0.5 0.02
M08_PURU1773_01_SE_C08.indd 253 2/12/2013 8:02:08 AM

tP ( hP ) (Cv )Q
2
=
Now, tQ ( h )2 (Cv )P
Q
⎛ 35 ⎞⎟2
⎜⎜ ⎟
300 ⎜⎝ 2 ⎟⎠ ( C v )Q
i.e., = ×
100 ⎛ 25 ⎞ ( C v ) P
2
⎜⎜ ⎟⎟
⎜⎝ 2 ⎟⎠
( C v )Q ( 17.5 ) 2 1
= × = 6.53.
( v )P ( 12.5 )
C 2
0 .30
k
We know, Cv =
mr γ ω
kP ( C v )P ( m v )P × γω
∴ =
k Q ( C v )Q ( m v )Q × γ ω
= 6.53 × 2.64
= 17.25.
POINTS TO REMEMBER
8.1 A soil mass subjected to a change in the stress system resulting in a volume decrease is
termed the compressibility or compression of soil. The compression may be immediate
and/or time dependent.
8.2 The time-dependent volume decrease is attributed to compression of soil matter,
compression of water and air within the voids, and escape of water or air from the voids.
8.3 One-dimensional consolidation is based on the following considerations: (i) the
displacements are vertical, (ii) there are no lateral strains, (iii) flow of water from the
soil layer is in a vertical direction only, and (iv) the change in void ratio is a direct
function of the vertical component of effective stress.
8.4 The pressure (σ′)–void ratio (e) curve, obtained from a consolidation test, depicts the
consolidation history of the soil. The initial flatter portion of the pressure–void ratio
curve is called the recompression curve, and the steeper portion after the break in the
curve is called the virgin compression curve.
8.5 The pre-consolidation pressure (pc), the maximum stress ever experienced by the soil,
is reflected by the point of transition in the first loading portion of the pressure–void
ratio curve.
M08_PURU1773_01_SE_C08.indd 254 2/12/2013 8:02:08 AM

8.6 Compression index (Cc), a dimensionless parameter, is the slope of the linear portion
of the pressure–void ratio curve on a semi-log plot.
8.7 Coefficient of compressibility (av) is the second slope for a given pressure increment
of the pressure–void ratio curve.
8.8 Coefficient of volume compressibility (mv) is the compression of a soil layer per unit
of original thickness due to a given increase in pressure.
8.9 Degree of consolidation (Uz) is the ratio expressed as a percentage of the amount of
consolidation at a given time within a soil mass to the total amount of consolidation
obtainable under a given stress condition.
8.10 If the present effective overburden pressure in a deposit is the maximum pressure to
which the deposit has ever been consolidated at any time in the past, such a deposit is
called a normally consolidated clay deposit.
8.11 A clay deposit that has been fully consolidated under a pressure pc, in the past, longer
than the present overburden pressure p0 is called an over-consolidated (pre-consoli-
dated or pre-compressed) clay deposit.
8.12 Rapid natural deposition or deposits under recent fillings may not be fully consolidated
under the present overburden pressure; such a clay deposit is called under-consoli-
dated clay deposit.
8.13 Terzaghi’s one-dimensional consolidation theory predicts the time rate of consolida-
tion, i.e., it predicts the water pressure at any elapsed time and at any location in a soil
deposit.
8.14 Time factor is a dimensionless factor containing the physical constants of a soil stratum
influencing its time rate of consolidation.
8.15 Methods used to determine the coefficient of consolidation are called time-fitting
methods. They are time -fitting and log(time)-fitting methods.
8.16 Consolidation does not cease even after dissipation of excess pore water pressure but
continues. This continued time-dependent compression is referred to as the secondary
compression.
8.17 Total settlement is a combination of immediate settlement (Si), settlement due to
consolidation (Sc), and settlement due to secondary compression (Ss).
8.18 Settlement in coarse-grained soils occurs immediately, which is attributed to the
volume changes caused by lateral yielding or shear strain which occur in the soil.
QUESTIONS
Objective Questions
8.1 Identify the true and false statements from the following:
1. The coefficient of volume compressibility (mv) of a soil is always less than its
coefficient of compressibility (av).
2. For an intensive normally consolidated soil, the compression index can also be
found from remoulded samples.
M08_PURU1773_01_SE_C08.indd 255 2/12/2013 8:02:09 AM

3. Settlement resulting from the dissipation of the excess pore water pressure is
referred to as primary consolidation.
4. The rate of consolidation for a given soil deposit is a constant.
5. The time factor is a non-dimensional parameter and is independent of the degree
of consolidation.
8.2 When consolidation of saturated soil sample takes place, the degree of saturation
(a) Decreases (b) Increases
(c) Remains constant (d) Decreases initially and then increases
8.3 The liquid limit of a saturated normally consolidated soil is 50%. The compression
index of the soil for the virgin compression curve will be
(a) 0.36 (b) 0.505
(c) 0.605 (d) 0.705
8.4 Consolidation time
(a) Increases with increasing compressibility
(b) Increases rapidly with decreasing size of soil mass
(c) Decreases with increasing permeability
(d) Is dependent on the magnitude of the stress change
8.5 The value of volume compressibility is not constant but varies with the
(a) Thickness of the compressible layer (b) Level of effective areas
(c) Coefficient of permeability (d) Time factor
8.6 Which soil parameter is considered as a measure of the degree of over-consolidation?
(a) Pre-consolidation pressure (b) Compression index
(c) Over-consolidation ratio (d) Coefficient of consolidation
8.7 Compression in saturated soils may be caused by
1. Rearrangement of particles 2. Seepage of water
3. Crushing of particles 4. Elastic distortions
(a) 1, 2, and 3 are correct (b) 2, 3, and 4 are correct
(c) All are correct (d) 3, 4, and 1 are correct
8.8 Indicate the incorrect statement. In a conventional consolidation test, the following
factors do not significantly affect the result:
(a) Moderate variation in room temperature
(b) Variation in pore fluid properties
(c) Variation in specimen size
(d) Variation in duration of load increment
8.9 Assertion A: Both mv and av decrease with increasing stress.
Reason R: Both mv and av are influenced by the stress history.
Choose the correct codes.
M08_PURU1773_01_SE_C08.indd 256 2/12/2013 8:02:09 AM

8.10 A sand exhibits considerable secondary compression due to

(a) High stresses (b) High pore water pressure
(c) Presence of less fines (d) Delayed consolidation
8.11 Explain why the measured values of time and rate of settlement do not agree with the
predicted values.
8.12 How do you estimate the field e–p curve of an over-consolidated clay ?
8.13 Briefly explain the physical meaning of the coefficient of consolidation cv.
8.14 Settlements computed are normally higher than the actual settlements. Discuss this
statement critically.
8.15 Explain why there is a significant time lag in the settlement of clay soils but not in
sandy soils.
8.16 Why does it take infinite time for complete consolidation to occur?
8.17 Explain in sequence the occurrences that take place when a saturated clay deposit
undergoes consolidation.
8.18 The water table in a lake has been lowered by 20 m. Will this cause consolidation
settlement of the lake-bed sediments? Explain.
EXERCISE PROBLEMS
8.1 An oedometer test was made on a sample of saturated marine clay. The dial readings
corresponding to the final equilibrium condition under each load are given below:
Pressure (kN/m2) 13.3 26.6 53.3 106.4 213.0 426.0 852.0
Dial reading (mm×10−3) 275 540 965 1,880 3,340 5,000 6,600
The diameter and initial thickness of the sample were 60 and 38 mm, respectively. The
final water content of the sample was 35.2% and G = 2.72.
1. Compute the void ratio and plot the e–log p curve.
2. Estimate the maximum overburden pressure by the Casagrande method.
3. Draw the field curve and determine the compression index.
8.2 Compute the compression index of the soil from the following experimental data:
Load (kN/m2) 0 10 50 100 200 400 0
Δh (mm) 0 −0.7 −2.18 −0.70 −0.72 −0.85 +0.75
The initial thickness of the sample was 0.20 mm, the specific gravity of soil particles
2.70, and the final water content of the soil 38.2%. Also determine the pre-consolida-
tion pressure of the soil.
8.3 The thickness of a saturated specimen of clay under a consolidation pressure of
100 kN/m2 is 22.12 mm, and its water content is 20%. On increase of the consolidation
pressure to 200 kN/m2, the specimen’s thickness decreases by 1.28 mm. Determine
the compression index for the soil. G = 2.7.
M08_PURU1773_01_SE_C08.indd 257 2/12/2013 8:02:09 AM

8.4 A clay stratum 2.5 m thick lies over a sandy stratum and has drainage on both sides.
Calculate the values of the pore water pressure and the effective stress in the middle
of the clay stratum when 60% of consolidation has taken place under an increment of
load from 500 to 2,500 kN/m2.
8.5 A layer of soft clay is 6 m thick and lies under a newly constructed building. The pres-
sure due to the sand overlying the clay layer is 250 kN/m2, and the new construction
increases the overburden by 100 kN/m2. If the compression index is 0.5, compute the
settlement taking the natural water content of the clay layer to be 45% and the specific
gravity of soil grains to be 2.7.
8.6 A clay layer whose settlement under a given loading is expected to be 16 cm settles
4 cm at the end of 2 months. How many months will be required to reach a settlement
of 8 cm? How much settlement will occur in 10 months?
8.7 The following are the time dial readings in a consolidation test on a clay sample when
the load is increased from 100 to 200 kN/m2. The thickness of the sample at the start
of the load increment is 23 mm, and the void ratio is 0.95.
Elapsed time Dial reading (division)

(min) 1 div. = 0.0025 mm
0.00 1,012
0.25 890
1.00 850
2.25 833
4.00 818
6.25 802
9.00 788
12.25 774
16.00 760
20.25 746
25.00 732
30.25 718
36.00 704
49.00 676
64.00 651
81.00 630
100.00 610
144.00 572
270.00 490
520.00 441
1250.00 408
1440.00 371
Determine the coefficient of consolidation using Taylor’s method.
M08_PURU1773_01_SE_C08.indd 258 2/12/2013 8:02:09 AM

8.8 The following readings were obtained in a consolidation test:
Pressure (kPa) 0 50 100 200 400 800

Thickness of sample after 24 hours (mm) 19 18.98 18.95 18.80 18.31 17.76
The final water content was 30.1% and G = 2.75. Plot a graph of void ratio versus
effective pressure, and from the curve find the coefficient of compressibility when the
pressure is 300 kN/m2. If the coefficient of permeability is 20×10−6 mm/s, find the
coefficient of consolidation of the soil.
8.9 A consolidation test on a sample of clay gave the following results:
Effective pressure (kN/m2) 0 20 50 100 200 400

Void ratio 2.10 2.07 1.98 1.84 1.60 1.34
The layer of clay from which the consolidation sample was taken has a thickness of
1.2 m and is overlaid by 6 m of sand. The porosity and specific gravity of the sand
layer are 40% and 2.65, respectively. There is another sand layer present beneath the
clay layer. The groundwater table is 2 m above the top of the clay. The clay layer has
a specific gravity surface, which is estimated to increase the pressure by 110 kN/m2
at the top and by 75 kN/m2 at the bottom of the clay layer. Compute the settlement of
the structure due to the consolidation of the clay layer.
If the coefficient of permeability of the clay layer is 25 x 10−6 mm/s, how long will
it take for 50% of the settlement to take place?
8.10 In a consolidation test on a soil, the void ratio of the sample decreases from 1.24 to 1.12
when the pressure is increased from 200 to 400 kN/m2. Calculate the coefficient of
consolidation (in m2/year) given that the coefficient of permeability of the soil during
this pressure increment is 8.5×10−8 cm/s.
8.11 The time to reach 60% consolidation is 35 seconds for a sample of 1.2 cm thickness
tested in the laboratory under the conditions of double drainage. How long will the
corresponding layer in nature require to reach the same degree of consolidation if it is
11 m thick and drained on one side only?
8.12 A clay layer has a drainage arrangement as shown in Fig. 8.26. The coefficient
of consolidation of the soil is 10−1mm2/s. A fill is placed on the soil, which induces a
pressure of 50 kN/m2 at the surface.
1. Make a neat drawing showing
(a) The distribution of pore pressure before the fill is placed
(b) The distribution of pore pressure immediately after the fill is placed
2. How long will it take for the average consolidation of the layer to exceed 90%?
M08_PURU1773_01_SE_C08.indd 259 2/12/2013 8:02:09 AM

G.S
W.T. 2m
γ ′ = 11.2 kN/m3
Sand 6m
G = 2.65
γ ′ = 10.0 kN/m
3
Clay
G = 2.70 8m
Hard shale
Fig. 8.26
M08_PURU1773_01_SE_C08.indd 260 2/12/2013 8:02:09 AM

9
Shear Strength of Soils
CHAPTER HIGHLIGHTS
Stress–strain curve – Mohr–Coulomb failure criterion – Peak and residual
shear strength – Strain- and stress-controlled tests – Laboratory measurement
of shear strength: drainage condition, direct shear, triaxial shear, unconfined
compression, ring shear tests, choice of rate of deformation – Field measure-
ment of shear strength: vane shear test, bore hole shear device – Shear strength
of saturated cohesive soil – Pore pressure coefficients – Sensitivity of cohesive
soils – Shear strength of granular soils
9.1 INTRODUCTION
The capability of a soil to support a loading from a structure, or to support its own
overburden, or to sustain a slope in equilibrium is governed by its shear strength.
The shear strength of a soil is of prime importance for foundation design, earth and
rockfill dam design, highway and airfield design, stability of slopes and cuts, and lat-
eral earth pressure problems. It is highly complex because of various factors involved
in it, viz., the heterogeneous nature of the soil, the water table location, the drainage
facility, the type and nature of construction, the stress history, time, chemical action, or
environmental conditions.
The shearing strength and related deformations of a foundation or a construction soil
can be obtained from carefully extracted field samples. Field situations may demand
an in situ determination of shear strength for reasons of expediency and economy. This
chapter encompasses both theoretical and experimental approaches for determination
of shear strengths.
9.2 STRESS–STRAIN CURVE

Consider an element of soil subjected to a varying shear stress under a constant normal
stress. Figure 9.1 shows a typical shear stress–shear strain curve. Initially, when the
M09_PURU1773_01_SE_C09.indd 261 2/12/2013 5:37:14 PM

Yield
Point of failure
Peak
strength
= tf
Shear stress
Residual
strength
= tr
Strain
Fig. 9.1 Typical shear stress–shear strain curve
shear stress is low, the soil behaves like an elastic material depicting a linear stress–strain
curve. An increase in shear stress causes an increase in shear strain, and at a particular stress
level, significant plastic shear strain starts to develop, and that point is referred to as yield.
The shearing resistance of the soil increases with the plastic shear strain, and the material is
said to work hardening or strain hardening. The strain hardening can only increase the resis-
tance to a particular maximum shear stress, and that maximum resistance is called the peak
shear strength or simply the shear strength of the soil (τf), i.e., the shear strength of a soil is the
maximum shear stress that can be resisted by the soil. This yield level is considered to be
unstable, and the soil is said to fail. In some soils, the maximum shearing resistance decreases
after this point, and the soil is said to be strain softening or work softening. After a continued
large strain, the shearing resistance attains a constant level, and the corresponding shearing
resistance is called the residual shear strength or simply the residual strength (τr).
9.3 MOHR–COULOMB FAILURE CRITERION

It has been shown above that on any point within a soil mass, when the shear stress reaches
the shear strength of the soil, the soil fails at that point. Coulomb, in 1773, suggested a simple
relationship of shear strength of soil being a linear function of the normal stress as follows:
τ f = c ′ + σn′ tan φ ′ (9.1)
where τf is the shear strength of soil, c the cohesion intercept, σn the normal stress acting on
the plane of failure, and φ the angle of shearing resistance.
Equation 9.1 is given in terms of the total stress. After Terzaghi’s classical effective stress
concept, the above equation is modified in terms of effective stresses as
τ f = c ′ + σn′ tan φ ′ (9.2)
where c ′ and φ ′ are the shear strength parameters in terms of the effective stress and σ′ n =
σn – uw is the effective normal stress. The shear strength parameters c ′ and φ ′ are not
M09_PURU1773_01_SE_C09.indd 262 2/12/2013 5:37:15 PM

Shear Strength of Soils 263
Shear stress Mohr – Coulomb failure envelope

Angle of
shearing resistance
f
Cohesion
=c
(s3′)1 (s3′)2 (s1′)1 (s3′)3 (s1′)2 (s1′)3

Normal stress/principal stress
Fig. 9.2 Mohr–Coulomb failure envelope
constants for a given soil but depend on many factors such as void ratio, initial structure,
and pore pressure, and hence these parameters are referred to the type of test and conditions
under which they were measured. Equations 9.1 and 9.2 are referred to as Coulomb’s failure
envelopes or Coulomb’s failure loci.
The state of stresses in a soil mass at the time of failure may be represented in terms of the
effective major and minor principal stresses. Figure 9.2 represents three Mohr circles for three
sets of stresses at the time of failure in a soil mass. A tangent drawn to these circles is termed
Mohr’s envelope, and this satisfies Eq. 9.1. Any combination of stresses that falls within this
envelope represents a stable condition. A state of stresses plotting above the failure envelope
cannot exist. Thus, another way of stating Coulomb’s failure condition is that if the Mohr
circle for a state of stresses at a point is tangential to Coulomb failure loci, then that point is
said to be in a state of failure. This is generally known as the Mohr–Coulomb failure crite-
rion, and the failure loci are called Mohr–Coulomb failure envelopes. Because of its simplic-
ity, the Mohr–Coulomb failure criterion is widely used. In reality, the failure envelope may
not be a straight line but for the range of normal stresses involved in soil, the straight line
approximation is justified, and the shear strength parameters are determined for that stage.
9.4 PEAK AND RESIDUAL SHEAR STRENGTHS

A knowledge of shear stress at and after failure and the shape of the stress–strain curves are
of paramount importance in studies of stability problems and, more particularly, in materi-
als experiencing progressive failure* (Hvorslev, 1960).
Figure 9.3 represents typical stress–strain curves for normally consolidated (NC) and over-
consolidated (OC) clays tested very slowly. At a given effective normal stress perpendicular
to the plane of shear, the peak strength in both cases corresponds to the maximum shear
stress the clay can resist. Continued straining after the strain at peak strength decreases the
*The processes that cause a gradual reduction in strength to the residual value are termed progressive
failures.
M09_PURU1773_01_SE_C09.indd 263 2/12/2013 5:37:15 PM

Peak f¢
tf
tr f ¢r
Residual
Normally consolidated
Shear strength
Shear strength
sn′
Peak
tf f′
tr f¢r
Residual
Over-consolidated
C′
Cr′
Displacement Effective normal stress sn′
Fig. 9.3 Simplified shear strength parameters of clay
resistance, and further, at large strains the residual strength stage is attained. Hence, peak
and residual strength parameters can be expressed as
τ f = c ′ + σn′ tan φ ′ (9.3)
τ r = cr′ + σn′ tan φr′

where c′ is the cohesion intercept corresponding to peak strength, c′r the cohesion intercept
corresponding to residual strength, φ′ the angle of shearing resistance corresponding to peak
strength, and φ′r the angle of shearing resistance corresponding to residual strength.
Some of the main features which are revealed from Fig. 9.3 (Skempton, 1964) are as
follows:
1. The post-peak drop off in strength is more pronounced in over-consolidated clays (e.g.,
stiff clays) and less in normally consolidated clays (e.g., soft clays). With high clay contents,
the drop in both cases is appreciably increased.
2. At the residual condition, the strength at a given effective stress is independent of past
stress history.
3. The strength envelope for residual condition is generally located at a little lower than that
for the peak strength of a normally consolidated clay but extremely lower than that of
over-consolidated clay.
The residual cohesion intercept c′r is relatively low and can be taken to be zero. In general,
the value of φ′r decreases with increasing clay content. The complete shape of the stress–dis-
placement curve up to displacement at residual condition clearly depicts the degree of brit-
M09_PURU1773_01_SE_C09.indd 264 2/12/2013 5:37:16 PM

IB = 0.1
b
Shear stress
IB = 0.6
c
Displacement
Fig. 9.4 Stress–displacement curves showing different degrees of brittleness
tleness of the clay. Figure 9.4 represents three typical stress–displacement curves. Bishop
(1967) introduced a factor called brittleness index* IB relating the peak and residual strengths as
τf − τr
IB =
τf
In Fig. 9.4, Curve a represents a plastic non-brittle clay, whereas Curve b represents a
highly brittle clay. Skempton and Hutchinson (1969) showed that the shape of the stress–
displacement curve is also significant since two different stress–displacement curves yield
the same brittleness index (e.g., Curves b and c). This anomaly may be taken into account
by considering the energy parameter introduced by Bishop (1967), which is expressed as the
ratio of the work done in shearing from peak to residual to the work done if the clay contin-
ued shearing at peak strength.
In over-consolidated clay slopes, Skempton (1964) observed that field shear strength
(τ ), associated with a slip, lies somewhere between τf and τr. Skempton (1964) related these
strengths by a factor called residual factor, R:
τf − τ
R=
τf − τr
In physical terms, R is the proportion of the total slip surface in the clay along which
its strength has fallen to the residual value. A value of R = 0 signifies no reduction in
strength to have occurred, and R = 1 represents a complete reduction to residual strength.
Although it is known that R increases with time, no method is yet available to determine
the same.
The peak shear strength parameters depend on factors such as per cent clay content,
drainage condition, type of loading, consolidation history, stress level, anisotropy, and other
environmental factors. The residual shear strength is independent of many of the factors
listed above, but φ′r decreases with increasing clay content.
*
This is equal to (1 – λR), where λR is Haefeli’s residual coefficient (Haefeli, 1965).
M09_PURU1773_01_SE_C09.indd 265 2/12/2013 5:37:17 PM

9.5 LABORATORY MEASUREMENT OF SHEAR STRENGTH

The theoretical presentation made above is quite satisfactory in defining failure in a soil mass,
but its validity and usefulness have to be established from direct and indirect measurements.
Shear strength is measured both in the laboratory and in the field. In the laboratory test
samples, the field conditions have to be simulated to arrive at reasonable strength param-
eters. If the strength of natural soil deposits is needed, only undisturbed samples have to be
used. This is particularly important for soft and sensitive clays, which will be disturbed sub-
stantially due to re-moulding. The strengths of proposed compacted earth embankments are
obtained from samples compacted at the required water content and density, adopting a par-
ticular compaction method. It is difficult to obtain undisturbed samples in non-cohesive soils.
The shear strength of such samples is obtained from disturbed samples that are re-moulded
in the laboratory to field densities. There are direct and indirect methods for determining the
shear strength in the field.
9.5.1 Drainage Condition

The drainage condition during the test significantly influences the measured values, and the
shear strength parameters c and φ are referred to test conditions. One of the following three
basic sets of drainage conditions may be used.
Unconsolidated–Undrained. When no drainage of water is allowed during the entire test
process, the test is referred to as the undrained test. Thus, an increase in the total stress pro-
duces a corresponding stress increase in pore water pressure. As no account of pore water
pressure is taken, the shear strength parameters are called total shear strength parameters.
τ f = cu + σn tan φu (9.4a)
where cu and φu are called undrained shear strength parameters. For a saturated soil under
undrained condition φu = 0°, then
τ f = cu (9.4b)
Consolidated–Undrained. The specimen is tested in two stages. During the first stage,
called the consolidation stage, the specimen is allowed to consolidate under conditions of
constant applied stress and full drainage. During the second stage, called the shear stage, the
specimen is subjected to a shear under no drainage condition with or without pore pressure
measurements. Accordingly, the shear strength parameters are referred to as effective shear
strength parameters, c′cu and φ′cu, and total shear strength parameters, ccu and φcu.
′ + σn′ tan φcu
τ f = ccu ′ (9.5a)
and
τ f = ccu + σn tan φcu (9.5b)
Consolidated–Drained. The first stage of the test is similar to that of the consolidated–
undrained test. When the consolidation stage is complete, the specimen is sheared with full
drainage so that the rate of loading does not increase the pore water pressure. Since there
is no increase in pore water pressure, the increase in total stress is equal to the increase in
effective stress. The shear strength parameters are referred to as c′d and φ′d.
τ f = cd′ + σn′ tan φd′ (9.6)
M09_PURU1773_01_SE_C09.indd 266 2/12/2013 5:37:18 PM

9.5.2 Shear- and Strain-Controlled Tests

Shearing can be accomplished by controlling (Fig. 9.5) either the rate of strain or the rate of
stress (Jumikis, 1962). In the strain-controlled apparatus, the shearing deformation is continu-
ously applied at a constant speed, and shear force is measured by means of a proving ring or
a load transducer. Such a test can be continued even after the failure of the specimen. In the
stress-controlled apparatus, the magnitude of the shearing force is the controlled variable. The
stress is increased uniformly or in increments. The applied shear load is continuously kept
till the deformation ceases. This test cannot be continued after the failure of the specimen.
Generally, the strain-controlled test is preferred, as a better control on the rate of loading can
be had and the test can be run even after failure. The stress-and strain-controlled direct shear
tests are shown in Fig. 9.5.
9.5.3 Direct Shear Test

The direct shear test is a simple and widely used test. The apparatus is a shear box with
separated lower and upper halves. The lower half is fixed to the frame, and the upper half is
capable of moving horizontally relative to the lower half. A rectangular prism of soil with a
cross section 60 mm×60 mm is placed in the box with approximately half of the specimen
within either section (Fig. 9.6). In case of a non-cohesive soil, the specimen has to be pre-
pared in the box itself at the required void ratio. Suitable porous discs may be placed one on
top and the other at the bottom of the specimen, depending on the drainage condition.
A pressure pad is placed on the top and a vertical load applied.
Normal load
Forced shear plane Proving ring
Displacement T
guage
(a) Strain-controlled Arrangement for

Normal constant rate of strain
load
Forced shear plane
Displacement guage T
Pulley
Dead
weight
(b) Stress-controlled
Fig. 9.5 Two types of direct shear testing devices
M09_PURU1773_01_SE_C09.indd 267 2/12/2013 5:37:18 PM

Normal
Dial gauge
load
to measure
Dial gauge Loading plate volume change
to measure
shearing displacement
Porous stone Shearing
force, S
Soil specimen Shearing plane
Porous stone
Fig. 9.6 Schematic diagram of direct shear apparatus
The upper half of the apparatus is then moved laterally, forcing the sample to shear across
the plane between the two halves of the apparatus. The magnitude of the shearing force
is measured. Changes in sample thickness that occur during the shearing process are also
recorded so that volume change versus shearing stress or shearing strain can be studied. The
test is repeated on three or four specimens taken from the same soil sample. The detailed test
procedure and typical results are given in Chapter 10.
Test Curves. Typical shear stress–displacement and volume change–displacement curves
are shown in Fig. 9.7. The slope of the curve is the stress–strain modulus of the soil Es. This
modulus is an indication of the stiffness or resistance to deformation of the material. The higher
the value of Es, the stiffer or stronger is the soil. The failure stress is represented by the maxi-
mum shear stress value (i.e., peak value) if the stress–displacement curve is of work-softening
type. Otherwise, the shear stress corresponding to 20% strain is taken as the shear strength.
Failure Envelope. Normal and shear stresses are calculated and plotted in Fig. 9.8. A best
possible straight line is fitted to these points and this line is Coulomb’s failure envelope. The
shear strength parameters c′ and φ′ are measured.
100
t1
Shear stress, kN/m2
80
Shear stress–
displacement curve
60
sn = 150 kN/m2
40
20
1 2 3 4 5 6
Displacement, mm
Volume change–
displacement curve
Fig. 9.7 Typical shear stress and volume change curves for direct shear test
M09_PURU1773_01_SE_C09.indd 268 2/12/2013 5:37:19 PM

120
Shear stress, kN/m2

100
80
60
40
f′ = 24°
20
c ′ = 20 kN/m2
0 20 40 60 80 100 120 140 160 180 200
Normal stress, kN/m2
Fig. 9.8 Coulomb’s failure envelope from direct shear test results
Shear stress, kN/m2
Failure plane Coulomb's failure envelope

80
s n′ = 100 kN/m2
Origin of planes, Op
60
Major principal plane
90⬚
40 Minor
principal
20 tf = 60 kN/m2
plane
0 20 40 60 80 100 120 140 160 180 200

s3′ Normal stress, kN/m2 s′
Fig. 9.9 Mohr diagram for a direct shear test at failure
Mohr Stress Circle. For one normal stress, the Mohr circle is plotted in Fig. 9.9, and the
pole, the major and minor principal stresses, and planes are shown.
Merits. The following are some of the merits of direct shear test: (i) as the draining time
is relatively less, consolidated–undrained and consolidated–drained tests may be con-
ducted even on soils of low permeability; (ii) this is the only test wherein both the shear-
ing stress and the normal stress on the plane of failure are measured directly; (iii) volume
changes during the test can be measured easily; (iv) the shear strength parameters c and
φ obtained from this test are as reliable as triaxial values; (v) using a reversing technique,
the residual shear strength of the soil can be determined; (vi) the direct shear test appara-
tus is much more adaptable to electronic read-out equipments, and hence, a long duration
test can be done conveniently; and (vii) as the displacements in a direct shear test take
place in two directions, vertically and parallel to the direction of shear, this test is also
referred to as plane strain test. This condition is similar to the strains developed in many
field problems such as very long foundations or long embankments representing a two-
dimensional case.
Demerits. The following are the demerits of direct shear test: (i) the shear–stress distribu-
tion over the plane of failure is non-uniform and (ii) the drainage cannot be controlled, and
so the pore pressure behaviour cannot be obtained from the test.
M09_PURU1773_01_SE_C09.indd 269 2/12/2013 5:37:20 PM

9.5.4 Reversal Direct Shear Test

Residual strengths are determined from the shear box test by adopting the reversal technique.
Suitable arrangements can be made in the conventional apparatus for the reversing tech-
nique. In general, the top box is moved in the forward direction at a slow rate so that no
excess pore water pressure develops. After completion of the forward motion, the top box is
pulled back to its original position, preferably with the same rate of strain, and the test
repeated. Forward and reverse cycles are repeated till a constant value of residual strength
is obtained. This technique was suggested by Skempton (1964). While attempting to
standardize this procedure, Cullen and Donald (1971) brought out the following advantages:
(i) thin samples with reasonable rapid drainage time can be used; (ii) specimens can be read-
ily oriented in the correct direction of sliding – this is valuable when testing specimens with
defects, such as faults, fissures, and pre-existing slip planes; (iii) large deformations can be
obtained by continuous reversal of shear direction; (iv) reasonably large areas of failure
surface can be tested; (v) specimens can be readily wire cut (sometimes samples are cut to
reduce the number of cycles) after consolidation if desired; and (vi) the apparatus is widely
available and needs only minor modification.
The serious disadvantage of the apparatus is that the normal stress cannot be kept constant
throughout the test. Because of the serious disturbance at the failure plane due to reversal, the
stress–strain curve is not reliable at all strains. However, the stress–strain curves in the first
and last forward shears are acceptable. Figure 9.10 shows typical stress ratio–displacement
curves for different cycles.
9.5.5 Triaxial Shear Test

The triaxial shear test is a versatile apparatus suitable for all types of soil and test conditions.
Cylindrical soil specimens having a height–diameter ratio of 2:1 are used, and the conven-
tional size of the test specimen is 76 mm height and 38 mm diameter. The general arrangement
of the apparatus is shown in Fig. 9.11.
Normal stress, 100 kN/m2

0.5
0.4
t/s
0.3
0.2
0.1
0 4 8 12 16 20 24 28 32 36 40 44
Displacement, mm
Fig. 9.10 Stress ratio–displacement curves
M09_PURU1773_01_SE_C09.indd 270 2/12/2013 5:37:20 PM

Proving ring
Air release valve
Loading ram
Loading cap Rubber "O" ring

Perspex cylinder
Rubber membrane
Soil specimen Cell filled with water
Base cap Porous disc
Pedestal Rubber "O" ring
Valve Valve
Cell pressure
To pore pressure measurement
Fig. 9.11 Schematic diagram of triaxial test
The soil specimen is covered by a thin membrane which extends over a top cap and a
bottom pedestal. The perspex cell is filled with water, and the required cell or confining
pressure (σ3) is applied. This confining pressure acts horizontally on the cylindrical surface
of the sample and vertically on the top of the specimen. An additional vertical stress called
deviator stress (σ1− σ3) is then applied and steadily increased until failure of the specimen
occurs. The test procedure is repeated on different specimens taken from the same sample
for different confining pressures and the corresponding deviator stresses calculated.
Detailed test procedures for undrained, consolidated–undrained, and drained tests are
given in Chapter 10.
Typical Stress–Strain Curve. Figure 9.12 shows typical stress–strain curves for undrained,
consolidated–undrained, and drained shear tests.
Over-consolidated clay –
CU test
Normally consolidated clay –
CD test
Stress, s1 – s3
Re-moulded sensitive clay –

UU test
Strain, %
Fig. 9.12 Typical stress–strain curves from triaxial shear tests
M09_PURU1773_01_SE_C09.indd 271 2/12/2013 5:37:21 PM

Failure Envelope. Mohr’s circle for each test is plotted. The tangent to the resulting circles
gives Mohr’s envelope (Fig. 9.13). The shear parameters are measured from the plot.
Mohr’s Stress Circle. Mohr’s circle representing the state of stress for one specimen is
shown in Fig. 9.14. Point P represents the stress condition at the time of failure on a failure
plane inclined at an angle θf from the plane on which the major principal stress acts.
The radius of Mohr’s circle is (σ′1 – σ′3)/2, and its centre is located at a distance of (σ′1 + σ′3)/2
from the origin. From Fig. 9.14,
Over-
consolidated
Normally
consolidated
CD envelope
f d′
CU envelope
fcu
Shear stress
UU envelope
fu = 0°
Mohr's circle Mohr's circle Principal stress

for for
CU strength test UU strength test
Mohr's diagram for
CD strength test
Fig. 9.13 Typical strength envelops for CD, CU, and UU tests on clay
s1′
Failure
envelope
Minor
s3′ s3′
Shear stress
principal plane
qf
P
s1′
s1′ – s3′
tf f′ =
2
Failure Major
c′ plane principal plane
f′ Op qf
Q O s3′ s1′ + s3′ S s1′
c′ cot f′ Principal stress
2
sn′
Fig. 9.14 Stress conditions at failure
M09_PURU1773_01_SE_C09.indd 272 2/12/2013 5:37:22 PM

1/ 2(σ1′ − σ 3′ )
Rearranging, sin φ ′ = (9.7)
c ′ cot φ ′ + 1/ 2(σ1′ + σ3′ )
or σ1′ = σ3′ tan 2 ( 45° + φ ′ / 2) + 2 c ′ tan 2 ( 45° + φ ′/ 2) (9.8a)
σ1′ = σ3′ Nφ + 2 c ′ Nφ (9.8b)

where
Nφ = tan 2 ( 45° + φ ′ / 2) (9.9)
The quantity Nφ is referred to as the flow value.

This is the relationship between the effective major and minor principal stresses in terms
of effective shear strength parameters.
θf = 45° + φ ′ / 2 (9.10)
The coordinates of point P are
τ f = 12 (σ1′ − σ3′ )sin 2 θf (9.11)
and
σn′ = 12 (σ1′ + σ3′ ) − 12 (σ1′ − σ3′ )cos 2 θf (9.12)
The pole and the major and minor principal planes and stresses are shown in Fig. 9.14.
Modified Failure Envelope. Any state of stress can also be represented by a stress point
instead of a Mohr circle, the point being represented by 1 (σ′1− σ′3) and 1 (σ′1 + σ′3). Such
2 2
a plot is shown in Fig. 9.15 for the Mohr circles represented in Fig. 9.2. A modified failure
envelope is obtained with the equation of the form
1
2
(σ1′ − σ3′ ) = c∗+ 12 (σ1′ + σ3′ ) tan φ∗ (9.13)
where c∗ and φ∗ are the modified shear strength parameters. The parameters c′ and φ′ are
then given by
Modified
failure envelope
– s3′)
(s1′)3 + (s3′)3
2 Stress
(s1′)2 + (s3′)2 point
– (s1′
(s1′)1 + (s3′)12
1
2
(s1′)3 – (s3′)3
f°(s ′) – (s ′) (s1′)2 – (s3′)2 2
1 1 3 1 2
c′ 2
45° 45°
(s3′)3 1
(s1′)3
– (s1′ + s3′)
2
Fig. 9.15 Alternative representation of stress conditions
M09_PURU1773_01_SE_C09.indd 273 2/12/2013 5:37:24 PM

φ ′ = sinⴚ1 (tan φ∗ ) (9.14)
c∗
c′ = (9.15)
cos φ ′
Lines drawn from the stress point at angles of 45° (as shown in Fig. 9.15) intersect the
horizontal axis at σ′3 and σ′1.
Types of Failure. The diameter of the sample increases due to the shortening of the sample
under load, and the middle of the sample bulges. Certain specimens fail with predominant
failure planes, while in others, multiple parallel shear planes of the type shown in Fig. 9.16
are formed. In some saturated specimens of low shear strength, no shear planes are formed,
but the sample shortens, bulges, and has the appearance of a barrel. In such cases, because of
the continued increase of area, no definite pronounced peak will appear in the stress–strain
curve, and the deviator stress corresponding to 20% strain is taken as the deviator stress at
failure.
Triaxial Test for Residual Strength. In general, the triaxial test is to be preferred
for measuring shear strength along discontinuities, as here the kinematic restraints are
minimal (Skempton and Petley, 1967). But the limited maximum strain which can be
applied to the specimen in the triaxial test is usually too small to permit the determi-
nation of residual strength in an originally intact clay. But where a pre-existing fissure
surface is to be tested, or where an inclined plane has been made in the clay specimen,
the triaxial test can be used, provided the effects of horizontal components of the load are
taken into account.
Merits. The following are the merits of triaxial shear test: (i) pore pressure may be controlled
and measured accurately in the test; (ii) suitable confining pressures can be applied (isotro-
pic, anisotropic, or cyclic) to initially consolidate the sample to some pre-determined state;
and (iii) special strain tests, such as increasing (compression) or decreasing (extension) axial
load can be performed.
Demerits. The following are the demerits of triaxial shear test: (i) preparation of sand
specimen and soft clay is cumbersome and (ii) testing requires a skilled person.
q1
(a) Complete shear (b) Partial shear (c) Barrelled or plastic

failure failure failure
Fig. 9.16 Types of failure in the triaxial test
M09_PURU1773_01_SE_C09.indd 274 2/12/2013 5:37:25 PM

9.5.6 Unconfined Compression Test

This is essentially a special case of the triaxial compression test where the minor principal
stress is zero. The test may be carried out in the laboratory or in the field. In the laboratory,
the standard triaxial apparatus can be used without a perspex cylinder. Generally, a mem-
brane is not needed, but it is sometimes used to prevent the sides of the specimen from dry-
ing. A field apparatus is also available to obtain the unconfined compressive strength from
calibrated charts. The apparatus is shown diagrammatically in Fig. 9.17.
The soil specimen is placed between the central fixed plate and the lower moving plate.
The upper moving plate is raised by turning the handle. The movement of the upper plate
causes an extension in the spring, and a force proportional to the extension is transmitted to
the specimen. A chart is attached to the upper moving plate, and the movement is recorded
by a pencil attached to the lower moving plate. As the vertical movement corresponds to the
extension of the spring, the lateral movement of the pencil corresponds to the compression
of the specimen.
Thus, the maximum extension of the spring corresponds to the maximum load on the
sample (P), and the unconfined compressive strength is given as
P(1 − ε)
σ1 = (9.16)
A0
where ε is the longitudinal strain and A0 the original cross-sectional area of the specimen.
This test is one of the easiest and simplest for determining the shear strength of cohesive
soils. It can be used for both undisturbed and re-moulded samples. The detailed test proce-
dure is given in Chapter 10.
Test Curves. The stress–strain curve for an undisturbed specimen and the curve after re-
moulding are shown in Fig. 9.18. The ratio of undisturbed strength to re-moulded strength
is called the sensitivity of the soil (discussed in Section 9.9).
Lead screw
Upper fixed plate
Drawing plate Upper moving plate

Pencil
Calibrated spring
Moving tie rod Central moving plate

Pencil lever arm Central fixed plate
Soil sample
Lower moving plate Plane seatings
Adjustable arm
Fig. 9.17 Unconfined compression test field apparatus
M09_PURU1773_01_SE_C09.indd 275 2/12/2013 5:37:25 PM

Undisturbed
Compressive stress
Re-moulded
Axial strain, %
Fig. 9.18 Typical stress-strain curve for undisturbed and re-moulded insensitive clay
Failure Envelope and Mohr Stress Circle. The Mohr envelope is shown in Fig. 9.19. Since
σ3 = 0, only one circle can be drawn. The pole and the major and minor principal stresses
and planes are also shown in Fig. 9.19. This test is applicable only for saturated unfissured
clays. The shear strength τf = cu. If the failure occurs along a well-defined shear plane, the
angle θf is measured, and then φ and c are calculated as shown below.
θf = 45° + φ / 2
s1
s3 = 0 s3 = 0
qi
t Probable failure envelope s1

Failure
envelope
(as per test) fu=0°
Minor
s principal plane
t f = cu = 1
2
Major
principal plane
Op s1 = qu Principal stress
Fig. 9.19 Mohr diagram for unconfined compression test at failure
M09_PURU1773_01_SE_C09.indd 276 2/12/2013 5:37:26 PM

or
φ = 2 θf − 90° (9.17)
and
c = σ1 / 2 tan θf
Merits. The following are the merits of the unconfined compression test: (i) this is the
simplest test for determining the undrained shear strength of soil and (ii) a quick test and
field apparatus gives fast determination of undrained strength.
Demerits. The following are the demerits of unconfined compression test: (i) fissured
samples may not give reliable results because there is no confining pressure to close the fis-
sures and (ii) samples with high sand content give unreliable results.
9.5.7 Ring Shear Test

The first determination of residual strength of an undisturbed clay under drained condition
using a ring shear apparatus was proposed by Tiedemann (1937). Hvorslev (1960) also used
a similar ring shear apparatus. In general, the two obvious advantages of any type of torsion
or ring shear test are that there is no change in the area of the cross-section of the shear plane
as the test proceeds and that the sample can be sheared through a large displacement of any
magnitude.
A simple test of this type is one in which a laterally confined normally loaded annular
specimen is twisted with controlled rate of displacement or torque (Fig. 9.20). The rate of
rotation must be slow enough such that no excess pore water pressure develops in the speci-
men and the specimen remains under drained condition. The shear displacement may be
continued to a desired level, and there is no restriction to the magnitude. Shear stress, which
is calculated from the applied torque, is plotted against shear displacement, as shown in
Fig. 9.21. For purposes other than the measurement of residual strength, it has all the disad-
vantages of the shear box, such as high local concentrations of strain, uncertainty about the
direction of principal stresses as the test proceeds, and the enormous time taken for the test.
9.5.8 Choice of Rate of Deformation

The rate of deformation of a test is based on the drainage condition, the purpose of the test,
and the convenience for accurate booking of readings. For the unconsolidated–undrained
s n′
Fig. 9.20 Schematic representation of ring shear test
M09_PURU1773_01_SE_C09.indd 277 2/12/2013 5:37:27 PM

Peak
Shear stress
Residual
tf
tr
Shear deformation
Fig. 9.21 Typical shear stress and shear deformation curve from ring shear test
test, the percentage of strain varies from 0.5 to 2 per minute in triaxial and unconfined
compression tests, and the strain rate is about 1.5 mm/min in direct shear test. In the
consolidated–undrained triaxial tests with pore water pressure measurements, the rate of
strain ranges from 0.05 to 0.1% per minute. Drained tests have to be conducted still at a slow
rate such that during the entire test, no excess pore water pressure develops. The conven-
tional strain rate is about 0.01 to 0.2 mm/min.
Although suitable rates for carrying out drained tests may be found by a process of trial
and error, they can be estimated from the consolidation data of the soil. The time required for
failure (tf) when 95% dissipation is ensured may be obtained from the expression
b2 20 h 2
tf = =
nd cv (1 − U z ) nd cv
where
2 h = initial thickness or height of the specimen
nd = a constant for drainage
= 0.75 – drainage from one end only
= 3.00 – drainage from both ends
= 32 – drainage from radial boundary only
= 35 – drainage from both ends and radial boundary
cv = coefficient of consolidation
Uz = degree of consolidation
From a knowledge of time to failure, the rate of strain for the test may be calculated.
9.6 FIELD MEASUREMENT OF SHEAR STRENGTH

It is not always feasible and practical to attempt to duplicate the in situ conditions in the
laboratory tests. Further sampling, handling, and sample preparation disturb the soil
structure and hence the shear strength of the soil. It is thus desirable to test the soil in the in
situ condition.
M09_PURU1773_01_SE_C09.indd 278 2/12/2013 5:37:27 PM

9.6.1 Vane Shear Test

Vane shear test is one field test that is coming into wider use for determining the in situ
strength of cohesive soils. In particular, the test is very suitable for soft clays and may not
give reliable results if the clay contains sand or silt.
The equipment consists of thin-bladed vanes that can be pushed into the soil with a mini-
mum of disturbance (Fig. 9.22). The length of the vane is equal to twice its overall width. A
torque is applied gradually to the upper end of the rod. Shear failure takes place over the
surface and the ends. The shear strength is obtained from the expression
⎛ d 2 b ad 3 ⎞⎟
T = π cu ⎜⎜⎜ + ⎟⎟ (9.18)
⎜⎝ 2 4 ⎟⎠
where
T = torque at failure and a = 2/3 for uniform end shear
d = overall vane width = 3/5 for parabolic end shear
b = vane length = 1/2 for triangular end shear
cu = shear strength (undrained)
It is assumed that the shear strength is isotropic. Re-moulding can be done by rotating
several revolutions, and then the torque is measured, which corresponds to the re-moulded
shear strength. A detailed description of the test is given in Chapter 10.
For measurements of low undrained shear strength of saturated clays, a laboratory vane
shear apparatus can be used (IS: 2720 – Part 30, 1980). The apparatus may be hand operated
or motorized with an arrangement to rotate the vane at the rate of 0.1°/s. The penetrating
edge of the vane blade is sharpened to enable easy insertion into the soil without distur-
bance. A hand-operated vane is illustrated in Fig. 9.23.
Uniform end shear

Parabolic Triangular end
end shear shear
Central
torque rod Uniform
side shear
d Vane
blades h
(a) Vane shear blade (b) Shear stress distribution
Fig. 9.22 Scheme diagram of vane shear test
M09_PURU1773_01_SE_C09.indd 279 2/12/2013 5:37:27 PM

Gear bracket
Lead screw handle Torque applicator handle
Support pillar Slow-motion bevel and work gears
Gear bracket clamp screw
Bracket
Torque spring
Strain indicating pointer
Nut Maximum pointer
Locating pins
Lead screw Vane fixing screw
Shear vane
Base
Fig. 9.23 Laboratory vane shear apparatus (Source: IS: 2720 – Part 30, 1980)
9.6.2 Borehole Shear Device

A borehole shear device (Fig. 9.24) is a field method which has been developed by Wineland
(1975). This device is inserted into the bore and expanded into the soil with a pressure pv. The
shear is caused by a pull pb, as in a direct shear device. The test device operates by applying
the direct shear device technique in situ. This test approximates a consolidated–undrained
test. The reader may get additional details from Wineland (1975).
Ph Ph
Ground at hole
provides reaction
to develop the
equivalent of P h of
direct shear test
Pv is produced using
compressed air and
hydraulic cyclinder
Pv
Fig. 9.24 Borehole shear device
M09_PURU1773_01_SE_C09.indd 280 2/12/2013 5:37:28 PM

9.7 SHEAR STRENGTH OF SATURATED COHESIVE SOILS

The shear strength of cohesive soils depends on the degree of saturation, effective consolidation
pressure (stress history), and loading and drainage conditions. In order to understand its
behaviour, it is customary to treat cohesive soils in a saturated state, as this is the worst condi-
tion that can prevail in the field. Thus, the parameters estimated will be generally conserva-
tive. Because of the heterogenic nature of the soil, researchers concentrated many of their
studies on re-moulded cohesive soils in evolving the basic concepts concerning the strength
characteristics of saturated cohesive soils. From the results obtained, the strength behaviour
of in situ soils has been analysed.
9.7.1 Undrained Strength

If a saturated sample of re-moulded soil is subjected to a consolidation pressure in a triaxial
test (or a normal stress in a direct shear test), the effective stress remains unchanged for no
drainage condition. Any subsequent change in all-round pressure (or normal stress) simply
results in a corresponding change in pore water pressure. Similarly, the subsequent change
in σ1 will result in the same change in pore water pressure. Therefore, the undrained strength
is constant and independent of all-round pressure (Fig. 9.25). Thus, in intact clays the und-
rained strength can be determined simply from an unconfined compression test. The und-
rained shear strength is represented by cu, the undrained cohesion, and φu = 0. If in a slip
plane, θf > 45°, then the undrained cohesion will be less than cu (for which φ > 0°). As the
initial condition is not changed, there will be only one effective stress circle, and the same
can be obtained by measuring the pore water pressure at failure.
The undrained shear strength may be applied in field problems where the change in total
stress is immediately compensated by a change in pore water pressure, because of non-
availability of sufficient time for dissipation of pore water pressure. The typical examples in
saturated clays are initial bearing capacity of footings or embankment foundations, initial
stability of slopes or cuts, and initial stability of a braced excavation.
Unconfined
compression Effective stress Failure envelope
test circle
fu=0°
cu
Fig. 9.25 UU strength of saturated clays
M09_PURU1773_01_SE_C09.indd 281 2/12/2013 5:37:29 PM

The undrained shear strength of in situ samples can be obtained from the unconfined
compression test if the clay is intact and if fissured undrained triaxial test can be used.
In such cases, the in situ samples should have the same void ratio. In fissured clays, the
failure envelope is curved at low values of all-round pressure. However, at large pressures,
the fissures close and behave like an intact sample. Proper judgement should be exercised
in using such undrained strength results of fissured clays in practical problems. Figure 9.26
shows failure envelopes for normally consolidated and over-consolidated intact soils. For
sampling in sensitive clays, thin-walled tube samplers can be used. For extra-sensitive and
quick clays, in situ vane shear tests are highly desirable.
In uniform, normally consolidated clays, the undrained shear strength increases approxi-
mately linearly with depth. That is, the ratio of undrained strength, cu, to the effective over-
burden pressure, po, is approximately constant. Skempton (1957) proposed a correlation
between the ratio cu/po and the plasticity index as
cu
= 0.11 + 0.0037 I p (for Ip > 10% and scatter of ± 0.05) (9.19a)
po
and
cu
= 0.45wL (for wL > 0.4 and scatter of ± 0.10) (9.19b)
po
Equations 9.19a and b may be used to estimate the value of cu for normally consolidated clays.
Based on the nature of deposition and the subsequent consolidation, clay particles in
cohesive soils have a tendency to orient in a direction perpendicular to the major principal
stress. Such orientations may cause the cohesive soil to show varied strength in different direc-
tions, or in other words, the clay may be anisotropic with respect to strength. A Casagrande
and Carrillo (1944) proposed an expression (Eq. 9.20) for the directional variation of the und-
rained shear strength as
(cu )i = (cu )H +[(cu )V − (cu )H ] sin 2 i (9.20)
where (cu)i is the undrained shear strength of a specimen whose axis is inclined at an angle
i with the horizontal, (cu)H the undrained shear strength of a specimen taken horizontally
(i = 0°), and (cu)V the undrained shear strength of a specimen taken vertically(i = 90°).
t
Failure envelope
Envelope for for
fissured clay over-consolidated clay
(cu)OCC Failure envelope

for
normally consolidated clay
(cu)NCC
s
Fig. 9.26 Failure envelope for NC and OC intact clays
M09_PURU1773_01_SE_C09.indd 282 2/12/2013 5:37:29 PM

Table 9.1 Typical shear strength values for clays
Consistency of clay Shear strength (N/mm2)
Very soft <0.120

Soft 0.120–0.240
Medium 0.240–0.480
Stiff 0.480–0.960
Very stiff 0.960–1.920
Hard >1.920
Source: Lambe (1951).
The ratio of (cu)V and (cu)H is called coefficient of anisotropy. For natural deposits, this
coefficient varies from 0.75 to 2.0.
A classification of saturated clays based on the undrained shear strength obtained from
unconfined compressive strength is presented in Table 9.1 (Lambe, 1951).
9.7.2 Consolidated–Undrained Strength

The consolidated–undrained test in a triaxial or in a direct shear test consists of testing a soil
under undrained condition after the initial value of void ratio has been changed by consoli-
dation. Thus, the undrained strength forms a function of the confining pressure in triaxial test
(or normal stress in direct shear test). If pore pressures are measured at the time of failure,
then the effective stresses can be determined. Figure 9.27 shows the plots of both the total and
the effective stress circles for a re-moulded soil from a consolidated–undrained triaxial test.
The shear strength parameters ccu and φcu are obtained from total stress circles, and c′ or c′cu
and φ′ or φ′cu are obtained from effective stress circles. The consolidated–undrained total
stress parameters (ccu and φcu) should be regarded as a rough guide to the undrained shear
strength of the soil.
Total stress envelope

t ′=ccu+sn tanfcu
Effective stress envelope
tf = cn′ + sn′ tan f′
fcu
ccu f′
c′
(uw)1 (uw)2 s ′, s
Fig. 9.27 Effective and total CU shear test plots for saturated clays
M09_PURU1773_01_SE_C09.indd 283 2/12/2013 5:37:30 PM

t Normally consolidated t Over-consolidated

clay clay
Effective stress Total stress

envelope envelope
f ′cu fcu
Total stress Effective stress ′
f cu
envelope fcu envelope
ccu c¢cu
s, s ¢ s, s ¢
Consolidation pressure Consolidation pressure
sc (a) sc (b)
Fig. 9.28 Effective and total stress CU test plots for NC and OC clays
The consolidated–undrained shear tests in terms of total and effective stresses for normally
and over-consolidated clays are shown in Fig. 9.28, and the corresponding deviator stress and
pore water pressure variations with strain are shown in Fig. 9.29. In normally consolidated
clays, as a result of positive excess pore water pressure during shear (with no drainage), σ1
and σ3 are greater than σ′1 and σ′3, respectively, and hence, φ′cu > φcu. However, (σ1 – σ3) and
(σ′1 – σ′3) are equal; hence, the Mohr circles have the same diameter, but the effective stress
circles are shifted to the left, reflecting higher φ′cu than φcu. In the over-consolidated case,
because of negative pore water pressure, the effective stress circles are shifted to the right.
In this case, φcu may be slightly greater or lesser than φ′cu but ccu, is always greater than c′cu .
In situ clays have been consolidated anisotropically, i.e., the effective vertical and horizon-
tal pressures are unequal. In the laboratory test, the consolidation is effected isotropically.
s1 – s3
s1 – s3 s1 – s3
Normally consolidated
clay Over-consolidated
uw uw clay
s1 – s3
uw
uw
Increasing
over-consolidation ratio
Axial strain, % Axial strain, %
uw
(a) (b)
Fig. 9.29 Deviator stress and pore water pressure variation with strain from CU test on NC and OC clays
M09_PURU1773_01_SE_C09.indd 284 2/12/2013 5:37:30 PM

This isotropic consolidation leads to a lower void ratio than the in situ one, and hence, the
laboratory undrained strength overestimates the field value. As this is an unsafe situation,
the specimen should be anisotropically reconsolidated in the laboratory.
9.7.3 Strength in Terms of Effective Stresses

The drained shear and the consolidated–undrained shear tests (in terms of effective stresses)
should result in the same Mohr failure envelope for re-moulded samples. So parameters c′
and φ′ can be obtained from drained triaxial tests (or direct shear tests). The strain rate adopted
should be such that full dissipation of excess pore water pressure is maintained throughout
the test. Thus, at any time of the test, the total and effective stresses are equal. The volume
change during the application of principal stress difference is measured to correct the area of
cross-section of the specimen. The strength in terms of effective stresses can also be obtained
from the consolidated–undrained shear test with pore water pressure measurement at the
time of failure. Typical results of failure envelopes for normally and over-consolidated sam-
ples are shown in Fig. 9.30, and typical results of deviator stress and volume change are given
in Fig. 9.31. In drained tests, normally consolidated specimens fail at a low strain. A decrease
in volume in normally consolidated clays occurs after failure, and in over-consolidated clays,
an initial decrease in volume is followed by an increase at and after peak failure. With strain
increase, a normally consolidated clay hardens, whereas an over-consolidated clay softens.
In saturated re-moulded clays, the difference between φ′cu and φ′ and c′cu and c′ is suf-
ficiently small. It is necessary in practical problems to consider effective stress parameters
whenever pore water pressure can change independently of the total stresses. Typical exam-
ples are earth dams under steady-state seepage conditions and natural earth slopes without
excess pore water pressure but in equilibrium with the water table. The c′ and φ′ parameters
may be obtained from drained shear tests or from consolidated–undrained tests, with pore
water pressure measurements. The values thus obtained are different, unlike in re-moulded
clays. Natural slopes which are subjected to progressive failure have to be analysed with
residual shear strength parameters, c′r (≈0) and φ′r. The values of c′r and φ′r may be obtained
from a ring shear test or from a reversible direct shear box test.
Normally consolidated Over-

t clay t consolidated clay
f′d or f′ f′cu
f′d or f′ and f′cu
c′d or c ′ or c ′cu
(a) s′ (b) s′
Fig. 9.30 Failure envelopes from drained tests on NC and OC clays
M09_PURU1773_01_SE_C09.indd 285 2/12/2013 5:37:31 PM

Normally consolidated Over-

s1 – s3 clay s1 – s1 consolidated clay
ΔV ΔV
Increasing
ΔV over-consolidation ratio
Strain, % Strain, %
ΔV
(a) (b)
Fig. 9.31 Deviator stress and volume change variation with strain from drained test on NC
and OC clays
9.8 PORE PRESSURE COEFFICIENTS

9.8.1 Theory
In many practical problems involving deformation of soil masses, it is essential to estimate
the magnitude of the changes in pore water pressure resulting from changes in the state of
stress. In a saturated soil, changes in the principal stresses of σ1, σ2, and σ3 result in a change
in pore water pressure, Δuw, for no drainage condition. Let the volume change be ΔV. The
changes in effective stresses will be
Δσ1′ = Δσ1 −Δuw
Δσ 2′ = Δσ2 −Δuw (9.21)
Δσ 3′ = Δσ3 −Δuw
Consider an element of soil mass (Fig. 9.32). And let ε1, ε2, and ε3 be the principal strains. Then,
1
ε1 = [Δσ1′ − v(Δσ2′ + Δσ3′ )]
E
1
ε2 = [Δσ2′ − v(Δσ1′ + Δσ3′ )] (9.22)
E
1
ε3 = [Δσ3′ − v(Δσ1′ + Δσ2′ )]
E
Considering small strains,
ΔV
= ε1 + ε2 + ε3 (9.23)
V
M09_PURU1773_01_SE_C09.indd 286 2/12/2013 5:37:32 PM

Δs 1
Δs 3
Δs 2
Fig. 9.32 Three-dimensional stress system on a soil element
Substituting for ε1, ε2, and ε3 from Eq. 9.22, we have
ΔV 1 − 2v
= (Δσ1′ + Δσ2′ + Δσ3′ ) (9.24)
V E
Let the compressibility of soil skeleton be Cs and be represented as
3(1 − 2v)
Cs = (9.25)
E
Now representing the change in effective stresses in terms of the change in total stresses and
pore water pressure, we have
ΔV ⎛ Δσ + Δσ2 + Δσ3 ⎞
= Cs ⎜⎜ 1 −Δuw ⎟⎟⎟ (9.26)
V ⎜
⎝ 3 ⎠
As only change in volume causes consequent change in pore water pressure, the compress-
ibility of the pore fluid Cf is given as
1 ΔVw
Cf = (9.27)
Vw Δuw
Since the soil is saturated, Vw = nV and ΔVw = ΔV. Therefore,
1 ΔV
Cf = (9.28)
nV Δuw
or
ΔV
= Cf nΔuw (9.29)
V
Equating Eq. 9.26 and Eq. 9.29, we have
⎛ Δσ + Δσ2 + Δσ3 ⎞
Cs ⎜⎜⎜ 1 −Δuw ⎟⎟⎟ = Cf nΔuw (9.30)
⎝ 3 ⎠
M09_PURU1773_01_SE_C09.indd 287 2/12/2013 5:37:33 PM

Rearranging,
⎛ 1 ⎞⎟⎛ Δσ1 + Δσ2 + Δσ3 ⎞
Δuw = ⎜⎜⎜ ⎟⎜ ⎟⎟ (9.31)
⎜⎝ nCf / Cs + 1⎟⎟⎠⎜⎜⎝ 3 ⎟⎠
Equation 9.31 represents the change in pore water pressure, Δuw, due to changes in the
total principal stresses for undrained conditions. In a triaxial test, it is customary to take
Δσ2 = Δσ3 and, hence, Eq. 9.31 reduces to
1 ⎛ Δσ1 + 2Δσ3 ⎞⎟
Δuw = ⎜⎜ ⎟⎟
(nCf / Cs ) + 1 ⎜⎝ 3 ⎠
1 ⎛ Δσ1 −Δσ3 + 3Δσ3 ⎞⎟
= ⎜⎜ ⎟⎟
(nCf / Cs ) + 1 ⎜⎝ 3 ⎠
or
1 ⎡ 1 ⎤
Δuw = ⎢Δσ3 + (Δσ1 −Δσ3 )⎥ (9.32)
(nCf / Cs ) + 1 ⎢⎣ 3 ⎥⎦
As the soil mass is not an elastic and isotropic material, the coefficients have been replaced
by two parameters, A and B, referred to as pore pressure coefficients by Skempton (1954).
Thus,
Δuw = B[Δσ3 + A(Δσ1 −Δσ3 )] (9.33)
The change in pore water pressure, Δuw, is due to an isotropic stress increase Δσ3 together
with an axial stress increase (Δσ1 − Δσ3), as happens in a conventional triaxial test.
An overall coefficient B can be obtained by dividing Eq. 9.33 by Δσ1. Thus,
Δuw ⎡ Δσ ⎛ Δσ3 ⎞⎟⎤

= B ⎢⎢ 3
+ A ⎜⎜⎜1 − ⎟⎥
Δσ1 ⎢⎣ Δσ1 ⎝ Δσ1 ⎟⎟⎠⎥⎥⎦
⎡ ⎛ Δσ3 ⎞⎟⎤
= B ⎢⎢1 − (1 − A)⎜⎜⎜1 − ⎟⎥
⎢⎣ ⎝ Δσ1 ⎟⎟⎠⎥⎥⎦
or
Δuw
=B
Δσ1
where
⎡ ⎛ Δσ3 ⎞⎟⎤
B = B ⎢⎢1 − (1 − A)⎜⎜⎜1 − ⎟⎥ (9.34)
⎢⎣ ⎝ Δσ1 ⎟⎟⎠⎥⎥⎦
Now,
1
B= (9.35)
(nCf / Cs ) + 1
Water is an incompressible pore fluid, and hence, in a saturated soil Cf Cs. Hence,
B ≈ 1.0. Thus, for a saturated soil, a uniform increase in total stress results in an equal rise in
M09_PURU1773_01_SE_C09.indd 288 2/12/2013 5:37:35 PM

1.0
0.8
0.6
B
0.4
0.2
0 0.2 0.4 0.6 0.8 1.0

Sr
Fig. 9.33 Typical Variation of B with the degree of saturation Sr
pore water pressure, and the effective stress remains unchanged. That is, Δuw = Δσ3 during
application of all-round pressure and Δuw = (Δσ1 − Δσ3) during application of deviator stress.
The presence of air in the voids increases the compressibility of the pore fluid in partially
saturated soils. Thus, Cf Cs, and hence, B<1.0. Figure 9.33 shows a typical variation of B
with the degree of saturation.
The pore pressure coefficient A is not a constant but depends on the stress history, stress
level, strain level, strain rate, and mode of application of stress (viz., increasing or decreas-
ing). The value of A at the time of shear failure is denoted as Af and is often quoted as a guide
value. Values of Af for different soils are shown in Table 9.2.
9.8.2 Measurement of Pore Pressure Coefficients

The pore pressure coefficients, A and B, may be measured in a consolidated–undrained test.
The coefficient B is determined during the consolidation stage and A during the shear stage
of the test.
Soon after the application of all-round pressure, and before permitting drainage, the
change in pore water pressure due to change in σ3 can be measured. Then, B = Δuw/Δσ3.
After allowing for drainage and at the end of completion of consolidation, the drainage valve
Table 9.2 Values of pore pressure coefficient Af
Soil type Af
Dense fine sands –0.3 to 0.0

Medium fine sands 0.0 to 1.0
Very loose fine sands 2.0 to 3.0
Heavily pre-compressed clays –0.5 to 0.0
Lightly pre-compressed clays 0.3 to 0.7
Normally consolidated clays 0.7 to 1.30
Very sensitive to quick claysa 1.2 to 2.5
Compacted sandy clays 0.25 to 0.75
Compacted clay gravels –0.25 to 0.25
a
Refer Section 9.9.
M09_PURU1773_01_SE_C09.indd 289 2/12/2013 5:37:35 PM

is closed and a deviator stress is applied. The pore water pressure change (Δuw) resulting
from the change in the principal stress difference (Δσ1 − Δσ3) is measured; then,
Δuw
A=
(Δσ1 −Δσ3 )
9.8.3 Application of Pore Pressure Coefficients

The knowledge of anticipated change in pore water pressure due to change in total stress
provides a means whereby the margin of uncertainty in construction work can be greatly
reduced. The following are some of the uses of pore pressure coefficients in practice
(Bishop, 1954).
1. The initial pore water pressure caused due to the stressing of a natural stratum forming a
foundation or an earth dam, which is necessary to estimate the stability during and after
construction.
2. The pore water pressure induced due to the stressing of compacted impervious fill of an
earth dam during construction, which is needed to control the stability and amount of
settlement after construction.
3. The change in pore water pressure caused by the removal of water load on the impervi-
ous fill of an earth dam due to rapid draw-down.
4. The pressure changes resulting from the removal of the weight of soil in the formation of
slopes and cuttings in natural strata.
9.9 SENSITIVITY OF COHESIVE SOILS

Changes in stress and environment with time may result in the soil having a higher strength
in the undisturbed state than in the re-moulded state. The term sensitivity (St) is used to
describe this difference in strength. This is given by the ratio of the undisturbed strength to
the re-moulded strength, where the strength may be undrained, consolidated–undrained, or
drained. The undrained strength value is most commonly used. The re-moulded strength is
obtained by re-moulding the specimen used for the undisturbed strength, and both the tests
should have the same water content. The sensitivity may vary from 1 to as high as 100, and
thus, accordingly, they are classified as insensitive, medium sensitive, sensitive, extra sensi-
tive, and quick. Table 9.3 gives the sensitivity classification. Figure 9.34 shows typical stress–
strain curves before and after re-moulding, for sensitive clays.
Highly over-consolidated clays are insensitive. This is partly due to the low natural water
content in the soil deposits. Glacial till clays are medium sensitive, whereas a few glacial
clays and most freshwater deposits are very sensitive or extra sensitive.
A few of the freshwater deposits and many marine deposits are quick. The sensitivity of
a majority of cohesive deposits ranges from 2 to 8.
A part of the disturbance caused due to re-moulding is attributed to the disturbance of
adsorbed water in clay layers. If the re-moulded soil is left undisturbed at the same water
content, the lost strength will be regained. This phenomenon is called thixotrophy.
M09_PURU1773_01_SE_C09.indd 290 2/12/2013 5:37:36 PM

Table 9.3 Sensitivity classification
Description Sensitivity, St
Insensitive <2
Medium sensitive 2–4
Sensitive 4–8
Very sensitive 8–16
Slightly quick 16–32
Medium quick 32–64
Quick >64
Source: Bjerrum (1954).
Undisturbed
Undisturbed
Compressive stress
Compressive stress
Re-moulded
St = 8.8
Re-moulded
St = 2.2

(a) (b)
Fig. 9.34 Typical stress–strain curves for sensitive clays
9.10 THIXOTROPHY OF CLAYS

Clays with an initial flocculant structure may lose strength due to disturbance or re-mould-
ing. However, with passage of time, the clay may gain the original strength due to thixotropy.
Thixotropy is defined as an isothermal, reversible, time-dependent process which occurs
under constant composition and volume. This phenomenon is attributed to a process of soft-
ening caused by re-moulding followed by a time-dependent regain of the original hard state.
The extent of strength gain by the thixotropic effect depends on the type of clay. The clay
mineral that absorbs large quantities of water into their lattice structure, e.g., montmoril-
lonite, may have high thixotropic gain. Figure 9.35 shows the gain in strength of a soil due
to thixotropic effect.
Most of the soils partly regain the lost strength with time after remoulding. Such a
material is referred to as partially thixotropic. The nature of the strength gain and the time
for partially thixotropic material is shown in Fig. 9.35b. The difference in strength, that the
undisturbed strength and the thixotropic strength gain, is attributed to the destruction of the
clay-particle structure.
M09_PURU1773_01_SE_C09.indd 291 2/12/2013 5:37:36 PM

qu (undisturbed)
St =
qu (remolded)
qu (undisturbed) Initial
undisturbed
Strength
strength
Remolding
Remolding
Remolding
ing
ing
ing
en
en
rd
en
rd
Ha
Ha
rd
Ha
Remolded
qu (remolded) strength
Time
(a)
Initial
undisturbed
strength
Strenght after
thixotropic
Strength
hardening
Remolding
Remolding
Remolding
ing
ing
ing
en
en
rd
en
rd
Ha
Ha
rd
Ha
Remolded
strength
Time
(b)
Fig. 9.35 Behaviour of (a) thixotropic material; (b) partially thixotropic material
9.11 SHEAR STRENGTH OF GRANULAR SOILS

The shear strength of granular soils is a relatively simple phenomenon unlike that of clayey
soils. Particle orientation and shape, surface roughness, grain-size distribution, and surface
forces are significant factors. The strength of granular material is affected largely by the
initial void ratio and the effective stresses. The c′ value of uncemented granular soil is zero.
The parameters necessary to completely define the shear strength of a granular soil, tested
under slowly applied shear deformation, are φ′ and σ′n.
Hence,
τ f = σn′ tan φ ′ (9.36)
M09_PURU1773_01_SE_C09.indd 292 2/12/2013 5:37:36 PM

and φ′ is influenced slightly by the presence of water. Only the drained strength of a granular
soil is useful in practice because of high permeability. Undrained strength is insignificant
except during an earthquake.
The relevant shear strength parameter (Eq. 9.36) may be obtained from the results of direct
shear tests or drained triaxial tests. The characteristics of dry and saturated granular soils are
the same, provided all are under effective stress conditions and in the case of saturated soils
no excess pore water pressure exists.
The principal difficulty is in the evaluation of σ′n. Because of shear strains in granular
soils, significant volume change occurs under drained and undrained shear conditions. The
volume change mechanism is complex and is associated with random displacements of soil
particles in the shear zone (Leonards, 1962).
Typical stress–strain curves and the respective volume change diagrams with different
initial void ratios are shown in Fig. 9.36. Such relationships between shear stress and shear
displacement under the same confining pressure can be obtained. The dense sample shows
a definite peak in the stress–strain curve, followed by a decrease in the deviator stress with
increase in strain until it levels out to a residual value. The loose sample does not have a
peak point, and the failure criterion is governed by the strength at 20% strain, while at large
strains the curve merges with the dense sample.
In dense samples there will be a considerable inter-locking between particles. Before
complete shear failure can take place, the inter-locking must be overcome in addition to
the frictional resistance at the points of contact. The degree of inter-locking decreases after
the peak, and the shear stress necessary to continue shear displacement is correspondingly
reduced. In a loose sample, there is no inter-locking, and hence, after the peak, the shear
stress necessary to continue the shear displacement remains unchanged.
In a loose sample the void decreases rapidly at first and then attains a minimum value.
It then increases slightly until the peak stress is reached and continues to show less or no
change in volume. In a dense sample after an initial adjustment, there is a rapid increase in
void ratio, which then gradually decreases and approaches a constant value. At sufficiently
large strains, a void ratio is reached at which continued shear deformations will occur at
constant volume. This void ratio has been referred to as critical void ratio, ecr . This shows
Dense
Loose Critical
void ratio
Shear stress
ecr
Loose Dense
Void ratio
sn – Constant sn – Constant
Deformation Deformation
(a) (b)
Fig. 9.36 Typical stress–strain and volume change for granular soils
M09_PURU1773_01_SE_C09.indd 293 2/12/2013 5:37:37 PM

that if the initial void ratio (e0) of a sample is at the critical void ratio, the shear deformation
would take place at constant volume; if e0 is higher than ecr, then volume reduction would
occur; and if e0 is lower than ecr, then volume increase would occur.
The appropriate strength envelope (peak or residual) for a given problem depends on the
field conditions that are to be simulated. In most practical situations, large strains cannot be
tolerated and the peak stress is used to define failure. Typical values of the effective angle of
shearing resistance, φ′, for granular soils are given in Table 9.4.
It is very rare to test undisturbed non-cohesive samples using direct or triaxial shear tests. As
the direct shear test is simple and rapid, it is preferred for tests on sand. Further, it is believed
that the direct shear test is a plane strain test. Both triaxial and direct shear tests give values of
φ′ as 2° to 8° smaller than those given by the plane strain test. Lade and Lee (1976) related the
angles of shearing resistance of coarse-grained soils from the plane strain test, φ′ps , as
′ = 1.5φtr′ − 17°
φps (φtr′ > 34°)
′ = φtr′
φps (φtr′ ≤ 34°)
Table 9.4 Typical effective angle of shearing resistance, φ′, for

coarse-grained soils
Soil φ′°
Loose sample Dense sample
Gravel 34–40 40–50

Uniform 27 33
Well-graded sand 33 45
Gravel 35 50
Silty sand 25–35 30–36
WORKED EXAMPLES
Example 9.1 The following results were obtained from a direct shear test on a sandy
clay sample:
Normal load (N) Shear load providing ring reading (division)
360 13
720 19
1,080 26
1,440 26
If the shear box is 60 mm square and the proving ring constant is 20 N per division, esti-
mate the shear strength parameters of the soil. Would failure occur on a plane within this
soil at a point where the normal stress is 320 kN/m2 and the corresponding shear stress is
138 kN/m2?
M09_PURU1773_01_SE_C09.indd 294 2/12/2013 5:37:37 PM

Solution
Normal load Normal stress PR dial reading Shear stress
(N) (kN/m2) (kN/m2)
360 360/[(0.06)2×1000] = 100 13 (13×20)/[(0.06)2×1000] = 72.2

720 200 19 105.6
1,080 300 26 144.4
1,440 400 32 177.7
400
c = 34 kN/m2
f = 20°
Shear stress, kN/m2
320
240
160
(320, 138)
80
c = 34 kN/m2
0 80 160 240 320 400 480 560 640
Fig. 9.37
The shear stresses are plotted against the corresponding normal stresses as shown in
Fig. 9.37. The straight line having the best fit to the plotted points is drawn. The shear
strength parameters taken from the plot are given as
c = 34 kN / m 2 φ = 20°
The stress state τn = 138 kN/m2 and σn = 320 kN/m2 plots below the failure envelope and
therefore would not produce failure.
Example 9.2 A specimen of fine dry sand, when subjected to a triaxial compression test,
failed at a deviator stress of 400 kN/m2. It failed with a pronounced failure plane with an
angle of 24° to the axis of the sample. Compute the lateral pressure to which the specimen
would have been subjected.
Solution
The failure angle θf = 90° – 24° = 66°. From Eq. 9.8,
σ1 = σ3 tan 2θf + 2c tan θf
M09_PURU1773_01_SE_C09.indd 295 2/12/2013 5:37:38 PM

Since the soil is dry sand, c = 0.Therefore,

σ1 = σ3 tan 2θf
Deviator stress Δσ = σ1− σ3
or
σ1 = Δσ + σ3 = 400 + σ3
or
400 + σ3 = σ3 tan 2 (66°)
or
400
σ3 = = 98.9 kN / m 2
tan 2 (66°) − 1
Example 9.3 Samples of a dry sand are to be tested in triaxial and direct shear tests. In the
triaxial test the sample fails when the major and minor principal stresses are 980 and
280 kN/m2, respectively. What shear strength would be expected in the direct shear test
when the normal stress is 240 kN/m2?
Solution
The relationship between σ1 and σ3 is given as
σ1 = σ3 tan 2 θf + 2c tan θf
As the soil is dry sand, c = 0. Therefore,
980 = 280 tan 2 θf
or
θf = 45° + φ / 2 = 61.88°
or
φ = (61.88°− 45°)× 2 = 33.83°
In the direct shear test, σn = 240 kN/m2.

τ f = σn tan φ = 240 tan 33.83° = 160.7 kN / m 2
Example 9.4 A boring log reveals that a thin layer of silty clay exists at a depth of 15 m
below the natural ground surface. The soil above this layer is a silt having γd = 15.5 kN/m3
and w = 28%. The groundwater table is found to exist approximately near the ground sur-
face. Triaxial shear tests on the undisturbed silty clay samples give the following results:
ccu = 48.3 kN / m 2 , φcu = 13° and cd′ = 41.4 kN / m 2 , φd′ = 23°
Estimate the shearing resistance of the silty clay on a horizontal plane (i) when the shear
stress builds up rapidly and (ii) when the shear stress builds up very slowly.
Solution
Total unit weight of silt γ sat = γd (1 + w)
⎛ 28 ⎞⎟
= 15.5 ⎜⎜⎜1 + ⎟ = 19.84 kN/m
3
⎝ 100 ⎟⎠
M09_PURU1773_01_SE_C09.indd 296 2/12/2013 5:37:39 PM

Submerged unit weight γ′ = γsat – γw = 19.8 – 9.807 = 10.03 kN/m3

Effective pressure at a depth of 15 m σ′n = 15×10.03 = 150.45 kN/m2
Total pressure at the depth of 15 m = 15×19.84 = 297.6 kN/m2
For a rapid build-up of stresses there is no time for dissipation of pore water pressure, and
the total stress parameters are used. Therefore,
Shear strength τf = ccu + σn tan φcu

= 48.3 + 297.6 tan 13° = 117.0 kN/m2
For a slow build-up of stresses, there is no excess pore water pressure, and the effective stress
parameters are used. Therefore,
Shear strength τf = c′d + σ′n tan φ′d
= 41.4 + 150.45 tan 23° = 105.3 kN / m 2
Example 9.5 Triaxial compression tests were conducted on a specimen from a large sample
of undisturbed clay. Tests 1 to 4 were run slowly, permitting complete drainage, and Tests
5 to 8 were run without permitting drainage. Plot Mohr’s modified strength envelope, and
determine the shear strength parameters for both kinds of tests.
1 2 3 4 5 6 7 8
2
(σ1− σ3) at failure (kN/m ) 447 167 95 37 331 155 133 119
σ3 (kN/m2) 246 89 36 6 481 231 131 53
Solution
Compute the points for plotting modified Mohr envelope as shown in the table below:
Test no. (σ1 – σ3) σ′3 σ′1 (σ′1 – σ′3)/2 (σ′1 + σ′3)/2
1 447 246 693 223.5 469.5

2 167 89 256 83.5 172.5
3 95 36 131 47.5 83.5
4 37 6 43 18.5 24.5
(σ1 – σ3) σ3 σ1 (σ1 – σ3)/2 (σ1 + σ3)

5 331 481 812 165.5 646.5
6 155 231 386 77.5 308.5
7 133 131 264 66.5 197.5
8 119 53 172 59.5 112.5
Mohr’s modified strength envelopes are plotted as shown in Figs. 9.38 and 9.39 for Tests
1 to 4 and Tests 5 to 8, respectively.
From Fig. 9.38, the modified parameters are obtained as
c∗ = 10 kN / m 2 and φ∗ = 23°
M09_PURU1773_01_SE_C09.indd 297 2/12/2013 5:37:40 PM

c∗ = 10 kN/m2
400
(s′1 – s′3)/2, kN/m2 f∗ = 23°
300
23°
200
100
0 100 200 300 400 500 600 700

(s′1 + s′3)/2 kN/m2
Fig. 9.38
c∗ = 30 kN/m2
f∗ = 10°
(s 1′ – s 3′ )/2, kN/m2
300
200
100 10°
0 100 200 300 400 500 600 700

(s 1′ + s ′3)/2, kN/m2
Fig. 9.39
Therefore, φ′ = sin–1 (tan 23°) = 25.1°

and
10
c′ = = 11.04 kN/m 2
cos(25.1°)
From Fig. 9.39, the modified parameters are obtained as
c∗ = 30 kN / m 2 and φ∗ = 10°
φu = sin−1 (tan 10°) = 10.16°
M09_PURU1773_01_SE_C09.indd 298 2/12/2013 5:37:41 PM

and
cu = 30 /cos(10.16°) = 30.48 kN / m 2
Example 9.6 A vane of 80 mm diameter and 160 mm height has been pushed into an in situ
soft clay at the bottom of a bore hole. The torque required to rotate the vane was 76 N-m.
Determine the undrained shear strength of the clay. After the test, the vane was rotated several
times, and the ultimate torque was found to be 50 N-m. Estimate the sensitivity of the clay.
Solution
T
cu =
π(1/ 2bd + 1/ 6 d 3 )
2
76 ×10−3
= = 40.5 kN / m 2
π ⎢⎡1/ 2× 0.160 ×(0.08)2 + 1/ 6(0.08)3 ⎤⎥
⎣ ⎦
Therefore, the undisturbed undrained strength = 40.5 kN/m2.
After re-moulding, the undrained shear strength is obtained as
50 ×10−3
cu = = 26.65 kN / m 2
π ⎡⎢1/ 2× 0.160 ×(0.08)2 + 1/ 6(0.08)3 ⎤⎥
⎣ ⎦
Therefore, the re-moulded undrained strength = 26.65 kN/m2
Undisturbed undrained strength 40.50
Sensitivity St = = = 1.52
Re-moulded undrained strength 26.65
Example 9.7 A normally loaded deposit of undisturbed clay extends to a depth of 15 m

from the ground surface, with the groundwater level at 5 m depth from the ground surface.
Laboratory test on the clay shows a plasticity index of 68% saturated and dry unit weights
as 19.2 and 14.5 kN/m3, respectively. An undisturbed specimen for unconfined compressive
strength is taken at 10 m depth. What unconfined compressive strength is it likely to exhibit?
Solution
The effective overburden pressure at a depth of 10 m p0 = 5 × 14.5 + (19.2 – 9.807) ×
5 = 119.45 kN/m2
Now,
cu / p0 = 0.11 + 0.0037 I p
or
cu = p0 (0.11 + 0.0037 × I p )
= 119.45(0.11 + 0.0037 × 68) = 43.2 kN / m 2
Unconfined compressive strength = 43.2× 2 = 86.4 kN / m 2
M09_PURU1773_01_SE_C09.indd 299 2/12/2013 5:37:42 PM

Example 9.8 A triaxial sample was subjected to an ambient pressure of 200 kN/m2, and the
pore pressure recorded was 50 kN/m2. In this state, the sample was found to be fully satu-
rated. Then, the cell pressure was raised to 300 kN/m2. What would be the value of pore
pressure? Then, a deviator stress of 150 kN/m2 was applied to the sample. Assuming the
pore pressure parameter A to be 0.50, determine the pore pressure value.
Solution
Change in cell pressure Δσ3 = 300 – 200 = 100 kN/m2
We know that the pore pressure parameter B is given as
Δuw
B=
Δσ3
As the soil is saturated, B = 1. Therefore,
uw = Δσ3 = 100 kN / m 2
Pore pressure after increase of cell pressure = uw
= (uw )0 + Δuw
= 50 + 100 = 150 kN / m 2
The pore pressure change due to deviator stress change is given as
Δuw = A(Δσ1 −Δσ3 )

= 0.5×150 = 75 kN / m 2
Pore pressure after application of deviator stress = (uw)0 + Δuw
= 150 + 75 = 225 kN / m 2
Example 9.9 A sample of stiff clay was tested in a triaxial shear test and found to have a
cohesion c of 200 kN/m2 and angle of shearing resistance of 37°. What will be its effective
compressive strength if a horizontal hole is made with zero confining stress and a water
pressure of 220 kN/m2?
Solution
Plot the Coulomb envelope taking c = 200 kN/m2 and φ = 37°, as shown in Fig. 9.40.
From the origin, draw a line with angle θf = 45° + φ/2(63.5°) to intersect the envelope
at D.
At D, erect a perpendicular to cut the x-axis at E. With E as centre and ED as radius, draw
a circle which will pass through the origin and intersect the x-axis at F. Point F represents the
total compressive stress.
From the plot, σ1 = 760 kN/m2. The pore water pressure uw = 220 kN/m2. Therefore,
Effective compressive strength = σ′1 = σ1 – uw
= 760 − 220 = 540 kN / m 2
M09_PURU1773_01_SE_C09.indd 300 2/12/2013 5:37:43 PM

600
Shear stress, kN/m2

f = 27°
400
D
200
c = 200
kN/m2 qf E F
0 200 400 600 800 1,000
Fig. 9.40
Example 9.10 In a triaxial test on a saturated clay, the sample was consolidated under a cell
pressure of 160 kN/m2. After consolidation, the cell pressure was increased to 350 kN/m2,
and the sample was then failed under undrained condition. If the shear strength parameters
of the soil are c′ = 15.2 kN/m2, φ′ = 26°, B = 1, and Af = 0.27, determine the effective major
and minor principal stresses at the time of failure of the sample.
Solution
Pore pressure at the time of failure (uw)f = (uw)0 + Δuw.
Pore pressure soon after increase in cell pressure
= BΔσ3 = 1×(350 − 160) = 190 kN / m 2
Therefore,
(uw )f = 190 + A(σ1 − σ3 )
= 190 + 0.27(σ1 − σ3 )
Now,
σ3′ = σ3 − (uw )f = 350 − 190 − 0.27(σ1 − σ3 )
Since σ1− σ3 = σ′1− σ′3
σ3′ = 160 − 0.27(σ1′ − σ3′ ) = 160 − 0.27σ1′ + 0.27σ3′
or
160 − 0.27 σ1′
σ3′ = = 219.2 − 0.37 σ1′
1 − 0.27
We know from Eq. 7.8 that
σ1′ = σ3′ tan 2 θf + 2c ′ tan θf
or
φ′ 26°
θf = 45 + = 45° + = 58°
2 2
M09_PURU1773_01_SE_C09.indd 301 2/12/2013 5:37:44 PM

Therefore,
σ1′ = (219.2 − 0.37 σ1′ ) tan 2 58° + 2×15.2 tan 58°
= 561.4 − 0.95σ1′ + 48.55
σ1′ = 312.8 kN / m 2
and
σ3′ = 219.2 − 0.37 × 312.8 = 34.96 kN / m 2
Example 9.11 Coulomb failure envelope of soil is τf = c′ + σ′ tan φ′. For the same soil, the
modified failure envelope in a q′ − p′ plot can be expressed as q′ = m + p′ tan α. Express α as
a function of φ′ and m as a function of c′ and φ′.
Solution
From Fig. 9.41
(σ1′ − σ3′ )
AB AB 2
sin φ ′ = = =
AC CO + OA
c ′cotφ ′+
(σ1′ + σ3′ )
2
Rearranging in a straight-line equation form, we get:
(σ1′ − σ3′ ) (σ1′ − σ3′ )

= c ′cos φ ′+ sin φ ′
2 2
This can be represented as:
q′ = m + p′ tan α
Then m = c′ cos φ′
and tan α = sin φ′
i.e., α = tan−1(sin φ′).
t f = c⬘ + σ⬘ tanf⬘
f⬘
Shear
stress
s1⬘ − s3⬘
2
c⬘
C f
O s3⬘ A s1⬘ Normal stress
c⬘ cotf⬘ s1⬘ + s3⬘
2
Fig. 9.41
M09_PURU1773_01_SE_C09.indd 302 2/12/2013 5:37:45 PM

POINTS TO REMEMBER
9.1 Peak shear strength of a soil is the maximum shear stress that can be resisted by the
soil.
9.2 In a strain-hardening soil the peak shear strength is referred to the point at which
significant shear strain starts. In a strain-softening soil the peak shear strength is
well defined and after a continued large strain, the shearing resistance attains a
constant level, and the corresponding shearing resistance is called the residual shear
strength.
9.3 Coulomb suggested a simple linear relationship of shear strength (τf = c′ + σ′ tan φ′)
controlled by the shear strength parameters c′ (cohesion intercept) and φ′ (angle of
shearing resistance). These parameters, c′ and φ′, are not constants for a given soil but
depend on factors like void ratio, initial stress, pore pressure, drainage conditions,
and type of test.
9.4 Coulomb’s failure condition is stated in another form – that if the Mohr’s circle for a
state of stress at a point is tangential to Coulomb’s failure loci, then that point is said
to be in a state of failure. This is known as Mohr–Coulomb failure criterion, and the
failure loci is called Mohr–Coulomb failure envelope.
9.5 Peak shear strength parameters depend on per cent clay contact, drainage condition,
type of loading, consolidation history, stress level, anisotropy, and other environmental
factors. Residual shear strength is independent of many of the above factors, but φ′r
decreases with increasing clay content and c′r is almost zero.
9.6 Shear strength parameters are designated, based on the drainage conditions of the
test: (i) when no drainage is allowed during both the stages (i.e., consolidation stage
and shear stage) of the test, then the parametres are referred to as undrained shear
strength parameters (cu and φu); (ii) drainage is allowed during the consolidation
stage and no drainage during the shear stage and the parameters are referred to as
consolidated–undrained shear strength parameters cu and φu; and (iii) drainage is
allowed during both the stages, and the shear strength parameters are effective or
drained shear strength parameters c′d and φ′d.
9.7 Undrained shear strength parameters may be applied in field problems where
the change in total stress is immediately compensated by a change in pore water
pressure. The consolidated–undrained total stress parameters may be taken as a
rough guide to the undrained shear strength of the soil. Whenever pore water pres-
sure can change independently of the total stresses, effective shear strength param-
eters should be used.
9.8 Change in pore water pressure resulting from changes in the state of stress (which
occurs in many practical problems involving deformation of soil masses) are esti-
mated using Skempton’s pore pressure parameters A and B.
9.9 Changes in stress and environment with time may result in cohesive soil having
a higher strength in the undisturbed state than in the re-moulded state. The term
sensitivity of cohesive soils is used to describe this difference in strength, which is given
by the ratio of the undisturbed strength to the re-moulded strength.
M09_PURU1773_01_SE_C09.indd 303 2/12/2013 5:37:45 PM

9.10 Shear strength of granular soils depends on particle shape, orientation, surface rough-
ness, grain-size distribution, initial void ratio, and effective stresses. Only the drained
strength parameter (φ′d) is useful in practice because of high permeability. Undrained
strength is insignificant except during an earthquake.
QUESTIONS
Objective Questions
9.1 State whether the following statements are true or false. Justify your choice.
1. Pore pressure parameter B is a function of strain at failure.
2. In a partially saturated soil, the χ-parameter is always greater than unity.
3. Mohr’s failure theory does not consider the effect of intermediate principal stress.
4. The shear strength of granular material is affected largely by the initial void ratio.
5. Consolidated–undrained and drained tests on normally consolidated clays show
zero cohesion.
9.2 When a saturated soil mass is loaded under undrained conditions, the load according
to Terzaghi’s concept is
(a) Borne entirely by water
(b) Borne entirely by soil solids
(c) Shared equally by soil solids
(d) Shared between soil solids and water proportional to their volumes
9.3 The unconsolidated–undrained strength of an intact saturated clay does not depend on
(a) Major principal stress (b) Maximum shear stress
(c) Minor principal stress (d) Maximum prinicipal stress ratio
9.4 Cohesionless soils whose natural void ratios are above the critical will ______ in
volume during shear.
(a) Decrease (b) Remain constant
(c) Increase (d) Initially increase and then remain constant
9.5 For a very heavily over-consolidated clay sample, the probable value of pore pressure
parameter A at failure is likely to be
(a) 0.85 (b) 0.35
(c) 0.0 (d) 0.20
9.6 Consider the following statements.
1. Volume change is considered usually as three-dimensional effect.
2. Plastic flow is the mass movement of soil laterally.
3. Shear failure occurs where part of the soil mass moves as a single unit along a
defined surface of rupture.
(a) 1 and 2 are correct (b) 2 and 3 are correct
(c) 3 and 1 are correct (d) 1 alone is correct
M09_PURU1773_01_SE_C09.indd 304 2/12/2013 5:37:45 PM

9.7 The Mohr theory of rupture implies that there is no influence on failure by
(a) Minor principal stress (σ3) (b) Intermediate principal stress (σ2)
(c) Major principal stress (σ1) (d) Principal stress difference (σ1− σ3)
9.8 Identify the incorrect statement. Effective stress shear parameters of a clay can be
obtained from
(a) Drained triaxial shear test
(b) Drained direct shear test
(c) Consolidated–undrained triaxial shear test with pore water pressure measure-
ments
(d) Unconsolidated–undrained triaxial shear test with pore water pressure measure-
ments
9.9 Both the shear stress and the normal stress on the plane of failure are measured
directly in
(a) Triaxial shear test (b) Vane shear test
(c) Direct shear test (d) Unconfined compression test
9.10 The unconfined compression test is a special type of
(a) Vane shear test
(b) Unconsolidated–undrained triaxial test
(c) Unconsolidated–undrained direct shear test
(d) Drained triaxial test
9.11 Explain why the angle of shearing resistance of a soil is not always the same as the
angle of internal friction.
9.12 Discuss the type of laboratory triaxial test you would recommend to be carried out for
the following field problems:
1. The initial stability of a foundation on saturated clay
2. The stability of a clay foundation of an embankment, the rate of construction being
such that some consolidation of the clay occurs
3. The long term-stability of a slope in stiff fissured clay
9.13 Explain why the angle of the failure plane observed in a shear test might differ more
often from that predicted from a Mohr diagram at failure.
9.14 Define critical void ratio. Explain the shear behaviour of a soil whose void ratio is less
than the critical void ratio.
9.15 An undrained triaxial shear test is conducted on a fully saturated cohesionless soil
specimen. How does this shearing resistance compare with that from a drained test if
the initial condition of the specimen was dense?
9.16 How are the drainage conditions adopted in a triaxial shear test realized in the field?
9.17 Explain how a negative pore water pressure develops in a consolidated–undrained
test on a over-consolidated clay.
9.18 Discuss at least three factors which govern the shear strength of cohesionless soils.
M09_PURU1773_01_SE_C09.indd 305 2/12/2013 5:37:45 PM

EXERCISE PROBLEMS
9.1 Describe the state of samples A to D when the Mohr circles describing their state of
stresses are as follows: For A the Mohr circle is a dot on the normal stress axis, for B
the Mohr circle is too small to touch the failure envelope, and for C the Mohr circle is
so large that part of the circle is above the failure envelope.
9.2 The following observations were taken in a series of tests in a 60 mm×60 mm direct
shear box.
Normal load (N) Max. shear load (N)
100 150
200 230
300 308
400 380
Compute the values of c and φ for the soil. Also, find the orientation of principal
planes for Test 1.
9.3 A soil sample taken from a sand deposit is tested in a direct shear test and found to
have an angle of shearing resistance of 32° at a unit weight of 19.8 kN/m3. Estimate
the shear strength of the soil in a horizontal plane, at a depth of 4.5 m below the
ground surface. A structure proposed to be built on the site will cause the vertical and
shear stresses to increase by 65 and 50 kN/m2, respectively, at the same depth. Check
whether the shearing stress exceeds the shear strength of the soil at that depth. Will
the structure be stable if the groundwater rises to the ground surface?
9.4 A clay stratum of 10 m depth is just sheared due to an adjacent structure leaning
against it. The lateral pressure at 10 m depth is estimated to be 115 kN/m2. If the clay
is completely saturated and the failure might be under undrained condition, what is
the shear strength of the clay? The saturated unit weight of the clay is 22.5 kN/m3.
9.5 A series of unconsolidated–undrained triaxial tests on saturated clay yielded the fol-
lowing results:
Lateral stress (kN/m2) 20 40 60

2
Deviator stress at failure (kN/m ) 22.2 22.0 22.2
Determine the shear strength parameters cu and φu.

9.6 In an undrained triaxial test on a sample of saturated clay, the confining pressure is
maintained at 100 kN/m2. At what vertical applied pressure in addition to the confin-
ing pressure should the sample fail? Cohesion c for the soil = 50 kN/m2. What is the
additional vertical pressure required for failure if the confining pressure is 200 kN/m2
instead of 100 kN/m2?
M09_PURU1773_01_SE_C09.indd 306 2/12/2013 5:37:46 PM

9.7 Two consolidated–undrained triaxial tests are performed on undisturbed silty clay
samples. One sample is consolidated under an all-round pressure of 170 kN/m2, and
when subjected to shear, it fails at an added axial stress of 124 kN/m2. The pore water
pressure at the time of failure is found to have a positive value of 110 kN/m2. The sec-
ond sample is consolidated under a pressure of 430 kN/m2 and fails at an added axial
stress of 310 kN/m2. The corresponding pore water pressure at the time of failure is
270 kN/m2. Find the total and effective shear strength parameters of the soil. Also,
compute the pore pressure parameter A at the time of failure. Take B = 1.
9.8 The following results were obtained during a consolidated–undrained triaxial test
with pore pressure measure:
Test no. 1 Test no. 2 Test no. 3
Chamber pressure (kN/m2) 100 200 300

Principal stress difference (kN/m2) 150 190 240
Pore pressure at failure (kN/m2) 50 75 135
Estimate the effective shear strength parameters by plotting a modified Mohr–

Coulomb plot.
9.9 A consolidated–undrained test yields shear strength parameters as c′ = 12.8 kN/m2
and φ′ = 28°. The pore pressure parameters at the time of failure were B = 1.0 and
Af = –0.18. If a specimen with identical initial condition fails at 165 kN/m2 in an
unconfined compression test, estimate the initial value of the suction pore pressure
in the soil.
9.10 A direct shear test conducted on identical soil specimens gave the following results:
Normal stress (kN/m2) Shear stress (kN/m2)
50 40
100 70
Determine the shear strength parameters. If an undrained triaxial test was conducted
on the same soil and at the same density and water content with a cell pressure of
75 kN/m2, estimate the deviator stress at failure.
9.11 An undisturbed sample with a unit weight of 16 kN/m3 has been extracted from a
depth of 7 m below the ground surface for shear testing in the laboratory. What stress
condition would you apply to the specimen prepared from this sample in (i) the direct
shear apparatus and (ii) the triaxial equipment before shearing the specimen to simu-
late the conditions in the ground? During sampling, no groundwater table is encoun-
tered but the groundwater rises to the ground surface during the rainy season.
9.12 A sample of dense sand is tested in the following tests:
1. Direct shear with a normal stress of 150 kN/m2
2. Triaxial shear with a confining pressure of 150 kN/m2
M09_PURU1773_01_SE_C09.indd 307 2/12/2013 5:37:46 PM

Find the maximum shear stress at failure in both the cases if the angle of internal
friction of the sand is 36°. Explain your results with the Mohr–Coulomb envelope.
9.13 A 7 m high embankment is constructed with a soil whose effective shear strength
parameters are c′ = 62 kN/m2, φ′ = 22°, and γ = 15.8 kN/m3. The pore pressure
parameters as determined from triaxial tests are A = 0.39 and B = 0.94. Find the shear
strength of the soil at the base of the embankment just after the fill has been raised
from 7 to 10 m. Assume that the dissipation of pore water pressure during this stage
of construction is negligible and that the lateral pressure at any point is held at half the
vertical pressure.
9.14 A soil specimen measuring 85 mm in length and 40 mm in diameter fails at a load of
90 N when subjected to the unconfined compression test. The axial deformation at the
time of failure is found to be 6 mm. What is the shear strength of the soil sample?
9.15 In an unconfined compression test on a saturated clay, the unconfined compressive
strength was found to be 160 kN/m2, It is known that the same soil showed an angle
of shearing resistance of 10° in a consolidated–undrained test. What is the percentage
of error, and is it conservative or unconservative to use cu = qu/2? Give reasons.
9.16 A vane shear test was performed on a uniform normally consolidated clay at a certain
depth and the following data obtained:
T = 65 N-m d = 65 mm b = 110 mm
wL = 68.4 % wp = 34.1%
Estimate the effective overburden pressure present at the depth of testing.

9.17 An attempt has been made to measure the anisotropic undrained shear strength
of a soft clay by conducting two vane shear tests with different sizes of vanes. The
heights of the vanes were 150 and 300 mm, and they had the same diameter of 75 mm.
The measured torques were 65 N-m and 150 N-m for 150 mm and 300 mm height
vanes, respectively. Determine the undrained cohesion in the vertical and horizontal
directions. Assume the shear stress distribution on the ends as parabolic.
M09_PURU1773_01_SE_C09.indd 308 2/12/2013 5:37:46 PM

10
Laboratory Measurement of
Soil Properties
CHAPTER HIGHLIGHTS
Preparation of dry soil samples – Specific gravity of soil solids: density bot-
tle method, pycnometer or gas jar method – Water content – In-place density:
core-cutter method, sand replacement method – Grain-size distribution: sieve
analysis, pipette method and hydrometer method – Liquid limit – Plastic limit
– Shrinkage factors – Linear shrinkage – Permeability: constant head, falling
head – Free swell – Proctor compaction – Density index – Consolidation –
Unconfined compression – Direct shear – Triaxial shear – California Bearing
Ratio (CBR)
10.1 INTRODUCTION
All geotechnical engineering problems in civil engineering are solved by a combination
of theoretical knowledge and practical knowledge of the geology and history of the site
under consideration and of the knowledge of geotechnical properties of the soil or rock
obtained from laboratory and field tests. The problems associated with the construction
of structures may have different aspects, such as settlement predictions, strength
requirements, stability, and effects of groundwater. The data needed to evaluate the
salient features are obtained from a site investigation and testing programme and are
interrelated. Quite often pilot site investigations, involving sampling, are carried out to
establish the type and characteristics of the soil to be studied. From such a pilot investi-
gation, a laboratory or field testing programme is then decided upon, which considers
the size of samples, the quality of samples, and the frequency of sampling with respect
to variations in the soil. After deciding the appropriate laboratory or field tests, the
major site investigation is undertaken with the distinct purpose of fulfilling a considered
M10_PURU1773_01_SE_C10.indd 309 2/12/2013 7:59:39 AM

test programme. It is the responsibility of the engineer in charge to present the test results
and findings in a report form with all relevant details in a sensible and concise way.
In this chapter, the essentials of laboratory testing of soils, including methods of sample
collection, sample preparation, testing methods, data collection, and presentation of results,
are given in a lucid form. The testing techniques explained in this chapter follow quite
closely the Indian Standards for testing soils, and the relevant facilities are found in most
institutions in India. Further, only testing methods which are relevant to an undergraduate
course in geotechnical engineering are dealt with. However, reference has been made to
certain tests which have some bearing on the main tests.
10.2 TEST NO.1: PREPARATION OF DRY SOIL SAMPLES FOR VARIOUS

TESTS
Scope
To prepare dry soil samples from bulk soil samples received from the field
Apparatus
Wooden mallet
Trays
Pulverizing apparatus – mortar and rubber-covered pestle
Sampler – a suitable riffle-sampler
Sieves – sizes 19, 9.5, 4.75, and 2 mm and 425 μm
Oven with accurate temperature control in the range 105 to 110°C or other suitable
apparatus
Balance of 10 kg capacity, 100 g sensitivity
Balance of 1 kg capacity, 1 g sensitivity
Balance of 250 g capacity, 0.01 g sensitivity
Procedure
1. Dry the bulk field soil samples in air or in the sun. In wet weather, use a drying apparatus
with temperature not exceeding 60°C. Break the clods with a wooden mallet to hasten
drying.
2. Remove organic matter, tree roots, pieces of bark, and shells, and make a note of them. But
for the organic content or lime content test use the whole sample.
3. Dry in ovens with temperature not exceeding 10°C. Depending on the type of test
(Table 10.1) the temperature of the oven is adjusted. Do not resort to chemical drying of
samples.
4. Break the big clods with a wooden mallet, and do further pulverization in the mortar with
the pestle.
5. Sieve the pulverized soil through a specified sieve depending on the type of test
(Table 10.1). Repeat the pulverization till the required quantity is collected. Take care not
to break up the individual soil particles.
6. Decide the actual quantity of the soil sample and the bulk field sample (Table 10.1) based
on the type of test. When a small representative quantity is required from a bigger soil
mass, obtain the same by quartering or riffling.
M10_PURU1773_01_SE_C10.indd 310 2/12/2013 7:59:39 AM

Laboratory Measurement of Soil Properties 311
10.3 TEST NO.2: SPECIFIC GRAVITY OF SOIL SOLIDS

Scope
To determine the specific gravity of soil solids
Apparatus
Density bottle with stopper – 50 ml capacity
Gas jar with rubber bung – 1,000 ml capacity
Ground glass plate or a plastic slip cover
Mechanical shaking apparatus
Balances of 0.001 and 0.2 g accuracies
Hot water bath
Oven with accurate temperature control in the range 105 to 110°C
Thermometer – 0 to 50°C range with 0.1° graduation
Vacuum desiccator – 200 to 250 mm
Table 10.1 Quantity of soil sample required for various tests
Sl. no. Name of test Type of Amount of soil sample Degree of pulverization
drying required for test (passing IS sieve size)
1. Specific gravity Oven 50 g for fine-grained soils, 2 mm

400 g for others
2. Water content Oven As given in test –
3. Grain size analysis Air As given in test –
4. Liquid limit Air 270 g 425 μm
5. Plastic limit Air 60 g -do-
6. Shrinkage factors Air 100 g -do-
7. Linear shrinkage Air 450 g -do-
8. Permeability Oven 2,500 g for 100 mm dia 9.5 mm
mould
5,000 g for 200 mm dia
mould
9. Free swell index Oven 20 g 425 μm
10. Compaction Air 6,000 g 19 mm
11. Consolidation Air/oven 500 g –
12. Unconfined Oven 1,000 g –
compressive strength
13. Vane shear Air/oven 250 g –
14. Direct shear Air/oven 120 g 4.75 mm
15. Triaxial compression Oven 1,000 g to 5,000 g –
16. Swelling pressure Air/oven 2 kg 2 mm
Note: All oven drying is done for 24 hours at 110 ± 5°C except for tests 1, 2, 8, 9, and 16, which are
dried at 105°C to 110°C.
Source: IS: 2720 – Part 1, (1983).
M10_PURU1773_01_SE_C10.indd 311 2/12/2013 7:59:39 AM

Source of vacuum – good filter pump or vacuum pump

Spatula or glass rod
Wash bottle
Procedure
(a) Laboratory method: Density bottle method for fine-grained soils
1. Weigh a clean and dry (dried at 105°C to 110°C and cooled in the desiccator) density
bottle with the stopper to the nearest 0.001 g (M1).
2. Transfer about 5 to 10 g of soil sample, oven dried at 105°C to 110°C and passing 2 mm
IS test sieve, to the bottle and weigh to the nearest 0.001 g (M2).
3. Add air-free distilled water into the bottle such that the soil in the bottle is fully covered.
4. Keep the bottle (without stopper) and contents in a vacuum desiccator and evacuate
gradually till the pressure is reduced to 20 mm of mercury and allow it to remain for at
least 1 hour until no further loss of air is apparent. Alternatively, remove the entrapped
air by heating the bottle and contents in a water bath or sand bath.
5. Release the vacuum and remove the bottle from the desiccator.
6. Stir the soil in the bottle using a spatula or glass rod and carefully transfer the particles
adhering to the blade or rod by washing off with a few drops of air-free water.
7. Put back the bottle and contents in the desiccator and once again evacuvate.
8. Repeat Steps 4 to 7 till no more air is released from the soil.
9. Remove the bottle and contents from the desiccator and add air-free water till the bottle
is full. Insert the stopper and immerse the bottle up to the neck in a water bath till the
contents of the bottle attain the constant temperature of the bath. If there is a decrease in
the volume, add more water and keep it in the water bath again. Repeat the procedure
till the volume remains constant.
10. Remove the stoppered bottle from the bath, wipe the outside, and weigh to the nearest
0.001 g (M3).
11. Clean the bottle, top it with air-free water, fix the stopper, and immerse it in the water
bath till it attains the constant temperature of the bath. Make good if there is any decrease
in volume. Repeat the procedure till the volume remains constant.
12. Remove the stoppered bottle from the bath, wipe the outside, and weigh to the nearest
0.001 g (M4).
13. Repeat Steps 2 to 12 for two more samples.
(b) Field method : Gas jar method for all soils

1. Weigh a clean and dry gas jar and ground glass plate or plastic slip cover to the nearest
0.2 g (M1).
2. Add 200 g of fine-grained soil or 400 g of medium- or coarse-grained soil into the glass
jar and weigh the jar along with the slip cover to the nearest 0.2 g (M2).
3. Add 500 ml of water at room temperature (±2°C) to the soil. Set aside the jar and its
contents for 4 hours in case of medium- or coarse-grained soils.
4. Push a rubber stopper into the mouth of the jar and shake by hand until the particles are
separated and are in suspension. Shake again in a shaking apparatus for a period of 20
to 30 minutes.
5. Remove the stopper and wash off the stopper and the top of the jar carefully into the jar.
Also disperse any froth with a fine spray of water.
M10_PURU1773_01_SE_C10.indd 312 2/12/2013 7:59:39 AM

6. Add some water to the gas jar, allow the soil to settle, and then fill the jar up to the brim
with more water.
7. Place the slip cover on the top, taking care not to trap any air under the plate.
8. Carefully dry the outside of the jar and weigh the jar and contents along with the slip
cover to the nearest 0.2 g (M3).
9. Clean the gas jar, fill completely with water up to the brim, place the slip cover, dry the
outside, and weigh to the nearest 0.2 g (M4).
10. Repeat Steps 2 to 9 for two more samples.
Computations
Specific gravity at T°C,
M2 − M1
G=
( M4 − M1 ) − ( M3 − M2 )
If a liquid other than water is used, then the specific gravity is calculated as follows:
GL ( M2 − M1 )
G=
( M4 − M1 ) − ( M3 − M2 )
where GL is the specific gravity of the liquid used at T°C.

The specific gravity is usually reported at 27°C (unless otherwise specified) and is calcu-
lated thus:
(G)27°C = K(G)T°C
where
Relative density of water at T °C
K=
Relative density of water at 27 °C
The mean value based on three samples is reported to the nearest 0.01. Tests are repeated
if the results differ by more than 0.03 from the mean value.
Typical observations of data and test results of specific gravity from the density bottle
method are shown in Table 10.2.
Discussion
The major source of error is the complete removal of air from the sample. To ensure accurate
results, the soil should be left in vacuum for several hours. Soils often contain a substantial
Table 10.2 Data and results of specific gravity test from density bottle method
1. Mass of density bottle including stopper (M1) g 36.632

2. Mass of density bottle + stopper + oven-dried soil (M2) g 45.842
3. Mass of density bottle + stopper + oven-dried soil + air-free water (M3) g 72.013
4. Mass of density bottle + stopper + air-free water (M4) g 66.273
5. Specific gravity at 40°C 2.654
6. Specific gravity at 27°C 2.644
M10_PURU1773_01_SE_C10.indd 313 2/12/2013 7:59:39 AM

proportion of heavy or light particles. Such soils may give erratic values of specific gravity,
and the tests have to be repeated a sufficient number of times to obtain a reasonable average.
Presence of organic matter may decrease the specific gravity.
For soils containing soluble salts, kerosene or white spirit may be preferred in place of
water. Conventionally, oven-dried soil is used. If there is a possibility of loss of water of
hydration at the oven temperature, the soil may be dried at a temperature less than 80°C
(IS: 2720 – Part 3/Sec 1, 2, 1980, 1981).
10.4 TEST NO.3: WATER CONTENT DETERMINATION

BY OVEN-DRYING METHOD
Scope
To determine the water content of a given soil by the oven-drying method
Apparatus
Non-corrodible air-tight moisture cups
Balance of 0.01 g sensitivity
Oven with accurate temperature control of 110 ± 5°C
Desiccator with a suitable desiccating agent
Procedure
1. Weigh a dry and clean moisture cup with lid (M1).
2. Take a representative sample of wet soil in the cup, replace the lid, and weigh (M2).
3. Keep it in the temperature-controlled oven with the lid removed and allow it to dry for
24 hours.
4. When the sample has dried to a constant weight, replace the lid and cool the cup in a
desiccator.
5. Weigh the moisture cup with the lid with dried soil (M3).
Computations
The water content w of a sample is given as
M2 − M3
w= ×100%
M3 − M1
Results
The water content of the soil is represented as a percentage to two significant figures. Typical
data and results of water content determination are given in Table 10.3.
Table 10.3 Data and result of water content determination
Moisture cup No. 9

Mass of cup (M1) g 19.99
Mass of cup and wet soil (M2) g 52.31
Mass of cup and dry soil (M2) g 49.33
Mass of moisture (M2 – M3) g 2.98
Mass of dry soil (M3 – M1) g 29.34
Water content w = [(M2 – M3)/(M3 – M1)] × 100% 10.16
M10_PURU1773_01_SE_C10.indd 314 2/12/2013 7:59:40 AM

Discussion
The factors which are essential for accurate determination of water content are the mass of
the wet representative sample, and the temperature and duration of the drying of sample.
As per IS recommendations (IS: 2720 – Part 2, 1973), the following masses of soil have to
be used to provide reasonable results.
Size of particles more than Minimum mass of soil specimen

90% passing to be taken for test (g)
425 μm IS sieve 25
2 mm IS sieve 50
4.75 mm IS sieve 200
10 mm IS sieve 300
20 mm IS sieve 500
40 mm IS sieve 1,000
The effects of temperature and duration are discussed in Chapter 2.

The oven-drying method is recommended by Indian Standards as the standard method.
Other methods are the sand bath method, alcohol method, infrared lamp method, torsion
balance method and calcium carbide method (IS: 2720 – Part 2, 1973). The latter two methods
are rapid methods.
10.5 TEST NO. 4: IN-PLACE DRY DENSITY OF SOIL

BY CORE-CUTTER METHOD
Scope
To determine the in-place dry density of soil by the core-cutter method
Apparatus
Cylindrical core-cutter of steel with steel dolly (Fig. 10.1a)
Steel rammer (Fig. 10.1b)
Balance of 1 g sensitivity
Steel rule
Palette knife
Straight edge
Apparatus for water-content determination
Apparatus for extracting samples from the cutter
Procedure
1. Measure inner dimensions (nearest to 0.25 mm) of the core-cutter and find its volume (Vc).
2. Weigh the cutter without the dolly (Mc).
3. Clear and level a small area of about 300 mm2 where the in-place density is to be
determined.
4. Place the cutter on the levelled surface. Keep the dolly on the cutter and advance the
cutter into the subsoil, using the rammer until about 15 mm of the dolly protrudes above
the surface.
M10_PURU1773_01_SE_C10.indd 315 2/12/2013 7:59:40 AM

25
100 Corner rounded off
6 108
105
Dolley
900 approx.
25 mm solid
mild steel staff
Mild steel foot
130
75 All dimensions in mm
10 100 Hardened
106 cutting edge 1400
(a) Cutter (b) Rammer
Fig. 10.1 Apparatus for core-cutter method (Source: IS: 2720 – Part 29, 1975)
5. Dig the soil around the cutter using a spade or pickaxe and bodily remove the cutter
with soil allowing some soil to project from the lower end of the cutter. Trim the top and
bottom of the cutter by means of a palette knife and straight edge.
6. Weigh the cutter with soil, and without dolly (Msc).
7. Remove the soil from the cutter and determine the water content of the soil.
Computations
The bulk density ρ is given as
Msc − Mc
ρ= g / cc
Vc
and the dry density ρd is given as

ρ
ρd = g / cc
w
1+
100
The bulk unit weight γ is given as

γ = 9.807ρ kN/m3
and the dry unit weight γd is given as
γd = 9.807ρd kN/m3
Results
The dry density (in g/cc) and the dry unit weight (in kN/m3) of the soil are reported to the
second decimal place and the water content of the soil (per cent) to two significant figures.
A knowledge of the specific gravity of the soil-solids of the soil will enable us to find the void
ratio and the degree of saturation of the soil. A typical data sheet with relevant results is
shown in Table 10.4.
M10_PURU1773_01_SE_C10.indd 316 2/12/2013 7:59:40 AM

Table 10.4 Data and test results of in-place density
Length of core-cutter 130 mm

Diameter of core-cutter 100 mm
Volume of core-cutter (Vc) 1,021 ml
Mass of core-cutter (Mc) 1,335 g
Mass of core-cutter with wet soil (Msc) 3,056 g
Mass of wet soil (Msc – Mc) g 1,721
Bulk density [ρ = (Msc– Mc)/Vc] 1.69 g/cc
Bulk unit weight (γ = 9.807ρ) 16.6 kN m2
Moisture cup No. 306
Mass of cup 30.85
Mass of cup and wet soil 49.67 g
Mass of cup and dry soil 46.15 g
Water content 23.0% (wt.)
ρ
Dry density ρd = 1.37 g/cc
1 + ( w / 100 )
Dry unit weight (γd = 9.807ρd) 13.48 kN/m3
Discussion
The core-cutter method is convenient and quick; it works best on fine-grained soils but cannot
be used on stony or non-cohesive soils. For the purpose of this test, a soil is a fine-grained soil
if not less than 90% of it passes a 4.75 mm IS sieve. The Indian Standards (IS: 2720 – Part 29,
1975) recommend repeat determinations (at least three) and averaging out of results. Further,
the number of determinations should be such that an additional test will not alter the average
significantly. This method is less accurate than the sand replacement method.
For determination of the bearing capacity of soils, for calculation of the overburden pressure
in settlement computations, and for stability analysis of natural slopes, the in- place den-
sity of natural soil is needed. In all earth dam and embankment projects, the in-place
density is used to check the compaction criterion, and hence this test is usually referred to
as the control test.
10.6 TEST NO. 5: IN-PLACE DRY DENSITY OF SOIL BY THE SAND

REPLACEMENT METHOD
Scope
To determine the in-place dry density of soil by the sand replacement method.
Apparatus
Sand pouring cylinder (Fig. 10.2a)
Tools for excavating holes
Cylindrical calibration can (Fig. 10.2b)
Balance of 1 g sensitivity
Glass plate – 450 mm2, 9 mm thick
Metal tray or container
M10_PURU1773_01_SE_C10.indd 317 2/12/2013 7:59:41 AM

115 200
Handle Flat
100 surface
Shutter cover
plate 5
Shutter
Flange
380 13
150
85 75
115 5
200
(a) Sand pouring cylinder (b) Calibration can
Fig. 10.2 Apparatus for sand replacement method (Source: IS: 2720 – Part 28, 1974)
Metal tray with hole – 300 mm2, 40 mm deep, with a 100 mm hole in the centre
Clean and closely graded natural sand passing the 1 mm IS sieve but retained on the 600 μm
IS sieve
Apparatus for water content determination
Procedure
1. Fill the pouring cylinder with clean sand till the level of sand is about 10 mm from
the top, and weigh (M1). Maintain this mass constant throughout the test, for which
the pouring cylinder has to be calibrated.
2. Place the pouring cylinder on a glass plate and close the tap when the conical portion has
been filled.
3. Collect the sand on a glass plate carefully and weigh the sand. Repeat Steps 1 to 3 at least
three times and take the average mass of sand filling the cone (M2).
4. Measure the internal dimensions of the calibration can and find its volume. Fill the can
with water up to the brim and find the mass. From this mass of water, find the volume
and check the previous value obtained based on the measurement of internal dimen-
sions. Let the volume be V.
5. Place the pouring cylinder concentrically on top of the calibration can with initial mass
M1. Open the shutter and allow the sand to fill it. Tap the cylinder to ensure that the can
and the conical portion are completely filled with sand. Weigh the cylinder.
6. Repeat Step 5 at least thrice and record the average mass M3 of the cylinder after filling
the cone and the can.
7. Clean and level an area of 450 mm2 of the soil to be tested.
8. Place the square tray with a central hole on the prepared surface, excavate a circular hole
of 100 mm diameter and 150 mm depth. Carefully collect all the excavated soil, and find
its mass (Ms). In fine-grained soils push a core-cutter into the soil until its edge is flush
with the levelled surface. Remove the soil within the core-cutter approximately up to a
depth of 100 mm and collect and weigh the soil (Ms). Keep the core-cutter in position
during the rest of the test procedure.
M10_PURU1773_01_SE_C10.indd 318 2/12/2013 7:59:41 AM

9. Take some representative soil for water content determination.

10. Remove the tray and place the pouring cylinder concentrically on the hole with
initial mass M1. Open the shutter and allow the sand to fill it. Tap the cylinder to
ensure that the hole and the conical portion are completely filled with sand. Weigh
the cylinder (M4).
Computations
Mass of sand filling calibration can Ma = (M1 – M3 – M2) g
Ma
Bulk density of sand ρsd = g / cc
V
Mass of sand required to fill the excavated hole Mb = (M1 – M4 – M2) g
Bulk density of soil ρ = (Ms/Mb) × ρsd g/cc
Bulk unit weight γ = 9.807ρ kN/m3
⎛ ρ ⎞⎟
Dry density of soil ρd = ⎜⎜ ⎟ g / cc
⎜⎝ 1 + (w / 100) ⎟⎟⎠
Dry unit weight, γd = 9.807ρd kN/m3

Results
The dry density (in g/cc) and unit weight (in kN/m3) of the soil are reported to the second
decimal place and the water content of the soil (per cent) to two significant figures. A typical
data sheet with relevant results is shown in Table 10.5.
Discussion
As the in-place dry density of a soil varies from point to point, Indian Standards (IS: 2720 –
Part 28, 1974) recommend repeat tests at different locations close to each other and taking the
mean value. However, the number of determinations should be such that an additional test
should not make a significant difference in the mean value.
In granular soils with little or no fines, there is a possibility of error because of the
slumping of the sides of the excavated hole. It is customary to check the calibration bulk
density during each day’s work as there is a possibility of bulking of sand due to atmospheric
humidity.
For fine- and medium-grained soils, a pouring cylinder of 3 litre capacity is used, and for
fine-, medium-, and coarse-grained soils a greater capacity of 16.5 litres is used. Here soils
with particles less than 2 mm in size are considered as fine-grained soils. Further, if the depth
of excavation exceeds 150 mm for some reason or the other, a different calibration can with
a depth comparable to the depth of the hole has to be chosen.
The sand replacement method is relatively slow but can be used on any type of soil.
Two more methods of determining the in-place density are available, viz., the ring and
water replacement method (IS: 2720 – Part 3, 1971) and the rubber-balloon method (IS: 2720 –
Part 34, 1972). The ring and water replacement method is suitable for coarse-grained soils,
including gravels, cobbles, boulders, and rocks, while the rubber-balloon method is suitable
for compacted or firmly bonded soils.
M10_PURU1773_01_SE_C10.indd 319 2/12/2013 7:59:42 AM

Table 10.5 Data and test results of in-place density determination
(a) Calibration
Mass of pouring cylinder with sand before pouring (M1) 11,686 g
Average mass of sand filling cone only (M2) 1,093 g
Volume of calibration can (V) 1,178 ml
Average mass of pouring cylinder and sand after filling can and cone (M3) 8,850 g
Mass of sand filling calibration can (Ma = M1 – M3 – M2) g 1,743 g
Ma 1.48 g/cc
Bulk density of sand ρsd =
V
(b) Measurement of soil density
Mass of excavated soil from the hole (Ms) 2,807 g
Mass of pouring cylinder and sand after filling hole and cone (M4) 8,245 g
Mass of sand required to fill the excavated hole Mb = (M1 – M4 – M2) 2,348 g
Ms 1.77 g/cc
Bulk density of soil ρ = × ρsd
Mb
Bulk unit weight, γ = 9.807ρ 17.35 kN/m3
Moisture cup No. 18
Mass of cup 16.95 g
Mass of cup and wet soil 29.96 g
Mass of cup and dry soil 29.30 g
Water content 5.34% (wt.)
⎛ ρ ⎞⎟
Dry density of soil ρd = ⎜⎜ 1.68 g/cc
⎜⎝ 1 + ( w + 100 ) ⎟⎟⎠
Dry unit weight γd = 9.807ρd 16.48 kN/m3
10.7 TEST NO.6: GRAIN-SIZE DISTRIBUTION BY SIEVE ANALYSIS
Scope
To determine the grain-size distribution of a soil by sieve analysis
Apparatus
Sieves – 100 mm, 63 mm, 20 mm, 10 mm, 4.75 mm, 2.4 mm, 1.2 mm, 600 μm, 300 μm, 150 μm,
and 75 μm IS sieves
Oven with accurate temperature control in the range from 105 to 110°C
Trays and buckets
Brushes for cleaning sieves
Mortar with a rubber-covered pestle
Mechanical sieve shaker
Reagents – sodium hexametaphosphate or a mixture of sodium hydroxide and sodium car-
bonate, or any other dispersing agent
M10_PURU1773_01_SE_C10.indd 320 2/12/2013 7:59:42 AM

Procedure
1. Prepare the soil sample received from the field, as suggested in Test No. 1.
2. Take a certain quantity of soil* (IS: 2720 – Part 4, 1975) and separate the soil fraction
passing and retained on the 4.75 mm sieve.
3. Conduct a separate sieve analysis test for each fraction.
4. Sieve the soil retained on the 4.75 mm sieve by hand sieving through the following set of
sieves: 100 mm, 63 mm, 20 mm, 10 mm, and 4.75 mm. Agitate the sieve while sieving such
that the soil sample rolls in an irregular motion over the sieve. Rub the sample with a rub-
ber pestle, if necessary, and re-sieve to ensure that only individual particles are retained.
Maximum size of material present in Mass of soil to be

substantial quantities (mm) taken for test (kg)
80 60
40 25
25 13
20 6.5
12.5 3.5
10 1.5
6.3 0.75
4.75 0.40
5. Record the mass of material retained on each sieve. If the soil contains more than about
20% of gravel particles with cohesive particles adhering to them, then wash the gravel on
the 4.75 mm sieve with sodium hexametaphosphate solution and record the correct mass
of soil retained on the 4.75 mm sieve, and thereby record the correct mass of soil passing
the 4.75 mm sieve.
6. Sieve the soil passing the 4.75 mm sieve through the following sieves: 2 mm, 1 mm, 600 μm,
300 μm, 150 μm, and 75 μm. Arrange the sieves in descending order of sieve openings with
the 2 mm sieve at the top. Place the cover and a receiver at the top and bottom of the sieves,
respectively. Keep the entire set of sieves on a sieve shaker and allow the sample to be sieved
for a minimum period of 10 minutes.
7. Record the mass of material retained on each sieve.
Computations
Mass of soil retained
1. Percentage retained on any sieve = ×100
Total soil mass
2. Cumulative percentage retained on any sieve = sum of percentages retained on all coarser
sieves
3. Percentage finer N = 100 – (cumulative percentage retained)
*If more than 500 g of soil passes the 4.75 mm sieve, take about 500 g of soil and calculate the combined
percentage finer (N) for the second sieve analysis from the relation N = N′ × (M2/M1), where N′ is the
percentage finer for the second sieve analysis based on the soil taken for the second sieve test, M1 is the
mass of soil taken for the entire sieve analysis (as taken in Step 2), and M2 is the mass of soil passing
4.75 mm sieve.
M10_PURU1773_01_SE_C10.indd 321 2/12/2013 7:59:42 AM

Results
The grain-size distribution curve is plotted by taking the percentage finer on the arithmetic
scale and the sieve opening on the logarithmic scale. Typical test results are presented in
Table 10.6, and the grain-size distribution curve is given in Fig. 10.3.
Discussion
A wet sieve analysis has to be performed if the material passing the 4.75 mm sieve contains
more clay-size particles. For this, the soil is soaked in a dispersing agent. The dispersing
agent is prepared by mixing 2 g of sodium hexametaphosphate or 1 g of sodium hydroxide
with 1 g of sodium carbonate in 1 litre of water. The soaked soil specimen is washed through
the nest of sieves. The collected material in each sieve is dried and weighed.
10.8 TEST NO. 7: GRAIN-SIZE DISTRIBUTION BY PIPETTE METHOD

Scope
To determine the grain-size distribution of a soil by the pipette method
Apparatus
Sampling pipette – as illustrated in Fig. 2.5 with a capacity of approximately 10 ml and fit
enough to arrange to a required depth as shown in Fig. 10.4
Glass sedimentation tubes – 50 mm in diameter, 350 mm long, marked at 500 ml volume,
with rubber bungs to fit a minimum of two numbers
Weighing bottles – fitted with round stoppers or crucibles with suitable lids, approximately
25 mm in diameter and 50 mm high. Mass of bottles is found to the nearest 0.001 g
Constant temperature bath – capable of being maintained at 27 ± 0.1°C with provision to
immerse the tube up to the 500 ml mark
Stirring apparatus – mechanical stirrer with a speed of 8,000 to 10,000 rpm when loaded and
with dispersion cups with baffle rod
Sieves – 2 mm, 425 μm, 75 μm IS sieves
Oven with an accurate temperature control in the range from 105 to 110°C
Stopwatch
Desiccator
Evaporating dish
Conical beaker – 650 ml or 1 litre capacity
Funnel – Buchner or Hirch about 70 mm in diameter
Filter flask – 500 ml capacity
Measuring cylinder – 100 ml capacity
Pipette – 25 ml capacity
Glass filter funnel – about 100 mm in diameter
Wash bottle
Filter paper
Blue litmus paper
Glass rod – 4 to 5 mm in diameter and 150 to 200 mm long
Thermometer – 0 to 50°C, accurate to 0.5°C
Reagents – hydrogen peroxide – 20 volume solution
M10_PURU1773_01_SE_C10.indd 322 2/12/2013 7:59:42 AM

Table 10.6 Sieve analysis test results
IS sieve no. Particle size Mass of soil Percentage Cumulative Percentage

(mm) retained (g) retained percentage retained finer
100 mm 100.0 0.0 0.0 0.0 100.00

63 mm 6.3 5.5 1.1 1.1 98.90
20 mm 20.0 4.5 0.9 2.0 98.00
10 mm 10.0 5.8 1.16 3.16 96.84
4.75 mm 4.75 29.0 5.80 8.96 91.04
2.0 mm 2.0 70.2 14.04 23.0 77.00
1.0 mm 1.0 124.6 24.92 47.92 52.08
600 μm 0.6 69.1 13.82 61.74 38.26
300 μm 0.3 58.2 11.64 73.38 26.62
150 μm 0.15 82.0 16.40 89.78 10.22
75 μm 0.075 19.3 3.86 93.64 6.36
Pan – 31.8
Note: Total mass of soil taken is 500 g.
Indian soil classification
Gravel Sand Silt Clay

80 4.75 0.075 0.002
100
Percentage finer
80
60
40
20
0
100 10.0 1.0 0.1 0.01 0.001
Diameter of particle, mm
Fig. 10.3 Grain size distribution curve from sieve anlaysis
M10_PURU1773_01_SE_C10.indd 323 2/12/2013 7:59:43 AM

Scale graduated
in cm and mm
A and B
C
D
E
F
A and B – 125-ml bulb
G funnel with stopcock
C – Safety bulb suction
Sliding inlet tube
panel
D –Safety bulb
H E –Three–way stopcock
F –Outlet tube
Constant G –Sampling pipette
temperature H –Sedimentation tube
bath
Note: D, F, and G are
jointed to three-way
stopcock E.
Fig. 10.4 Pipette set-up (Source: IS: 2720 – Part 4, 1975)
Hydrochloric acid, approximately 1 N solution – 89 ml of concentrated hydrochloric acid

(G = 1.18) diluted with distilled water to make 1 litre of solution
Sodium hexametaphosphate solution – dissolve 33 g of sodium hexametaphosphate and 7 g
of sodium carbonate in distilled water to make 1 litre of solution
Procedure
(a) Calibration of sampling pipette
1. Clean the sampling pipette thoroughly and immerse the nozzle in distilled water.
2. Close tap B and keep tap E open (Fig. 10.4). Attach a rubber tube to C and suck up water
in the pipette until it rises above E. Close tap E and remove the pipette from the water.
3. Pour surplus water in the cavity above E through F into a small beaker by opening the tap E.
4. Discharge the water contained in the pipette and tap E into a glass weighing bottle of
known mass and determine the mass of water. From the mass determine the internal
volume (Vp) of the pipette and the tap, to the closest 0.05 ml.
(b) Pre-treatment of soils

5. Determine the percentage of soluble solids in the soil and wash with water before further
treatment if the soluble solids percentage is more than 1.
6. Take two samples of soil, 50 g each, and the soil passing the 4.75 mm IS sieve. Determine
the water content of one of the samples.
7. Out of the other sample, use the entire 50 g if the soil contains more fine sand or use
20 g if the soil is clayey. Find the mass of soil accurately to 0.001 g (Ma) and place it in a
650 ml conical beaker.
M10_PURU1773_01_SE_C10.indd 324 2/12/2013 7:59:43 AM

8. Add 50 ml of distilled water and gently boil the soil suspension till the volume is reduced
to about 40 ml. Add 75 ml of hydrogen peroxide and allow the sample to stand overnight
covered with a cover glass.
9. Heat the sample gently, taking care to avoid frothing over. Agitate frequently either by
stirring or by shaking the beaker. When vigorous frothing has subsided on addition of
fresh hydrogen peroxide, reduce the volume to 30 ml by boiling.
10. If the soil contains calcium carbonate, add 10 ml of hydrochloric acid after cooling the
solution obtained in Step 9. Stir the solution with a glass rod for a few minutes and allow
it to stand for about 1 hour or for longer periods. Continue the treatment till the solution
gives an acid reaction to litmus.
11. Filter the solution, pre-treated with peroxide and acid alone, using the Buchner of Hirch
funnel and wash with warm water until the filtrate shows no acid reaction to litmus.
Transfer the wet soil to an evaporating dish and wash the funnel and filter paper with
minimum water. Dry the contents of the evaporating dish, cool in a desiccator, and
weigh accurately. Record the mass of soil remaining after pre-treatment (Mb).
12. Omit pre-treatment of the soil if it does not contain calcium compounds or soluble solids
and has a low (less than 2%) organic content.
(c) Dispersion of soil
13. Add about 25 ml of sodium hexametaphosphate solution to the mixture, warm gently
for about 10 minutes, and then transfer the mixture to the cup of a mechanical stirrer
using a jet of water.
14. Stir the soil suspension for 15 minutes.
15. Transfer the suspension through a 75 μm IS sieve placed on a receiver and wash off all
traces of suspension adhering to the dispersion cup.
(d) Sedimentation
16. Transfer the suspension, that has passed through the sieve, to a sedimentation tube and
make the volume to 500 ml by adding distilled water.
17. Add 25 ml of sodium hexametaphosphate solution in a 500 ml sedimentation tube
(comparison tube), and add water to make the level exactly 500 ml.
18. Immerse the sedimentation tube with soil suspension in a constant temperature bath (if
used) and note down the temperature of the bath. Fix a rubber bung on the mouth of the
sedimentation tube and allow the suspension to attain the temperature of the bath.
19. Remove the sedimentation tube from the constant temperature bath and shake it
thoroughly by inverting the tubes several times. Replace the tube in its position in the
apparatus and remove the rubber bung carefully without disturbing the tubes.
20. Keep tap E closed and lower the pipette vertically into the suspension until the end is
100 ± 1 mm below the surface of the suspension. Take care to lower the pipette into the
suspension about 15 seconds before collecting the sample.
21. Open tap E and draw up a sample (Vp ml) into the pipette till the pipette and the bore
in the tap E are filled and then close tap E. Complete this operation within a time of
10 seconds.
M10_PURU1773_01_SE_C10.indd 325 2/12/2013 7:59:43 AM

22. Withdraw the pipette and wash with distilled water the surplus suspension drawn
above the bore of the tap E through the outlet tube F, by opening the tap E in such a way
as to connect D and F. Allow distilled water to run from bulb funnel A into D and out
through F until no solution remains in the suspension. Repeat this operation during each
time of sampling.
23. Keep a tared weighing bottle under the end of the pipette and open the tap E so that the
contents of the pipette are delivered into the bottle. Wash the inner walls by allowing
distilled water to run from bulb A, through E, into the pipette.
24. Repeat Steps 17 to 20 after expiry of a particular time approximately corresponding to
particle diameters 0.02 mm, 0.06 mm, 0.002 mm, and 0.001 mm. Take the time of settling
to a depth of 100 mm of particles of various diameters for a given temperature from
Table 10.7.
25. Place the weighing bottles along with the contents in the oven. After drying, cool in a
desiccator and weigh to the nearest 0.001 g. Find the mass of the solid materials in the
sample (M1, M2, M3, and M4 for each respective sampling time).
26. Also take a sample of volume Vp from the comparison tube and find the mass of the solid
material (Ms) in the sample tube.
27. Determine the specific gravity of soil solids from Test No. 2.
Computations
1. Loss in mass after pre-treatment is given as
M b (100 + w)
P = 100 −
Ma
where P is the loss in mass in percentage, Mb the mass of soil after pre-treatment, w the
air-dry moisture content of the soil taken for analysis, and Ma the mass of air-dry soil.
2. The diameter of the particle is given as
30ηw He
D=
980(ρs − ρw ) t
3. The mass of solid material in 500 ml of suspension for each sampling time is given as
Mi or Ms
Mi′ or Ms′ = × 500
Vp
where M′i is the mass of material in 500 ml from respective samplings (e.g., M′1, M′2 M′3,
etc.), M′s the mass of sodium hexametaphosphate in 500 ml of solution, Mi the mass of
material in Vp ml from respective samplings (e.g., M′1, M′2, M′3, etc.), and Ms the mass of
sodium hexametaphosphate in Vp ml of suspension.
4. Percentage finer is given as N′ = (M′i – M′s)/Mb ×100
5. Combined gradation may be calculated based on the total soil sample taken for analysis.
Results
Typical test results are presented in Table 10.8 and the gradation curve is shown in Fig. 10.5.
M10_PURU1773_01_SE_C10.indd 326 2/12/2013 7:59:44 AM

Table 10.7 Rate of settling of particles at various temperatures
Diameter (mm)
0.001 0.002 0.003 0.004 0.005 0.006 0.007 0.008 0.009 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.01
M10_PURU1773_01_SE_C10.indd 327
Temp. (°C) (Time for a fall of 100 mm)
Hours Minutes Seconds
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16) (17) (18) (19) (20)
10 40.80 10.90 4.53 2.55 98.0 68.0 53.0 38.3 30.2 24.5 367 163 92 59 40.8 31.8 22.9 18.1 14.7
11 39.64 9.91 4.40 2.48 95.2 66.47 51.5 37.2 29.3 23.8 357 158 89 57 39.6 30.9 22.3 17.6 14.3
12 38.55 9.63 4.28 2.41 92.6 64.2 53.1 36.1 28.5 23.1 347 154 87 55 38.5 30.0 21.7 17.1 13.9
13 37.48 9.37 4.16 2.34 90.0 62.5 48.7 35.1 27.7 22.5 337 150 84 54 37.5 29.2 21.1 16.7 13.5
14 36.39 9.10 4.04 2.28 87.6 60.8 47.3 34.2 27.0 21.9 328 146 82 52 36.4 28.4 20.5 16.2 13.1
15 35.45 8.86 3.93 2.22 85.1 59.2 46.1 33.3 26.2 21.3 319 142 80 51 35.4 27.6 19.9 15.8 12.5
Laboratory Measurement of Soil Properties
16 34.49 8.62 3.83 2.16 82.8 57.6 44.9 32.4 25.5 20.7 310 138 78 50 34.5 26.9 19.4 15.3 12.4
17 33.64 8.41 3.73 2.10 80.8 56.1 43.7 31.5 24.9 20.2 302 134 76 48 33.6 26.2 18.9 14.8 12.1
18 32.73 8.18 3.64 2.04 78.6 54.6 42.5 30.7 24.2 19.6 294 131 74 47 32.7 25.5 18.4 14.5 11.8
19 31.89 7.98 3.54 1.99 76.6 53.2 41.4 29.9 23.6 19.1 287 127 72 46 31.9 24.8 17.9 14.1 11.5
20 31.10 7.77 3.45 1.94 74.6 51.8 40.4 29.1 23.0 18.6 280 124 70 45 31.1 24.2 17.5 13.8 11.5
21 30.28 7.57 3.36 1.89 72.7 50.6 39.4 28.4 22.4 18.2 273 121 68 44 30.3 23.6 17.0 13.4 10.9
22 29.55 7.38 3.28 1.85 70.9 49.3 38.4 27.7 21.8 17.7 266 118 66 42 29.5 23.0 16.6 13.1 10.6
23 28.81 7.21 3.20 1.80 69.2 48.1 37.5 27.0 21.3 17.3 259 115 65 41 28.8 22.4 16.2 12.8 10.4
24 28.12 7.03 3.12 1.76 67.5 46.9 36.6 26.4 20.8 16.9 253 113 63 40 28.1 21.9 15.8 12.5 10.1
25 27.78 6.86 3.05 1.72 65.9 45.8 35.7 25.8 20.3 16.5 243 110 62 39 27.8 21.4 15.5 12.2 9.9
26 26.81 6.71 2.98 1.68 64.4 44.7 34.8 25.2 19.8 16.1 241 107 60 39 26.8 20.9 15.1 11.9 9.6
27 26.19 6.54 2.91 1.64 62.9 43.7 34.0 24.6 19.4 15.7 236 105 59 38 26.2 20.4 14.7 11.6 9.4
28 25.6 6.40 2.84 1.60 61.4 42.7 33.2 24.0 19.0 15.4 231 102 58 37 25.6 19.9 14.4 11.4 9.2
29 25.04 6.25 2.78 1.56 60.1 41.7 32.5 23.4 18.5 15.0 226 100 56 36 25.0 19.4 14.1 11.1 9.0
30 24.46 6.12 2.72 1.53 58.8 40.8 31.8 22.9 18.1 14.9 221 98 55 35 24.5 19.0 13.8 10.8 8.8
31 23.95 5.98 2.66 1.50 57.5 39.9 31.1 22.4 17.7 14.4 216 96 54 34 23.9 18.6 13.5 10.6 8.6
32 23.44 5.86 2.60 1.47 56.3 39.1 30.4 21.9 17.3 14.1 211 94 53 34 23.4 18.2 13.2 10.4 8.5
33 22.95 5.74 2.55 1.44 55.1 38.3 29.8 21.5 17.0 13.8 206 92 52 33 22.9 17.8 12.9 10.2 8.4
34 22.50 5.62 2.56 1.41 54.0 37.5 29.2 21.1 16.6 13.5 202 90 51 32 22.5 17.5 12.6 10.1 8.1
35 20.01 5.50 2.45 1.38 52.9 36.7 28.6 20.7 16.3 13.2 198 89 50 32 22.0 17.2 12.4 9.8 7.9
Source: IS: 2720 – Part 4 (1975).

327
2/12/2013 7:59:45 AM
Table 10.8 Data and test results of the pipette method
Specific gravity of solids 2.66

Mass of pre-treated soil passing 75-μm sieve taken for analysis (Mb) 25 g
Volume of suspension 500 ml
⎛ 40 ⎞⎟ 500
⎜⎜
⎜⎝ 1000 ⎟⎟⎠ 25 × 500 = 1 g
Mass of dispersing agent in 500 ml suspension (M′s)
Temperature 20°C
Assumed Corresponding time Actual diameter M′i (g) M i′ − M s′

diameter (mm) from Table 10.7 (mm) N′ = × 100%
Mb
0.020 280 seconds 0.02010 18.5 70.0

0.006 51.8 minutes 0.00604 16.2 60.8
0.002 7.77 hours 0.00201 14.3 53.2
0.001 31.10 hours 0.00101 13.1 48.4
4.75 0.075 0.002
100
Percentage finer
80
60
40
20
100 1.0 0.10 0.01 0.001 0.0001

Fig. 10.5 Grain-size distribution curve from pipette method
M10_PURU1773_01_SE_C10.indd 328 2/12/2013 7:59:45 AM

Discussion
This method is not recommended if less than 10% of the material passes the 75 μm IS sieve.
If a constant temperature bath is not available, the test may be performed at room temperature.
Note down the temperature and incorporate necessary corrections in the expression for
determining the diameter of the particle.
10.9 TEST NO. 8: GRAIN-SIZE DISTRIBUTION

BY THE HYDROMETER METHOD
Scope
To determine the grain-size distribution of soil by the hydrometer method
Apparatus
Hydrometer – range 0.995 to 1.03 with an accuracy of 0.0005
Glass measuring cylinders – two of 1,000 ml capacity, 70 mm diameter, and 330 mm height
Thermometer – 0 to 50°C with an accuracy of 5°C
Water bath – maintained at constant temperature
Stirring apparatus
Oven with accurate temperature control in the range from 105 to 110°C
Stopwatch
Desiccator
Evaporating dishes
Wide-mouth conical flask or conical beaker of 1,000 ml capacity
Buchner or Hirch funnel – 100 mm diameter filter flask
Measuring cylinder – 100 ml capacity wash bottle with distilled water
Filter papers
Reagents – hydrogen peroxide, hydrochloric acid, sodium hexametaphosphate
Blue litmus paper
Procedure
(a) Calibration of hydrometer
1. Immerse the hydrometer in a graduated jar and note down the increase in volume as
read on the graduation, or weigh the hydrometer to the nearest 0.1 g and record the mass
in grams as the volume of the hydrometer (Vh) in millilitres.
2. Obtain the cross-sectional area of the jar (Af) by dividing the volume between two
calibration marks by the distance between the same two marks.
3. Record the distance from the lowest calibration mark on the stem of the hydrometer to
each of the major calibration marks [Rh = 1000(rh – 1), where rh is the actual reading on
the hydrometer stem].
4. Record the distance from the neck of the bulb to the nearest calibration mark.
5. Compute the distance H1 corresponding to a reading Rh as the sum of the distances
measured in Steps 3 and 4.
6. Record the distance (h) from the neck to the bottom of the bulb.
7. Compute the effective depth He corresponding to each of the major calibration marks Rh
from the expression
M10_PURU1773_01_SE_C10.indd 329 2/12/2013 7:59:46 AM

Vh
He = He′ −
2 Af
where
h
He′ = H1 +
2
8. Obtain a graphical relationship between He and Rh by plotting a smooth curve between
them. Use this calibration curve for readings beyond 4 minutes. For readings at timings
of 0.5, 1, 2, and 4 minutes, use the expression H′e = H1 + h/2 to obtain the calibration
curve.
9. Insert the hydrometer in a 1,000 ml measuring cylinder with 700 ml water and note the
readings corresponding to the upper and lower limits of the meniscus. Record the differ-
ence between the two readings as the meniscus correction Cm.
(b) Pre-treatment of soil

10. Determine the percentage of soluble solids. If it is more than 1, wash the soil with water
before further treatment.
11. From the air-dried sample passing the 4.75 mm sieve, obtain two samples of mass 50 or
100 g.
12. Use 50 g of soil for a clayey soil and 100 g for a soil with a little sand along with fines.
13. Determine the water content of one sample and place the other in a wide-mouth conical
flask after finding the mass of the soil (Ma) accurately.
14. Add 150 ml of hydrogen peroxide and stir the mixture gently with a glass rod and allow
it to stand overnight.
15. Now gently heat the conical flask carefully to avoid frothing over.
16. Reduce the volume to about 50 ml by boiling after subsidence of the frothing.
17. Cool the mixture and add about 50 ml of hydrochloric acid if the soil contains calcium
compounds. Stir the solution for a few seconds and then allow it to stand for some more
hours. Add more acid if the soil still contains a considerable amount of calcium. Check
the solution for acid reaction to litmus.
18. Omit pre-treatment of the soil if it does not contain calcium compounds or soluble solids.
19. Filter the mixture prepared at Step 17 and wash with warm water until no acid reaction
is noticed.
20. Transfer the damp soil to an evaporating dish of known mass, wash the filter paper and
funnel, and transfer them to the dish.
21. Dry the evaporating dish with its contents in an oven at 105 to 110°C.
22. After drying, cool in a desiccator and find the mass of the soil. Record it as mass of soil
after pre-treatment (Mb).
(c) Dispersion of soil

Add 50 ml of sodium hexametaphosphate solution to the soil and follow the steps given
in pipette method (Test no. 7) for dispersion.
M10_PURU1773_01_SE_C10.indd 330 2/12/2013 7:59:46 AM

(d) Sedimentation analysis

23. Transfer to a 1,000 ml measuring cylinder the suspension that has passed through the
sieve and add distilled water to make the volume of the suspension exactly 1,000 ml.
24. Add 50 ml of sodium hexametaphosphate solution in a 1,000 ml measuring cylinder, add
distilled water to make 1,000 ml, and maintain it at the same temperature as that of the
soil suspension. This cylinder has to be used when the hydrometer is not in use. The com-
bined correction C of temperature and dispersing agent is taken from this cylinder.
25. Close the mouth of the measuring cylinder containing the suspension, shake vigorously,
and finally invert end to end.
26. After shaking, immediately make the measuring cylinder stand in the constant tempera-
ture bath (if used) and start a stopwatch.
27. Insert the hydrometer, take readings after periods of 0.5, 1, 2, and 4 minutes, and there-
after rinse it and place it in the distilled water cylinder (prepared in Step 24).
28. Re-insert the hydrometer and take readings at intervals of 8, 15, and 30 minutes and
1, 2, and 4 hours after starting the test. Remove the hydrometer, rinse it and place it in
the distilled water cylinder after every reading. After 4 hours take hydrometer readings
once or twice within 24 hours. Take one reading at the end of 24 hours.
29. Record the temperature once during the first 15 minutes and then after every subse-
quent reading. Take hydrometer readings in the distilled water cylinder (prepared in
Step 24) corresponding to these temperatures and calculate the combined correction, C.
The correction for temperature is positive for temperatures greater than the calibrated
temperature of the hydrometer. The dispersion agent correction is negative. The com-
bined correction is negative for the range of laboratory temperatures.
Computations
1. Compute the loss in mass in pre-treatment (as in Test No. 7).
2. The diameter of the particles in the suspension, at any time t, is given as
30ηw He
D=
980 (ρs − ρw ) t
where t is the time elapsed between the beginning of sedimentation and the taking of the
hydrometer reading, in minutes.
The hydrometer reading corrected for the meniscus is given as
Rh = Rh′ + Cm
where R′h is the hydrometer reading at the upper rim of the meniscus.
3. The percentage finer N′, based on the mass Mb, is given as
100 G
N′ = Rh
M b (G − 1)
where Rh = Rh + C .
M10_PURU1773_01_SE_C10.indd 331 2/12/2013 7:59:47 AM

Results
The results of the grain-size analysis are presented in a graph by taking the diameter of the
particle in a log scale and the percentage finer in an arithmetic scale. Details of calibration
data and calibration curve are shown in Table 10.9 and Fig. 10.6, respectively. Test data and
results are given in Table 10.10, and the graduation curve is presented in Fig. 10.7.
Discussion
As per Indian Standards, this method is considered as a subsidiary method. It is not
recommended if less than 10% of the material passes the 75 μm IS sieve.
The hydrometer and pipette methods give fairly accurate results, but both are time-
consuming. A new device called the plummet balance is in use in different laboratories
(Malhotra and Chandra, 1982). A plummet balance is nothing but a specific gravity balance.
It is based on the principle that the depth of immersion of a plummet in a suspension is
Table 10.9 Data for calibration of hydrometer

Volume of hydrometer (Vh) = 102.3 ml
Area of cross-section of jar (Aj) = 42.3 cm2
Height of bulb (h) = 11.8 cm
Sl. no. Hydrometer Rh = 1000(rh – 1) H1 H′e = H1 + h/2 H e = H e′ −(Vh / 2 Aj )

reading
1. 0.995 –5.00 8.90 14.80 13.59

2. 1.000 0.00 6.45 12.35 11.14
3. 1.005 5.00 4.05 9.95 8.74
4. 1.010 10.00 1.65 7.55 6.34
22
No immersion 20
correction
18
He or He'
16
14
He' = H1 + h/2
With immersion 10
correction
8
Vh
6 He = He' –
2Aj
4
–10 –8 –6 –4 –2 0 2 4 6 8 10
Rh = 1000 (rh – 1)
Fig. 10.6 Hydrometer calibration curve
M10_PURU1773_01_SE_C10.indd 332 2/12/2013 7:59:47 AM

M10_PURU1773_01_SE_C10.indd 333
Table 10.10 Data and test results from the hydrometer test
Mass of pre-treated soil passing 75 μm sieve taken for analysis (mb) = 20 g
Specific gravity of soil particles of minus 75 μm (G) = 2.65
Meniscus correction (Cm) = +0.05
Lapsed time Hydrometer R′ = 1000 (r′h – 1) Temperature Rh = R′h + Cm C H′e or He D (mm) Rh = rh + C Percent
(min) reading (r′h) (°C) (cm) finer, N′
Laboratory Measurement of Soil Properties
0.25 1.00925 9.25 27.5 9.75 –0.55 7.6 0.0686 9.20 73.9
0.50 1.00875 8.75 27.5 9.25 –0.55 7.8 0.0491 8.70 69.9
1.00 1.00725 7.25 27.5 7.75 –0.55 8.4 0.0361 7.20 57.8
2.00 1.00600 6.00 27.5 6.50 –0.55 9.0 0.0264 5.95 47.8
4.00 1.00500 5.00 27.5 5.50 –0.55 9.6 0.0193 4.95 39.8
8.00 1.00300 3.00 27.5 3.50 –0.55 10.2 0.0103 2.95 23.7
15.00 1.00150 1.50 27.5 2.00 –0.55 10.2 0.0103 1.45 10.0
30.00 1.00100 1.00 28.0 1.50 –0.55 10.4 0.0073 0.95 7.6
60.00 1.00075 0.75 28.0 1.25 –0.55 10.5 0.0052 0.70 5.6
75.00 1.00050 0.50 28.0 1.00 –0.55 10.7 0.0018 0.45 3.6
140.00 1.00025 0.25 27.5 0.75 –0.55 10.8 0.0011 0.20 1.6
333
2/12/2013 7:59:47 AM
4.75 0.075 0.002
100
Percentage finer
80
60
40
20
10.0 1.0 0.1 0.01 0.001 0.0001

Fig. 10.7 Grain size distribution curve from hydrometer method
countered by the movement of a needle-shaped beam on a graduated scale, and the reading
of the beam on the scale represents the percentage fraction of a particular size in a given time
of fall (Marshall, 1956). Malhotra and Chandra (1982) used this apparatus on six different
fine-grained soils and found it more suitable in clayey soils. This method has the advantage
of being quick, but it is yet to be standardized.
10.10 TEST NO. 9: LIQUID LIMIT OF SOIL

Scope
To determine the liquid limit of a soil using a mechanical liquid limit device
Apparatus
Mechanical liquid limit device – as illustrated in Fig. 10.8a
Grooving tools – as illustrated in Fig. 10.8b
Evaporating dish
Spatula
Palette knives
Wash bottle or beaker
M10_PURU1773_01_SE_C10.indd 334 2/12/2013 7:59:48 AM

Clamping nuts
Screw adjustment between
10 mm clear with cup in cam and follower
rigid position Brass cup
28 mm 45 mm
51 mm Rubber Base 51 mm
150 mm 125 mm
20
15 (a) Liquid limit apparatus
50
75
50
8
20 30 All dimensions in mm
R
22
11
11 10 53
40 59
Type A Type B Type C (b) Grooving tools
Fig. 10.8 Liquid limit device and tool (Source: IS: 2720 – Part 5, 1970)
Procedure
1. Clean and check the liquid limit device to see that it is in working order. Also clean the
grooving tools.
2. Use a gauge or the handle of the grooving tool and the adjusting plate of the liquid
limit device such that the cup falls exactly 10 mm for one revolution of the handle.
After adjustment, secure the plate by tightening the screw.
3. Weigh about 120 g of the soil sample passing the 425 μm IS sieve and transfer it to an
evaporating dish or on to the flat glass plate.
4. Mix the soil with distilled water to form an uniform paste.
5. Take a portion of the paste in the cup (of the liquid limit device) above the spot where
the cup rests on the base, squeeze down, spread into position, and level to a depth of
10 mm at the point of maximum thickness.
6. Divide the soil in the cup by firmly running the grooving tool (Type A) diametrically
such that a sharp groove is formed. Use Type B or C grooving tools for non-adhesive
soils.
7. Turn the crank at the rate of two revolutions per second until the two parts of the soil
come in contact with the bottom of the groove along a distance of about 12 mm and
record the number of drops needed.
8. Add a small quantity of soil from the evaporating dish, mix it thoroughly, and repeat
Steps 6 and 7 until two consecutive runs give the same number of drops for closure of
the groove.
9. Take a representative slice of the soil sample, about the width of the spatula, at right
angles to the groove, including that portion of the groove in which the soil flowed
together, for water content determination.
M10_PURU1773_01_SE_C10.indd 335 2/12/2013 7:59:48 AM

10. Transfer the remaining soil to the evaporation dish and add more water or soil to change
the consistency of the paste. Repeat Steps 5 to 9. Repeat the test 4 to 5 times and record
the number of drops required to close the groove in the range of 15 to 35.
Computations and results
Plot the number of drops on a logarithmic scale and water content on an arithmetical scale
and join them by a straight line. Such a curve is called a flow curve. Read the moisture content
corresponding to 25 drops from the curve and report it to the nearest whole number as the
liquid limit (wL of the soil).
Extend the flow curve on either side and find the slope of the line as the difference in
water content at N2 and N1 drops and report it as the flow index If, thus
w1 − w2
If =
log10 ( N 2 / N1 )
where w1 is the water content corresponding to N1 drops and w2 the water content corre-
sponding to N2 drops.
Typical test data and results are shown in Table 10.11 and Fig. 10.9.
Discussion
In general, natural soils used for liquid and plastic limit tests (given elsewhere) should not
be oven dried. Drying causes the particles to sub-divide and also causes the removal of
absorbed water. It is reported (Lambe, 1951) that oven-dried organic soils tend to show a
lower wL value than those of soils that have not been dried.
Thus, sometimes natural soils are directly used for a test without oven drying if all the
particles are less than 425 μm in size. In case some stones are present, the wet soil is rubbed
through the 425 μm IS sieve till a sufficient quantity of soil is collected to run the test.
A soil with a low clay content has to be tested immediately after thorough mixing with
water. In case of tearing of the sides of the groove or slipping of the soil, the groove may be
cut in stages. Instead of flowing, some soils tend to slide; in such cases discard the result and
report that the liquid limit could not be obtained (IS: 2720 – Part 5, 1970).
Another method of finding the liquid limit is the use of the cone penetrometer method,
which was discussed in Chapter 2. The mechanical liquid limit device has been recognized as
a routine test. The cone penetrometer has been reported to have more advantages compared
with the mechanical device (IS: 2720 – Part 5, 1970); yet it has not been accepted as a routine test.
Table 10.11 Test data and results of liquid limit test
Determination no. 1 2 3 4 5
Number of drops 16 19 22 27 31
Moisture cup no. 36 63 54 90 81
Mass of cup (g) 17.33 17.82 15.06 17.40 15.71
Mass of cup with wet soil (g) 23.40 24.83 21.13 22.96 22.40
Mass of cup with dry soil (g) 21.45 22.67 19.29 21.38 20.62
Water content (%) 47.33 44.54 43.50 39.70 36.25
Liquid limit from plot (%) 4
Flow index from plot 32.6
M10_PURU1773_01_SE_C10.indd 336 2/12/2013 7:59:48 AM

48
40–36
If = = 32.6%
log 35
10
46
Water content, % 44
42
40
38
36
1 10 100
Number of drops
Fig. 10.9 Liquid limit flow curve
10.11 TEST NO. 10: PLASTIC LIMIT OF SOIL

Scope
To determine the plastic limit of a soil
Apparatus
Evaporation dish or flat glass plate
Palette knife or spatula
Surface for rolling-ground – glass plate about 200 mm × 150 mm
Rod – 3 mm in diameter and about 100 mm long
Procedure
1. Take about 20 g of soil passing through the 425 μm IS sieve in an evaporating dish or glass
plate. Add distilled water and thoroughly mix such that the soil mass becomes plastic
enough to be easily moulded with the fingers. For clayey soils, allow sufficient time for
moisture equilibrium.
2. Take about 8 g of this wet soil, make a ball out of it, and roll it on the glass plate with
the palm of the hand to form a thread of uniform diameter. Continue the rolling till the
thread is of 3 mm diameter. Knead the soil together to form a uniform mass and roll
again. Continue the process of rolling and kneading until the thread just crumbles at
3 mm diameter.
M10_PURU1773_01_SE_C10.indd 337 2/12/2013 7:59:48 AM

Table 10.12 Test data and results of plastic limit
Determination no. 1 2 3
Moisture cup no. 117 171 126

Mass of moisture cup (g) 15.82 16.17 16.26
Moisture cup with wet soil thread (g) 30.54 30.42 30.41
Moisture cup with dried soil (g) 27.95 27.88 27.99
Water content (%) 21.35 20.8 20.63
Average plastic limit (%) 20.93
3. Repeat Steps 1 to 3 to obtain two more determinations of the plastic limit. Also, find the
natural water content of the soil (wn).
Computations and results
The mean water content obtained from three trials is the plastic limit of the soil (wp). The
indices (as discussed in Chapter 2) are calculated.
Discussion
For sandy soils, first determine the plastic limit; if it cannot be determined, report the
plasticity index as Np (non-plastic). When the plastic limit is equal to or greater than the
liquid limit, the plasticity index is reported as zero.
Typical data and results of a plastic limit test are shown in Table 10.12.
10.12 TEST NO. 11: SHRINKAGE FACTORS OF SOIL

Scope
To determine the shrinkage limit, shrinkage ratio, shrinkage index, and volumetric shrink-
age of soils
Apparatus
Evaporating dishes
Spatula
Shrinkage dish – 45 mm diameter and 15 mm height
Straight edge – 150 mm in length
Glass plates, plain and with metal prongs – 75 mm × 75 mm, 3 mm thick
Glass cup – 50 to 55 mm in diameter and 25 mm in height
Sieve – 425 μm IS sieve
Mercury
Desiccator – with any desiccating agent other than sulphuric acid
Procedure
1. Clean the shrinkage dish and weigh it (M1). Fill the dish with mercury. Remove the excess
mercury by pressing a glass plate over the top of the dish. Weigh the mercury, divide it by
the density of mercury, and obtain the volume of the dish which is the volume of the wet
soil (V).
M10_PURU1773_01_SE_C10.indd 338 2/12/2013 7:59:48 AM

2. Take about 30 g of soil, passing the 425 μm IS sieve, in an evaporating dish and thoroughly
mix with water of an amount slightly greater than the liquid limit. The soil–water mixture
should be capable of flowing if allowed to drop.
3. Coat the inside of the shrinkage dish with a thin layer of grease and fill one-third of the
dish with soil–water mixture. Tap the dish on a firm surface, cushioned by several layers
of blotting paper, rubber sheet, or similar material. In three operations, completely fill the
dish. Strike off the excess soil paste with a straight edge, clean the outside surface, and
weigh (M2).
4. Dry the soil pat in air until the colour changes from dark to light, and then dry it in a
temperature-controlled oven. After drying, cool it in air and weigh the shrinkage dish and
dry the soil pat (M3).
5. Fill the glass cup with mercury and remove the excess mercury by pressing the glass plate
with three prongs firmly over the top of the cup.
6. Place the glass cup with mercury in a large evaporating dish and place the dry soil pat on
the surface of the mercury.
7. Force the soil pat under the mercury carefully by means of the glass plate with the prongs,
so that the soil pat is completely submerged in mercury (Fig. 10.10). Collect the displaced
mercury, weigh it, and find its volume, which is the volume of the dry soil pat (V0).
Computations
Moisture content of wet soil pat
M − M0
w= ×100
M0
where
M = M2 − M1
M0 = M3 − M1
Shrinkage limit (re-moulded soil)
⎡ ⎛ V − V0 ⎞⎟ ⎤
ws = ⎢⎢ w − ⎜⎜⎜ ⎟⎟×100⎥ %
⎥
⎢⎣ ⎜⎝ M0 ⎟⎠ ⎥⎦
Wet soil Dry soil

Shrinkage
dish Shrinkage dish
Glass plate with prongs
Mercury
Before After
shrinkage shrinkage Glass cup
Evaporating dish
Top of glass cup Mercury displaced

Dry soil by soil pat
ground surface pat
Fig. 10.10 Liner shrinkage mould (Source: IS: 2720 – Part 20, 1966)
M10_PURU1773_01_SE_C10.indd 339 2/12/2013 7:59:49 AM

Shrinkage index
I S = I p − ws
Shrinkage ratio
M0
R=
V0
Volumetric shrinkage (or volume change)
Vs = (w1 − ws )R
where w1 is the given moisture content in percent.
Shrinkage limit of re-moulded soil when the specific gravity is known,
⎛1 1⎞
ws = ⎜⎜⎜ − ⎟⎟⎟×100
⎝R G⎠
Results
Shrinkage limit (re-moulded soil) tests are repeated at least three times, and the average
value is reported, and if any test shows a variation of 2% against the mean, the test is
repeated. The shrinkage limit for a typical re-moulded soil is given in Table 10.13.
Discussion
In order to determine the shrinkage limit of undisturbed soils, prepare a wet soil pat of
dimensions 45 mm diameter and 15 mm height and round off its edges to prevent the
entrapment of air during mercury displacement (IS: 2720 – Part 6, 1972). Air-dry and then
Table 10.13 Data and results of shrinkage limit test
Mass of shrinkage dish (M1) (g) 7.56 7.75 7.67

Mass of shrinkage dish with wet soil pat (M2) (g) 78.12 78.01 78.16
Mass of shrinkage dish with dry soil pat (M3) (g) 57.30 55.70 56.35
Mass of dry soil pat M0 = (M3 – M1) (g) 49.74 47.95 48.68
Mass of wet soil M = (M2 – M1) (g) 70.56 70.26 70.49
Water content (%, wt.) 41.85 46.53 44.80
Mass of shrinkage cup with mercury (g) 537.10 548.60 540.20
Mass of mercury only (g) 529.54 540.85 532.53
Volume of shrinkage dish = volume of wet soil (V) ml 38.94 39.77 39.16
Mass of displaced mercury (g) 316.10 313.20 315.50
Volume of dry soil pat (V0) (ml) 23.24 23.03 23.20
⎡ ⎛ V − V0 ⎞⎟ ⎤
Shrinkage limit (re-moulded soil) ws = ⎢⎢ w − ⎜⎜⎜ ⎟⎟× 100⎥ (%)
⎥ 10.29 11.62 12.01
⎣⎢ ⎝⎜ M0 ⎠⎟ ⎦⎥
Average shrinkage limit = 11.31%
M10_PURU1773_01_SE_C10.indd 340 2/12/2013 7:59:50 AM

oven-dry the pat; cool it and weigh it (M0s). Using the mercury displacement procedure, find
the volume of the pat (V0s). Then shrinkage limit of undisturbed soil,
⎛v 1⎞
wsu = ⎜⎜⎜ 0s − ⎟⎟⎟×100%
⎜⎝ M0s G ⎟⎠
The shrinkage limit test is useful in obtaining a quantitative indication of how much
volume change can occur with changes in the water content.
10.13 TEST NO. 12: LINEAR SHRINKAGE OF SOIL

Scope
To determine the linear shrinkage of re-moulded soil
Apparatus
Two palette knives
Flat glass plate or evaporating dish
Cast iron or brass mould (as in Fig. 10.11)
Oven with accurate temperature control in the ranges 60 to 65°C and 105 to 110°C
Callipers
Silicone or any grease
Procedure
1. Thoroughly clean the mould and measure its length to get the initial length of the
specimen (L1).
2. Apply a thin layer of grease to the inner walls of the mould so as to prevent the soil from
adhering to the sides of the mould.
3. Take 150 g of the soil sample passing through a 425 μm sieve in an evaporating dish.
4. Add sufficient quantity of distilled water (as a rough measure, this may be about 2%
above the limit of the soil) and thoroughly mix. Set it aside for 24 hours for moisture
equilibrium to be attained.
5. Place the thoroughly mixed soil in the mould such that it is slightly above the sides of the
mould.
6. Remove the entrapped air bubbles by gently tapping the mould on a soft pad.
25 mm
40 mm
125 mm
140 mm 12.5 mm
Plan
20 mm
Elevation End view
Fig. 10.11 Liner shrinkage mould (Source: IS: 2720 – Part 20, 1966)
M10_PURU1773_01_SE_C10.indd 341 2/12/2013 7:59:51 AM

Table 10.14 Linear shrinkage and result of data
Initial length of specimen (Li) (mm) 140 140 140

Length of oven dry specimen (Ld) (mm) 121.12 120.8 121.4
Linear shrinkage = [1 − (Ld / Li )]× 100% 13.43 13.71 13.29
Average linear shrinkage = 13.48%
7. Level the soil along the top mould with the palette knife.
8. Dry the mould in three stages, viz., in air, in the oven at a controlled temperature of 60 to
65°C and finally in the oven at a controlled temperature of 105 to 110°C. The time needed
to dry during each stage depends on the type of soil. However, as a general guide,
24 hours may be allowed during each stage.
9. Remove the mould from the oven, cool, and measure the mean length of the soil bar (Ld);
if the specimen has curved, measure along the mean arc.
10. Repeat the test for two more specimens.
Computations
The linear shrinkage of the soil is given as
⎛ Length of oven − dry specimen ⎞⎟
⎜⎜1− ⎟⎟×100%
⎜⎜⎝ Initial length of the specimen ⎟⎠
Results
The liner shrinkage of the soil is represented as a percentage to the nearest whole number.
Test data and results are shown in Table 10.14 for a typical case.
Discussion
Soil of low plasticity may not show cracks when subjected to rapid drying and in such soils
the drying may be done directly at 110°C. For a highly colloidal clay, the drying process may
have to be slowed down to prevent cracking. In soils of varying particles size, segregation of
larger particles to the bottom of the mould may be avoided by reducing the soil–water
wetness (IS: 2720 – Part 20, 1966).
10.14 TEST NO. 13: PERMEABILITY TEST

Scope
To determine the permeability of a given soil using a falling or constant head permeameter
Apparatus
Permeameter mould – 1,000 ml capacity, 100 mm diameter, and 12.73 mm height
Compaction equipment – suitable dynamic or static compaction equipment
Drainage base – with porous disc of 12 mm thickness and dummy plate of 12 mm thickness
to suit the mould, provided with water inlet/outlet connection
Drainage cap – with porous disc, 12 mm thick, and water inlet/outlet connection to constant
head tank
M10_PURU1773_01_SE_C10.indd 342 2/12/2013 7:59:51 AM

Water inlet
Stand
Overflow
Glass stand pipes
Scales
Valves
Permeameter
mould
Porous
Soil sample
stone
Fig. 10.12 Permeability test set-up
Set of stand pipes – glass stand pipes of diameter 5 to 20 mm, suitably mounted on stand
(Fig. 10.12)
Miscellaneous apparatus – IS sieves, mixing tray, graduated cylinder, metric scale, stopwatch,
75-gauge wire, thermometer, and source of water
Procedure
1. Take 2.5 kg of sample (as suggested in the standard compaction test) and the desired water
content (may be field water content or optimum moisture content depending on the dry
density requirement), spread uniformly, and allow moisture equilibrium to be attained.
2. Weigh the empty permeameter. Attach the extension collar, grease the inside of the
mould and collar, and keep the assembly on a firm base.
3. Choose the type of compaction and compactive effort to suit the field condition, and
complete the compaction process.
4. Remove the collar, level the soil, detach the base plate, and weigh.
5. Assemble the mould, drainage base, and cap along with porous discs (saturate the
porous discs before use).
6. Saturate the specimen, by allowing water to flow with a sufficient head through it or by
immersion for a high-permeability specimen and by subjecting it to a high head (for a
day or two) for permeable specimen.
(a) Falling head test
7. Connect the specimen through the top inlet to a selected stand pipe of inside area (a).
Open the bottom outlet and note down the interval (t) required for the water level to fall
from the initial head (h1) to a known final head (h2), the heads being measured above the
centre of the outlet.
8. Fill the reservoir again to a higher h1 and note the time taken for the water level to fall to
h1 h2 and then to h2 again.
M10_PURU1773_01_SE_C10.indd 343 2/12/2013 7:59:51 AM

9. The time taken to fall from h1 to h1 h2 and then from h1 h2 to h2 should be the same.
Otherwise, repeat the test after re-filling the stand pipe.
10. Report the test and take three observations.
(b) Constant head test
11. Connect the specimen through the top inlet to the constant head water reservoir. Open
the bottom outlet and ascertain that the flow has attained a steady state.
12. Collect the quantity of flow for a convenient time interval (t) and repeat this for the same
time interval thrice.
13. Find the mass of wet soil in the mould.
14. Keep samples for water content determination.
Computations
The coefficient of permeability (k) for
1. The falling head test,
2.303 aL h
k= log10 1
At h2
2. The constant head test,

qL
k=
Ah
The permeability at 27°C is given as
ηωT
k 27° C = kT
ηω 27° C
Results
The coefficient of permeability is reported in mm/s or m/s at 27°C. Typical test results are
given in Tables 10.15 and 10.16 for falling and constant head permeabilities, respectively. The
void ratio, degree of saturation, and dry density are presented in Table 10.17.
Table 10.15 Data and test results of falling head test

Length of specimen (L) = 127 mm
Area of specimen (A) = 7,854 mm2
Volume of specimen (V) = 9,97,458 mm3
Area of stand pipe (a) = 113 mm2
Specific gravity of soil (G) = 2.65
Temperature of water = 30°C
Sl. no. Initial head, h1 Final head, h2 Time, t log10(h1/h2) kT (mm/s) k27 (mm/s)
(mm) (mm) (seconds)
1. 1,200 550 122 0.339 0.0117 0.0110

2. 1,200 400 173 0.477 0.0116 0.0109
3. 1,200 250 244 0.681 0.0118 0.0110
Average: 0.0117 0.0110
M10_PURU1773_01_SE_C10.indd 344 2/12/2013 7:59:52 AM

Table 10.16 Data and test results of constant head test

Length of specimen (L) = 127 mm
Area of specimen (A) = 7,854 mm2
Volume of specimen (V) = 9,97,458 mm3
Specific gravity of soil (G) = 2.65
Temperature of water = 30°C
Sl. no. Time, t (s) Head, h Hydraulic Quantity, Q3 q = Q/t kT (mm/s) k27 (mm/s)
(mm) gradient, h/L (mm3) (mm3/s)
1. 150 300 2.36 32,400 214.8 0.0116 0.0109

2. 300 300 2.36 66,000 217.0 0.0117 0.0110
3. 450 300 2.36 96,300 216.5 0.0117 0.0110
Average: 0.0117 0.0110
Table 10.17 Void ratio, degree of saturation, dry density

(same for both tests)
Mass of saturated soil (M) = 2,087 g

Mass of moisture cup = 18.3 g
Mass of cup with wet soil = 37.09 g
Mass of cup with dry soil = 34.10 g
Water content = 18.92%
⎛ M ⎞⎟
Mass of dry soil in the mould ⎜⎜ Ms ⎟g
⎜⎝ 1 + ( w / 100 ) ⎟⎠
Ms
ρd = = 1.76 g / cc
V
G ρw
e= − 1 = 0.51
ρd
ωG
Sr = = 98.3%
e
Discussion
Permeability tests can also be conducted on undisturbed specimens. Prepare carefully a
specimen 85 mm in diameter and 127 mm in height to suit the permeameter. Place the spec-
imen centrally over the porous disc and fill the annular gap with a cement slurry or benton-
ite sand mix in the ratio 1:9. Fix the drainage cap. Now a falling or constant head test may be
conducted, depending on the type of soil.
The constant head test is usually preferred for sandy soils and the variable head test for
silty and clayey soils. A separate constant head method for granular soils has been recom-
mended by Indian Standards (IS: 2720 – Part 36, 1975). This method is suitable for disturbed
granular soils containing less than 10% soil passing 75 μm IS sieve. This range of particle
sizes is used for construction of embankments and base courses under pavements. Granular
soils with a particle size up to 20 mm can be tested using this method under laminar flow
conditions.
M10_PURU1773_01_SE_C10.indd 345 2/12/2013 7:59:53 AM

Although these two laboratory methods are routinely used in various laboratories, they
do not provide a reliable value for the following reasons:
1. A soil specimen in the laboratory is always disturbed to some extent and does not exist in
the same state as in the field.
2. A laboratory specimen does not simulate the orientation of an in situ stratum to the flow
of water.
3. Boundary conditions are not the same as simulated in the laboratory, e.g., smooth walls of
the mould do not exist in the field.
4. There is a difference between the field and laboratory hydraulic gradients.
5. Complete saturation conditions are not possible in a laboratory sample, and the effect of
entrap-ped air bubbles on the coefficient of permeability may be severe.
6. Prediction of the behaviour of a large formation in situ from the test results for a small
sample is highly unreliable.
It is apparent that the laboratory determined k is not representative and is hence not reli-
able. But tests on undisturbed samples might improve this situation.
10.15 TEST NO. 14: FREE SWELL INDEX OF SOILS

Scope
To determine the free swell index of a soil
Apparatus
Graduated glass cylinders of 100 ml capacity
Sieve – 425 μm IS sieve
Procedure
1. Take 10 g of oven-dry soil passing through the 425 μm sieve and pour it into a 100 ml
graduated jar. Similarly, prepare another cylinder with the same weight of soil.
2. Fill one with kerosene oil and the other with distilled water up to the 100 ml mark.
3. Remove the entrapped air from both the cylinders by shaking and/or stirring with a glass rod.
4. Allow both the cylinders to settle down for 24 hours.
5. Read out the level of the soil in the kerosene-filled graduated jar (Vk). Kerosene, being a
non-polar liquid, does not cause swelling of the soil.
6. Also read out the level of soil in the distilled water-filled graduated jar (Vd).
Computations
The free swell index of the soil can be calculated from the expression
Vd − Vk
Free swell index = ×100%
Vk
Results
The free swell index is expressed as a percentage to two significant figures. Typical data and
results are given in Table 10.18.
M10_PURU1773_01_SE_C10.indd 346 2/12/2013 7:59:53 AM

Table 10.18 Data and results from a free swell index test
Volume of soil sample read from kerosene-filled graduated cylinder (Vk) 43 ml

Volume of soil sample read from distilled water–filled graduated jar (Vd) 56 ml
Vd − Vk
Free swell index × 100 = 30%
Vk
Discussion
To get accurate results for highly swelling soils, the quantity of sample taken may be reduced
to 5 g or the volume of cylinder may be increased to 250 ml (IS: 2720 – Part 40, 1977).
10.16 TEST NO. 15: MOISTURE CONTENT – DRY DENSITY RELATIONSHIP

(STANDARD PROCTOR COMPACTION TEST)
Scope
To determine the relation between the moisture content and the dry density of a soil
Apparatus
Cylindrical metal mould – 1,000 ml diameter, with detachable base, and height of 127.3 mm
with extension collar (Fig. 10.13)
Metal rammer – 2.6 kg with 310 mm fall (Fig. 10.14)
Sample extruder
120 mm Removable
extension
Three lugs
brazed on
60 mm
Approx.
10 mm
5 mm 10 mm
Three pins to form
Catch for Extension
127.3 mm
Two lugs
brazed on
Detachable
base plate
15
10 mm
150 mm
180 mm
Fig. 10.13 Mould for compaction (Source: IS: 2720 – Part 7, 1974)
M10_PURU1773_01_SE_C10.indd 347 2/12/2013 7:59:54 AM

65
27 20
6 ,4
holes
Guide Rammer adjusted
length to have a total
of travel 335 weight of 2.6 kg
of
361.5 rammer
310 mm 25
60
12 Holes
1.5 thick
25 rubber gasket
25 50
13 All dimensions
52 in mm
60 50
(a) Sleeve (b) Metal rammer
Fig. 10.14 Metal rammer and sleeve (Source: IS: 2720 – Part 7, 1974;)

Balance of 10 kg capacity with 1 g sensitivity
Steel straight edge
Sieves – 50 mm, 20 mm, and 4.75 mm IS sieves
Mixing tools
Procedure
1. Weigh about 25 kg of air-dried soil passing through the 50 mm IS sieve. Sieve the soil
through the 20 mm and 4.75 mm sieves and find both the fraction passing and that
retained in the each sieve. Reject the fraction retained on the 20 mm sieve.
2. From the soil passing the 20 mm IS sieve, find the ratio of the soil fraction retained on the
4.75 mm IS sieve to the soil fraction passing the 4.75 mm sieve.
3. If the fraction retained on the 4.75 mm IS sieve is more than 20%, maintain the ratio of this
material to the material passing the 4.75 mm IS sieve. Take about 20 kg of the material in
the calculated proportion, as mentioned above. If the fraction retained on the 4.75 mm IS
sieve is less than 20%, then directly take about 20 kg of soil passing the 20 mm IS sieve.
4. Add enough water to bring its moisture content to about 7% (for sandy soils) or 10% (for
clayey soils) less than the estimated optimum moisture content. Keep the processed soil
in an airtight container for about 18 hours for moisture equilibrium.
5. Clean and dry the empty mould, measure its dimensions, and weigh it to the nearest
gram (Mm). Fit in the base plate and the extension collar.
6. Divide the processed soil – water mix into eight equal parts.
7. Take one part (about 2.5 kg) of the processed soil and compact it into the mould in three
equal layers, each layer, being given 25 blows to be distributed uniformly. Score each
layer with a spatula before putting in the soil for the succeeding layer.
M10_PURU1773_01_SE_C10.indd 348 2/12/2013 7:59:54 AM

8. Remove the collar and carefully level off to the top of the mould by means of a straight
edge. Weigh the mould (M).
9. Eject the soil from the mould, cut at the middle, and take representative samples for
water content determination.
10. Repeat Steps 7 to 9 for 5 or 6 samples, using a fresh part of the soil specimen each time,
after adding a higher water content than in the preceding specimen, so that at least two
readings, one below and above the optimum moisture content, are available.
Computations
Compute the volume (Vm) of the mould from its height and diameter.
⎛ M − Mm ⎞⎟
Bulk density ρ = ⎜⎜ ⎟ g / cc
⎜⎝ Vm ⎟⎟⎠
⎛ ρ ⎞⎟
Dry density ρd = ⎜⎜
⎜⎝ 1 + w/ 100 ⎟⎟⎠
g / cc
⎛ Gρ ⎞
Void ratio e = ⎜⎜⎜ w − 1⎟⎟⎟
⎜⎝ ρd ⎟⎠
⎛ ρ ⎞
Porosity n = ⎜⎜⎜1 − d ⎟⎟⎟×100%
⎜⎝ Gρw ⎟⎠
Results
Plot a curve of water content versus dry density. The dry density (rounded to two decimal
places) corresponding to the maximum point of the curve and the corresponding mois-
ture content (rounded to the first decimal place) shall be reported as the maximum dry
density (ρd max) and the optimum moisture content (OMC), respectively. A typical test
results on a soil is presented in Table 10.19. The moisture content–dry density curve is
plotted in Fig. 10.15.
Discussion
Instead of a 1,000 ml capacity mould, the Indian Standards (IS: 2720 – Part 7, 1974) also rec-
ommend a 2,250 ml mould to be used; in that case, for each layer 56 blows are given with the
standard hammer. The Standard/Proctor Test is also termed a light compaction test.
As the material retained on the 20 mm IS sieve has been rejected for the test, a correction
is applied to get the corrected maximum dry density and OMC.
ρ0 ×ρd max
Corrected maximum dry density =
n1ρd max + n2 ρ0
Corrected OMC = n1A0 + n2w0
where ρ0 is the density of over-size particles (i.e., G0ρw, where G0 is the specific gravity of
the over-size particles), ρd max the maximum dry density obtained in the test, in g/cc, n1
the fraction by weight of over-size particles in the total soil expressed as a ratio, n2 the
fraction by weight of portion passing 20 mm IS sieve (or 4.75 mm IS sieve) expressed as a
fraction of the total soil, A0 the water absorption capacity of over-size material, if any,
M10_PURU1773_01_SE_C10.indd 349 2/12/2013 7:59:55 AM

Table 10.19 Data and test results from compaction test

Type of test: Standard/Proctor test
Volume of mould (Vm) = 1,000 ml
Mass of the rammer = 2.6 kg
Fall of the rammer = 310 mm
Specific gravity = 2.65
Percentage of material:
(i) Retained on 20 mm IS sieve =5
(ii) Passing 20 mm IS sieve and retained on 4.75 mm IS sieve = 0
(iii) Passing 4.75 mm IS sieve = 90
Ratio of (ii) to (iii) = 1:9
Determination no. 1 2 3 4 5 6
Mass of mould (M) (g) 6,245 6,245 6,245 6,245 6,245 6,245
Mass of mould with 8,130 8,211 8,260 8,279 8,268 8,220
compacted soil (Mm) (g)
Mass of compacted soil (Mm – M) (g) 1,885 1,966 2,015 2,034 2,023 1,955
Wet density ρ = ( Mm − M ) / Vm 1.885 1.966 2.015 2.034 2.023 1.995
Moisture cup no. 9 18 27 45 54 36
Mass of cup and wet soil (g) 41.20 37.12 40.47 40.35 39.46 40.48
Mass of cup and dry soil (g) 39.35 35.18 38.18 37.89 36.82 37.71
Mass of cup (g) 22.67 20.74 22.84 22.65 21.84 22.37
ρ 1.697 1.733 1.753 1.751 1.720 1.690
Dry density ρd = ( g / cc )
1 + w / 100
⎛ Gρ ⎞ 0.562 0.529 0.512 0.513 0.541 0.568
Void ratio e = ⎜⎜⎜ w − 1⎟⎟⎟
⎜⎝ ρd ⎠⎟
⎛ ρ ⎞
Porosity n = ⎜⎜⎜1 − d ⎟⎟⎟× 100 36.0 34.6 33.9 33.9 35.1 36.2
⎜⎝ Gρw ⎟⎠
From the plot ρd max = 1.755 g/cc and OMC = 15.45%.
expressed as the percentage of water absorbed, and W0 the OMC obtained in the test in
per cent.
This formula is based on the assumption that the volume of the compacted portion
passing the 20 mm IS sieve (or 4.75 mm IS sieve) is sufficient to fill the voids between the
over-size particles (IS: 2720 –Part 7, 1974).
With field compacting equipment becoming heavier and more efficient, it has become
necessary to increase the amount of compacting energy in the laboratory test, and hence a
standard test for heavier compaction (Modified Proctor Test) has been suggested (IS: 2720 –
Part 8, 1983).
The procedure for conducting the test with heavier compaction is similar to that of light
compaction, but with slightly modified equipment. In this case, a rammer with a mass of
4.89 kg and a fall of 450 mm is used.
The soil is compacted in five layers, with each layer being given 25 blows for a 1,000 ml
mould or 56 blows for a 2,250 ml mould.
M10_PURU1773_01_SE_C10.indd 350 2/12/2013 7:59:56 AM

1.78
1.76
1.755
ρd max
Dry density, g/cc 1.74
1.72
1.70
OMC =
15.45%
1.68
10 12 14 16 18 20
Moisture content, %
Fig. 10.15 Moisture content–dry density curve
Another method which uses a constant weight of soil is available for determination of the
moisture–density relation for the soil passing through the 4.75 mm IS sieve (IS: 2720 – Part
9, 1971). This is a rapid method which can be used as a field control method. It may also be
used as a rapid laboratory test. However, it cannot be used as a substitute for the standard
tests discussed earlier.
The compaction tests (both standard and modified) are satisfactory for cohesive soils.
Clean sands and gravels which are displaced easily during the rammer blows do not indi-
cate proper compaction characteristics. A knowledge of the maximum dry density and OMC
obtained from this test suggests that the maximum density is obtainable in the field using a
suitable roller and adopting a moulding water content almost equal to the OMC. A check can
be made on the field-compacted soil by adopting field control tests.
10.17 TEST NO. 16: DENSITY INDEX OF NON-COHESIVE SOILS

Scope
To determine the density index (relative density) of non-cohesive free draining soils
Apparatus
Graduated cylinder – 1,000 ml
Large glass funnel and glass rod
Compaction mould
Wooden hammer
Needle vibrator
Straight edge
Hand scoop
M10_PURU1773_01_SE_C10.indd 351 2/12/2013 7:59:56 AM

Procedure
1. Fill the graduate cylinder up to 50% of its capacity.
2. Place the large funnel on the top of the graduated cylinder such that the tip of the funnel
is in the water.
3. Take a known mass of dry sand and slowly pour it into the cylinder through the funnel
such that every particle settles down independently.
4. Stop pouring sand when the cylinder is two-thirds full and note down the volume (V1)
and the mass of sand added (M1).
5. Repeat Steps 1 to 4 three times and use the minimum value of M1/V1.
6. Take sufficient sand to fill the compaction mould and add sufficient water to saturate it
completely.
7. Fill the compaction mould one-thirds full and compact with the wooden hammer such
that the voids are minimum. Use a needle vibrator to obtain the required condition.
8. Place more soil and repeat Step 6 such that the mould is filled and about 50% of the collar
is full.
9. Remove the collar and level the soil with a straight edge.
10. Find the mass of wet sand (M2) and the volume of the compaction mould (V2).
11. Keep a certain quantity of soil for water content determination.
12. Repeat Steps 6 to 10 three times and use the maximum value of M2/V2.
13. Find the filled density of the soil using the sand replacement method.
14. Find the field void ratio e from the field density, knowing the specific gravity of soil solids.
Computations
Minimum dry density
M1
ρd min =
V1
Maximum void ratio

Gρw
emax = −1
ρd min
Maximum dry density

M2
ρd max =
V2
Minimum void ratio

Gρw
emin = −1
ρd max
Density index
emax − e
Dr = ×100%
emax − emin
Results
The density index is expressed as a percentage. Typical observations and test results are
given in Table 10.20.
M10_PURU1773_01_SE_C10.indd 352 2/12/2013 7:59:57 AM

Table 10.20 Density index test results
Inital mass of sand (g) 2,000 2,000 2,000

Mass of sand after pouring in graduated jar (g) 1070.5 1002.5 854.4
Mass of sand (M1) (g) 929.5 997.5 1145.6
Volume of sand (V1) (ml) 650 700 800
M
Minimum dry density ρd min = 1 ( g / cc) 1.43 1.425 1.432
V1
Gρw
Average maximum void ratio (G = 2.65) emax = − 1 = 0.86
ρd min
Minimum void ratio
Mass of mould (g) 6,245 6,245 6,245
Mass of mould + wet soil (g) 8,331 8,342 8,339
Mass of mould (V2) (ml) 1,000 1,000 1,000
Mass of wet soil (M2) (g) 2,086 2,097 2,094
Mass of cup (g) 22.67 2,0.74 22.84
Mass of cup + wet soil (g) 42.95 40.72
Mass of cup + dry soil (g) 39.62 37.45
Water content (%) 19.65 19.55
M2 / V2
Maximum dry density ρd max = ( g / cc ) 1.743 1.754
1 + w / 100
Gρw
Average minimum void ratio emin = − 1 = 0.51
ρd max
Void ratio infield e = 0.62
e −e
Density index Dr = max × 100% = 68.6%
emax − emin
Discussion
This procedure is just sufficient for obtaining a fairly accurate value, provided the loosest
density is obtained carefully (Prakash, 1969). A more comprehensive method has been given
in IS: 2720 – Part 14 (1983).
10.18 TEST NO. 17: CONSOLIDATION TEST

Scope
To determine the consolidation properties of soil
Apparatus
Consolidation ring – a rigid ring with a smooth and polished inner surface and provided
with a cutting edge to facilitate preparation of specimens. The minimum diameter of the ring
should be 60 mm with a diameter–height ratio of 3.0 (Fig. 10.16)
Porous stone – shall be of silicon carbide, aluminium oxide, or other porous materials with
high porosity such that free drainage is assured throughout the test. The diameter of the
M10_PURU1773_01_SE_C10.indd 353 2/12/2013 7:59:58 AM

porous stone is 0.2 to 0.5 mm less than that of the inside diameter of the ring. The stone size
varies depending on the type of ring (Fig. 10.16)
Consolidation cell – a container to house the consolidation ring has a provision to hold water
and allow measurement of the change in height of the specimen at its central axis
Dial gauge – has a length of travel of 50% of the specimen height with an accuracy of at least
0.001% of the specimen height
Loading device – capable of taking axial loads in suitable increments with a suitable lever
ratio and of maintaining this for a large duration of time with an admissible variation of
±1% of the applied load (Fig. 10.17). There should be no significant impact during load
application. It should be located in an area free from vibrations
Sample extruder
Trimming equipment
Equipment for water content determination
Stopwatch with least count of 1 second
Increase in pore
water pressure
Friction Ring Friction
Δuw
Porous stone Porous stone

Sample Sample
Porous stone Porous stone
(a) Floating ring (c) Fixed ring
Fig. 10.16 Types of consolidometers
Drain tap
Dial gauge Reservoir
Yoke
Consolidometer
Loading frame
Level
tube
Weight hanger
Counter- Weights
balance
weight
Fig. 10.17 Consolidation test set-up
M10_PURU1773_01_SE_C10.indd 354 2/12/2013 7:59:59 AM

Procedure
1. Find the mass of the empty consolidation ring (M1).
2. Coat the inside surface with silicone grease or oil. Trim a sample carefully to fit the con-
solidation ring and weigh the mass along with the ring (M2). Keep a small quantity of
the soil from the trimmings for water content determination.
3. Record the thickness of the specimen. In case of difficulties in measuring the thickness,
take the thickness of the ring as the initial thickness.
4. Depending on the type of ring (fixed or floating), choose the correct size of the porous
stone. Place the ring and the specimen centrally on the saturated bottom porous stone
and place the upper saturated porous stone, followed by the loading cap.
5. Place the consolidometer in the loading device and attach the dial gauge. Fill the
consolidometer with water, apply a seating load of 5 kN/m2, and allow it to reach
moisture equilibrium in 24 hours.
6. Apply the first load increment and simultaneously take deformation readings at elapsed
times of 0.25, 0.50, 1, 2, 4, 8, 15, 30, and 60 minutes and 2, 4, 8, and 24 hours.
7. After 24 hours apply the increment load, keeping mind the fact that the applied pres-
sure at any loading stage should be double that at the preceding stage. Apply the
following loading sequence: 10, 20, 40, 80, 160, 320 kN/m2. Each time repeat Step 6.
8. On completion of the final loading, unload the specimen with pressure decrements which
decrease the load to one-fourth the previous load. Take dial gauge readings during each
stage of unloading. If desired, the same time interval as adopted during loading may be
adopted. Keep the last unloading at a pressure of 5 kN/m2 for 24 hours to minimize the
swelling during disassembly.
9. Remove the ring, wipe the water on the outside of the ring, and find the mass (M3).
10. After drying, weigh the specimen with the ring and find the mass (M4).
Computations
(a) Coefficient of consolidation
Plot dial gauge reading versus t or versus log t for each load increment, and find the coef-
ficient of consolidation from the following expressions.
(i) Square root of time method:
0.848 ( Hav / 2)2
Cv =
t90
(ii) Logarithm of time method:
0.197 ( Hav / 2)2
Cv =
t50
where Hav is the average thickness of the specimen for that load increment.
(b) Coefficient of compressibility
Ms
Volume of solids, Vs =
Gρw
where Ms = M4 − M1.
M10_PURU1773_01_SE_C10.indd 355 2/12/2013 8:00:00 AM

The equivalent height, Hs = Vs / A

where A is the area of specimen.
The void ratio e, at the end of each pressure increment, is given as
H
e= −1
Hs
where H is the height of specimen at the end of each pressure increment.

The coefficient of compressibility av, with units inverse of those for stress is given as
Δe
av =
Δp
(c) Compression index Cc

Plot the void ratio e versus log p. The slope of the straight line portion of the curve is the
compression index Cc which is given as
Δe
Cc =
log ( p2 / p1 )
where p2 and p1 are the successive values of pressure and Δe is the change in the void ratio
over the above range of pressures.
Results
The consolidation test results are presented in the form of the following curves:
1. e versus log p
2. Dial reading versus log t for different stress ranges
3. Dial reading versus t for different stress ranges
4. av versus log p
5. cv versus log p
Some typical test results are presented in Tables 10.21 to 10.23, and the corresponding
plots are given in Figs. 10.18 and 10.19.
Discussion
As the effects of sample preparation are the same for any size of sample, larger samples pro-
vided more reliable results. The floating ring reduces the frictional loss along the sides of the
sample between the soil and ring, and hence, the test rate is about four times faster. The fixed
ring has the advantage of measuring the k value of the sample as it is tested. The curve fitting
methods are discussed in Chapter 6. Other details of the test can be obtained from IS: 2720
– Part 15 (1986).
M10_PURU1773_01_SE_C10.indd 356 2/12/2013 8:00:00 AM

Table 10.21 Data and some details of a consolidation test
Height of consolidation ring 25 mm

Diameter of consolidation ring 60 mm
Area of cross-section of consolidation ring (A) 28.27 cm2
Mass of empty consolidation ring (M1) 162.50 g
Mass of wet soil + ring (M2) 182.70 g
Initial thickness of sample 20 mm
Initial water content 20.9% (wt.)
Specific gravity of soil solids (G) 2.68
Initial void ratio (e0 = wG) 0.560
Mass of wet soil + ring (after completion of consolidation) (M3) 180.96 g
Mass of dry soil + ring (M4) 259.48 g
Mass of dry soil (Ms = M4 − M1) 96.98 g
⎛ Ms ⎞⎟
Volume of soil solids ⎜⎜Vs = ⎟⎟ 36.19 ml
⎜⎜⎝ Gρw ⎟⎠
Final water content 19.03%
Table 10.22 Dial gauge reading versus time for three loadings
Elapsed time (minutes) Time (minutes)1/2 Consolidation pressure (kN/m2)

(Dial gauge reading least count = 0.01 mm)
10.0 20.0 40.0
0 0.00 195.0 175.40 150.60

0.25 0.50 191.0 173.25 148.00
0.50 0.71 190.25 172.00 147.00
1.00 1.00 189.75 170.75 145.50
2.00 1.41 189.00 169.00 143.25
4.00 2.00 187.75 166.50 140.25
8.00 2.83 186.50 163.75 137.25
15.00 3.87 185.25 161.75 134.75
30.00 5.48 184.50 159.50 133.00
60.00 7.75 183.25 158.25 131.75
120.00 10.96 182.75 157.75 129.25
180.00 13.42 182.50 157.25 129.00
1440.00 37.95 175.40 150.60 123.03
M10_PURU1773_01_SE_C10.indd 357 2/12/2013 8:00:01 AM

Table 10.23 Final results from consolidation test
Applied Final Compression, e =(H/Hs–1) Δe Δp av = Δe/Δp Hav t90 cv (mm2/

pressure dial ΔH (mm) 2 2
(kN/m ) (m /kN) (mm) (minutes) min)
(kN/m2) reading
195.0 0 20 0.560 0 0 0 20 – –
175.4 0.196 18.80 0.547 0.013 10 1.30 × 10–3 19.90 17.64 4.76
150.6 0.248 19.55 0.548 0.019 10 1.90 × 10–3 19.68 12.25 6.70
123.0 0.276 19.27 0.506 0.022 10 2.20 × 10–3 19.41 10.24 7.80
Time (minutes)½
2 4 6 8 10 12 14
200
t 90 = 4.2
192
0–10 kN/m2
184
a
1.15a
176
t 90 = 3.5
10–20 kN/m2
168
Dial gauge reading
160
a
1.15a
152
t 90 = 3.2
144 20–40 kN/m2
136
128 a
1.15a
120
Fig. 10.18 Curve fittings by time method
M10_PURU1773_01_SE_C10.indd 358 2/12/2013 8:00:01 AM

0.57
e0
Cc = 0.0664
0.55
Void ratio
0.53
Δe = 0.02
0.51
p1 = 15 kN/m2 p2 = 30 kN/m2
0.49
1 10 100
Pressure, kN/m2
Fig. 10.19 e–log p plot
10.19 TEST NO. 18: UNCONFINED COMPRESSION TEST

Scope
To determine the compressive strength and sensitivity of a cylindrical sample of cohesive
soil
Apparatus
Unconfined compression testing machine (strain controlled)
Sampling tube
Sample ejector
Deformation dial gauge – 0.01 mm graduations and specific travel to permit 20% axial strain
Vernier callipers – of least count 0.1 mm
Timer
Oven with accurate temperature control in the range 110 ± 5°C
Miscellaneous equipment, such as specimen trimmers, carving tools, re-moulding appara-
tus, moisture cups, etc.
Procedure
(a) Preparation of test specimen
Undisturbed, compacted, or re-moulded specimens may be prepared, depending on the
case.
1. Prepare undisturbed cylindrical specimens (38 mm diameter, 76 mm length) from large
undisturbed field samples using a lathe or trimmer. Alternatively, directly obtain field
samples in thin sampling tubes of the same diameter as that of the specimen. Obtain the
required length by ejecting the sample through a split mould.
M10_PURU1773_01_SE_C10.indd 359 2/12/2013 8:00:02 AM

2. Prepare a re-moulded compacted specimen, of any predetermined water content and

density, in a large mould and then cut it using the sampling tube. Or prepare a re-
moulded specimen from a failed undisturbed specimen by pushing the soil inside a split
mould, with the same void ratio and natural water content.
3. In both the cases, the wet density and water content of the specimens are determined.
(b) Compression test

4. Measure the dimensions of the specimen. Weigh the specimen and keep representative
samples for water content determination.
5. Place the specimen on the bottom plate of the loading device and adjust the upper plate
to make contact with the specimen.
6. Adjust the deformation and proving ring dials to zero and apply the axial load with a
strain rate of 0.5% to 2% per minute.
7. Record the force and deformation readings at suitable intervals, with closer spacing dur-
ing initial stages of the test.
8. Apply the load till the failure surfaces have definitely developed or until an axial strain
of 20% is reached.
9. Carefully sketch the failure pattern, and if the specimen has failed with a pronounced
failure plane, measure the angle of the failure surface with the horizontal.
10. Take water content representative samples from the failure zone of the specimen.
Computations
Stress–strain values are calculated as
ΔL
Axial strain ε =
L0
where ΔL is the the change in the specimen length (mm) and L0 the initial length of the
specimen (in mm).
The average cross-section area A at a particular strain is given by
A0
A=
1− ε
where A0 is the initial average area of cross-section of the specimen.
P
Compressive stress σ1 =
A
where P is the compressive force.
Plot σ1 versus ε and obtain the maximum stress which gives the unconfined compressive
strength qu. In case no pronounced peak is observed, take the strength corresponding to 20%
strain as the unconfined compressive strength.
For φ = 0 condition, the shear strength or cohesion of the soil may be taken to be equal to
half the unconfined compressive strength.
M10_PURU1773_01_SE_C10.indd 360 2/12/2013 8:00:02 AM

Table 10.24 Data and test results from unconfined compression test
Type of specimen (undisturbed, compacted, or compacted and re-moulded)
Initial length (L0) = 83 mm
Initial diameter (D0) = 38 mm
Initial area (A0) = 1,134 mm2
Initial mass of specimen = 75.8 g
Initial density = 1.76 g/cc
Initial water content = 15.5%
Rate of strain adopted = 1.27 mm/min
Sl. no. Elapsed time Load (N) Deformation (mm) Strain (%) Area A = A0/(1 – ε) Stress
(minutes) (mm2) (N/mm2)
1. 0.50 30.97 0.10 0.1205 1135.4 0.027

2. 1.00 61.94 0.30 0.3615 1138.1 0.054
3. 1.50 89.65 0.60 0.7229 1142.3 0.079
4. 2.00 118.90 1.00 1.2048 1147.8 0.104
5. 3.00 128.77 1.20 1.4458 1150.6 0.112
6. 4.00 140.18 1.40 1.6868 1153.5 0.122
7. 5.00 146.70 1.50 1.8072 1154.9 0.127
8. 6.00 148.33 1.60 1.9277 1156.3 0.128
9. 8.00 144.71 1.70 2.0482 1157.7 0.125
10. 10.00 143.44 1.80 2.1687 1159.1 0.124
Final water content = 15.5%

Unconfined compressive strength (qu) = 0.128 N/mm2
⎛ q ⎞
Undrained shear strength or cohesion ⎜⎜cu = u ⎟⎟⎟ = 0.064 N / mm 2
⎜⎝ 2⎠
Results
The observations made during the test are recorded as shown in the data sheet for a typical
case (Table 10.24).
Discussion
The strain-controlled unconfined compression test is used universally. The test is somewhat
sensitive to the strain rate, and can be performed only in a strain-controlled machine. But a
stress-controlled test may show an erratic strain response due to incremental changing of loads.
The shear strength obtained from the unconfined compressive strength is not very reli-
able for at least three reasons:
1. The lateral restraint present in the field is not properly simulated in the laboratory.
2. There is no control on the internal soil conditions (degree of saturation, pore water pres-
sure, etc.).
3. The end platens because of lateral restraint alter the internal stresses.
These three factors have been properly taken care of in modern triaxial shear equipment
(discussed elsewhere).
M10_PURU1773_01_SE_C10.indd 361 2/12/2013 8:00:02 AM

The unconfined compressive strength test is a quick test and gives the approximate shear
strength of a cohesive soil. Another advantage of the unconfined compression test is that the
failure occurs along the weakest portion of the clay and hence provides a conservative shear
strength value. Keeping in view the deficiencies of the test, a reasonable interpretation has
to be made. The unconfined compression test gives misleading results with heterogeneous
soils because of the boundary condition (IS: 2720 – Part 10, 1973).
10.20 TEST NO. 19: DIRECT SHEAR TEST

Scope
To determine the shear strength of a soil using direct shear apparatus
Apparatus
Shear box – size 60 mm2 and about 50 mm deep to suit particles with size less than 4.75 mm–
grid plates, porous stones, base plate, loading pad
Container for shear box (Fig. 10.20)
Loading frame with proving ring/load cell
Micrometer dial gauges – 0.01 mm accuracy
Sample trimmer or specimen cutter
Stopclock
Balance of 1 kg capacity with 0.1 g sensitivity
Spatula and a straight edge
Procedure
(a) Preparation of specimen
1. Prepare undisturbed specimens of the required size by trimming from suitable large
undisturbed samples.
2. For re-moulded cohesive soils, compact the soil to the required density at the appropri-
ate water content, extract the sample from the mould, and then trim to the required size
or use a standard specimen cutter. Alternatively, compact the soil in the shear box itself
to the desired density.
Loading pad Pin or screw to hold the

two halves of the shear box
Shear box
60 mm × 60 mm × 50 mm
Grid plate
50 mm
Base plate
U arm
Container for
shear box
Fig. 10.20 Assembly of shear box and container (Source: IS: 2720 – Part 13, 1972)
M10_PURU1773_01_SE_C10.indd 362 2/12/2013 8:00:03 AM

3. For re-moulded non-cohesive soils, place the soil in the shear itself and tamp it till the
required density is reached. Use a porous stone, if necessary, depending on the type of test.
4. Weigh the cut or trimmed specimen and record the mass of soil used in the case of a non-
cohesive soil to find the bulk density of the specimen.
(b) Shear tests
(i) Undrained test
5. Keep plain grid plates, one on either side of the specimen (with serrations of grid plates
at right angles to the direction of shear), and place the specimen with grid plates on the
base plate.
6. Place the loading pad on the top grid plate and add water in the shear box container so
as to prevent drying of the specimen. Apply the required normal stress, depending on
the field condition or design requirement.
7. Choose a suitable strain rate such that no drainage takes place during the test. Raise the
upper part of shear box such that a gap of about 1 mm is left between the two parts of
the box.
8. Apply the shear load at the chosen strain rate till failure or to 20% longitudinal displace-
ment, whichever occurs first.
9. Record the shear load reading and longitudinal displacement. Ensure that no drainage
has taken place by noting the vertical compression dial.
10. Remove the soil specimen and keep it in the oven for water content determination.
11. Repeat the test on three more separate specimens with the same initial conditions.
(ii) Consolidated undrained test
12. Follow Step 5 but use perforated grid plates and saturated porous stones at the top and
bottom of the specimen.
13. Follow Step 6 and record the vertical compression (caused due to consolidation) and the
time elapsed. Ensure that the consolidation is complete.
14. Follow Steps 7 to 11.
(iii) Consolidated drained test
15. Follow Steps 12 and 13.
16. Adopt a slow rate of strain during load application such that complete drainage occurs
with 95% pore pressure dissipation.
17. Follow Steps 10 and 11.
Computations
Calculate the proving ring constant and hence the load at different displacements. Calculate
the shear stress using the corrected area, which is given as
⎛ δ⎞
Corrected area = A0 ⎜⎜⎜1 − ⎟⎟⎟
⎝ 3⎠
where A0 is the initial area of specimen and δ the displacement.
Results
Plot the shear stress versus the longitudinal displacement readings and note down the
maximum shear stress and the corresponding longitudinal displacement for the particular
M10_PURU1773_01_SE_C10.indd 363 2/12/2013 8:00:03 AM

normal stress. Plot the normal stress versus the maximum shear stress and obtain the shear
strength parameters, c and φ.
For the consolidated undrained and consolidated drained tests, report the consolida-
tion pressure and the consolidation characteristics. Typical data and results are shown in
Tables 10.25 to 10.28. Figures 10.21 and 10.22 represent the shear stress–displacement and
Coulomb’s strength envelope, respectively.
Discussion
The procedure outlined above is for soils with particle size not greater than 4.75 mm (IS: 2720 –
Part 13, 1972). For a test procedure for soils containing gravel, the reader may refer to IS:
2720 – Part 39/Sec. 1 (1977).
Failure in direct shear may be considered to occur at maximum shear stress or at maxi-
mum obliquity of the Mohr failure envelope. The angle of shearing resistance obtained con-
sidering the maximum shear stress is less than the other one, and the error is on the safe side.
The error involved is much more important in sands than in clays.
Table 10.25 Data of direct shear test
Type of test Consolidated undrained test

Rate of strain 1.27 mm/min
Soil specimen Undisturbed
Size = 60 mm × 60 mm Area = 36 cm2
Height = 25 mm Volume = 90 cm3
Initial wet weight = 186.07 g Initial water content = 48.6%
Bulk density = 2.07 g/cc
Table 10.26 Consolidation details

Normal stress = 0.01 N/mm2
Time Vertical dial Vertical dial Thickness of

reading difference specimen (mm)
(Hour) (Minutes)
0 0 0 25.0
15 20 20 24.8
30 23 3 24.77
1 00 24 1 24.76
1 30 27 3 24.73
2 00 30 3 24.70
2 30 32 2 24.68
3 00 36 4 24.64
3 30 37 1 24.63
4 00 38 1 24.62
5 00 38 0 24.62
6 00 38 0 24.62
7 00 38 0 24.62
M10_PURU1773_01_SE_C10.indd 364 2/12/2013 8:00:03 AM

Table 10.27 Shearing details

Displacement Displacement δ Area Corrected area Stress dial Shear force Shear stress
dial reading (cm) correction (mm2×100) reading (5)×1.108 (N) (N/mm2)
20 0.02 0.993 35.75 29 32.14 0.0089

40 0.04 0.987 35.53 53 58.72 0.0165
60 0.06 0.980 35.28 71 78.67 0.0223
80 0.08 0.973 35.03 85 94.18 0.0269
100 0.10 0.967 34.81 97 107.48 0.0309
120 0.12 0.960 34.56 106 117.45 0.0340
140 0.14 0.953 34.31 114 126.31 0.0368
160 0.16 0.947 34.09 117 129.64 0.0380
180 0.18 0.940 33.84 119 131.85 0.0390
200 0.20 0.933 33.59 121 134.07 0.0399
220 0.22 0.927 33.37 122 135.18 0.0405
240 0.24 0.920 33.12 121.5 134.62 0.0407
260 0.26 0.913 32.87 120 132.93 0.0405
280 0.28 0.907 32.65 117 129.64 0.0397
Table 10.28 Shear parameters
Test no. Normal stress Shear stress Cohesion Angle of shearing

(N/mm2) (N/mm2) (N/mm2) resistance (°)
1 0.04 0.0397
2 0.06 0.0510 0.018 28
3 0.08 0.0590
4 0.12 0.0810
10.21 TEST NO. 20: TRIAXIAL SHEAR TEST

Scope
To determine the shear strength of a soil using triaxial shear apparatus
Apparatus
Triaxial compression machine – 50 kN capacity (strain-controlled) with a strain rate range of
0.05 to 7.5 mm/min (Fig. 10.23)
Triaxial cell to accommodate 38 mm diameter sample
Constant cell pressure system with a capacity of 1,000 kN/m2
Volume measuring device
Pore pressure apparatus
Rubber membrane
Membrane stretcher
Sample trimming device
M10_PURU1773_01_SE_C10.indd 365 2/12/2013 8:00:04 AM


0.05
0.04
Shear stress, N/mm2
0.03
0.02
0.01
0
0.04 0.08 0.12 0.16 0.20 0.24 0.28
Displacement, cm
Fig. 10.21 Shear stress versus displacement
0.10
0.08
Shear stress, N/mm2
0.06
0.04
φ = 28°
0.02
0 0.02 0.04 0.06 0.08 0.10 0.12 0.14 0.16

Normal stress, N/mm2
Fig. 10.22 Coulomb’s strength envelope
M10_PURU1773_01_SE_C10.indd 366 2/12/2013 8:00:04 AM

Load frame
Burette Pore pressure

measuring device
Proving ring Loading
or load cell ram Pressure gauge
Volume Deformation dial
measuring Scale
Cell
device Manometer
Mercury
Constant cell
pressure
system
Pump
Null indicator
Fig. 10.23 Triaxial test assembly
Split mould
Trimming knife
Stopwatch
Procedure
(a) Preparation of specimen
(i) Cohesive soil
1. Undisturbed, compacted, or re-moulded cohesive soil specimens may be prepared as
explained for the unconfined compression test (Test no. 18).
(ii) Non-cohesive soil
2. Attach a rubber membrane to the base platen (Fig. 10.24) using rubber O-rings. Place a
porous stone on the base of the platen.
3. Take a known mass of dry sand so that the sample density can be obtained and approxi-
mately duplicated for successive tests.
4. Place a split mould around the membrane and fold the top portion of it over the mould.
5. Carefully transfer the sand to the membrane in two or three layers and tamp each layer
with a glass rod to obtain the shape and density. If the test is to be conducted under satu-
rated conditions, the sand may be placed in water and then transferred to the membrane.
6. Place a porous stone on the top of the sample and then place the top platen. Apply
silicone grease to the sides of the platen to obtain a better leak-proof seal. Roll the mem-
brane on to the top platen and seal it with rubber O-rings.
7. Attach a tube from the top platen to the vacuum outlet and apply a vacuum of 200 to 250
mm of mercury to the sample, or if the test is to be carried out in a saturated condition,
attach a tube to the base, connect it to the U-tube manometer of the pore pressure appa-
ratus, and apply a small negative pore water pressure to keep the specimen straight.
8. Remove the split mould and check for holes and leaks.
9. Take the average height and diameter to obtain the density.
M10_PURU1773_01_SE_C10.indd 367 2/12/2013 8:00:04 AM

Glass rod
Water
Sand
Funnel
Stopper
O–rings Rubber bung
Membrane
Clamp
Metal or plastic
split mould
Circlip
Porous stone
Split
O-rings
Base of cell
Fig. 10.24 Preparation of saturated cohesionless soil specimen
(b) Shear tests

(i) Undrained test
10. Measure the dimensions of the specimen. Weigh the specimen and keep a representative
sample for water content determination.
11. Place a solid Perspex platen over the specimen, which in turn is placed over another
Perspex platen. Place the loading cap on the top platen.
12. Insert a rubber membrane using a membrane stretcher and fix two O-rings, one at the
bottom and the other on the top of the platen or loading cap. (Steps 10 to 12 are not
needed for non-cohesive soil specimens prepared following Steps 2 to 9.)
13. Place one cell on the triaxial cell base and transfer the same to the compression machine,
and just make load contact of the loading ram.
14. Close the drainage valve, fill the cell with water, and apply the pre-determined chamber
pressure.
15. Adjust the deformation and proving ring dials to zero, and apply the axial load with a
strain rate of 0.5% to 2% per minute.
16. Record the force and deformation readings at suitable intervals with a closer spacing
during the initial stages of the test.
17. Apply the load till the proving ring dial recedes backwards or until an axial strain of 20%
is reached.
M10_PURU1773_01_SE_C10.indd 368 2/12/2013 8:00:05 AM

18. Unload the specimen and drain off the cell fluid. Dismantle the cell and carefully remove
the membrane and note down the mode of failure.
19. Weigh the specimen and take representative water content samples from the failure zone
of the specimen.
20. Repeat the test on three or more identical specimens under increased cell pressures.
(ii) Consolidated undrained test
21. Follow Steps 1 to 10. Place the specimen over a saturated porous stone which in turn is
placed on top of the specimen, and then place a loading cap with a drainage outlet.
22. To quicken the process of consolidation of the specimen, place a series of small threads
or a strip of filter paper around the surface of the specimen. Now insert the rubber mem-
brane and fix the O-rings, one at the bottom and another at the top.
23. Follow Step 13. Connect the drainage valve to a volume-measuring device or to a burette.
Fill the cell with water and apply the pre-determined chamber pressure.
24. Open the drainage valve and note the volume change during consolidation. Ascertain
the completion of the volume change by noting down the constant water level in the
burette or volume change device. Close the drainage valve.
25. If the pore water pressure is to be measured, connect a pressure transducer or a null
pressure indicator device to the saturation line.
26. Adjust the null-indicator to the initial position or the transducer output to the initial
reading. Adjust the deformation and proving ring dials to zero and apply the axial load
at a slow rate such that the pore pressure readings can be taken conveniently.
27. Record the force, deformation and pore pressure readings at suitable intervals with a
closer spacing during the initial stages of the test.
28. Follow Steps 17 to 20.
Computations
1. The axial strain

ΔL
ε=
L0
2. The average cross-sectional area A at a particular strain is calculated as done in the uncon-
fined compression test.
3. The deviator stress
PRR × PRC
Δσ =
A
Plot the deviator stress versus the strain and obtain the stress at the peak point unless the
stress at 20% strain occurs first. Compute the major principal stress for each test as
σ1 = σ3 + Δσ
Also compute the pore water pressure corresponding to the maximum deviator stress.
Compute the effective major and minor principal stresses as
σ3′ = σ3 − uw and σ1′ = σ1 − uw
M10_PURU1773_01_SE_C10.indd 369 2/12/2013 8:00:05 AM

370
Table 10.29 Data and test results for dry non-cohesive soil from the triaxial test
Length of specimen = 76 mm
Diameter of specimen = 38 mm
Proving ring constant (PRC) 1 div = 3.13 × 10–3 kN
Vertical ε = ΔL/L0 1–ε A = A0/(1 – ε) Proving ring Deviator stress σ’1 = σ′3 + Δσ′
M10_PURU1773_01_SE_C10.indd 370
deformation (m3 × 10–3) reading (PRR) PRR × PRC (kN/m2)
(mm × 0.01) Δσ =
A
Sample 1 0 0 0.1 1.134 0 0 140.0
σ3′ = 140 kN / m 2 38 0.005 0.995 1.140 116 318.5 458.5
143 0.019 0.981 1.156 220 595.7 735.7
287 0.038 0.962 1.179 247 655.7 795.7
430 0.057 0.943 1.202 250 651.0 791.0
776 0.102 0.898 1.263 258 639.4 779.4
1,346 0.177 0.823 1.378 265 601.9 741.9
1,742 0.229 0.771 1.471 272 578.8 718.8
Sample 2 0 0 1 1.134 0 0 210.0
σ3′ = 210 kN / m 2 73 0.010 0.990 1.145 199 544.0 754.0
238 0.031 0.969 1.170 256 684.9 894.9
412 0.054 0.946 1.200 333 868.6 1078.6
500 0.066 0.934 1.214 342 881.8 1091.8
685 0.090 0.910 1.246 352 884.2 1094.2
1,082 0.142 0.858 1.322 355 840.5 1050.5
1,515 0.199 0.801 1.416 360 795.8 1005.8
Sample 3 0 0 1 1.134 0 0 280.0
σ3′ = 280 kN / m 2 88 0.012 0.988 1.147 311 848.7 1128.7
272 0.036 0.964 1.176 369 982.1 1262.1
472 0.062 0.938 1.209 402 1040.7 1320.7
572 0.075 0.925 1.226 410 1046.7 1326.7
772 0.102 0.898 1.262 421 1044.2 1324.2
1,215 0.160 0.840 1.350 425 985.4 1265.4
1,447 0.190 0.810 1.401 433 967.4 1247.4
2/12/2013 8:00:06 AM
Plot Mohr’s circles for both the total and effective principal stresses and obtain the shear
strength parameters (c, φ and c′cu, φ′cu).
Results
The results are presented in the form of a stress–strain curve, strain–volume change curve,
stress–pore pressure curve, and Mohr–Coulomb plot. Some typical test results are presented
in Tables 10.29 and 10.30 and in Figs. 10.25 to 10.27.
Discussion
For certain special field conditions, the samples have to be consolidated anisotropically. This
is done using a dead-load frame and applying vertical pressure in conjunction with the cell
pressure σ3 to develop the desired stress ratio
σ h′
K=
σ v′
Pore pressure and volume measuring devices should be perfectly de-aired before use
to obtain accurate results. Pore pressure measurements can be performed using a pressure
transducer. In such cases, connect the pressure transducer to the saturation line and in turn
connect the output of the transducer to a voltmeter. Pressure transducers should be very
Table 10.30 Data and test results for dry non-cohesive soil from the triaxial test
Length of specimen = 76 mm
Diameter of specimen = 38 mm
PRC 1 div = 3.13 × 10–3 kN
Vertical ε = ΔL/L0 1 – ε A=A0/(1–ε) PRR Pore Deviator σ′3=σ3 – σ′1 = σ′3 +

deforma- (m2 × 10–3) pressure stress, Δσ′ = uw Δσ′
tion (mm) uw PRR×PRC/A (kN/m2) (kN/m2)
(kN/m2) (kN/m2)
0 0 1 1.134 0 0 0 420 420

3.05 0.040 0.960 1.181 680 110 461 310 571
3.81 0.050 0.950 1.194 699 112 468 308 580
5.08 0.067 0.933 1.215 835 131 550 289 681
6.35 0.084 0.916 1.238 998 172 645 248 817
7.62 0.100 0.900 1.260 1,101 218 699 202 917
8.89 0.117 0.883 1.284 1,168 241 728 179 969
10.16 0.134 0.866 1.310 1,221 249 746 171 995
11.43 0.150 0.850 1.334 1,266 255 759 165 1,014
12.70 0.167 0.833 1.361 1,281 252 753 168 1,005
13.97 0.184 0.816 1.390 1,291 249 743 171 992
15.24 0.201 0.799 1.419 1,298 248 732 172 980
16.51 0.217 0.783 1.448 1,306 248 722 172 970
M10_PURU1773_01_SE_C10.indd 371 2/12/2013 8:00:07 AM

σ3′ kN/m2
1,000
280
800
210
Deviator stress, kN/m2
600
140
400
200
0 4 8 12 16 20 24 28
Axial strain, %
Fig. 10.25 Deviator stress–strain curves
800 Mohr’s envelope

Shear stress, kN/m2
φ ′=35°
600
400
200
Apparent
cohesion
50 kN/m2
0 200 400 600 800 1000 1200 1400
Effective principal stress, kN/m2
Fig. 10.26 Mohr–Coulomb plot for dry cohesionless soil
M10_PURU1773_01_SE_C10.indd 372 2/12/2013 8:00:07 AM

1,000
800 Effective major

principal stress, σ 1′
Stress, kN/m2
600
Effective minor
principal stress, σ 3′
400
200
Pore water
pressure, uw
0 4 8 12 16 20 24
Axial strain, %
Fig. 10.27 Stress plots with strain
sensitive, even to small volume displacements. Special-type loading rams have to be used
for testing sensitive clays.
For more details of the triaxial equipment and procedure, the reader may refer to IS: 2720 –
Part 11 (1971), Part 12 (1981), and Part 35 (1974).
Other special tests which can be performed using the triaxial apparatus are the extension
test, the decreasing σ3 test, the constant volume test, etc. (Bishop and Henkel, 1962).
10.22 TEST NO. 21: CALIFORNIA BEARING RATIO (CBR) TEST

Scope
To determine the California bearing ratio
Definition
The California bearing ratio (CBR) is expressed as the percentage of force per unit area
required to penetrate a soil mass with a circular plunger of 50 mm diameter at a rate of 1.25
mm/min compared with that required for the corresponding penetration in a standard
material. The ratio is usually determined for penetration values of 2.5 and 5 mm. In general
the penetration value at 2.5 mm is greater than at 5 mm and the penetration value corre-
sponding to 2.5 mm is taken as the design CBR value. However, if the ratio at 5 mm is con-
sistently higher than that at 2.5 mm, the ratio at 5 mm is used.
Apparatus
Cylindrical mound – 150 mm inner diameter and 175 mm height
Collar – 50 mm height and 150 mm diameter
Base plate – 10 mm height
Metal spacer disc – 148 mm diameter and 47.7 mm in height
M10_PURU1773_01_SE_C10.indd 373 2/12/2013 8:00:08 AM

Metal rammer – standard metal rammer for preparation of re-moulded samples

Annular metal rings – 147 mm with a central hole of diameter 53 mm and weight 2.5 kg
Metal penetration plunger – 50 mm diameter and not less than 100 mm long
Dial gauges – accuracy 0.01 mm – 2 nos.
Sieves – 4.75 mm IS sieve and 19 mm IS sieve
Loading machine – capacity of 50 kN (5,000 kg approximately) with a rate of strain of 1.25
mm/min
Expansion measuring apparatus – adjustable stem and perforated plates
Miscellaneous apparatus – mixing bowl, straight-edge scales, soaking tank or pan, drying
oven, filter paper, dishes, and calibrated measuring jar
Procedure
(a) Preparation of test specimen
The test may be performed on (i) undistributed specimens or (ii) re-moulded specimens
which may be compacted either statically or dynamically.
(i) Undistributed specimen
This is obtained by fitting a cutting edge of 150 mm diameter to the mould and pushing the
mould as gently as possible into the ground. As the mould is pushed in, the soil is dug from
the outside. When the mould is full of soil, it is removed by under-digging. Then the top and
bottom surfaces are trimmed flat so as to get a specimen of the required length ready for
testing. If the soil is hard and the mould cannot be pressed, then a large undisturbed lump
of soil is cut out, from which the required specimen for the mould is made. If the specimen
is loose in the mould, the annular cavity is filled with paraffin wax. The density of the soil
and the water content are determined so as to determine the dry density.
(ii) Re-moulded specimen
The dry density to be determined for the re-moulded specimen may be the field density,
maximum dry density, or any other density. The water content required for preparation of
the specimen may be the OMC or the field moisture context.
The soil for the re-moulded specimen shall pass a 19 mm sieve. Allowance for larger size
particles shall be made by replacing the soil by an equal amount of material which passes a
19 mm IS sieve but is retained in a 4.75 mm sieve.
The required quantity of wet soil is prepared and compacted in the mould either statically
or dynamically as required.
In both cases of compaction, if the specimen is to be soaked, the water content of the soil
before and after compaction is determined. If the specimen is not soaked, a representative
sample of material from one of the pieces of the material cut after penetration shall be taken
to determine the water content.
(b) Test for swelling
(i) Place a filter paper over the specimen and the adjustable stem. Place the perforated
plate on the compacted soil specimen in the mould.
(ii) Place weights on the compacted soil specimen to produce a surcharge equal to the
weight of the base material and pavement to the nearest 2.5 kg.
(iii) Immerse the whole mould and weights in a tank of water allowing free access of water
to the top and bottom of the specimen.
M10_PURU1773_01_SE_C10.indd 374 2/12/2013 8:00:08 AM

(iv) Mount the tripod of the expansion mounting device on the edge of the mould and
record the initial dial gauge reading.
(v) Keep the set-up for 96 hours without disturbance and note the readings every day
against the time of reading. Maintain a constant water level throughout the period.
(vi) At the end of the soaking period, note the change in the dial gauge reading and remove
the tripod and the mould from the water tank.
(vii) Allow the specimen to drain for 15 minutes downwards.
(viii) Remove the weights, the perforated plate, and the filter plate.
(ix) Weigh the mould with the soaked soil specimen.
(c) Penetration test
(i) Place the mould containing the specimen, with the base plate in position and the top
face exposed, on the lower plate of the testing machine (Fig. 10.28).
(ii) Place on the specimen the required number of surcharge weights to simulate the inten-
sity of loading equivalent to the base material and pavement.
(iii) In order to prevent upheaval of soil into the holes of the surcharge weights, place 2.5
kg of annular weights on the surface prior to seating the penetration plunger and then
the balance surcharge weights.
(iv) Apply a seating load of 4 kg so that free contact is established between the surface of
the specimen and the plunger.
(v) Set the load and deformation gauges to zero.
(vi) Apply the load on the plunger into the soil at a rate of 1.25 mm/min.
(vii) Note the readings of the load at penetrations of 0.5, 1.0,1 .5, 2.0, 2.5, 4.0, 5.0, 7.5, 10.0,
and 12.5 mm.
(viii) Note the maximum load and penetration if the maximum load occurs at a penetration
less than 12.5 mm.
(ix) Raise the plunger and detach the mould from the loading machine.
(x) Determine the water content of the soil sample taken from the top 30 mm layer of the
specimen.
(xi) Find also the average water content of the specimen by taking samples from the entire
depth of the specimen.
(xii) In case of undisturbed specimens from the field, carefully examine the presence of any
oversize soil particles which may affect the results if they happen to be located directly
below the penetration plunger.
(xiii) As a check, the penetration test may be repeated on the rear side of the specimen.
(d) Expansion ratio

The expansion ratio is calculated from the expression
df − ds
Expansion ratio = ×100
h
where df is the final dial gauge reading in mm, ds the initial dial gauge reading in mm, and h
the initial height of the specimen in mm.
The expansion ratio is used to identify qualitatively the potential expansiveness of the
soil.
M10_PURU1773_01_SE_C10.indd 375 2/12/2013 8:00:08 AM

Load applied
Proving ring for
measuring load
Dial gauge for
penetration
measurement
50 mm penetration
plunger Surcharge
weight
125 mm
Soil specimen
Fig. 10.28 Set-up for CBR test
(e) Load–penetration curve

Plot the load–penetration curve. This curve is usually convex upwards although the initial
portion of the curve may be convex downwards due to surface irregularities. A correction
shall then be applied by drawing a tangent to the point of greatest slope and then transpos-
ing the axis of the load so that zero penetration is taken as the point where the tangent cuts
the axis of penetration. The corrected load–penetration curve will then consist of the tangent
from the new origin to the point of tangency on the re-shifted curve and then the curve itself.
(f) The CBR
The CBR values are usually calculated for penetrations of 2.5 and 5.0 mm. Corresponding to
the penetration value at which the CBR value is desired, the corrected load value should be
taken from the load–penetration curve and the CBR calculated as follows:
PT
CBR = ×100
Ps
where PT is the corrected unit (or total) test load corresponding to the chosen penetration
from the load–penetration curve and Ps the unit (or total) standard load for the soil depth of
penetration as for PT taken from the table given in Fig. 10.29.
M10_PURU1773_01_SE_C10.indd 376 2/12/2013 8:00:09 AM

Generally, the CBR value at 2.5 mm penetration will be greater than that at 5.0 mm pen-
etration and in such a case, the former shall be taken as the CBR value for design purposes.
If the CBR value corresponding to a penetration of 5.0 mm exceeds that for 2.5 mm, the test
shall be repeated. If identical results follow, the CBR corresponding to 5.0 mm penetration
shall be taken for the design.
(g) Presentation of results
(i) Weight of mould with base plate = 7,445 g
(ii) Weight of mould with base plate + wet soil = 12,495 g
(iii) Weight of wet soil = 5,050 g
(iv) Weight of mould + wet soil after soaking = 12,820 g
(v) Weight of water absorbed = 325 g
(vi) Percentage of water absorbed = 6.44%
(vii) Moisture content = 15.90%
100
90
No correction
required
80
70
Load on piston, kg/cm2
60
50
Corrected 5.0 mm
penetration
40
30
Corrected 2.5 mm
penetration
20
Corrected for concave
upward shape
10
0
0 2.5 5.0 7.5 10.0 12.5
Penetration, mm
Penetration depth Unit standard pressure Total standard load

(mm) (kg/cm2) (kgf)
2.5 70 1,370
5.0 105 2,055
Fig. 10.29 Correction load penetration curves
M10_PURU1773_01_SE_C10.indd 377 2/12/2013 8:00:09 AM

Penetration values versus load on plunger is given in Table 10.31 and presented in
Fig. 10.30.
From Fig. 10.30, the load on plunger at 2.5 mm = 155 kg
5.0 mm = 255 kg
Then
155
CBR value at 2.5 mm penetration = ×100 = 11.3
1370
255
CBR value at 5.0 mm penetration = ×100 = 12.4
2055
Result
The CBR value at 5.0 mm penetration is greater than the CBR at 2.5 mm penetration. As per
rule, the test has to be repeated. Since the difference between the two values is small for all
practical purposes a CBR value of 12.0 may be taken.
Discussion
The CBR test is an empirical one and not based on any mathematical reasoning. It is
only of use when the data available show the results of a known intensity of traffic on a
pavement.
It has been reported that CBR values are higher when the compacted densities are high
and when the clay content, liquid limit, and plasticity index are low. Further, the results of
the penetration test on both compacted and soaked specimens show that the results are not
reproducible.
CBR values are extremely sensitive to changes in moulding water content and density.
The presence of gravel and coarse particles in undisturbed specimens influence greatly the
CBR value.
CBR values cannot be accurately related to any other fundamental property of soil. How-
ever, the deformation of a soil specimen is predominantly shear deformation, and the CBR
values can be regarded as an indirect measure of the shearing strength.
Table 10.31 Penetration versus load on plunger
Sl. no. Penetration (mm) Total load on plunger (kg)
1 0.64 55
2 1.27 100
3 1.91 125
4 2.54 150
5 5.08 250
6 7.62 350
7 10.16 400
8 12.17 450
M10_PURU1773_01_SE_C10.indd 378 2/12/2013 8:00:10 AM

500
450
400
350
300
Load, kg
255
250
200
155
150
100
50
0 2.5 5.0 7.5 10.0 12.5 15.0 17.5

Penetration, mm
Fig. 10.30 Load – penetration curve
POINTS TO REMEMBER
10.1 Dry soil samples have to be prepared for laboratory tests whenever needed. Drying
may be done in air or oven as the case may be. Oven drying is generally done for 24
hours at 110 ± 5°C.
10.2 The density bottle method is a laboratory method for determination of the G of
fine-grained soils. The pycnometer or gas jar method is used for all soils. In the
test the major source of error is the complete removal of air from the sample. For
soils containing soluble salts, kerosene or white spirit may be preferred in place of
water.
10.3 The factors which are essential for accurate determination of water content are the
mass of the wet representative sample, the temperature, and the duration of drying of
the sample. The ovendrying method of water content determination is recommended
by Indian Standards as the standard method.
M10_PURU1773_01_SE_C10.indd 379 2/12/2013 8:00:11 AM

10.4 The in-place density determined by the core-cutter method is convenient and quick
and suitable for fine-grained soils. Sand replacement method is relatively slow but
can be used for any type of soil.
10.5 Sieve analysis is suitable for coarse-grained soils. A wet sieve analysis has to
be preferred if the material passing the 4.75 mm sieve contains more clay-size
particles.
10.6 Sedimentation methods and the pipette and hydrometer methods are suitable for
fine-grained soils. These methods are not recommended if less than 10% of the mate-
rial passes the 75 mm IS sieve. Both the methods give fairly accurate results, but both
are time consuming.
10.7 The mechanical liquid limit device has been recognized as usable in a routine test.
For accurate results, natural soils have to be used and should not be oven dried. Soils
with low clay content have to be tested immediately after thorough mixing with
water.
10.8 Plastic limit test should be conducted on natural soils for accurate results. If the plas-
tic limit cannot be conducted on some soils like sandy soils, then the plasticity index
is reported as non-plastic soil (Np). When the plastic limit is greater than or equal to
liquid limit, Ip is reported as zero.
10.9 The shrinkage limit indicates the extent of volume change which can occur with
changes in water content.
10.10 The constant head permeability test is usually preferred for coarse-grained soils, and
the variable head permeability is preferred for silts and clays. Laboratory k determina-
tion does not represent the real field conditions and hence is not reliable. But silts on
undisturbed samples might give better results.
10.11 Compaction tests (both standard and modified) are satisfactory for cohesive soils.
A knowledge of the maximum dry density is obtainable in the field using a suitable
roller and adopting a moulding water content almost equal to the OMC.
10.12 In a consolidation test, a floating ring reduces the frictional loss along the sides and
the test is faster. A fixed ring has the advantage of providing the k value of the sample.
Larger samples provide more reliable results.
10.13 The unconfined compressive strength test is a quick test and gives the approximate
shear strength of a cohesive soil. Further, in the test, failure occurs along the weakest
portion and hence provides a conservative shear strength value.
10.14 Failure in a direct shear test may be considered to occur at the maximum shear
stress or at the maximum obliquity of the Mohr failure envelope. The φ angle
obtained considering the maximum shear stress is less than the others, which is on
the safe side.
10.15 In order to obtain accurate results from the triaxial test, the pore pressure and volume
measuring devices should be perfectly air dried. A special type of loading ram has to
be used for testing sensitive clays. For certain field conditions, the samples have to be
consolidated anisotropically.
10.16 The CBR is expressed as the percentage of the force per unit area required to penetrate
a soil mass with a standard plunger compared with that required for the correspond-
ing penetration in a standard material.
M10_PURU1773_01_SE_C10.indd 380 2/12/2013 8:00:11 AM

QUESTIONS
Objective Questions
(1) The size of particles smaller than 0.075 mm is generally obtained from a wet
mechanical analysis.
(2) The rate of secondary compression is dependent on the specimen thickness.
(3) Changes in the laboratory temperature affects the permeability, which in turn
affects the coefficient of consolidation.
(4) The vane shear test cannot be used where the apparent angle of internal friction
(φu) is not equal to zero.
(5) In an unconfined compression test, the inclination of the failure plane is always 45°.
10.2 Increase in permeability of a soil results due to change from
(a) Large to small size particles for the same void ratio
(b) High to a low viscous fluid at the same temperature
(c) Dry side of optimum to wet side in a compacted specimen at the same porosity
(d) Flocculated to dispersed structure at the same dry density
10.3 For a highly fissured clay the best method of finding the undrained shear strength is
(a) The direct shear test
(b) The triaxial shear test with σ3 = 0
(c) The field vane shear test
(d) The unconfined compression test
10.4 The effective shear strength parameters of a sand can be obtained by conducting
(a) Consolidated undrained tests on saturated samples in triaxial shear
(b) Unconsolidated undrained tests on saturated specimens with pore water pressure
measurement in triaxial shear
(c) The field vane shear test with the a low rate of loading
(d) Consolidated undrained tests on dry sand in direct shear
10.5 In a laboratory consolidation test, will the coefficient of consolidation alter (answer
yes or no)
(a) If the pore fluid is replaced by salt water?
(b) If the rate of loading is changed?
(c) If the room temperature is increased?
(d) If the load duration is reduced?
10.6 State the maximum and minimum sizes of particles which may be determined by
hydrometer analysis and give reasons for these limitations.
10.7 What are the inherent errors in using Stokes law to determine the grain-size distribu-
tion of fine-grained soils?
M10_PURU1773_01_SE_C10.indd 381 2/12/2013 8:00:11 AM

10.8 Explain the reasons for plotting a grain-size distribution curve on a semi-logarithmic
plot rather than on an arithmetic scale.
10.9 Why should only distilled water be used in running the tests for limits?
10.10 In an Atterberg limit test, the drop of the cup was found to be 0.95 cm. If the liquid
limit as obtained was 72%, is the true liquid limit is greater or lesser than 72%?
10.11 During the determination of the volume of a soil pat, a certain quantity of air was
entrapped between the plate and the pat. How will this affect the shrinkage limit
result?
10.12 Explain why the dry density is used instead of the wet density in describing the den-
sity of a soil mass.
10.13 How are we justified in using laboratory methods for determining the coefficient of
permeability of soils?
10.14 How do you ensure that saturation is complete in a variable head permeability test
specimen?
10.15 What are the effects of the friction on the loading ram in a triaxial test on the shear
strength of a soil? How will you eliminate the friction on the loading ram?
10.16 How will you ensure when failure occurs in a soil specimen tested in the direct or
triaxial shear apparatus?
10.17 Explain the reasons for the loss in strength of clay as a result of re-moulding.
10.18 Explain the influence of the end restraint on the triaxial shear test on saturated
specimens.
10.19 It is always preferable to obtain samples for the consolidation test from strata under-
lying a building site. Why?
10.20 In conventional laboratory compression testing, what is the cause of soil volume
decrease?
M10_PURU1773_01_SE_C10.indd 382 2/12/2013 8:00:11 AM

11
Lateral Earth Pressure
CHAPTER HIGHLIGHTS
Limit analysis and limit equilibrium methods – Earth pressure at rest –
Rankine’s states of plastic equilibrium – Rankine’s earth pressure theory –
Coulomb’s earth pressure theory – Culmann’s graphical method – Poncelet’s
graphical method
11.1 INTRODUCTION
In the field of civil engineering there are many problems associated with lateral earth
pressure. Some of the structures which require an estimation of lateral pressure for their
design are retaining walls, sheet pile walls, buried pipes, basement walls, braced exca-
vations, cofferdams, thrust blocks, and others.
Lateral pressures most typically develop against structures supporting soil or water.
While designing retaining structures for waterfront facilities, such as cofferdams, quay
walls, etc., one must consider the effects of both soil and water pressure. Lateral pres-
sure depends on several factors, such as physical and time-dependent behaviour of soil,
soil deformation, surface roughness of wall, and movement of retaining structure and
imposed loading.
The state of stress in the backfill of a retaining structure depends on the movement
of structure with reference to the backfill. The backfill material is said to be in a state
of elastic equilibrium when the stress involved and the corresponding strain are within
elastic limits. This generally occurs for no or very little movement of the wall. Fur-
ther increase in stresses develops shear stresses at some point in the body, reaching the
shear strength of the soil. Subsequent increase in stresses causes a substantial increase
in strain, producing a condition known as plastic flow. The soil mass prior to the onset
of the plastic flow condition is said to be in a state of plastic equilibrium, and the load
M11_PURU1773_01_SE_C11.indd 383 2/12/2013 7:58:26 AM

or stress in this condition is referred to as the collapse load. The determination of the collapse
load, adopting plasticity theory, is rather complex. However, plasticity theory also provides
simplified analyses (as discussed below).
11.2 LIMIT ANALYSIS AND LIMIT EQUILIBRIUM METHODS

All the earth pressure problems, viz., earth-retaining structures, bearing capacity of founda-
tions, and slope stability, may be solved by limit analysis or limit equilibrium methods.
The limit analysis method is based on a yield criterion and its associated flow rule, which
considers the stress–strain relationship (Drucker and Prager, 1952). It can be used to cal-
culate lower and upper bounds to the true collapse load. By a suitable choice of stress and
velocity fields, the analysis may produce the same result, which would then be the exact
value of the collapse load.
The lower-bound theorem states that if an equilibrium distribution of stress can be found
which balances the applied loads and the boundary conditions and nowhere violates the
yield criteria, which includes c and φ, the soil mass will not fail or will be just at the point of
failure.
The upper-bound theorem states that the collapse will occur if, for a compatible plastic
deformation, the rate at which the external forces do work on the body equals or exceeds the
rate of internal dissipation of energy.
Limit equilibrium analysis considers a limiting value that can be reached when the forces
acting to cause failure balance the forces resisting failure. This method adopts the following
basic elements:
1. An assumed failure surface of a simple shape (e.g., planar, circular, or log-spiral) is con-
sidered.
2. A reasonable assumption about the stress distribution along the failure surface is made.
3. An estimation of mobilized shear strength is made, and the same is assumed to act simul-
taneously along the failure surface.
Based on the above basic elements, an overall equilibrium equation is developed, and the
problem is solved by simple statics. Thus, the limiting values, viz., earth pressure on retain-
ing structures, bearing capacity of foundations, and factor of safety of slopes, are computed.
Most practical problems are statically indeterminate and need assumptions regarding force
systems and directions of their applications.
The application of limit analysis to practical problems has not yet been completely suc-
cessful because of difficulties in obtaining a proper stress–strain relationship. The limit
equilibrium method has been in wide use because of its simplicity. However, sufficient
judgement has to be exercised while making assumptions about the shape of the slip surface
and stress distribution.
11.3 EARTH PRESSURE AT REST

During the formation of a soil deposit, the soil mass at a point is acted on by the vertical
geostatic stress, σv, of the overburden. This vertical stress causes a vertical compression of
the soil and at the same time produces a lateral strain. This lateral strain is completely
M11_PURU1773_01_SE_C11.indd 384 2/12/2013 7:58:26 AM

Lateral Earth Pressure 385
restrained due to formation of all-round lateral stress of equal magnitude. With time, the
vertical compression and lateral creep strains become zero, and a stable state of stress is cre-
ated. Because of zero strain, a situation of effective vertical and horizontal stresses is attained.
This state of equilibrium is called the at-rest condition or K0-condition.
Consider an element of soil in such a homogeneous and isotropic soil bounded by a level
ground surface. The effective horizontal and vertical stresses are shown in Fig. 11.1a. For the
at-rest condition, the ratio of horizontal to vertical stress is called the coefficient of lateral stress
at rest or lateral stress ratio at rest or coefficient of earth pressure at rest, K0, that is,
σ h′ 0
K0 = (11.1)
σ v′
where
σ v′ = γ z (11.2)
and σ h′ 0 is the effective lateral stress for the at-rest condition; that is,
σ h′ 0 = K0 σ v′ (11.3)
Level ground surface
Homogeneous and
z isotropic soil
σ v′= g z
σh = K0 σ v′
0
(a) Sub-surface stresses in the soil mass

Shear stress
At plastic
Failure envelope
equilibrium
At-rest
σh0 = K0 σ v′ condition
σ ′v = g z
Effective normal stress

(b) Stress related to failure envelope for the at-rest condition
Fig. 11.1 Subsurface stresses for the at-rest condition
M11_PURU1773_01_SE_C11.indd 385 2/12/2013 7:58:27 AM

These stresses are represented by a Mohr’s circle along with the shear strength envelope
in Fig. 11.1b. The location of Mohr’s circle well below the failure envelope indicates a stable
equilibrium condition. An increase in stresses would still keep the soil in an elastic equilib-
rium until the stresses are increased further to cause a failure, and the soil is then said to be at
plastic or limiting equilibrium. Mohr circles for these two conditions are shown in Fig. 11.1b.
In general, for many situations, K0 < 1, except in over-consolidated clays (OCC) where K0
may be as high as 3.0. For normally consolidated clays (NCC), K0 < 1, and for sand deposits,
K0 varies from 0.40 to 0.50.
It is impossible to determine K0 by measuring σ h′ 0 in situ. Therefore, certain correlations
have been suggested. Brooker and Ireland (1965) have suggested correlations (Eqs. 11.4 to
11.6) for K0 in terms of the plasticity index Ip and effective friction angle φ′; that is,
K0 = M − sin φ ′ (11.4)
where
M = 1 for NCC and non-cohesive soils
= 0.95 for OCC for over-consolidation ratio (OCR) > 2
For NCC, K0 is also given as
K0 = a + b I p (11.5)
where
a = 0.40 and b = 0.007 for 0% < Ip < 40%
a = 0.64 and b = 0.001 for 40% < Ip < 80%
and for OCC,
(K 0 )OCC ≈ (K 0 )NCC OCR (11.6)

The term K0 may be related to Poisson’s ratio, ν, based on the theory of elasticity. The
equation of lateral strain is given as
1⎡
εh = σ h′ − ν (σ h′ + σ v′ )⎤⎦⎥ (11.7)
E ⎣⎢
For the no lateral strain condition, εh = 0 and σ h′ = σ h′ 0. Then,
σ h′ 0 ν
K0 = = (11.8)
σ v′ 1− ν
11.4 RANKINE’S STATES OF PLASTIC EQUILIBRIUM

Suppose every part of a semi–infinite mass (say, sand) at the K0-condition is brought on the
verge of failure either by stretching or by compressing, then such a state is called the general
state of plastic equilibrium. General states of plastic equilibrium are not possible in normal
practical problems except when a geological process is involved. Normal practical problems
of interest cause deformations only to a limited extent (for example, yielding of a retaining
wall), and hence a local state of plastic equilibrium is produced.
M11_PURU1773_01_SE_C11.indd 386 2/12/2013 7:58:28 AM

Suppose a rigid, frictionless, infinite wall, backfilled with a dry non-cohesive soil, is
allowed to move a slight distance away from the retained soil mass (Fig. 11.2a). The soil
starts to expand or stretch in the direction following the movement of the wall, resulting
in the decrease of horizontal stress from the initial at-rest condition. When adequate lateral
movement has occurred, the horizontal stress is decreased to a certain magnitude such that
the full shear strength of the soil is mobilized. There is no possibility for a further reduction
in the horizontal stress, and such a stress condition is called the active stress, σ h′ a , and the ratio
Direction of
Settlement
wall movement
σ ′v z
σ h′
a
Expanded configuration
Original configuration
(a) Wall and soil movement for the active case
Failure envelope
Shear stress
At active stage
Slip plane At-rest conditon
φ′ θf
σ ′3 = σh′ Effective normal stress
a
σ ′3 = σh′
0
σ 1′ = σ v′
(b) Mohr’s circle related to the active state
θf θ f = 45° + φ /2
(c) Inclination of slip planes for active state
Fig. 11.2 Rankine’s state of plastic equilibrium – active state
M11_PURU1773_01_SE_C11.indd 387 2/12/2013 7:58:28 AM

of horizontal to vertical stress is referred to as the coefficient of active stress or coefficient of active
earth pressure, Ka.
σ h′ a
Ka = (11.9)
σ v′
We know that
⎛ φ′ ⎞ ⎛ φ′ ⎞
σ3′ = σ1′ tan 2 ⎜⎜ 45°− ⎟⎟⎟ − 2c ′ tan ⎜⎜ 45°− ⎟⎟⎟ (11.10)
⎜⎝ 2⎠ ⎜⎝ 2⎠
Here, c ′ = 0, σ3′ = σ h′ a, and σ1′ = σ v′ . Then,

⎛ φ ′ ⎞ 1 − sin φ ′
Ka = tan 2 ⎜⎜ 45°− ⎟⎟⎟ = (11.11)
⎜⎝ 2 ⎠ 1 + sin φ ′
This condition is indicated by the Mohr’s circle in Fig. 11.2b. Figure 11.2c also shows
the inclination of slip planes for Rankine’s active state in the laterally expanding soil.
In Fig. 11.2a, the expanded configuration of the soil element is also shown.
Now let us suppose that the soil is compressed due to inward wall movement. This causes
an increase in the horizontal stress from the at-rest condition (Fig. 11.3a). When sufficient
lateral movement occurs, the maximum shear strength of the soil is mobilized and the hori-
zontal stress is maximum. This state of failure condition is called Rankine’s passive state, the
horizontal stress is called the passive stress, σ h′ p , and the ratio of horizontal to vertical stress is
referred to as the coefficient of passive stress or coefficient of passive earth pressure, Kp:
σ h′ p
Kp = (11.12)
σ v′
Substituting c′ = 0 and σ3′ = σ1′ = σ h′ p in Eq. 11.10,
⎛ φ ′ ⎞ 1 + sin φ ′
K p = tan 2 ⎜⎜ 45° + ⎟⎟⎟ = (11.13)
⎜⎝ 2 ⎠ 1 − sin φ ′
This condition of stress is represented by the Mohr’s circle in Fig. 11.3b, and the inclina-
tion of slip lines for Rankine’s passive state is shown in Fig. 11.3c. Figure 11.3a also shows
the compressed configuration of a soil element.
Now we recognize three lateral stresses depending on the strain or displacement experi-
enced by the backfill soil, viz.,
σ h′ 0 = K0 σ v′ At-rest condition Zero displacement

σ h′ a = Ka σ v′ Active condition Movement causing expansion
σ h′ p = K pσ v′ Passive condition Movement causing compression
The horizontal displacement required to attain the active state is substantially less than
that required to obtain the passive state.
The active state is a condition of loosening strains, where the frictional resistance is mobi-
lized to reduce the force necessary to hold the soil in position. As the soil cannot stretch
M11_PURU1773_01_SE_C11.indd 388 2/12/2013 7:58:30 AM

Direction of Heave
wall movement
Original
σ ′v configuration z
σ h′ p
Compressed
configuration
(a) Wall and soil movement for the passive case
Failure envelope
Shear stress
At passive state
At rest
Slip plane
conditon
f′ θf
σ ′3 = σh′ 0 Effective normal stress
σ ′3 = σ v′
σ ′1 =σ h′ p
(b) Mohr’s circle related to the passive state
θ f = 45°– f/2
(c) Inclination of slip planes for the passive state
Fig. 11.3 Rankine’s state of plastic equilibrium – passive state
more, the magnitude of this strain is less. On the other hand, a passive state is a condition of
densifying the soil by a lateral strain, where the frictional resistance is mobilized to increase
the force to cause more strain. Figure 11.4 illustrates the relative movements and the order
of magnitude of lateral earth pressure coefficients. For example, when φ ′ = 30°, K a = 0.333,
and K p = 3.0; then K P = 10 Ka .
11.5 RANKINE’S EARTH PRESSURE THEORY

In the course of various attempts at designing of earth-retaining structures, several earth
pressure theories have been suggested since 1687. Coulomb’s and Rankine’s are perhaps the
two best-known theories and are frequently referred to as classical earth pressure theories.
M11_PURU1773_01_SE_C11.indd 389 2/12/2013 7:58:31 AM

Usual range of earth

pressure coefficients
Relatively
Small Cohesionless Cohesive
large soils soils
Δ Δ
Lateral pressure, σ h′
Passive
3–14 1–2
At rest
0.4–0.6 0.4–0.6
Active 0.33–0.22 1–0.5
Away from 0 Against Wall movement, Δ

backfill backfill
Fig. 11.4 Relative wall movements and earth pressure coefficients (Source: Bowles, 1982)
The concept of Rankine’s state of plastic equilibrium can be applied to evaluate the lateral
earth pressure that acts against various retaining structures. Rankine’s theory (1857) is based
on the assumptions that (i) a conjugate relationship exists between the vertical and lateral
pressures, (ii) the mass of soil is homogeneous and isotropic, (iii) the soil is dry and non-
cohesive, and (iv) the wall is vertical and smooth.
Let us consider the general case of a sloping, dry, non-cohesive backfill behind a smooth
vertical wall. The element of soil in Fig. 11.5a depicts this condition. It is evident that these
are conjugate stresses acting on conjugate planes where these planes are not principal planes
(Fig. 11.5b).
Consider Mohr’s circle in Fig. 11.5c for the active condition. Draw a line passing through
the origin with an inclination i, the slope angle, which cuts the Mohr circle at A and C. Now
OC and OA represent the vertical and lateral stresses, respectively. Drop a perpendicular DB
to the slope line from the centre D of the Mohr circle.
Now,
OA OB − AB
= (11.14)
OC OB + AB
OB = OD cos i
r = OD sin φ ′
BD = OD sin i
AB = r 2 − BD 2 = (OD sin φ ′)2 − (OD sin i)2
OC = γz cos i
Substituting the above in Eq. 11.14, we have

⎡
OA ⎢ (OD)cos i − (OD) sin φ ′ − sin
2 2
i ⎤⎥
=⎢ ⎥ (11.15)
OC ⎢ (OD)cos i + (OD) sin 2 φ ′ − sin 2 i ⎥⎦
⎣
M11_PURU1773_01_SE_C11.indd 390 2/12/2013 7:58:32 AM

i >φ
45° + φ /2
σ v′ = γ z cos i
z
90° + φ
H τ
σ h′ τ = γ z ′ sin i cos i
(a) Sloping granular backfill (b) Inclination of slip planes
τ φ′
C i
B
A
r
r
0 σ h′
σh′ cos i D
σv′ cos i
(c) Mohr’s circle for active state–sloping backfill
Fig. 11.5 Lateral pressure and slip planes in granular sloping backfill
Reducing after substituting,
sin 2 φ ′ = 1 − cos 2 φ ′ and sin 2 i = 1 − cos 2 i
we have
⎡
OA ⎢ cos i − 1 − cos φ ′ − 1 + cos
2 2
i ⎤⎥
=⎢ ⎥
OC ⎢ cos i + 1 − cos 2 φ ′ − 1 + cos 2 i ⎥⎦
⎣
or
⎡ ⎤
OA ⎢ cos i − cos i − cos φ ′ ⎥
2 2
=⎢ ⎥ (11.16)
OC ⎢ cos i + cos 2 i − cos 2 φ ′ ⎥
⎣ ⎦
or
⎡ cos i − cos 2 i − cos 2 φ ′ ⎤
σ h′ a = ⎢⎢ ⎥ γ z cos i
⎥ (11.17)
⎢ cos i + cos 2 i − cos 2 φ ′ ⎥
⎣ ⎦
Let pa = σ h′ a ; then,
pa = Ka γ z (11.18a)
M11_PURU1773_01_SE_C11.indd 391 2/12/2013 7:58:34 AM

where
⎡ cos ⎤
i − cos 2 i − cos 2 φ ′ ⎥
Ka = cos i ⎢⎢ ⎥ (11.18b)
⎢ cos i + cos 2 i − cos 2 φ ′ ⎥
⎣ ⎦
Therefore,
Pa = ½γ H 2 Ka (11.19)
For the passive case,
pp = K p γ z (11.20a)
where
⎡ ⎤
⎢ cos i + cos i − cos φ ′ ⎥
2 2
K p = cos i ⎢ ⎥ (11.20b)
⎢ cos i − cos 2 i − cos 2 φ ′ ⎥
⎣ ⎦
Therefore,
Pp = ½γ H 2 K p (11.21)
Equations 11.19 and 11.21 are Rankine’s expressions for the lateral pressures for a wall of
height H with a backfill of unit weight γ for the active and passive cases. These forces act at
a height of H/3 from the base inclined at an angle i to the normal of the wall.
11.5.1 Effect of Level Backfill Surface

Level ground surface is a simplified condition and is most often adopted in practice. Consid-
ering that all the conditions remain identical except for i = 0, the expressions Ka and Kp
reduce to those for Rankine’s fundamental states of plastic equilibrium. Setting i = 0 in Eqs.
11.18b and 11.20b, we have
1 − sin φ ′
Ka = (11.22)
1 + sin φ ′
and
1 + sin φ′
Kp = (11.23)
1 − sin φ′
For this situation, Mohr’s circle is re-drawn as in Fig. 11.6. Thus, the total active thrust Pa
is given as
Pa = ½γ H 2 Ka (11.24)
2
Pp = ½γ H K p (11.25)
where Ka and Kp are taken from Eqs. 11.22 and 11.23, respectively. The pressure distribution
and the point of application of the forces are shown in Fig. 11.7.
Further discussions are confined to a level backfill surface but with different backfill
materials and loading conditions.
M11_PURU1773_01_SE_C11.indd 392 2/12/2013 7:58:35 AM

φ′
σ 2′
σ 1′–
2
0
σ3′ = sa σ′
σ1′ = σ v′
Fig. 11.6 Mohr’s circle for active state – level backfill
H Kγ z 1 γ H 2K
P=
2
H/3 { P = Pa or Pp
K = Ka or Kp
KγH
Fig. 11.7 Lateral pressure distribution
11.5.2 Effect of Surcharge Load on Backfill Surface

Consider a dry, non-cohesive level backfill (Fig. 11.8a) with a uniform surcharge load q
applied all over the surface. It may be assumed that the vertical effective stress is increased
by the amount of surcharge. Then, at any depth z,
σ v′ = γ z + q (11.26)
Surcharge Backfill
q Kq
σ v′ = γz +q z z
Kγz
H Kq + = P
{ P = Pa or Pp
K = Ka or Kp x
Kq KγH Kq + Kγ H
(a) Backfill with surcharge (b) Lateral pressure distribution
Fig. 11.8 Lateral pressure due to uniform surcharge of a level backfill
M11_PURU1773_01_SE_C11.indd 393 2/12/2013 7:58:36 AM

Thus the lateral pressure is increased by an amount Kaq or Kpq as the case may be. There-
fore, for the active case,
pa = Ka γ z + Ka q (11.27)
and for passive case,
Q − Line load
Parallel to φ -line
C
B D
Parallel to failure
a surface
Failure surface
ab
3
δ
γ
φ -line
Pa
δ = Angle of wall friction

b
θf
φ
A
(a) Line load left of the slip plane
Q − Line load
C
B
a Parallel to φ -line
Failure surface
ab
3
γ
δ
Pa
φ-line
θf
b φ
(b) Line load right of the slip plane
Fig. 11.9 Procedure for estimating the line of action of the resultant active thrust Pa caused by a
line load (Source: Dunn et al., 1980)
M11_PURU1773_01_SE_C11.indd 394 2/12/2013 7:58:37 AM

pp = K p γ z + K p q (11.28)
This shows that the lateral pressure varies linearly with depth due to unit weight and
remains constant with depth due to surcharge load. Figure 11.8b illustrates the pressure dis-
tribution. The area of the entire diagram gives the active thrust, Pa, or the passive resistance,
Pp on the wall. The line of action of Pa and Pp can be determined by considering the moments
of the individual and total areas about the base.
Concentrated surcharge loads Q (Fig. 11.9) running parallel to the wall may be induced
on the backfill (e.g., continuous footing, railroad tracks, etc.). Increased stresses on the wall
due to concentrated surcharge may be computed based on Boussinesq’s equation. It is
a laborious procedure and generally not recommended. However, graphical methods
(discussed elsewhere) are more expedient for this purpose (refer Example 11.8).
The point of action of active thrust or passive resistance is obtained by following the
procedure detailed below. The failure surface is located using any graphical method. The
concentrated load may lie within or away from the failure wedge. If the concentrated load is
within the failure wedge, then lines Db and Da are drawn parallel to the failure surface and
the φ′ line, respectively (Fig. 11.9a), and points a and b are located. If the concentrated load
is away from the failure wedge (Fig. 11.9b), then the heel represents point b, and point a is
located as explained earlier. Then, the point of application of the active thrust is at a distance
of ab/3 from point a.
11.5.3 Effect of Water Table on a Backfill

Consider again a non-cohesive level backfill with the water table at the surface. The vertical
stress at any depth z can be split into two, viz., one due to soil grains and the other due to
water; that is,
σ v′ = γ ′z + γ w z (11.29)
Hence, the active and passive cases will be

Pa = Ka γ ′z + K w γ w z
and
Pp = K p γ ′z + K w γ w z (11.30)
where Kw is the lateral coefficient for water, which is always 1. Thus,
Pa = Ka γ ′z + γ w z (11.31)
and
Pp = K p γ ′z + γ w z (11.32)
The pressure distribution is shown in Fig. 11.10. As before, Pa and Pp can be determined
from the area of the diagrams.
M11_PURU1773_01_SE_C11.indd 395 2/12/2013 7:58:38 AM

Soil grains Water
H + =
Kγ ′z Kγ ′z−γw z
P
{ P = Pa–Pp
K = Ka–Kp
Kγ ′H γwH Kγ ′H γw H
Fig. 11.10 Pressure distribution when the water table is at the surface
Let us consider a partial submergence now. Let the water table be at a depth of H1 from
the level surface (Fig. 11.11). Since the soil is non-cohesive, it is reasonable to assume that the
pore pressure above the water table is everywhere zero. The dry unit weight is considered
for soil above the water table and submerged unit weight below. Below the water table the
unit weight of water is also taken into account. Thus,
pa = Ka γd z ⎪⎫⎪
⎬ for (0 < z ≤ H1 ) (11.33)
pp = K p γd z⎪⎪
⎭
and
pa = Ka γ ′z + γ w z + K a γd H1 ⎫⎪⎪
⎬ for [H1 ≤ z ≤ ( H1 + H 2 )] (11.34)
pp = K p γ ′z + γ w z + K p γd H1 ⎪⎪
⎭
The problem may also be solved by assuming the first layer to act as a surcharge load,
q (q = γd H1 ), on the second layer. The pressure diagram is drawn for the retaining wall with
height H1. The second layer is treated separately, as if loaded by a surcharge load q and the
water table at the surface, the height of the retaining wall being H2. The pressure distribution
is shown in Fig. 11.11. Values of Pa and Pp are obtained from the area of the diagrams.
z
H1
Kγ ′z
H KγdH1 z Kγ ′z γw z
H2 + + = P
P=Pa orPp
z
K=Ka orKp
Kγd H1 Kγ ′H2 γw H2 KγdH1 Kγ ′H2 γw H2
First layer Second layer Second

soil grains soil grains layer water
Fig. 11.11 Pressure distribution for partial submergence in a backfill
M11_PURU1773_01_SE_C11.indd 396 2/12/2013 7:58:39 AM

f1′, g1
H1 First layer K1 g1 H1 K1 g1 H1
K1 =Ka1 or Kp1
H f2′, g2
H2 P
Second layer + =
K2 = Ka2 or Kp2 x
P = Pa or Pp K2g1H1
K2g2 H2 K2g1 H1+K2g2 H2
(a) Layered soils when f1′ > f2′
f1′, g1
H1 K1 g1 H1
First layer
H f2′, g2 K2g1 H1
H2 P
Second layer
K2 = Ka2 or Kp2 x
P = Pa or Pp
K1g1 H1 +K2g2 H2
(b) Layered soils when f1′ < f2′
Fig. 11.12 Pressure distribution for layered soils in a backfill
11.5.4 Effect of Stratified Soils in the Backfill

Consider two dry, non-cohesive soils in the level backfill. This is similar to the case of partial
submergence. Here the angles of shearing resistances are different in the two layers, whereas
in the previous case they were the same in both the layers. If φ′1 and φ′2 are the angles of
shearing resistances, γ1 and γ2 the unit weights in the top and bottom layers of heights H1
and H2, and Ka1, Kp1 and Ka2, Kp2 the lateral coefficients for the respective layers, then
pa = Ka1 γ1 z and pp = K p1 γ1 z for (0 ≤ z ≤ H1 ) (11.35)
pa = Ka2 γ 2 z + K a2 γ1 H1 and
pp = K p2 γ 2 z + K p2 γ1 H1 for [H1 ≤ z ≤ ( H1 + H 2 )] (11.36)
Here again, for the second layer, the first layer acts like a surcharge. Similar methods may
be used if the number of layers is more than two. For a three-layer case, the top and middle
layers will act as surcharges on the bottom layer.
The pressure distribution is shown in Fig. 11.12, and the values of Pa and Pp are calculated
as usual.
11.5.5 Effect of c–φ Soils as Backfill

Rankine’s theory was originally proposed only for non-cohesive soils. It was extended by
Bell (1915) to include c–φ soils, and its application has also been widened. Consider a dry c–φ
soil level backfill. We know that
1 + sin φ ′ 1 + sin φ ′
σ1′ = σ3′ + 2c ′ (11.37)
1 − sin φ ′ 1 − sin φ ′
M11_PURU1773_01_SE_C11.indd 397 2/12/2013 7:58:40 AM

or
1 − sin φ ′ 1 − sin φ ′
σ3′ = σ1′ − 2c ′ (11.38)
1 + sin φ ′ 1 + sin φ ′
For the active condition, σ h′ a = σ3′ , σ v′ = σ1′ , and
1 − sin φ ′
Ka =
1 + sin φ ′
Thus,
σ h′ a = σ v′ Ka − 2c ′ Ka
Let pa = σ h′ a and σ v′ = γ z . Then,

pa = Ka γ z − 2c ′ K a (11.39)
For the passive condition, σ h′ p = σ1′ , σ v′ = σ3′ , and

1 + sin φ ′
Kp =
1 − sin φ ′
Substituting the values of σ h′ p , σ v′ , and Kp in Eq. 11.37, we have
σ h′ p = σ v′ K p + 2c ′ K p
Let pp = σ h′ p and σ v′ = γ z . Then,
p p = K p γ z + 2c ′ K p (11.40)
The pressure diagrams for these two cases are shown in Fig. 11.13a and b. In the active
case, there exists a tension up to a depth of z = z0, where the active pressure pa = 0. Further,
at the surface (z = 0), the active pressure has a value of pa = − 2c ′ K a .
The soil up to a depth of z0 will be in a state of tension and will neither impart any pres-
sure on the wall nor provide support. When the tension is released, tension cracks will
develop from the surface up to a depth of z0. From a practical point of view, the tension zone
is ignored, and the active thrust is calculated only for the height (H–z0) from the base (the
pertinent area is shown shaded in Fig. 11.13a); that is, let pa = pa′ at z = H ; then,
pa′ = Ka γ H − 2c ′ K a (11.41)
Thus,
Pa = 12 pa′ ( H − z0 ) (11.42)
is acting at a height of (H – z0)/3 from the base.
The depth of the tension zone, z0, is obtained by setting pa = 0 with z = z0 in Eq. 11.39:
2c ′
z0 = (11.43)
γ Ka
M11_PURU1773_01_SE_C11.indd 398 2/12/2013 7:58:43 AM

2c Ka
Unit weight Cohesion
z z = z0
H 2c Ka
– =
Ka γ z H –z0
Pa
H –z0
3
Ks γ H 2c Ka Ka γ H – 2c Ka
(a) Active case p′a
Unit weight Cohesion
z
2c Kp
H
+ =
Kp γ z
Pp
Kpγ H Kpγ H +2c Kp

2c Kp
(b) Passive case
Fig. 11.13 Active and passive pressure distributions: c–φ′ soil as backfill
The existence of a tension zone in c–φ soils suggests that an unsupported excavation
would be theoretically possible. The maximum unsupported depth of excavation, Hc, may
be theoretically taken as 2 z0, where the tensile stress is equal to the cohesive strength.
Hence,
4c ′
H c = 2 z0 = (11.44)
γ Ka
The application of this theoretical depth in practice should be done more cautiously.
Any factor which reduces the cohesion (e.g., the possibility of water entering the crack and
causing a reduction in shear strength) may affect the estimation drastically. However, such
depths may be adopted for short-duration works.
For many reasons, a cohesive backfill is not recommended in practice since changes in
water content may significantly alter the performance of a retaining wall due to frequent
swelling and shrinking of the soil. Further, placing, densifying, and maintaining a cohesive
backfill is extremely difficult.
M11_PURU1773_01_SE_C11.indd 399 2/12/2013 7:58:43 AM

Analytical solutions (for both active and passive conditions) based on Rankine’s theory
have been given by Babu Shankar (1980) for a c–φ soil sloping backfill with surcharge. These
general expressions readily reduce to a particular case depending on the problem.
11.6 COULOMB’S EARTH PRESSURE THEORY

Rankine’s earth pressure theory and its applications discussed in the previous section were
based on the assumption that the surface of the retaining wall is frictionless. But in reality,
the wall is more or less rough, and thus, in Rankine’s theory errors do occur which might be
on the safe side.
Coulomb, in 1776, developed an earth pressure theory which includes the effect of friction
between the backfill and the wall. The theory considers a dry, non-cohesive inclined backfill,
and the lateral earth pressure required to maintain the equilibrium of a sliding wedge with
a plane slip surface is calculated.
Consider a retaining wall (Fig. 11.14) with its backface inclined at an angle β with the
horizontal. Let i be the inclination of the backfill with the horizontal. The roughness of the
wall is represented by the angle of wall friction (i.e., the angle of friction between the soil
and the wall, δ).
It is presumed that the wall has moved sufficiently outwards such that the active-state
condition is created. The movement causes a wedge ABC1 to slide along the plane surface
BC1. The forces acting on the wedge are
1. weight of the wedge, W;
2. the resultant R on the surface BC1, inclined at angle φ′ to the normal (as the sliding takes
place between soil and soil); and
3. the active thrust Pa, inclined at an angle δ to the normal to the wall.
Among the three forces, the magnitude for one (i.e., W ) and the direction of all three are
known; the force polygon can be drawn, and the value of Pa, corresponding to the assumed
wedge ABC1, can be determined. Different trial wedges are taken and the corresponding Pa
values determined. The maximum value of Pa may be mathematically expressed as
Pa = ½γ H 2 Ka (11.45)
where Ka is Coulomb’s active earth pressure coefficient
sin 2 (β + φ ′)
= (11.46)
sin 2 β sin(β − δ ){1 + [sin(φ ′ + δ ) sin(φ ′ − i)/sin(β − δ ) sin(i + β )]}2
The mathematical treatment for Eq. 11.46 is beyond the scope of this book. The active
thrust Pa acts at a height H/3 from the base and is inclined at an angle δ to the normal of the
back surface.
For computing Coulomb’s passive resistance Pp, consider the retaining wall and other
details shown in Fig. 11.14b. Again, taking into account the forces which contribute to the
equilibrium, viz., W, R, and Pp, the force triangle is drawn and the passive resistance Pp
determined. The minimum value of Pp from the plot of Pp versus wedge locations represents
Coulomb’s passive resistance. Mathematically, this can be expressed as
Pp = ½γ H 2 K p (11.47)
M11_PURU1773_01_SE_C11.indd 400 2/12/2013 7:58:44 AM

Pa mass
active
c1
thrust
c2 c0
c3
A Failure surface
Sand
W
Pa
H N
β−δ
φ′
t
d R
H/3 Pa θ−φ′
b qf R
B
(a) Active case
Passive
resistance c c1
Pp min c3 c2 0
A
Failure surface
W τ
R R
W φ f + φ′
H φ′
Pp
N b-d
d
H/3 qf Pp
b
B
(b) Passive case
Fig. 11.14 Coulomb’s active and passive wedges
where Kp is Coulomb’s passive earth pressure coefficient

sin 2 (β − φ ′)
= (11.48)
sin 2 β sin(β + δ ){1 − [sin(φ ′ + δ ) sin(φ ′ + i)/sin(β + δ ) sin(β + i)]}2
The passive resistance Pp acts at a height of H/3 from the base at an angle δ to the normal.
Coulomb’s plane failure surface assumption omits to take into account the actual or true
nature of the failure surface (Fig. 11.15). Although the active pressure is not significantly
affected by the plane surface assumption, this is not the case for passive pressure. The esti-
mated value of Pp is on the unsafe side and increases with increase in wall friction. Therefore,
rigorous analyses have been carried out by researchers, assuming different shapes, such as
circle, ellipse, and log-spiral, for the slip surface.
Retaining walls are generally constructed with mass concrete or masonry. The wall fric-
tion angle, δ, depends on the type of backfill material and the type of wall, as shown in
Table 11.1. The value of δ may be assumed to be between φ′/2 and 2/3φ′.
The above treatment is applicable only for an unloaded backfill. But Babu Shankar (1981)
has provided analytical solutions for the active case based on Coulomb’s theory for inclined
backfill of a c′–φ′ soil with uniform surcharge load.
M11_PURU1773_01_SE_C11.indd 401 2/12/2013 7:58:45 AM

45°+ f/2 45°− f/2 45°− f/2
45
°+
f
H1
/2
H Pp
δ
Pa Failure surface δ Failure surface
(a) Active case (b) Passive case
Fig. 11.15 Nature of failure surface in a soil with wall friction
Table 11.1 Angles of wall friction for masonry

or mass concrete walls
Backfill material Range of δ (°)
Gravel 27–30
Coarse sand 20–28
Fine sand 15–25
Stiff clay 15–20
Silty clay 12–16
Source: Das (1984).
11.7 CULMANN’S GRAPHICAL METHOD

Culmann, in 1875, suggested a graphical procedure to determine the magnitude and loca-
tion of the resultant earth pressures, both active and passive, on retaining walls. This method
can be applied to cases where the backfill surface is level or sloped, regular or irregular and
where the backfill material is uniform or stratified. This method considers the effect of wall
friction and varied surcharge load conditions. Although this method was basically proposed
for non-cohesive soils, it can be extended to cohesive soils also. However, it demands a con-
stant angle of shearing resistance in the backfill.
Consider a retaining wall with a sloping non-cohesive backfill in the active state
(Fig. 11.16). The step-by-step graphical construction is as follows:
1. Choose a suitable scale, and draw the retaining wall along with the backfill.
2. Draw a line AD from the point A, making an angle φ above the horizontal.
3. Draw another line AE at an angle ψ (=β – δ) from the line AD.
4. Consider a wedge ABC1 with AC1 as the slip surface.
5. Determine the weight W1 of the wedge ABC1.
6. Select a convenient force scale and represent W1 on the line AD as AW1.
7. From W1, draw a line parallel to AE to meet the assumed slip surface AC1 at F1.
8. Choose another wedge ABC2 and repeat Steps 5 to 7 and find point F2.
9. Establish similar points, and connect these points of intersection with a smooth curve,
called Culmann’s curve.
10. Draw a tangent to Culmann’s curve parallel to AD. Point F represents such a tangent
point. An irregular curve may have more than one tangent.
M11_PURU1773_01_SE_C11.indd 402 2/12/2013 7:58:45 AM

C4
C3
C2 C
C1 Slip surface
B i
Tangent D
parallel to AD
F3 F4 Culmann’s
F curve
F2
H
F1 W4
δ
W3
W
Pa W2
H/3
W1
β φ′ θ f
ψ X
A
Pressure line
E
Fig. 11.16 Culmann’s graphical method – active case
11. Draw FW parallel to AE. The magnitude of FW, based on the selected scale, represents
the active thrust Pa. If several tangents to the curve are possible, the largest of them
becomes the value of Pa.
12. The failure surface is AFC and is inclined at θf to the horizontal.
Culmann’s procedure for the determination of passive resistance Pp is similar to that for
the active case, with some notable differences (Fig. 11.17): (i) line AD makes an angle φ below
Culmann’s curve
F2 F4
Tangent F F3
parallel to AD
c4
Slip surface c2 c c1
c1
B
H Pp
δ
H/3
β θf
A φ′ X
ψ W1 W2 W W3 W4
D
Pressure line
Fig. 11.17 Culmann’s graphical method – passive case
M11_PURU1773_01_SE_C11.indd 403 2/12/2013 7:58:46 AM

the horizontal and (ii) the pressure line makes an angle ψ (= β + δ) with line AD. Parallel
lines to AE are drawn from these points to meet the assumed slip surface. A Culmann line
is drawn connecting these intersection points. A tangent parallel to AD is drawn to the Cul-
mann curve with the passive resistance being the scaled value of line FW.
Surcharge loads and irregular backfills can be included in the procedure as discussed ear-
lier. Worked examples are given at the end of the chapter to further clarify these conditions.
11.8 PONCELET’S GRAPHICAL METHOD

Based on the principles postulated by Rebhann, Poncelet (1840) suggested a graphical method
to determine the earth pressure on a rough wall for a non-cohesive, homogeneous, and inclined
backfill. The method of construction for the active case is as explained below (Fig. 11.18).
1. Choose a suitable scale and draw the retaining wall along with the backfill.
2. Draw a line AD from the point A, making an angle φ′.
3. Draw another line AE at an angle ψ (=β – δ) from the line AD.
4. Draw a semicircle with AD as the diameter.
5. Draw BF parallel to the pressure line ψ to meet the φ′ line at F.
6. Draw a perpendicular FG at F to meet the semicircle at G.
7. With A as centre and AG as radius, draw an arc to meet AD at J.
8. From J, draw JC parallel to the pressure line to meet the backfill line at C.
9. With J as centre and JC as radius, draw an arc to meet AD at K.
10. Join AC and KC.
11. Find the area of the triangle JKC; then,
Pa = (Area of triangle) × γ
where γ is the unit weight of the backfill. AC represents the failure surface.
D
Slip plane
C
B
i
H
F
δ K
Pa
H/3
β θf
φ′
A ψ
Pressure line
E
Fig. 11.18 Poncelet’s graphical method – active case
M11_PURU1773_01_SE_C11.indd 404 2/12/2013 7:58:46 AM

Slip plane
B i
D
Pp
H
H/3 δ
F
θf
A φ′
ψ
G
J
Pressure line
E
K
Fig. 11.19 Poncelet’s graphical method – passive case
When the slope of the backfill surface i and the angle of shearing resistance φ′ are equal
or nearly equal, slight modifications are made in the procedure. The modifications are high-
lighted in the worked examples.
Poncelet’s procedure for the determination of passive resistance Pp is similar to that for
the active condition, with some notable differences (Fig. 11.19): (i) line AD makes an angle
φ′ below the horizontal and is projected backwards to meet the extended backfill surface at
D and (ii) the pressure line makes an angle ψ (= β + δ) with line AD. Other steps are similar
to those followed for the active case. For all angles of φ′ and i, the procedure is the same, as
these lines are the converging ones, and hence no modification is needed.
11.9 ARCHING OF SOILS

In a supported soil mass, when a certain part of the soil mass yields, then the soil adjoining
the yielding part also gets displaced from its original position. The deformation of the parted
soil is resisted by mobilization of shearing resistance along the zones of contact between the
yielding and non-yielding portions of the soil. As the direction of mobilization of shear
strength is opposed to the direction of deformation of the yielding soil, there is a reduction
in pressure on the yield part of the support and a consequent increase in the pressure of the
adjoining stationary parts. This phenomenon of the transfer of pressure from the yielding
part of a soil mass to the non-yielding part of the mass is referred to as arching.
Consider the yielding of a horizontal strip (Fig. 11.20). The actual failure surfaces will
be curved starting from the yielding point to the ground surface. Let the slip surfaces be
assumed to be vertical rising from the yielding strip to the surface.
M11_PURU1773_01_SE_C11.indd 405 2/12/2013 7:58:46 AM

q
G.S G.S
Actual
slip
surface
Assumed
z slip
surface
()
+ Δsz f
ΔW = g BΔsz
B
Fig. 11.20 Yielding of a horizontal strip
Consider the equilibrium of the yielding slice of width B and unit length; then,
B [σ z + Δσ z ] = Bσ z + ΔW − 2Δ z τ f
Substituting for
ΔW = γ BΔz
and
τ f = c ′ + σ x tan φ ′
and
σx = Kσz
where K is an empirical constant, we have

BΔσ z = γ BΔz − 2c ′Δz − 2Kσ z tan φ ′Δz
or
Δσ z γ − 2c ′ 2K tan φ ′
= − σz
Δz B B
Also, at z = 0, σ z = q.
A solution of the above equation yields the following expression:
B[γ − (2c ′/ B)] ⎡
1 − e(−2 Kz /B) tan φ ⎤⎥ + qe(−2 Kz /B) tan φ
′ ′
σz =
2K tan φ ′ ⎢⎣ ⎦
M11_PURU1773_01_SE_C11.indd 406 2/12/2013 7:58:48 AM

When c ′ = q = 0
Bγ ⎡1 − e(−2 Kz/B) tan φ ′ ⎤
σz =
2 K tan φ ⎢⎣
′ ⎥⎦
For a cohesionless soil c = 0, and for the no surcharge condition q = 0; the above equation
gives the intensity of vertical pressure on the yielding strip, considering that the mobiliza-
tion of shearing resistance takes place along the full length of the surface z.
For the condition of z = ∞, σz becomes constant and independent of z; that is,
Bγ
σ z =∞ =
2 K ′ tan φ ′
It has been reported in the literature that at distances of more than 2.5 B in sand, the yield-
ing of the strip has no effect on the state of stress in sand.
WORKED EXAMPLES
Example 11.1 Compute the total lateral force acting against a 10 m high, vertical, smooth,
unyielding wall, which retains a normally consolidated clay. The soil parameters are
γ = 21kN m 3, I p = 35%. The water table is at the ground surface.
Solution
Since the soil is normally consolidated, the coefficient of earth pressure at rest, K0, is obtained
from Eq. 11.5. Thus,
K 0 = a + bI p
For Ip = 35%, a = 0.40 and b = 0.007, K0 = 0.40 + 0.007×35 = 0.645.

2
Total vertical pressure σ v = γ H = 21×10 = 210 kN m
Therefore, total lateral pressure σ h0 = K0 σ v′ = 0.645× 210 = 135.5 kN m 2
Total lateral force Ph0 = ½ (K0 σ v′ ) H = ½ (135.5)×10 ×1
= 677.5 kN/m length of wall
Example 11.2 An 8 m high vertical, smooth retaining wall above the water table supports
a 15° soil slope. The retained backfill has a unit weight of 18.6 kN/m3, and the shear strength
parameters are c’ = 0 and φ′ = 35°. Compute the total active thrust on the wall, and also find
the directions of the two sets of failure planes relative to the horizontal.
Solution
Draw the failure envelope and the backfill slope inclined at 15° to the origin (Fig. 11.21).
The vertical stress at a depth of 8 m = σ z′ = γ z cos i
σ z′ = 18.6 × 8 × cos 15°
= 143.7 kN m 2
M11_PURU1773_01_SE_C11.indd 407 2/12/2013 7:58:50 AM

E
C
OC = 143.7 kN/m2
35° 60°
A
15° 65°
O σ n′
A′
45 kN
E′
Fig. 11.21
Choose a stress of 1 mm = 1.5 kN and set off this stress (distance OC) along the 15° line.
Draw a Mohr circle passing through this point C and tangential to the failure envelope. Then
measure the distance OA or OA′, which represents the active pressure to scale. Therefore,
pa = 45 kN m 2
Then,
Pa = ½ pa H = ½ × 45× 8 = 180 kN m
The failure planes are parallel to AE and AE′. The directions of these lines are measured as
60° and 65° (= 90° + φ′ = 90° + 35° = 125°).
Example 11.3 For the retaining wall shown in Fig. 11.22, make a sketch of the distribution
of active pressure on the wall, giving the principal values. Compute the thrust per metre
length of the wall neglecting cohesive and frictional forces on the back of the wall.
Solution
This is the same condition as given in Fig. 11.12b.
For the sand layer:
1 − sin 25°
Ka1 = = 0.406
1 + sin 25°
For the gravel layer:

1 − sin 33°
Ka2 = = 0.353
1 + sin 33°
Sand layer:
p1 = K a1 γ1 H1 = 0.406 ×18.2× 3 = 22.17 kN m 2
p1′ = K a2 γ1 H1 = 0.353 ×18.2× 3 = 19.27 kN m 2
M11_PURU1773_01_SE_C11.indd 408 2/12/2013 7:58:51 AM

Sand
22.17 kN/m2
3 m c1 = 0
φ′1 = 25° 22.17
γ ′1 = 18.2 kN/m3 p1
kN/m2
p1′ 19.27
8m Gravel kN/m2
c2 = 0
5m + =
φ′2 = 33° 225.8 kN/m
γ ′2 = 21.8 kN/m3
p2
19.27 38.48 19.27 38.48

kN/m2 kN/m2 kN/m2 kN/m2
Fig. 11.22
Gravel layer:
p2 = Ka2 γ 2 H 2 = 0.353 × 21.8 × 5 = 38.48 kN m 2
The lateral pressure diagram is shown in Fig. 11.22.

Pa = ½ × 22.17 × 3 ×1 + 5×19.27 ×1 + ½ × 38.48 × 5×1 = 225.8 kN m
Example 11.4 An 8 m high retaining wall supports a 5.5 m deep sand ( γd = 18.5 kN/m3,
3
φ = 34°) overlying a saturated sandy clay (γ sat = 20.3 kN / m , φ = 28°, c = 17 kPa). The
groundwater level is located at the interface of two layers. Sketch the lateral stress distribu-
tion up to a depth of 8 m for an active condition.
Solution
For the sand layer:
1 − sin 34°
Ka1 = = 0.283
1 + sin 34°
For the sandy clay layer:

1 − sin 28°
Ka2 = = 0.361
1 + sin 28°
Sand layer:
p1 = K a1 γd H1 − 2c Ka1
p1 = 0.283 ×18.5× 5.5 − 0 = 28.8 kN m 2
p1′ = K a 2 γd H1 = 0.361×18.5× 5.5 = 36.73 kN m 2
M11_PURU1773_01_SE_C11.indd 409 2/12/2013 7:58:53 AM

Sandy clay layer:

pa = Ka2 γ ′z + γ w z − 2c Ka2
At the interface, z = 0 , pa = p1′′. Thus,
p1′′ = 0 + 0 − 2×17 × 0.361 = −20.43 kN m 2
At the base, z = H2, pa = p2. Thus,
p2 = Ka2 γ ′H 2 + γ w H 2 − 2c Ka2
p2 = 0.361(20.3 − 9.8107 ) 2.5 + 9.807 × 2.5 − 2×17 × 0.361
= 9.47 + 24.53 − 20.43 = 13.57 kN m 2
( a) ( b) (c )
The components of p2, i.e., (a), (b), and (c), are shown in Fig. 11.23.
Resultant pressure at the ground surface = 0 kN/m2
Resultant pressure at the bottom of the first layer = 28.8 kN/m2
Resultant pressure at the top of the second layer is
p1′ + p1′′ = 36.73 − 20.43 = 16.3 kN m 2
Resultant pressure at the bottom of the second layer is
p1′ + p2 = 36.73 + 13.57 = 50.3 kN m 2
The lateral stress distribution is shown in Fig. 11.23.
Sand
5.5 m
γ =18.5 kN/m3
p2
φ′1 =34° 22.8
kN/m2 28.8
a b c
p1 kN/m2
Sandy p1′ 16.3
clay
2.5 m kN/m2
c =17 kPa + + – =
γsat =20.3 kN/m3
φ 2′ = 28°
36.73 9.47 24.53 20.43 50.3 kN/m2

kN/m2 kN/m2 kN/m2 kN/m2
Fig. 11.23
M11_PURU1773_01_SE_C11.indd 410 2/12/2013 7:58:55 AM

Example 11.5 A two-layer cohesive horizontal backfill is supported by a 10 m high vertical

smooth wall. Determine the Rankine active force per unit length of the wall both before and
after a tensile crack occurs in the top layer. Also, determine the line of action of the resultant
in both cases. The soil layer parameters are given below:
0–5 m, Top layer: cu = 12 kN m 2 , φu = 0°, γ = 17 kN m 3
5–10 m, Bottom layer: cu = 35 kN m 2 , φu = 10°, γ = 18 kN m 3
Solution
Top layer
1 − sin 0°
Ka1 = = 1.0
1 + sin 0°
Bottom layer
1 − sin 10°
Ka2 = = 0.704
1 + sin 10°
pa = Ka γ z − 2c Ka
Top layer. When z = 0,

Cohesion component p0 = −2c1 K a1
= −2×12×1 = −24 kN m 2
Weight component p0′ = 0
When z = H1,
Cohesion component p1 = −2×12×1 = −24 kN m 2
Weight component p1′ = K a1 γ1 H1 = 1×17 × 5 = 85 kN m 2
Bottom layer. For the bottom layer, the weight of the first layer acts as a surcharge q (= γ1H1).
Therefore, we consider the second layer separately, and the general equation is
pa = Ka2 γ 2 z − 2c2 Ka2 + Ka2 γ1 H1
When z = 0 (at the interface),

Cohesion component p2 = 2c2 K a2 = −2× 35× 0.704 = 58.73 kN m 2
Weight component p2′ = 0
Surcharge component p2′′ = Ka2 γ1 H1 = 0.704 ×17 × 5 = 59.84 kN m 2
When z = H2 (at the bottom of the wall),
Cohesion component p3 = −2× 35× 0.704 = −58.73 kN m 2
Weight component p3′ = Ka2 γ 2 H 2 = 0.704 ×18 × 5 = 63.36 kN m 2
Surcharge component p3′′ = 0.704 ×17 × 5 = 59.84 kN/m 2
The pressure distribution for the no-tension condition is shown in Fig. 11.24a.
The total thrust
Pa1 = 12 × 85.0 × 5 − 24 × 5 + 12 × 63.36 × 5 + 59.84 × 5 − 58.73 × 5
Pa1 = 212.5 − 120 + 158.4 + 299.2 − 293.65 = 256.45 kN
M11_PURU1773_01_SE_C11.indd 411 2/12/2013 7:58:57 AM

24 kN/m2
1.41 m
5 m P1 – P2
61 kN/m2
1.11 kN/m2 1.11 kN/m2

85 24 P2 273.5 kN
P3
kN/m2 kN/m2 – 256.5 kN
P1 +
5m
3.1 m 3.5 m
63.36 59.84 58.73 64.47 kN/m2 64.47 kN/m2

kN/m2 kN/m2 kN/m2
a b
Fig. 11.24
The line of action,
212.5×(5 / 3 + 5) − 120(5 / 2 + 5) + 158.4 ×(5 / 3) + 299.2×(5 / 2) − 293.65×(5 / 2)

x1 =
256.45
x1 = 3.1 m
Pressure distribution after the development of the tension crack is shown in Fig. 11.24b.
Here,
2c1 2×12
z0 = = = 1.41 m
γ Ka1 17 ×1
The total thrust

Pa2 = 12 × 61(5 − 1.41) + 158.4 + 299.2 − 293.65
Pa2 = 109.5 + 158.4 + 299.2 − 293.65 = 273.5 kN
109.5×[(3.59 / 3) + 5] + 158.4 ×(5 / 3) + 299.2×(5 / 2) − 293.65×(5 / 2)
x2 =
273.5
x2 = 3.5 m
Example 11.6 A vertical smooth-faced 8 m high retaining wall yields when rotated about
the bottom. Estimate the movement at the top of the retaining wall required to establish an
active case. The soil retained is a dry sand with angle of internal friction equal to 37°.
Solution
Figure 11.25 shows the active zone that would develop if the retaining wall AB yields by
rotating about the bottom to a position A′B.
M11_PURU1773_01_SE_C11.indd 412 2/12/2013 7:58:58 AM

Δ L
Slip surface
H
θ f = 45° + φ ′/2
Fig. 11.25
The required yield (Δ) at the top of the wall is specified in terms of the width (L) of the
active zone at the top of the wall. Therefore,
Δ
Yield strain ε =
L
Based on triaxial shear tests, the yield strain required to establish the active case is approx-
imately 0.005. Now,
φ′ 37°
θf = 45° + = 45° + = 63.5°
2 2
L = H tan(90°− θf° ) = 8 × tan(90°− 63.5°)
L = 3.99 m
Therefore, movement, Δ, at the top of the wall = 0.005×3.99 = 0.0199 m
Example 11.7 A retaining wall of 6 m vertical height has the pressure face inclined at 85° to
the horizontal and has a 20° angle of wall friction. The backfill is sloping at 15° to the hori-
zontal and has the following properties: c = 0, φ′ = 37°, γ = 17.2 kN/m3. Compute Coulomb’s
active force per unit length of the wall.
Solution
For the given problem, β = 85°, δ = 20°, φ′ = 37°, and γ = 17.2 kN/m3 and H = 6 m. The
active earth pressure coefficient is obtained from
sin 2 (β + φ ′)
Ka = 2
⎛ sin (φ ′ + δ ) sin (φ ′ − 1) ⎞⎟⎟
⎜⎜
sin β sin (β − δ )⎜1 +
2
⎟
⎜⎜
⎝ sin (β − δ ) sin (i + β ) ⎟⎟⎟⎠
M11_PURU1773_01_SE_C11.indd 413 2/12/2013 7:58:59 AM

Substituting the respective values, we get
sin 2 (85° + 37°)

Ka = 2
⎛ sin (37° + 20°) sin (37°− 15°°) ⎞⎟⎟
⎜⎜
sin 85 sin (85 20 )⎜1
2
° °− ° + ⎟
⎜⎜⎝ sin (85°− 20°) sin (15° + 85°) ⎟⎟⎠
Ka = 0.324
Therefore, active thrust Pa = 12 K a γ H 2 = 12 × 0.324 ×17.2× 6 2 = 100.3 kN.
Example 11.8 For the retaining wall shown in Fig. 11.25a, determine the active thrust on
the wall by Culmann’s construction. What will be the change in lateral thrust and the line of
action (i) if a line load of 100 kN/m acts at a distance of 2.9 m from the face of the wall and
(ii) a uniform surcharge of 36 kN/m2acts on the surface?
Solution
The retaining wall is drawn to a scale of 1mm = 100 mm.
No surcharge load. Different trial wedges are taken and their weights are computed and
tabulated in Column 2 in the following table:
Wedge no. No surcharge load Weight of wedge (kN)
Line load (Col. 2 + 100) Uniform surcharge

(1) (2) (3) (4)
W1/W′1 241 341 336

W2/W′2 359 459 550
W3/W′3 511 611 781
W4/W′4 654 754 999
Taking a force scale of 1 mm = 10 kN, the weights of the wedges are represented (as AW1,
AW2, etc.) on the φ-line. Establish points to plot the Culmann curve. This is shown in Fig.
11.26a by a solid line. Draw a tangent to the curve and measure the active thrust represented
by FW.
Active thrust Pa1 = (FW × Force scale)
Pa1 = 17 ×10 = 170 kN m
With line load. Locate the line load position. The weight of the line load, q, will be acting on
all the wedges, and the revised weights of the wedges are shown in Column 3 of the table.
Represent the weights of the wedges (as AW′1, AW′2, etc.) with the same scale. Again, plot
the Culmann curve (shown by broken lines), and find the active thrust represented by F′W′.
Therefore,
Active thrust on the wall with a line load on the backfill
= Pa2 = 23 ×10 = 230 kN m
M11_PURU1773_01_SE_C11.indd 414 2/12/2013 7:59:00 AM

Q = 100 kN/m
2.9 m
C1 C2 C C3 C4
Q
With
Culmann curve – no Q
Slip
plane No Q Culmann curve with Q
P1=230 kN ′
F3′ F4
F2′ F3
8m F
F1′ F3 F3 F2 No line load Pa1=170 kN/m
Pa1 = 170 b F2 x1=2.67 m
W4
x2 = 5.2 m kN F 1
W3 With line load Pa2=230 kN/m
W
W2 x2=5.20 m
x1=2.67 m
β = 90° W 1
θ f = 53° φ = 30° δ = 20° Scale:
A = φ = 30° γ = 18 kN/m3 1 mm =100 m

ψ
1 mm = 10 kN
β−
=
δ
90 (a) Construction with and without line load

°–
E 20
°=
70
°
C1′ C2′ C3′ C4′

B′ X
36
18 = 2 m
C
F4 Culmann curve
Slip plane
F3
F
F2 W4
8m W3
F1 W2
W
x 3 = 3.33 m
W1
θ f = 52°
A φ = 30°
ψ
=
70 (b) Construction with uniform surcharge
°
E
Fig. 11.26
As the line load is left of the slip plane, establish the line of action following the procedure
given in Fig. 11.9a. Thus, x2 is obtained as
x2 = 5.2 m
M11_PURU1773_01_SE_C11.indd 415 2/12/2013 7:59:01 AM

With uniform surcharge. Surcharge load may be represented by an equivalent height of

backfill; that is,
q 36
H′ = = =2m
γ 18
Draw the additional height of backfill as shown in Fig. 11.26b. Consider B′X as the new
level surface, and repeat the procedure as was done for the no surcharge load condition. The
weights of wedges for this condition are computed and presented in Column 4 of the table.
Draw the Culmann curve and find the active thrust.
Active thrust on the wall with a uniform surcharge
Pa3 = 26 ×10 = 260 kN m
x3 = 10 3 = 3.3 m
Example 11.9 The pressure surface of a retaining wall slopes up and away from the backfill
with a batter of 1 in 10. The backfill is a non-cohesive soil with a density of 19.2 kN/m3 and
angle of internal friction 35°. The angle of surcharge is 4°, the angle of wall friction is esti-
mated to be 20°, and the vertical height of the wall is 12 m. Compute the maximum active
thrust on the wall. Adopt Poncelet’s graphical method.
Solution
A linear scale of 1 mm = 200 mm is chosen, and the retaining wall along with the surcharge
is drawn, as shown in Fig. 11.27.
Here, γ = 19.2 kN/m3, φ = 35°, i = 4°, and δ = 20°.
β = 90°− tan−1 (1 2) = 85.24°
ψ = β − δ = 85.24°− 20° = 65.24°
Slip plane
D
C
B
4°
32 mm
10° J
F
12 m mm
36
20°
K
Pa
4m θf
β1 φ ′= 35°
A ψ = 65.24°
600 mm
Fig. 11.27
M11_PURU1773_01_SE_C11.indd 416 2/12/2013 7:59:02 AM

200 200
Area of triangle JKC = 12 × 36 × 32× × = 23.04 m 2
1000 1000
Active thrust Pa = (Area of ΔJKC) × γ = 23.04 × 19.2 = 442.4 kN/m
Indication of failure plane θf = 62°
Example 11.10 Rework Example 11.9 for the case of a vertical wall with (i) i = 25° and
(ii) i = 35°.
mm
39
B 25°
J
D′
Slip plane
mm
K 39
12 m
B´ J′
F′ 35° 55°
A 70°
G′
(a)
B 35° m
m
48
J
mm
12 m 52
K
35°
A 78°
(b) E
Fig. 11.28
M11_PURU1773_01_SE_C11.indd 417 2/12/2013 7:59:02 AM

Solution
Case I: For such cases, the backfill slope and φ-line may not meet within the space avail-
able on the paper. In such cases, choose an arbitrary point B′ on the pressure face, consider-
ing AB′ as the wall, and proceed. Establish point J’ (similar to J in Example 11.9).
Join B′J′ and draw a line BJ parallel to B′J′. From J, draw JC parrallel to AE. Make JC = JK.
Join C and K. Find the area of ΔJKC (as shown in Fig. 11.28a). Therefore,
Pa = (Area of ΔJKC)× γ
200 200
Pa = 12 × 39× 39× × ×19.2 = 584 kN m
1000 1000
Inclination of the slip plane, θf = 55°.

Case II: For this case, both the φ-line and backfill slope run parallel. Choose any point J on
the φ-line, draw JC parallel to the pressure line, and make JC = CK. Find the area of ΔJKC
(as shown in Fig. 11.28b).
Pa = (Area of ΔJKC)× γ
200 200
Pa = 12 × 52× 48 × × ×19.2 = 958.5 kN m
1000 1000
For such cases, the φ-line itself is the slip plane.
POINTS TO REMEMBER
11.1 Lateral pressures develop against structures supporting soil or water. They depend
on several factors, such as physical and time-dependent behaviour of the soil, soil
deformation, surface roughness, and movement of retaining structures and imposed
loading.
11.2 The backfill material is said to be in a state of elastic equilibrium when the stress
involved and the corresponding strain are within elastic limits. Subsequent increase
in stresses causes a substantial increase in strain, producing a condition known as
plastic flow. The soil mass prior to the onset of the plastic flow condition is said to be
in a state of plastic equilibrium.
11.3 When vertical compression and lateral creep strains become zero, a state of stable
equilibrium is attained, which is called the at-rest condition or K0 condition.
11.4 Suppose every part of a semi-infinite mass at K0-condition is brought on the verge of
failure either by stretching or by compressing, then such a state is called the general
state of plastic equilibrium.
11.5 When adequate lateral movement (stretching of backfill material) occurs, the horizon-
tal stress decreases to a certain magnitude such that the full shear strength of the soil
M11_PURU1773_01_SE_C11.indd 418 2/12/2013 7:59:03 AM

is mobilized. This horizontal stress condition is called Rankine’s active state, and the
stress is referred to as the active stress.
11.6 When sufficient lateral movement (compression of backfill material) occurs, then the
maximum shear strength of the soil is mobilized and the horizontal stress is at a maxi-
mum. This state of failure is called Rankine’s passive state, and the horizontal stress is
called the passive stress.
11.7 The ratio of horizontal stress to vertical stress in the active state is referred to as the
coefficient of active stress or coefficient of active earth pressure, Ka. The ratio of hori-
zontal stress to vertical stress in the passive state is called the coefficient of passive
stress or coefficient of passive earth pressure, Kp. The earth pressure coefficients may
vary from 0.14 to 14 from active to passive condition in cohesionless soils and 0.5 to 2
in cohesive soils.
11.8 Rankine’s theory assumes that (i) there is a conjugate relationship between vertical
and lateral pressures, (ii) the soil is homogeneous and isotropic, (iii) the soil is dry and
non-cohesive, and (iv) the wall is vertical and smooth.
11.9 The maximum unsupported depth of excavation, Hc, may be theoretically taken as
twice the depth of the tension zone (i.e., Hc = 2z0), where the tensile stresses equal
the cohesive strength. The application of this depth in practice should be done very
carefully.
11.10 Coulomb’s earth pressure theory includes the effect of friction between the backfill
and the wall, and a dry non-cohesive inclined backfill. The lateral earth pressure
required to maintain the equilibrium of a sliding wedge with a plane slip surface is
calculated.
QUESTIONS
Objective Questions
11.1 The lateral earth pressure coefficients Ka and Kp refer to
(a) Effective stresses (b) Total stresses
(c) Neutral stresses (d) None of the above
11.2 The active earth pressure caused by a cohesionless backfill on a smooth vertical sur-
face may be reduced by
(a) Saturating the backfill soil with water
(b) Compacting the backfill soil
(c) Reducing the effective stress of the backfill
(d) Providing surcharge load on the backfill
11.3 If for an inclined backfill, with the angle of backfill inclination i and angle of shearing
resistance φ are equal, then for the Rankine condition the active earth pressure coef-
ficient is
M11_PURU1773_01_SE_C11.indd 419 2/12/2013 7:59:03 AM

(a) cos2 i (b) sin2 i

(c) cos i (d) sin i
11.4 The presence of a water table in the backfill serves to increase the earth pressure due
to
(a) Decrease in cohesion (b) Increase in surcharge
(c) Increase in the unit weight (d) Increase in wall friction
11.5 Identify the incorrect statement. Lateral pressure can be developed under the follow-
ing conditions:
(a) Earthquake (b) Swelling pressure
(c) Ice formation (d) Over-consolidation
11.6 Assertion A: Earth pressure is not a unique property of a soil.
Reason R: Earth pressure is a function of backfill material, load on backfill, groundwater
condition, and deflection of retaining structure.
(a) Both A and R are true, and R is the correct explanation of A
(b) Both A and R are true, and R is not the correct explanation of A
(c) A is true, but R is false
(d) A is false, but R is true
11.7 The state of shear failure accompanying a minimum earth pressure is called the
(a) At-rest state (b) Active state
(c) Passive state (d) None of the above
11.8 Identify the incorrect statement. Culmann’s method is chiefly used under the follow-
ing conditions:
(a) The wall has an inclined or broken back
(b) The backfill surface is irregular
(c) Backfill carries a surcharge
(d) High seepage pressure is exerted on the wall
11.9 The amount of translation needed to produce an active pressure condition in a dense
cohesionless soil is
(a) 0.001H to 0.002H (b) 0.002H to 0.004H
(c) 0.01H to 0.02H (d) 0.02H to 0.05H
where H is the height of the wall.
11.10 A sandy loam backfill has a cohesion of 14 kN/m2, a friction angle of 18°, and unit
weight of 16.5 kN/m3. Then, the depth of the tension crack is
(a) 2.00 m (b) 3.33 m (c) 1.98m (d) 2.63 m
11.11 Explain with reasons the use of the at-rest lateral soil pressure condition for the design
of basement walls.
11.12 Explain the possible consequences of the over-compaction of a backfill material.
11.13 Give a critical comparison of the Coulomb and Rankine earth pressure theories.
M11_PURU1773_01_SE_C11.indd 420 2/12/2013 7:59:03 AM

11.14 Comment on the influence of wall friction on the passive earth pressure using
Coulomb’s method for granular soils.
11.15 How do tension cracks influence the distribution of active earth pressure in pure
cohesive soils?
11.16 State whether the following statements are true or false. Justify your choice with sup-
porting arguments.
(i) Rankine’s earth pressure analysis considers neither strains nor displacements.
(ii) “K0-condition” is when no lateral deformations occur in the soil mass.
(iii) The critical height for open cuts for brittle clay soil is directly proportional to the
unit weight and inversely proportional to the unconfined compressive strength.
(iv) The active earth pressure is decreased, while the passive earth pressure is
increased, due to the application of a uniform surcharge load.
(v) The lateral stress under the passive condition is due to compression of the backfill.
EXERCISE PROBLEMS
11.1 A retaining wall 6.5 m high supports an over-consolidated clay backfill with a plastic-
ity index of 32% and an over-consolidation ratio of 2.3. Determine the lateral force per
unit length of wall and the location if the yield of the wall is completely prevented.
The unit weight of the soil is 17.6 kN/m3.
11.2 A 4 m high smooth vertical wall retains a mass of dry loose sand. Compute the total
lateral force per metre acting against the wall if the wall is prevented from yielding.
The sand has a 30° angle of internal friction and unit weight of 14.8 kN/m3. Also, esti-
mate the lateral force per metre run of the wall if sufficient yield of the wall is permit-
ted so as to develop the active Rankine state.
11.3 A vertical frictionless pressure face of an 8 m high retaining wall supports a non-
cohesive 5° sloping backfill. The unit weight of the soil is 18 kN/m3, and the angle of
shearing resistance is 32°. Draw a Mohr circle representing the state of stresses, and
hence, compute the lateral passive resistance per linear length of the wall.
11.4 A wall 15 m high has to be designed so as to retain dry sand. Under loose condition
the sand has a void ratio of 0.65 and φ′ of 32°, and under dense condition the void
ratio and φ′ are 0.41 and 43°, respectively, and G = 2.65. Compute the resultant lateral
pressures for active and passive cases for both the density conditions. Recommend a
suitable resultant lateral force if the wall has to be designed for (i) the active case and
(ii) the passive case.
11.5 A dockside retaining wall 10 m high retains a non-cohesive backfill with a horizontal
surface level with the top of the wall. The properties of the backfill material are, G = 2.65,
e = 0.55, and φ = 32°. An additional superimposed load of 20 kN/m2 is induced at the
surface of the backfill due to construction of warehouses and dockyard traffic. Compute
the lateral thrust on the wall when the water table is (i) 2 m below the level surface, (ii) 5 m
below the level surface, and (iii) at the bottom of the wall. Neglect wall friction.
11.6 A dry granular level backfill of a 6.3 m high retaining wall weighs 16.2 kN/m3. The
active thrust on the wall is believed to be 75 kN/m length of the wall. It is intended
M11_PURU1773_01_SE_C11.indd 421 2/12/2013 7:59:03 AM

to increase the height of the wall and, at the same time, to keep the force on the wall
within permissible limits. The backfill to a depth of 2.8 m from the top is removed. The
removed portion is replaced by a material such as cinder with γ = 8.2 kN/m3. If the
portion of the additional height is also to be filled with cinder, estimate the additional
height of the wall without increasing the initial active thrust. Neglect the wall friction,
and assume that both the backfill soil and the cinder have the same friction angle.
11.7 Figure 11.29 represents a backfill behind a smooth vertical retaining wall. Estimate the
magnitude and line of action of the lateral active force per metre length of the wall.
What would be the reduction in the lateral force if drainage facility is provided to
lower the water table to the base of the wall?
q = 25 kN/m2
Saturated clay
4.5 m
γsat = 20.2 kN/m3, φu = 0
Unconfined comp, strength = 25 kPa
Saturated sand
γ sat = 19.8 kN/m3, φ = 30°
0.9 m
c=0
Fig. 11.29
11.8 A vertical wall 10 m high retains two horizontal layers of a saturated cohesive backfill
with a level surface. The top 4 m of the backfill has an undrained cohesion of 18.2 kPa and
a bulk unit weight of 18.6 kN/m3. The bottom clay layer has a bulk unit weight and an
undrained cohesion of 22 kN/m3 and 23.6 kPa, respectively. Estimate the likely depth
of the tension zone behind the wall. Compute the total active force if tension cracks
develop, and also locate the application of the resultant lateral force.
11.9 A 9.5 m high vertical, smooth retaining wall is supporting three layers of soil with the
following details:
Layer no. Depth Total unit weight Cohesion Angle of shearing
(m) (kN/m3) (kPa) resistance (°)
Top 3.5 19.2 17.5 0

Middle 3 20.2 0 30
Bottom 3 21.2 0 34
M11_PURU1773_01_SE_C11.indd 422 2/12/2013 7:59:03 AM

Compute the active thrust per metre run of the wall if the water table is located at the
interface of the top and middle layers.
11.10 It is intended to excavate a vertical unsupported cut of depth 5 m. The natural soil has
a unit weight of 17.5 kN/m3 and the shear strength parameters are c′ = 30 kPa and
φ = 6°. The groundwater table is deeper than the cut. Determine (i) the stress at the
top and bottom of the cut, (ii) the maximum depth of the potential tension crack, and
(iii) the maximum unsupported excavation depth.
11.11 A vertical retaining wall of height 6.5 m retains a non-cohesive level backfill weighing
19.2 kN/m3, with the angle of friction being 18°. Compute the total thrust on the wall
adopting Culmann’s graphical method. Later, it is planned to place a piece of machin-
ery weighing 30 kN on the surface, parallel to the crest of the wall. Find the minimum
horizontal distance from the back of the wall at which the machinery could be placed
without increasing the pressure on the wall. Take φ = 30°.
11.12 The front of a retaining wall slopes at an angle of 80° to the horizontal. The depth of
soil in front of the wall is 2.5 m. The soil surface is horizontal and the soil dry. The
other properties of the soil are c′ = 0, φ′ = 32°, δ = 18°, and γ = 18.2 kN/m3. Estimate
the total passive resistance developed at the front of the wall.
11.13 For the retaining wall shown in Fig. 11.30, determine the active lateral force per metre
length of the wall.
25 kN
2m 2m
c=0
φ = 32°
δ = 20°
γ = 19.2 kN/m3
5m
Fig. 11.30
11.14 For the retaining wall shown in Fig. 11.31, compute the lateral active force per metre
length of the wall using Poncelet’s graphical construction. Check the value using
Culmann’s graphical method.
M11_PURU1773_01_SE_C11.indd 423 2/12/2013 7:59:04 AM

20°
c=0
9m
φ = 25°
γ = 18.2 kN/m3
Fig. 11.31
M11_PURU1773_01_SE_C11.indd 424 2/12/2013 7:59:04 AM

12
Earth-Retaining Structures
CHAPTER HIGHLIGHTS
Gravity-type retaining walls: proportioning, earth pressure consideration,
stability requirements, backfill materials and drainage, joints – Sheet pile walls:
cantilever types, anchored types, wales, tie-rods and anchorages – Braced
excavations: earth pressure distribution, heave and stability
12.1 INTRODUCTION
In Chapter 11, the two basic earth pressure theories were discussed at length. These
theories are applied to the design of earth-retaining structures. In general, earth-
retaining structures are constructed when abrupt changes in the ground surface
elevation are needed to protect unstable slopes. The typical structures are various types
of retaining walls, sheet piles, braced sheeting of excavations, bulkheads or abutments,
basement or pit walls, etc. These may be self-supporting (e.g., gravity- or cantilever-
type walls) or they may be laterally supported by means of bracing of anchorages. The
retaining materials may be soil and water, coal or ore piles and water.
12.2 GRAVITY-TYPE RETAINING WALLS

Gravity-type walls provide slope and soil retention on account of their weight, which
can consist of masonry, concrete mass, concrete in combination with soil weight, or the
weight of earth mass alone. In addition to weight, they are aided by the passive resis-
tance developed in front of the wall. They are all free to deflect at the top and thereby
mobilize active earth pressure. Representative types of gravity-type walls are shown in
Fig. 12.1. Massive walls are uneconomical because of the large wall material used for
the dead weight. Reinforced concrete cantilever walls are more economical because the
backfill itself is aimed to provide most of the required dead weight.
M12_PURU1773_01_SE_C12.indd 425 2/12/2013 5:18:23 PM

Often with
counterforts
Rigid wall Reinforced rigid wall Cantilever wall
Fig. 12.1 Gravity-type retaining walls
12.2.1 Proportioning Retaining Walls

Simple gravity and cantilever walls are quite common. For the design of a retaining wall, a
preliminary dimension has to be assumed. This is referred to as proportioning, which enables
the engineer to decide the basic components of the wall for analysis. If unstatisfactory results
are obtained after the required stability checks, the sections are modified and re-checked.
Figure 12.2 represents the proportions of various wall components for preliminary selection
(Das, 1984). The top of the stem of any retaining wall should not be less than 0.3 m for
construction activities. The bottom of the base should be below the weathered soil due to
seasonal variation, and in no case less than 0.6 m. The counterfort walls may have similar
0.3 m min 0.3 m min
Backfill
20 mm Backfill
20 mm min min
Stem
H Stem
1m 1m
0.12–
0.17H
Df < 0.6 m Df < 0.6 m
0.1H 0.1H
Toe 0.12 Heel 0.1H

0.17H
0.5–0.7H 0.5–0.7H
(a) Gravity wall (b) Cantilever wall
Fig. 12.2 Proportioning retaining wall
M12_PURU1773_01_SE_C12.indd 426 2/12/2013 5:18:24 PM

Earth-Retaining Structures 427
dimensions to those of cantilever walls. The counterforts are 0.3 m thick and are spaced at
centre-to-centre distances of 0.3 to 0.7H, where H is the height of the retaining wall.
12.2.2 Earth Pressure Consideration

Cantilever and gravity walls (Fig. 12.3) are both liable to rotational and translational move-
ments, and hence Rankine or Coulomb theories may be used for the calculation of lateral
pressure.
If Rankine’s theory is to be applied, then it is assumed that the soil is retained by a verti-
cal face (shown by broken lines in Fig. 12.3) extending upwards from the heel to the ground
surface. Then, the active pressure is evaluated keeping AB as the face of the wall. While
checking for stability, the active thrust (Pa), the weight of soil above the heel (Ws), and the
weight of the concrete (Wc) should be taken into consideration. This assumption is theoreti-
cally correct as long as the zone bounded by the line BC (Fig. 12.3) is not obstructed by the
stem of the wall. However, Coulomb’s theory can be used directly on the real wall surface
without any assumption.
12.2.3 Stability Requirements

The retaining wall as a whole must satisfy the following stability requirements:
1. Safety against overturning
2. Safety against sliding
3. Safety against bearing capacity failure
4. Safety against overall stability
Check for Overturning. Figure 12.3 represents the forces acting on a cantilever retaining
wall for an active condition. The passive resistance (Pp) in front of the wall should not be
relied on unless the soil is firm and undisturbed. The passive resistance, Pp, is given as
A
c1
C φ1
γ1
X1
X2 H1
Pv Pa
X3
Ph
Ws
Wc
Df Pp B
D′
D pmin
pmax c2
Key φ2
B γ2
Fig. 12.3 Retaining wall with details of forces
M12_PURU1773_01_SE_C12.indd 427 2/12/2013 5:18:24 PM

Pp = 12 K p γ 2 Df2 + 2c2 K p Df
where c2, φ2, γ2, and Kp are the parameters related to the soil in the wall and the foundation.
The active thrust, Pa, is determined by applying Rankine’s theory on the vertical surface
AB. The weight of the wall (Wc) and the weight of the soil above the heel (Ws) are calculated.
It is generally assumed that if overturning were to occur, it would do so about the toe of
the wall. Thus, the factor of safety against overturning is defined as the ratio of resisting to
disturbing moments about the toe.
Let Ph and Pv be the components of the active force Pa, then, the sum of disturbing
moments (∑Md) is given as
⎛H ⎞
∑ Md = Ph ⎜⎜⎜ 1 ⎟⎟⎟ (12.1)
⎝ 3 ⎠
where Ph = Pa cos i.
The sum of stabilizing moments (∑Mr) is given as
⎛D ⎞
∑ Mr = Wc x1 + Ws x2 + Pv x3 + Pp ⎜⎜⎜ f ⎟⎟⎟ (12.2)
⎝ 3⎠
Therefore,
∑ Mr
FOT = (12.3)
∑ Md
where FOT is the factor of safety against overturning. This should not be less than 2.0.
Check for Sliding. The factor of safety against sliding along the base is defined as the
ratio of the resisting forces to the sum of the horizontal disturbing forces.
The only horizontal force causing the sliding is Ph, hence, the sum of driving forces (∑Fd)
is given as
∑Fd = Ph = Pa cos i (12.4)
The resisting forces (∑Fr) are the shearing resistance developed at the base (Sh = (∑V)tan
φ2 + Bc2) and the passive resistance (Pp)
∑ Fr = [( ∑ V )tan φ2 + Bc2 ] + Pp (12.5)

Then, the factor of safety with respect to sliding is
∑ Fr
FSL = (12.6)
∑ Fd
A minimum factor of safety of 1.5 is generally provided against shear. As discussed earlier
in many cases, the passive resistance is ignored. If adequate factor of safety is not achieved,
a key may be incorporated in the base (shown hatched in Fig. 12.3).
Check for Bearing Capacity Failure. The base pressure at the toe of the wall must not
exceed the allowable bearing capacity of the soil. The position of the resultant force R is
determined by dividing the algebraic sum of the moments of all forces about any point on
the base by the vertical component ∑V (Fig. 12.4).
M12_PURU1773_01_SE_C12.indd 428 2/12/2013 5:18:25 PM

A
c1
φ1
γ1
∑V H1
Ph
R
D1 Pp
D′
D E pmin
pmax c2
φ2
eb I γ2
B/2 B/2
Fig. 12.4 Forces to check for bearing capacity failure
In order to keep the base pressure compressive over the entire base width, the resultant R
must act within the middle third of the base, that is, the eccentricity (eb) of the base resultant
must not exceed B/6. Adequate safety against overturning of the wall will be ensured when
the resultant falls within the middle third of the base. Considering a linear variation of the
base pressure, the maximum and minimum pressures on the base are computed:
∑ V ⎛⎜ 6 e b ⎞⎟
pmax = ⎜1 + ⎟ (12.7)
B ⎜⎝ B ⎟⎠
∑ V ⎛⎜ 6 e b ⎞⎟
pmin = ⎜1 − ⎟ (12.8)
B ⎜⎝ B ⎟⎠
The value of pmax should be less than the allowable bearing capacity of the soil. The allowable
soil pressure considers both the safety against shear failure and the permissible settlement.
Generally, a factor of 3 is provided against shear failure.
The value of pmin becomes negative when eccentricity eb exceeds B/6. This should be
completely avoided as the tensile strength of the soil is very small.
Check Against Overall Stability. Apart from safety of the retaining wall against the three
factors explained above, the wall should also be safe against overall stability. Because of the
presence of a weak layer at an immediate depth below the base in cohesive soils, there is a
possibility of the entire soil slipping along with the wall due to inadequate strength, excess
pore water pressure, removal of resistance near the toe, etc. Some of the possible failures are
shown in Fig. 12.5. The factor of safety against the overall stability should not be less than 1.50.
12.2.4 Backfill Materials and Drainage

Backfill materials for retaining structures should be designed to minimize the lateral pressure.
A good backfill material should satisfy two important requirements, viz., high long-term
strength and free drainage.
M12_PURU1773_01_SE_C12.indd 429 2/12/2013 5:18:26 PM

Probable slip
surface
Centre of
Centre of rotation
rotation
Weak layer
Slip surface
Rigid base
(a) Shallow shear failure (b) Deep shear failure
Fig. 12.5 Check against overall stability
In general, granular materials make the best type of backfill since they maintain an
indefinite active state of stress and have free drainage. Clay backfills tend to creep and have
a very low permeability. They should be avoided as climatic changes are likely to cause
successive swelling and shrinkage of the soil. Swelling imposes unpredictable pressures on
the wall and its movements, and subsequent shrinkage may result in the formation of cracks
in the backfill surface.
Poorly graded to well-graded sands and gravels form an excellent backfill because of
their free-draining characteristics. Silty/clayey sand and gravels function as good backfill
materials provided they are kept dry or are provided with adequate drainage arrangement.
Low to high plastic clays and silts can be graded as poor backfill material. Organic silts
and clays and peat should not be used as backfill because of the swelling and shrinking
behaviour of such soils.
An important consideration is the control of the water table in the backfill. The easiest
way to control groundwater is to provide a free-draining backfill. Further, as a result of rain-
fall, or other reasons, the backfill may get saturated and increase the pressure on the wall,
creating an unstable condition. Weep holes and/or perforated drainage pipes are provided
to drain away such water and reduce the development of pore water pressure. Weep holes
should have a minimum diameter of 0.1 m and should be spaced adequately. Filter materials
are provided behind the weep holes and around the drainage pipes to prevent the possible
washing out of backfill materials into weep holes or drains (Fig. 12.6).
The backfill material has to be compacted to attain maximum strength and, hence, mini-
mum active thrust on the wall. However, over-compaction has to be avoided and sufficient
care taken not to disturb the wall while compacting the backfill.
12.2.5 Joints in Retaining Walls

A retaining wall is provided with construction, contraction, or expansion joints (Fig. 12.7).
Construction Joints. These are provided between two successive pours of concrete,
and are vertical and horizontal joints. Either each surface of the concrete is cleaned and
roughened before placing the next pour of concrete or a key is provided between the
joints.
M12_PURU1773_01_SE_C12.indd 430 2/12/2013 5:18:27 PM

Coarse filter
Weep hole
Filter fabric
(a) Weep holes (b) Weep holes with filter fabric
Drainage
blanket
Filter fabric
Drain
Coarse filter
Drain
(c) Lateral drain with filter (d) Lateral drain with drainage blanket
Fig. 12.6 Drainage arrangements
Wall Wall
Keys Roughened surface

Steel for added
shear
(a) Construction joint
Back of wall Back of wall

Construction Expansion joint
joint
Face of wall Face of wall
(b) Construction joint (c) Expansion joint

Plan
Construction joint
Expansion
joint
(d) Location of joints

Elevation
Fig. 12.7 Joints in retaining walls
M12_PURU1773_01_SE_C12.indd 431 2/12/2013 5:18:28 PM

Contraction Joints. These are vertical grooves or joints, 8 mm wide and 12 to 16 mm deep,
provided in the face of the wall (from the top of the base slab to the top of the wall). These
joints allow the concrete to shrink within permissible limits without harming it.
Expansion Joints. These joints are provided to withstand the effects of expansion due to
temperature changes. These are vertical joints extending from the base to the top of the wall
and are filled with flexible joint fillers.
12.3 SHEET PILE WALLS

Sheet pile walls are widely used for both small and large water front structures. These are
flexible structures compared to the gravity-type retaining walls discussed in the previous
section. Sheet pile walls primarily depend for stability on the passive resistance developed
by the soil in the front of the wall and on the lower part of the wall at the back. In certain
types, the stability is ensured by providing struts and anchorages.
A sheet pile wall may fail in any one of the following ways: (i) forward movement of the
base due to inadequate passive resistance in front of the wall, (ii) failure by bending, and (iii)
failure of anchors. Depending on the type of failure, the earth pressure distribution varies,
and it does not follow the conventional distribution adopted in rigid walls. However, simple
distributions are adopted, and the effects of these failures are examined wherever necessary.
Sheet piles are made out of different materials, such as wood, precast concrete, or steel.
Different types of sheet pile structures are shown in Fig. 12.8.
WL
Anchor Deadman
Sheet pile Sheet pile rod
Dredge line Dredge line
(a) Cantilever sheet piling (b) Anchored sheet piling
Sheet
pile
Fill
Sheet pile Strut
Anchor piles
(c) Anchored bulk head (d) Braced sheeting
Fig. 12.8 Sheet pile structures
M12_PURU1773_01_SE_C12.indd 432 2/12/2013 5:18:28 PM

12.3.1 Cantilever Sheet Pile Walls

Sheet piles of this type are constructed by driving the sheet pile to adequate depth into the
soil below the dredge line so that they develop a cantilever beam-type reaction to resist the
active pressures on the wall. These piles are economical only for a moderate height of up to
about 10 m. Because of the cantilever action, the lateral deflection of this type of wall is more.
As the stability of the wall depends primarily on the pressure developed in front of the wall,
any action of lowering the dredge line (e.g., by erosion or scour) should be controlled. The
use of such walls is primarily meant for temporary installations.
The embedment depth varies with different soils. Also, the pressure distribution varies
with the type of soil and water level conditions. The wall rotates about the point O, and the
development of active and passive conditions on either side of the wall is as shown in Fig.
12.9a. The probable actual pressure distribution is shown in Fig. 12.9b. However, for design
purposes, the distribution is simplified, as shown in Fig. 12.9c.
Cantilever Sheet Piling in Granular Soils. In this case, both the retained soil and that
below the dredge line are sands and are assumed to have the same properties. Appropri-
ate values of γ and φ should be used for a layered system. For non-level ground surfaces,
Coulomb earth pressure theory may be applied and for the rest, Rankine’s. In computing
earth pressures on the wall, it is not wise to rely on the vertical shearing forces between the
soil and the wall, and thus the application of Rankine’s theory is justified.
The positions of ground surface, dredge line, water table, and pressure diagram are
depicted in Fig. 12.10. Let φ, Ka, Kp, γ, γsat, and γ′ be the angle of shearing resistance, the
active and passive earth pressure coefficients, and the total, saturated, and submerged unit
weights, respectively. Then, the pressures p1 and p2 are given as
p1 = γ z1Ka (12.9)
and
p2 = (γ z1 + γ′z2)Ka (12.10)
At the dredge line the hydrostatic pressures on either side of the wall cancel each
other.
Sand Sand
Sand
Dredge
line Active state
Passive Active
Sand
state state Sand Sand
Active O Passive state
state
(a) Deflection of wall (b) Actual pressure (c) Simplified pressure

distribution distribution
Fig. 12.9 Cantilever sheet piling
M12_PURU1773_01_SE_C12.indd 433 2/12/2013 5:18:29 PM

z1 Sand
γ, φ
p1 C
γ sat, φ
H
z2
Dredge R
level
z
p2
D
z3 Sand
E z Mmax
D
z4
z5
p3 B p4
(a) Net pressure distribution (b) Moment diagram
Fig. 12.10 Cantilever sheet piling in granular soils
In order to determine the net lateral pressure below the dredge line and at the point of
zero pressure, consider any depth z from the dredge level. Again, hydrostatic pressures can-
cel each other, and the active passive pressure at depth z may be given as
pa = p2 + γ′zKa (12.11)
pp = γ′zKp (12.12)
The net lateral pressure, pz, is obtained as
pz = pa − pp = p2 + γ′zKa − γ′zKp
pz = p2 + γ′z(Ka − Kp)
At depth z = z3 , pz = 0
p2 + γ′z3(Ka − Kp) = 0
or
p2 p
z3 = = 2 (12.13)
γ ′(K p − Ka ) γ ′K ′
where K′ = Kp − Ka.
From similar triangles,
p3 p
= 2
z4 z3
or
p2
p3 = z 4 = z 4 γ ′K ′ (12.14)
z3
M12_PURU1773_01_SE_C12.indd 434 2/12/2013 5:18:30 PM

At the bottom of the sheet pile wall, the passive pressure acts from right to left and the
active pressure from left to right; hence,
pp = (γ z1 + γ ′z2 + γ ′D)K p (12.15)
and
pa = γ ′DKa (12.16)
Therefore,
pp − pa = p4 = (γ z1 + γ ′z2 )K p + γ ′DK ′ (12.17)
or
D = z3 + z4 (12.18)
Considering the equilibrium of all horizontal forces,
R − 12 p3 z4 + 12 z5 ( p3 + p4 ) = 0 (12.19)
where R is the area of the pressure diagram ACDE.
Considering the equilibrium of all the moments about the point B,
⎛z ⎞ ⎛z ⎞
R( z4 + z ) − ( 12 z4 p3 )⎜⎜⎜ 4 ⎟⎟⎟ + 12 z5 ( p3 + p4 )⎜⎜ 5 ⎟⎟⎟ = 0 (12.20)
⎝3⎠ ⎜⎝ 3 ⎠
From Eq. 12.19,
p3 z 4 − 2 R
z5 = (12.21)
p3 + p 4
On substituting Eq. 12.21 in Eq. 12.20 and rearranging, a fourth-order equation in z4 is
obtained:
z44 + c1 z43 + c2 z42 + c3 z4 + c4 = 0 (12.22)
where p0 ⎪⎫⎪
c1 = ⎪⎪
γ ′K ′ ⎪⎪
8R ⎪⎪
c2 = ⎪⎪
γ ′K ′ ⎪⎪
⎪⎪
6 R ⎡⎣ 2 z γ ′K ′ + p0 ⎤⎦ ⎪⎪
c3 = − ⎬ (12.23)
( γ ′K ′ ) 2 ⎪⎪
⎪⎪
⎪⎪
R(6 zp0 + 4 R) ⎪⎪
c4 = −
( γ ′ K ′ )2 ⎪⎪⎪
⎪⎪
p0 = (γ z1 + γ ′z2 )K p + γ ′z3 k ′⎪⎪⎪
⎭
A trial-and-error solution may be adopted to solve the equation for z4, whereby D is
obtained. The factor of safety is applied either by arbitrarily increasing the depth by 20 to
M12_PURU1773_01_SE_C12.indd 435 2/12/2013 5:18:31 PM

40% or reducing the Kp term by a factor (about 1.5 to 2.0). Generally, the former method is
preferred. Thus, the design depth
Dd = 1.2 to 1.4 D (12.24)
The variation of the bending moment with depth is shown in Fig. 12.10 b. The point of
zero shear corresponds to the point of maximum bending moment. Let z′ be the point of zero
shear from the point E. Then,
R = 12 ( z ′)2 k ′γ ′ (12.25)
or
2R
z′ = (12.26)
K ′γ ′
The maximum bending moment is obtained as
⎛ z′ ⎞
Mmax = R( z + z ′) − 12 γ ′K ′( z ′)2 ⎜⎜ ⎟⎟⎟ (12.27)
⎜⎝ 2 ⎠
The section modulus S of the sheet pile wall is
Mmax
S=
σfa
(12.28)
where σfa is the allowable flexural stress of the sheet pile.

Cantilever Sheet Piling in Cohesive Soils with Granular Backfill. In certain cases, sheet
piles have to be driven into cohesive soils with undrained cohesion (i.e., φ = 0° case). The
pressure diagram with other details is shown in Fig. 12.11. The values of p1 and p2 are the
A
Sand
z1 γ, φ
C
p1
z2 R
z Sand
E γ sat , φ
p5 p2 D
z Clay
D γ , c,φ = 0
sat
z4
B p6
Fig. 12.11 Cantilever sheet piling in cohesive soil with granular backfill
M12_PURU1773_01_SE_C12.indd 436 2/12/2013 5:18:32 PM

same as those determined by Eqs. 12.9 and 12.10. The net pressure below the dredge line is
calculated from pa and pp. At a depth z from the dredge line,
pa = [γ z1 + γ ′z2 + γ sc zK ac − 2c K ac ] (12.29)
where γsc is the saturated unit weight of clay and Kac is the active earth pressure coefficient
for clay (as φ = 0°, Kac = 1). Similarly,
pp = γ sc zK pc + 2c K pc (12.30)
where Kpc is the passive earth pressure coefficient for clay (as φ = 0°, γpc = 1). Thus, the net
pressure, p5, is given as
p5 = pp − pa = [γ sc z + 2c] − [γ z1 γ ′z2 +γ sc z]+2c

(12.31)
p5 = 4c − (γ z1 + γ ′z2 )
At the bottom of the sheet pile, the passive resistance from right to left is
pp = (γ z1 + γ ′z2 + γ sc D) + 2c (12.32)
Similarly, the active pressure from left to right is
pa = γ sc D − 2c (12.33)
Hence, the net pressure
p6 = pp − pa = 4c + (γ z1 + γ ′z2 ) (12.34)
Considering the equilibrium of horizontal forces,
R − [4c − (γ z1 + γ ′z2 )]D + 12 z4 [4c − (γ z1 + γ ′z2 ) + 4c + (γ z1 + γ ′z2 )] = 0 (12.35)
where R is the area of the pressure diagram ACDE.
Simplifying,
D[4c − (γ z1 + γ ′z2 )] − R
z4 = (12.36)
4c
Considering the equilibrium of moments about the point B,
D 1 ⎛z ⎞
R(D + z ) − [4c − (γ z1 + z2 )] + z4 (8c)⎜⎜⎜ 4 ⎟⎟⎟ = 0 (12.37)
2 2 ⎝3⎠
Combining Eqs. 12.36 and 12.37, we get
R(R + 12cz )
D2 [4c − (γ z1 + γ ′z2 )] − 2DR − =0 (12.38)
(γ z1 + γ ′z2 ) + 2c
Equation 12.38 is solved for D. This is increased by 20 to 40%, hence
Dd = 1.2 to 1.4 D (12.39)
M12_PURU1773_01_SE_C12.indd 437 2/12/2013 5:18:34 PM

The maximum bending moment occurs at a depth z′ from the dredge line, which is obtained as
R − p5 z ′ = 0
or
R
z′ = (12.40)
p5
Therefore,
′2
pz
Mmax = R( z ′ + z ) − 5 (12.41)
2
The section modulus is found from Eq. 12.28.
Cantilever Sheet Piling in Cohesive Soil with Cohesive Backfill. Sheet piling with cohe-
sive backfill is treated in the same way as granular backfill. However, additional consideration
with regard to consolidation of the clay layer, formation of tension crack, and the effect of
shrinking on stability is required. Because of the uncertainty of clay backfill, granular backfills
are generally preferred. Here, as both the soils are clay, φ = 0° and Kac = Kpc = 1.
The pressure diagram is given in Fig. 12.12. Now,
2c
z0 = (12.42)
γ ′ Kac
′ z2 − 2c Kac = Kac (γ sc z1 + γ sc
p2 = Kac γ z1 + Kac γ sc ′ z2 ) − 2c K ac
or
′ z2 − 2c
p2 = γ z1 + γ sc (12.43)
z1 z0
Clay
γ , c, φ = 0
C sat
p1
H
z2 R
Clay
z
γ , c,φ = 0
E sat
p5 p2 D
z
D
Clay
z4 γ , c, φ = 0
sat
B p6
Fig. 12.12 Cantilever sheet piling in cohesive backfill
M12_PURU1773_01_SE_C12.indd 438 2/12/2013 5:18:35 PM

where γ′sc is the submerged unit weight of clay. Other treatments are similar to that given to
granular soils, and p5 is modified as
′ z + 2c) − (γ z1 + γ sc
p5 = (γ sc ′ z2 − 2c) − γ sc z
or
′ z2 )
p5 = 4c − (γ z1 + γ sc (12.44)
12.3.2 Anchored Sheet Pile Walls

The category of flexible structures called anchored sheet pile walls or anchored bulkheads is
commonly used in water front construction. The construction of such walls consists of driv-
ing a sheet pile to the required depth, followed by dredging in front of the piling and back-
filling behind the piling. The upper end of the sheet piling is attached to the anchor block
through a tie rod. Such a provision of anchors reduces the depth of penetration and the
cross-section area of the sheet pile. The use of more than one anchor may be necessary to
reduce the lateral deflection and the bending moment. These walls achieve stability due to
the passive resistance developed in front of the wall and the resistive force offered by the
anchor force offered by the anchor system.
The behaviour of anchored bulk heads is highly complex, and therefore, considerable
simplifications have to be made in their design. There are two basic methods of analysis of
anchored bulkheads: (i) free earth support method and (ii) fixed earth support method.
The free earth support method assumes that the piling is rigid and may rotate at anchor
rod level. As the depth of embedment is considered less in this analysis, the toe of the pile
is not restricted and hence the bending moment near the toe is negligible (Fig. 12.13a).
The forces acting only on the sheet pile are from lateral soil pressures and the anchor pull,
and failure occurs by rotation about the anchor rod. The theoretical embedment depth is
Tie rod Tie rod
WT WT
Moment Moment
Dredge Deflection Deflection

line
(a) Free earth support (b) Fixed earth support
Fig. 12.13 Deflection and moment diagrams for anchored sheet piles
M12_PURU1773_01_SE_C12.indd 439 2/12/2013 5:18:36 PM

increased by 20% to 40% to arrive at the design depth or Kp may be reduced prior to the
computation of embedment depth.
The fixed earth support method considers the lowest section of the sheeting to be fixed
in the earth. In this case, the depth of embedment is considered more, and hence, the base
of the wall is assumed to be entirely restrained from rotation by the passive resistance of the
soil behind it. This passive resistance is in addition to the pressures considered in the free
earth support method (Fig. 12.13b). Since failure by forward movement of the toe is unlikely
in a wall designed in this way, no factor of safety is applied to the passive resistance of the
soil in front of the wall.
Free Earth Support Method for Penetration of Sandy Soil. The assumed pressure
diagrams and details of other terms are illustrated in Fig. 12.14. The values of p1 and p2 are
given as
p1 = γ z1 Ka (12.45)
p2 = (γ z2 + γ ′z2 )Ka (12.46)
The value of z3 is given by Eq. 12.13 as
p2
z3 =
γ ′K ′
At the bottom of the wall, the net pressure can be given as
p3 = γ ′K ′z4 (12.47)
Considering the equilibrium of the horizontal direction,
R − (area of EBF) − Fa = 0 (12.48a)
A
y1 F
z1 Sand
z0 y2 γ, φ
p1 C
Anchor rod
H
z2
R
Sand
z γ ,φ
sat
p2
z D
Sand
E γ ,φ
sat
D z4
B
F p3
Fig. 12.14 Anchored sheet pile wall penetrating sand
M12_PURU1773_01_SE_C12.indd 440 2/12/2013 5:18:37 PM

where R is the area of the pressure diagram ACDE and Fa is the tension in the rod per unit
length of the wall. Or
R − 12 p3 z4 − Fa = 0
Hence,
Fa = R − 12 p3 z4 (12.48b)
Taking the moment about the anchor rod,
−R[( H + z3 ) − ( z + y1 )] + 12 p3 z42 ( y 2 + z2 + z3 + 32 z4 ) = 0
or
3 R[( H + z3 ) − ( z + y1 )]
z43 + 1.5 z42 ( y 2 + z2 + z3 ) − =0 (12.49)
γ ′K ′
The solution for Eq. 12.49 is obtained by the trial-and-error method. The theoretical depth of
penetration,
D = z3 + z4
The design depth
Dd = 1.2D to 1.4D (12.50)
The point of zero shear, z, from the ground surface is obtained from
1
2
p1 z1 − Fa + p1 ( z − z1 ) + 12 Ka γ ′( z − z1 )2 = 0 (12.51)
From the knowledge of z, the magnitude of bending moment is obtained.

Free Earth Support Method for Penetration of Clay. The assumed pressure distribution
and other details are shown in Fig. 12.15. Here, the soil below the dredge line is saturated clay
A
y1 Anchor rod
Water level z1
y2 Sand
γ, φ
p1 C
H
z2 R
z
Dredge line Sand
γ ,φ
sat
Clay p2 D
Clay
D γ ,φ
sat
B
p3
Fig. 12.15 Anchored sheet pile wall penetrating clay
M12_PURU1773_01_SE_C12.indd 441 2/12/2013 5:18:39 PM

under undrained condition (i.e., φ = 0). The pressure distribution diagram and the values of
p1 and p2 are the same as in the previous case. The net pressure distribution diagram below
the dredge line (i.e., z = H to H + D) can be given as
p3 = 4c − (γ z1 + γ ′z2 ) (12.52)
Considering static equilibrium in the horizontal direction,
Fa = R − p3 D (12.53)
where R is the area of the pressure diagram above the dredge line. Taking the moment about
the anchor rod,
⎛ D⎞
R( H − y1 − z ) − p3 D ⎜⎜⎜ y 2 + z2 + ⎟⎟⎟ = 0
⎝ 2⎠
Simplifying,
p3 D2 + 2 p3 D( H − y1 ) − 2R( H − y1 − z ) = 0 (12.54)
From the above equation, the theoretical depth D is determined. This depth is increased,
and the design depth is determined as
Dd = 1.2 to 1.4D (12.55)
In this case, the maximum bending moment will occur at a depth of z1 < z < H, and hence,
the maximum bending moment is determined.
Rowe’s Moment Reduction Method. The hydrostatic earth pressure distribution is valid
only for rigid walls. As the sheet pile walls are flexible in nature, the conventional pressure
distribution is affected, and hence, the bending moment differs. Generally, this reduces the
bending moment. Thus the bending moment calculated based on the free earth support
method gives conservative results. Rowe (1952, 1957) proposed a method for reducing the
moments and thus suggested a more realistic design.
The factors on which Rowe’s charts are based are as follows:
1. The relative flexibility of the piling expressed in terms of the flexibility number
⎛ H ′ 4 ⎞⎟
⎜ ⎟⎟
ρ = 10.91×10−7 ⎜⎜
⎜⎝ Ep l ⎟⎟⎠ (12.56)
where H′ is the total depth of the sheet pile (m), E is the modulus of elasticity of the pile
material (MN/m2), and I is the moment of inertia of the pile section per metre of the wall
(m4/m of the wall).
2. The relative height of piling
H +D
Ha =
H
3. For cohesive soils the stability number Sn = [1.25c/(γz1 + γ′z2)] and for non-cohesive soils
relative density are considered.
Figure 12.16 represents the moment reduction curves for non-cohesive soils. The reduced
design moment Mr is obtained by noting down the values corresponding to the particular
M12_PURU1773_01_SE_C12.indd 442 2/12/2013 5:18:41 PM

1.0 Loose
sand
0.8 H
H+D
D
Mr 0.6
Dense sand
Mmax
and gravel
0.4
0.2
0
–4.0 –3.5 –3.0 –2.5 –2.0
Log ρ
(a) Sheet piles penetrating sand
1.0 Log ρ = –3.1

0.8
Mr Ha = 0.8
Mmax 0.7
0.6
0.6
0.4
1.0 Log ρ = –2.6
Mr 0.8
Ha = 0.8
Mmax
0.6
0.7
0.4 0.6
1.0 Log ρ = -2.0
0.8
Mr Ha = 0.8
Mmax
0.6
0.7
0.4 0.6
0 0.5 1.0 1.5 1.75
Stability number Sn
(b) Sheet piles penetrating clay
Fig. 12.16 Moment reduction charts (Source: Rowe, 1957).
log ρ and density for cohesionless soil and Ha, Sn, and log ρ for cohesive soil. Suitable inter-
polations may be made wherever necessary.
Fixed Earth Support Method for Penetration of Sandy Soil. As discussed earlier, the toe
of the pile is restrained (Fig. 12.13b). The assumed pressure distribution moment diagram
and identification of terms are illustrated in Fig. 12.17. Point C in the moment diagram is
the point of contraflexure. The pile may be assumed to be hinged at this point C. Thus,
the portion of the piling above the point C can be considered as a beam resisting the net
earth pressures through the anchor force and the shear force P. This is termed the equivalent
M12_PURU1773_01_SE_C12.indd 443 2/12/2013 5:18:41 PM

A
y1 Fa Fa
z1
Sand γ ,φ
p1 D p1
H
z2
Sand
E γsat,φ p2
z5 C p2 z5
z3 C P
p2′
F C P
z4 Sand D z3–z5
P′ γsat,φ
J z4
F′ P′
p2′′
B G
p2 ′′ = γ ′ (Kp − Ka) [z4]
(a) Pressure diagram (b) Moment diagram (c) Determination of z5
Fig. 12.17 Fixed earth support method: sheet pile wall penetrating sand
0.3
0.2
Z5
H
0.1
0
20 25 30 35 40
Angle of friction φ
Fig. 12.18 Chart to find point of contraflexure (Source: Blum, 1931)
beam method (Blum, 1931). Blum provided a chart (Fig. 12.18) relating the angle of shearing
resistance and the distance from the point of contraflexure to the dredge line, z5, as shown
in Fig. 12.17.
The knowledge of φ and H enables the determination of z5. Now, as discussed above,
the portion above point C is treated as a beam, and the shear force P is calculated using the
moments about the anchor rod. With P known, considering the moment equilibrium about
M12_PURU1773_01_SE_C12.indd 444 2/12/2013 5:18:42 PM

Wale
Sheet piling
Tie rod
Water Backfill side Anchor

front
Plan
Fig. 12.19 Arrangement of wale, tie rod, and anchor
the base yields an expression where the only unknown, D, can be determined. The depth is
increased by 20% to 40%, and hence the design depth Dd = 1.2 to 1.4D. The force Fa on the
anchor rod may be determined using the moments about the point of contraflexure, C. With
the value of design depth known, the calculation of the bending moment and, subsequently,
the selection of the section can be done.
12.3.3 Wales, Tie rods, and Anchorages for Sheet Piling

Wales are longitudinal members of a rolled channel section usually provided back to back
along the sheet pile length, as shown in Fig. 12.19. A wale is attached to the back of the wall
if a flush-front face is required, otherwise it is placed in a horizontal position in front of the
wall. Wales may be designed simply as supported beams with spans equal to the distance
between tie rods. A cable or a steel bar, threaded to allow vertical alignment and tension
adjustments, acts as a tie rod. Sufficient protection is provided against corrosion of tie rods
by treating them with a coat of paint or asphaltic material. The spacing of the rods depends
on the total anchor force to be provided and the capacity of each tie rod. Theoretical anchor
forces are increased by 20 to 30% for design purposes, particularly in cohesive soils, with an
allowable stress in steel approaching 80 to 90 % of the yield stress.
Anchors are basically classified as tie-back and deadman-type anchors. Figure 12.20
shows a variety of anchor schemes. The tie-back anchors are preferred in places where it is
possible to encroach on the adjacent ground to instal the anchor. This type permits an unob-
structed area in front of the wall for dredging and other installations. The only disadvantage
is in encountering underground utilities. Deadman-type anchors are constructed in place by
pouring concrete or by embedding a pre-cast beam. In order to develop sufficient passive
resistance, the deadman is constructed at an adequate distance from the sheet pile wall.
12.4 BRACED EXCAVATIONS

Excavations of soils to significant depths are laterally supported temporarily by braced
sheeting and permanently by retaining walls. Braced sheeting basically consists of a sheet
piling to support the sides of the excavation, with stability being maintained by means of
M12_PURU1773_01_SE_C12.indd 445 2/12/2013 5:18:42 PM

Anchor rod Original

Anchor rods ground
Elevation Tension pile
Varying length
Compression
pile
Plan
(a) Cast-in-place deadman (b) Piles used as anchors
Tie rod Anchor plate or Pressure grout

beam
Tie rod
or cable
(c) Anchor plate or beam (d) Tie-back
Fig. 12.20 Types of anchors
struts across the excavation. A variety of materials, methods, and procedures have been in
use. The choices are influenced by factors such as subsurface condition, excavation depth,
working space, climate and season, equipment, and labour available.
Two common techniques adopted for lateral bracing are illustrated in Fig. 12.21. In the
first method, generally referred to as lagging, wooden or steel soldier beams are driven into the
ground before excavation. As the excavation progresses, horizontal wooden sheeting or steel
plates (called laggings) are placed between the solider beams. When excavation reaches the
desired depth, wales and struts are carefully installed.
Wale Wale
Strut Strut
Soldier
beam
Lagging Sheet pile
Fig. 12.21 Techniques for lateral bracing
M12_PURU1773_01_SE_C12.indd 446 2/12/2013 5:18:43 PM

In the second method, interlocking sheet piles are driven to a depth greater than the antici-
pated depth of the excavation. As the excavation progresses, wales are installed horizontally
along the excavation at intermittent depths. These wales are supported by struts placed
horizontally at the required spacings.
12.4.1 Earth Pressure Distribution

The pressure distribution against bracing is different from the conventional distribution and
depends on the deformation condition from top to bottom. Because of less yielding at the
top, the lateral earth pressure will be close to earth pressure at rest, but the degree of yielding
increases with depth and the earth pressure at the bottom may be different from that of Ran-
kine’s active pressure. Typical deflection patterns of these two types of walls are shown in
Fig. 12.22. Thus, no theory can be directly applied; it depends on valid empirical methods.
Such suggestions were given by Peck (1969) and Tschebotarioff (1949) based on the results
from field tests. The pressure distributions on braced sheeting for sand and clay as recom-
mended by them are shown in Fig. 12.23. Further, the point of application of the resultant
pressure has been found at mid-height rather than at one-third height as in Rankine’s case.
12.4.2 Failure of Braced Cuts

Because of deformation changes with depth, failure of the soil of a braced excavation takes
the shape shown in Fig. 12.22b and c. This shows that the lower part of the soil is in a state
of plastic equil-ibrium whereas the upper part is in a state of elastic equilibrium. Initial fail-
ure of one of the struts leads to a progressive failure of the entire system. Since it is essential
that no individual strut should fail, the pressure distributions shown in Fig. 12.23 are repre-
sentative of random distributions obtained from field measurements. For medium to dense
sands, a uniform distribution of 0.65 times the Rankine active value has been found to be
appropriate.
The behaviour of a braced cut in clay depends on the value of γH/c. For γH/c <4, most
of the clays retained will be in a state of elastic equilibrium, and for γH/c >4, plastic zones
develop near the bottom of the cut. Hence, appropriate pressure diagrams for clay should be
used in the design to avoid a failure.
δ δ δ
Pp
(a) Retaining wall (b) Braced cut – lagging (c) Braced cut – sheet pile
Fig. 12.22 Nature of yielding of walls
M12_PURU1773_01_SE_C12.indd 447 2/12/2013 5:18:43 PM

0.25H 0.25H
H H H 0.50H
pa pa pa
0.75H
0.25H
pa = 0.65 γ HKa γH > γH

4 <4
Ka = tan2 (45° –φ/2) c c
pa = γ H [1 – (4c/γ H)] or 0.3γ H pa = 0.2γ H
(whichever is higher) to 0.4γ H
Sand Soft to medium clay Stiff clay
(a) Lateral pressure diagrams (Source: Peck, 1969)
0.1H
0.6H
0.75H
0.7H
H H H
0.25γ H
0.4H
0.2H 0.25H
0.375γ H
0.3γ H
0.5γ H 0.5γ H
Sand Medium clay Stiff clay
(permanent support) (temporary support)
(b) Lateral pressure diagrams (Source: Tshebotarioff, 1949)
Fig. 12.23 Lateral earth pressure diagrams against braced sheeting
12.4.3 Heave of the Bottom of a Clay Cut

In braced excavations of clay, there is a possibility that the base of the excavation will fail on
heaving when the stability number is greater than about 7. The stability against heave has to
be checked before the strut loads are considered. Due to base heave and inward deforma-
tion, the soil outside the excavation will move horizontally and vertically. Because of such
movements, damage to adjacent structures and services may be possible. In general, if the
flexibility of the wall system is large and the time for installation of struts or anchors is more,
movement has to be anticipated.
M12_PURU1773_01_SE_C12.indd 448 2/12/2013 5:18:45 PM

This factor of safety of braced excavation against bottom heave can be computed based on
Terzaghi’s (1943) analysis. Figure 12.24 shows the failure pattern of the bottom of the excava-
tion, in which surfaces CD and CF are arcs of the circles.
Consider the vertical load per unit length of the cut at the level of the bottom of the cut
along the line BD and AF as
Q = γ HB ′ − cu H (12.57)
where B′ = 0.7B and cu = cohesion (for the φ = 0° condition).

Based on Terzaghi’s bearing capacity theory (discussed in Chapter 14), treating B′ as the
width of a continuous foundation, the net ultimate load per unit length of the foundation is
given as
Qnu = 5.7 cu B ′ (12.58)
Hence, the factor of safety against heave is
Qnu 5.7 cu B ′ 1⎛ 5.7 cu ⎞⎟

Fh = = = ⎜⎜⎜ ⎟⎟ (12.59)
Q γ HB ′ − cu H H ⎝ γ − (cu / 0.7 B) ⎟⎠
⎜
The above equation is based on the assumption that the clay layer is homogeneous and
extends at least up to a depth of 0.7B below the bottom of the cut. If a hard rock or a hard
material is located at a depth D < 0.7B, the failure surface will make a tangent with that
surface. In such a case, Eq. 12.59 is modified as
1 ⎛⎜ 5.7 cu ⎞⎟
Fh = ⎜ ⎟⎟
H ⎜⎜⎝ γ − (cu / D) ⎟⎠
(12.60)
G E
cu cu
B′ B B′
cu
cu
F A 45° 45° B D
0.7 B
Failure
surface
C
Fig. 12.24 Failure mechanism of bottom heave
M12_PURU1773_01_SE_C12.indd 449 2/12/2013 5:18:46 PM

Sand
Flow of water
Fig. 12.25 Flow of water through the bottom of the cut in sand
12.4.4 Stability of the Bottom of the Cut in Sand

The bottom of the cut is stable as long as the water level inside the excavation is higher than
that outside. If the water level inside the cut is lowered by pumping such that the water level
inside is lower than that outside, instability may be created as a result of an upward seepage
of water into the cut. Such a situation is shown in Fig. 12.25. Here, the factor of safety against
piping has to be checked. After drawing the flow net, the maximum exit gradient is com-
pared with the critical gradient. Thus, the factor of safety against piping is given as
ic
Fp = (12.61)
ie
where ic is the critical hydraulic gradient and ie is the maximum exit gradient.
A factor of safety above 1.5 is desirable. If no groundwater table is encountered, then the
bottom of the cut in sand is generally stable.
WORKED EXAMPLES
Example 12.1 It is proposed to construct a gravity-retaining wall 6 m high, with a backfill

sloping at an inclination of 20° with the horizontal. The base of the wall is to be placed 1.5 m
below the ground surface. The properties of the backfill material are c1 = 0, φ1 = 36°, and
γ1 = 18.1 kN/m3, and the angle of wall friction is δ = 23°. The foundation soil is a cohesive
friction soil with c2 = 35 kN/m2, φ2 = 25°, and γ2 = 19.0 kN/m3. Neglect wall friction in the
front face of the wall. Unit weight of the wall material is 23.5 kN/m3. Proportion the dimen-
sions of the retaining wall and check for safety against overturning and sliding. The water
table is located at a greater depth.
M12_PURU1773_01_SE_C12.indd 450 2/12/2013 5:18:46 PM

0.8 m
20°
c1= 0
φ1 = 36°
γ 1 = 18.1 kN/m3
δ = 23°
6m
Pa
δ = 23°
10° – Pa cos 33°
2 3 0.5 m
1
1.5 m
0.4 m
4 0.9 m 0.75 m
D
0.7 c2 = 35 kN/m2 φ 2 = 25°
3.5 m
γ = 19.0 kN/m3
2
Fig. 12.26
Solution
Based on the details given in Fig. 12.2 for proportioning the wall, the dimensions of the
retaining wall for a height of 6 m are shown in Fig. 12.26.
As the back of the wall and the soil surface are both inclined and there is wall friction,
Coulomb’s theory can be applied to get Ka from Eq. 12.45. Here, β = 80°, δ = 23°, i = 20°, and
φ = 36°. Thus,
sin 2 (80° + 36°)
Ka = = 0.42
⎡ sin( 36 ° + 23 °) sin( 36 °− 20 °) ⎤
sin 80° sin(80°− 23°) ⎢⎢1 +
2 ⎥
⎢⎣ sin(80°− 23°) sin(20° + 80°) ⎥⎥⎦
Therefore,
Pa = 12 K a γ1 H 2 = 12 × 0.42×18.1× 6 2 = 13.87 kN/m
Based on Rankine’s theory, the passive resistance can be computed as the wall friction in
the front face equals zero. Therefore,
1 + sin 25°
Kp = = 2.47
1 − sin 25°
Pp = 12 (2.47 )(19)(1.5)2 + 2(35)( 2.47 )(1.5)
Pp = 52.80 + 165.02 = 217.8
8 kN/m
Considering the moments about point D, the following table is prepared:
M12_PURU1773_01_SE_C12.indd 451 2/12/2013 5:18:47 PM

Details Force per metre (kN) Moment arm (m) Moment (kN-m)
Pv Pa sin 33° = 74.55 2.4 178.8
Wall (Section 1) 1 24.0

2 × 0.4 × 5.25 × 23.5 = 24.68 2 × 0.40 + 0.70 = 0.97
3
Wall (Section 2) 0.8 × 5.25 × 23.5 = 98.7 0.4 + 0.4 + 0.7 = 1.50 148.1
Wall (Section 3) 1 1 92.1

2 × 0.7 × 5.25 × 23.5 = 43.18 3 × 0.7 + 1.9 = 2.13
Wall (Section 4) 3.5 × 0.75 × 23.5 = 61.69 3.5/2 = 1.75 108.0
Passive pressure 217.8 1.5/3 106.0
∑V = 520.6 ∑Mr = 657
⎛H⎞ ⎛H⎞
Overturning moment ∑ Md = Ph ⎜⎜ ⎟⎟ = Pa cos 33° ⎜⎜ ⎟⎟
⎜⎝ 3 ⎟⎠ ⎜⎝ 3 ⎟⎠
∑ Md = 136.87 cos 33°( 63 ) = 229.3 kN m
∑ Mr 657
For = = = 2.87 > 2
∑ Md 229.3
Hence, safe against overturning:

∑Fr = [(∑V) tan φ2 + Bc2] + Pp
∑Fr = 514.8 tan 25° + 3.5 × 35 + 217.8 = 577.26 kN/m
∑Fd = Pa cos a = 136.87 cos 33° = 114.8 kN/m
∑ Fr 577.26
FSL = = = 5.02 > 1.5
∑ Fd 114.80
Hence, safe against sliding also.
Example 12.2 For the cantilever retaining wall shown in Fig. 12.27, determine the maxi-
mum and minimum pressures under the base of the cantilever. The relevant shear strength
parameters of the backfill and foundation soil are c′ = 0, φ = 35°, and unit weight of the soil
γ = 17.5 kN/m3. The unit weight of the wall material is 23.5 kN/m3. Find also the factor of
safety against sliding, considering the reduced value of base friction as 2/3φ°.
Solution
Considering the vertical face A′B′, Rankine’s theory can be applied to determine the active
earth pressure. Thus, Ka is obtained from Eq. 11.18b:
⎡ cos 10°− cos 2 10°− cos 2 35° ⎤

⎢ ⎥
Ka = cos 10° ⎢ ⎥ = 0.28
⎢ cos 10°− cos 2 10° + cos 2 35° ⎥
⎣ ⎦
M12_PURU1773_01_SE_C12.indd 452 2/12/2013 5:18:49 PM

A′
0.5 m
10° 4
7.9 m
7m 1 3 Pa
10°
Ph
1m 2.63 m
0.7 m 2
c1 4.8 m B′
Fig. 12.27
Pa = 12 K a γ H12 = 12 × 0.28 ×17.5×7.92 = 152.9 kN/m

Pv = Pa sin 10° = 152.9 sin 10° = 26.55 kN/m
Pb = Pa cos 10° = 152.9 cos 10° = 150.58 kN/m
Considering the moments about point D, the following table is prepared:
Details Force per metre (kN) Moment arm (m) Moment (kN-m)
Pv Pa sin 10° = 26.55 4.8 127.4
Wall (Section 1) 0.5 × 6.3 × 23.5 = 74.03 1 + 0.5/2 = 1.25 92.5
Wall (Section 2) 0.7 × 4.8 × 23.5 = 78.96 4.8/2 = 2.40 189.5
Soil (Section 3) 3.3 × 6.3 × 17.5 = 363.83 3.3/2 + 1.5 = 2.6 946.0
2
Soil (Section 4) 1
2
× 3.3 × 0.9 × 17.5 = 25.99 3
× 3.3 + 1.5 = 3.7 95.8
∑V = 569.07 ∑Mr = 1451.2
⎛H ⎞ 7.9
Overturning moment ∑ Md = Ph ⎜⎜⎜ 1 ⎟⎟⎟ = 150.58 × = 396.5 kN-m
⎝ 3 ⎠ 3
Eccentricity e = B − ∑ Mr − ∑ Md = 4.8 − 1451.2 − 396.5

2 ∑V 2 569.07
4.8
e = 2.4 − 1.85 = 0.55 < = 0.80
6
∑ V ⎛⎜ e ⎞ 569.07 ⎛⎜ 6 × 0.55 ⎞⎟
Pressure at toe pmax = ⎜1 + 6 ⎟⎟⎟⎠ = ⎜1 + ⎟
B ⎜⎝ B 4.8 ⎜⎝ 4.8 ⎟⎠
M12_PURU1773_01_SE_C12.indd 453 2/12/2013 5:18:50 PM

pmax = 200.1 kN/m 2
∑ V ⎛⎜ e ⎞ 569.07 ⎛⎜ 6 × 0.55 ⎞⎟
Pressure at heel pmin = ⎜⎜⎝1 − 6 ⎟⎟⎟⎠ = ⎜⎜⎝1 + ⎟
B B 4.8 4.8 ⎟⎠
pmin = 37.05 kN/m 2
∑ Fr ∑ V tan(2/3φ) 569.07 tan(2/3 × 35)

FSL = = =
∑ Fd Pa cos i 150.58
FSL = 1.6 > 1.5
Hence, safe against sliding also.
Example 12.3 A cantilever sheet pile wall with a simplified pressure distribution is shown
in Fig. 12.28. Determine the depth of penetration, considering a factor of safety of 2 against
passive resistance.
Solution
Since wall friction is zero, Rankine’s theory can be applied. Thus,
1 − sin 38° 1
Ka = = 0.24 and K p = = 4.2
1 + sin 38° Ka
Kp 4.2
Reduced coefficient K p′ = = = 2.1
F 2
Pa = 1
2
× 0.24 ×19(D + 2)(D + 2) = 2.28 (D + 2)2
Pp = (2.1×19× D)× 12 × D = 19.95D2
Taking moments about B,
c ′= 0
H=2 m φ ′= 38°
γ = 19 kN/m3
δ = 0°
H′ Pa
D Pp /F
H ′/3
P
B
Fig. 12.28
M12_PURU1773_01_SE_C12.indd 454 2/12/2013 5:18:51 PM

D ⎛ D + 2 ⎞⎟
19.95D2 × = 2.28(D + 2)2 ⎜⎜⎜ ⎟
3 ⎝ 3 ⎟⎠
D3 = 0.114(D + 2)3
or
⎛ D ⎞⎟3
⎜⎜ = 0.114
⎜⎝ D + 2 ⎟⎟⎠
or
D = 1.18 m
This simplified distribution does not give the exact distribution of pressure near the base,
and the calculated driving depth may be increased by 20% in addition to the reduction made
in the passive resistance. Therefore,
Embedment depth Dd = 1.20 × 1.18 = 2.26 m
Example 12.4 A cantilever sheet pile is to retain 3.5 m of sand. Water table is at 0.5 m from
the top of the backfill. For the sand γ = 19 kN/m3, γ1 = 12.2 kN/m3, Ka = 0.2 and Kp = 5. Find
the depth of penetration for a factor of safety of 1.4.
Solution
Figure 12.10 is redrawn (Fig. 12.29) with the following data: z1 = 0.5 m, z2 = 1.5 m, Ka = 0.2,
Kp = 5, γ = 19 kN/m3 and γ′ = 12.2 kN/m3.
p1 = Kaγz1 = 0.2 × 19.0 × 0.5 = 1.9 kN/m2
p2 = Ka(γz1 + γ′z2) = 0.2(19.0 × 0.5 + 12.2 × 3) = 9.22 kN/m2
From Eq. 12.12,
p2 9.22
z3 = = = 0.158 m
′
γ (K p − Ka ) 12.2(5 − 0.2)
(1) ( 2) ( 3) ( 4)
R = 12 p1 × z1 + p1 × z2 + 12 ( p2 − p1 )z 2 + 12 p2 × z3
0.5 m 1
p1
H = 3.5 m
R = 17.88 kN
2 3
p2 4 z = 1.95 m
z3
D
z4
p3 p4
Fig. 12.29
M12_PURU1773_01_SE_C12.indd 455 2/12/2013 5:18:53 PM

(9.22 − 1.9)× 3 1
= 12 ×1.9× 0.5 + 1.9× 3 + + 2 × 9.22× 0.158
2
= 0.475 + 5.7 + 10.98 + 0.728 = 17.88 kN
0.475 ( 31 × 0.5 + 3.0 + 0.158) + {5.7 ×( 32 + 0.158) + 10.98 ×( 32 + 0.158)} + 0.728 × 32 × 0.158
z=
17.88
or
1.58 + 33.15 + 0.077
z= = 1.95 m
17.88
K ′ = K p − Ka = 5 − 0.2 = 4.8 and γ ′K ′ = 58.56 kN/m 3
p0 = (γ z1 + γ ′z2 ) K p + K ′γ ′z3
= (19× 0.5×12.2× 3)5 + 58.56 × 0.158 = 239.75 kN / m 2
p0 239.75
c1 = = = 4.09
γ ′K ′ 58.56
−8 R −8 ×17.88
c2 = = = −2.44
γ ′K ′ 58.56
−6 R(2 z γ ′K ′ + p0 ) −8 ×17.88(2×1.95× 58.56 + 239.75)
c3 = =
( γ ′ K ′ )2 (58.56)2
= −14.65
R(6 zp0 + 4 R) 17.88 (6 ×1.95× 239.75 + 4 ×17.88)
c4 = =
( γ ′ K ′ )2 (58.56)2
= −14.96
z44 + 4.09 z43 − 2.44 z42 − 14.63 z4 − 14.96 = 0
For z4 = 2, the above equation gives

16 + 32.72 − 9.76 − 29.26 − 14.96 = −5.26
which is less than zero. Try z4 = 2.1, then 19.45 + 37.88 − 10.76 − 30.72 − 14.96 = 0.887 = 0.
Hence, z4 can be taken equal to 2.1 m. Therefore,
D = z3 + z4 = 0.158 + 2.1 = 2.26 m
Hence, design depth, Dd = 1.40D = 3.16 m.
Example 12.5 An anchored sheet pile wall is constructed by driving a line of piling into a
saturated cohesive soil with shear strength parameters c = 20 kN/m2 and φ = 0°. Granular
backfill is placed behind the pile up to a depth of 5 m, with a saturated unit weight of
20 kN/m3 and a unit weight of 17 kN/m3, above the water table. The shear strength param-
eters are c′ = 0 and φ′ = 33°. Anchor rods are placed 1.0 m below the surface of the backfill.
The water levels in front of the pile as well as behind it are both 3 m below the surface of the
backfill. Determine the design depth of penetration of the piling. Also, find the tension in the
anchor rod. Use the free earth support method.
M12_PURU1773_01_SE_C12.indd 456 2/12/2013 5:18:54 PM

Solution
Figure 12.15 is redrawn (Fig. 12.30) with the following given data: H = 5 m, z1 = 2 m, z2 = 3 m,
y1 = 1 m, and y2 = 1 m.
1 − sin 33°
Ka = = 0.835
1 + sin 33°
γ′ = 20 − 9.81 = 10.19 kN/m3

p1 = Kaγz1 = 0.835 × 17 × 2 = 28.39 kN/m2
p2 = (γz1 + γ′z2)Ka = (17 × 2 + 10.19 × 3)0.835
p2 = 53.92 kN/m2
R = 21 × 28.39 × 2+ 21 ×(53.92 − 28.39 )× 3 + 28.39 × 3
(1) (2) ( 3)
R = 28.39+ 38.30 + 85.17 = 151.9 kN

28.39 (1/ 3 × 2 + 3) + 38.30 ×1/ 3 × 3 + 85.17 ×(3 / 2)
z= = 1.8 m
151.9
p3 = 4c − (γz1 + γ′z2) = 4 × 20 − (17 × 2 + 10.19 × 3) = 15.43 kN/m2
Taking the moments about the anchor rod,
R( H − y1 − z ) − p3 D[H + (D / 2) − y1 ] = 0
or
151.9 (5 − 1 − 1.8) − 15.43D[5 + (D/2) − 1] = 0
or
334.148 − 7.78D2 − 61.72D = 0
or
D = 3.7 m
y1 = 1 m Fa
z1 = 2 m
1
p1
H=5m
R = 151.9 kN
2 3 z– = 1.8 m
p2
p3
Fig. 12.30
M12_PURU1773_01_SE_C12.indd 457 2/12/2013 5:18:55 PM

The design depth Dd = 1.2 × 3.7 = 4.44 m. Force on anchor rod Fa = 151.9 − 15.43 × 3.7
= 94.81 kN.
Example 12.6 The cross-section of an anchored sheet pile is shown in Fig. 12.31a. Deter-
mine the design depth of penetration. Use the fixed earth support method.
Solution
1 − sin 30°
Ka = = 0.33 ; K p = 3 ; K ′ = K p − K a = 2.67
1 + sin 30°
γ′ = 21 − 9.81 = 11.2 kN/m3, y1 = 1.5 m, y2 = 1 m
2
p1 = γz1Ka = 17.2 × 2.5 × 0.33 = 14.19 kN/m
p2 = (γz1 + γ′z2)Ka = (17.2 × 2.5 + 11.20 × 3.5)0.33 = 27.13 kN/m2
p2 27.13
z3 = = = 0.91 m
γ ′K ′ 11.2× 2.67
z5/H for φ = 30° is obtained from Fig. 12.18 as z5/H = 0.08. Or
z5 = 0.080 × 6.0 = 0.48 m
p2 ( z3 − z5 ) 27.13(0.91 − 0.45)
p2′ = = = 13.71 kN / m 2
z3 0.91
To determine the unknown force P, taking moment of the pressure diagram ADEE′C
about the anchor rod,
A
Fa
1.5 m Fa
2.5 m 1
c=0
γ = 17.2 kN/m3, φ = 30° p1 D
3.5 m c=0 2 3
φ = 30°
p2
γsat = 21.0 kN/m3 E
z5 4 5
C p ′ P
2 E′
c=0 C P
φ = 30° z5 E′
D γsat = 21.0 kN/m3
F
z4
J p2 ′′ J′
(a) (b)
Fig. 12.31
M12_PURU1773_01_SE_C12.indd 458 2/12/2013 5:18:56 PM

⎛1 ⎞ ⎛ ⎞ ⎛ ⎞ ⎡ ⎤
⎜⎜ × p1 × z1 ⎟⎟×⎜⎜ y 2 − z1 ⎟⎟ + ( p1 × z2 )×⎜⎜ z2 + y 2 ⎟⎟ + ⎢ 1 ×( p2 − p1 )× z2 ⎥
⎜⎝ 2 ⎟⎠ ⎜⎝ 3 ⎟
⎠ ⎜
⎝ 2 ⎠⎟ ⎢⎣ 2 ⎥⎦
⎛ z ⎞ ⎛z ⎞ ⎡1 ⎤
×⎜⎜⎜ z2 × y 2 − 2 ⎟⎟⎟ + ( p2′ × z5 )×⎜⎜ 5 + z2 + y 2 ⎟⎟⎟ + ⎢ ( p2 − p2′ ) z5 ⎥
⎝ 3⎠ ⎜
⎝2 ⎠ ⎢⎣ 2 ⎥⎦
⎛z ⎞
×⎜⎜ 5 + z2 + y 2 ⎟⎟⎟ − P ×( z5 + z2 + y 2 ) = 0
⎜⎝ 2 ⎠
or
⎛1 ⎞ ⎛ ⎞ ⎛ ⎞
⎜⎜ × 14.19 × 2.5⎟⎟×⎜⎜1 − 2.5 ⎟⎟ + (14.19 × 3.5)×⎜⎜ 3.5 + 1⎟⎟
⎜⎝ 2 ⎟⎠ ⎜⎝ 3 ⎠ ⎟ ⎜
⎝ 2 ⎟⎠
⎡1 ⎤ ⎛ 3.5 ⎟⎞ ⎛ 0.48 ⎞
+ ⎢ ×(27.13 − 14.19)× 3.5⎥ ×⎜⎜3.5 + 1 − ⎟ + (13.71× 0.48)×⎜⎜⎜ + 3.5 + 1⎟⎟⎟
⎢⎣ 2 ⎥⎦ ⎜⎝ 3 ⎟⎠ ⎝ 2 ⎠
⎡1 ⎤ ⎛ 0.48 ⎞
+ ⎢ (27.13 − 13.71)× 0.448⎥ ×⎜⎜ + 3.5 + 1⎟⎟⎟ − P (0.48 + 3.5 + 1) = 0
⎢⎣ 2 ⎥⎦ ⎜⎝ 3 ⎠
2.96 + 136.58 + 75.48 + 31.19 + 15.01 − 4.98P = 0
or
261.22
P= = 52.45 kN
4.98
p2″ = γ ′ K ′ z4 = 11.2 × 2.67 z = 29.9 z4 kN/m 2
Taking the moment of the diagram CE′FJ′J about J′,

⎛1 ⎞⎛ 2 z ⎞ ⎛1 ⎞z
p ( z4 + z5 ) + ⎜⎜⎜ p2′ × z5 ⎟⎟⎟⎜⎜⎜ 5 + z4 ⎟⎟⎟ − ⎜⎜⎜ p2′′ z4 ⎟⎟⎟ 4 = 0
⎝2 ⎠⎝ 3 ⎠ ⎝2 ⎠3
or
⎛1 ⎞⎛ 2 ⎞ ⎛1 ⎞z
52.45 (0.48 + z4 ) + ⎜⎜⎜ 13.71× 0.48⎟⎟⎟⎜⎜⎜ × 0.48 + z4 ⎟⎟⎟ − ⎜⎜⎜ × 29.9× z4 × z4 ⎟⎟⎟ 4 = 0
⎝2 ⎠⎝ 3 ⎠ ⎝2 ⎠3
or
25.18 + 52.45z4 + 1.05 + 3.29z4 − 4.98 z43 = 0
or
z43 − 11.149 z4 − 26.23 = 0
Take z4 = 4.0; then,
64 − 44.76 − 26.23 = −6.99 < 0
Now try z4 = 4.18; then,
73.03 − 46.77 − 26.33 = −0.03 = 0
Therefore,
D = z3 + z4 = 0.91 + 4.18 = 5.09 m
Dd = 1.2 × 5.09 = 6.1 m
M12_PURU1773_01_SE_C12.indd 459 2/12/2013 5:18:58 PM

Example 12.7 It is proposed to construct a 5 m deep trench in a stiff clay with c = 40 kN/m2,
φ = 0°, and γ = 18.5 kN/m3 and to timber it with horizontal struts at 1, 2.5, and 4 m below
the top. Draw the earth pressure envelope and make reasonable assumptions to estimate the
load that each strut can carry per metre run of excavation.
Solution
γH 18.5× 5
= = 2.3 < 4
c 40
Hence, the earth pressure distribution suggested by Peck (1969) for stiff clay (Fig. 12.23a)
is considered. The earth pressure envelope along with the strut positions are shown in
Fig. 12.32a.
pa = 0.30γH = 0.3 × 18.5 × 5 = 27.75 kN/m2
Assuming hinges at the reaction points, the entire pressure distribution and the same in
split forms are shown in Fig. 12.32b and c. Taking the moments of the forces about R′2
⎛1 ⎞ ⎛1 ⎞ 1.25
R1 ×1.5 = ⎜⎜⎜ ×1.25× 27.75⎟⎟⎟×⎜⎜⎜ ×1.25 + 1.25⎟⎟⎟ + (1.25× 27.75)×
⎝2 ⎠ ⎝3 ⎠ 2
or
R1 = 33.76 kN/m run on the top strut
R2′ = 17.34 + 34.69 − 33.76 = 18.27 kN
⎛ 1.25 ⎞ ⎛ 27.75 + 22.9 ⎞ 0.25
R2′′×1.5 = (1.25× 27.75)×⎜⎜⎜ ×1.25 − 1.0⎟⎟⎟ + ⎜⎜⎜ × 0.25⎟⎟⎟×
⎝ 2 ⎠ ⎝ 2 ⎠ 2
or
R2′′ = 20.76 kN
R3′ = 34.69 + 6.33 − 20.76 = 20.26 kN
Therefore, the force on the central strut = R2′ + R2′′ = 18.27 + 20.7 = 38.97 kN / m run. Taking
moment about R4,
R3′′×1 = 12 ×1× 22.9× 32 ×1
R3′′ = 7.63 kN
The total force on the lower strut = R3′ + R3′′ = 20.26 + 7.63 = 27.89 kN / m run.
Reaction R4 = 12 (5 + 2.5)× 27.75 − 33.76 − 38.97 − 27.89
or
R4 = 3.44 kN/m
The reaction R4 is assumed to be provided by the soil.
Example 12.8 A long 5 m wide and 8 m high vertical channel has to be constructed in a
deep cohesive soil with c = 36 kN/m2 and γ = 18 kN/m3. Before protecting the sides using
sheet piles, it is intended to check the safety of the bottom of the channel against heave.
Consider the excavation to be completed rapidly and find the factor of safety against heave.
What will be the change in the factor of safety if a hard material is present at 2.5 m from the
bottom of the channel?
M12_PURU1773_01_SE_C12.indd 460 2/12/2013 5:19:00 PM

1m
0.25H
1.5 m Assumed
as hinges pa = 0.3γ H
H=5m
0.50H
1.5 m
1m 0.25H
(a)
27.75
kN/m2
1m 1.25 m
R1
1.5 m
1.25 m
R2′
R2′′
1.25 m
1.5 m
R3′
R3′′ 1.25 m
1m
(b)
27.75
kN/m2
R1
R2′
R2′′
R3′ 22.9 kN/m2
R3′′ 22.9 kN/m2
R4 (c)
Fig. 12.32
M12_PURU1773_01_SE_C12.indd 461 2/12/2013 5:19:00 PM

Solution
The vertical load affecting the stability is acting on a width
B′ = 0.7B = 0.7 × 5 = 3.5 m
Since the excavation is to be done rapidly, the φ = 0 condition prevails. Therefore,
φ = γ HB ′ − cu H
= 18 × 8 × 3.5 − 36 × 8 = 216 kN
The net bearing pressure for a long footing for the φ = 0 condition is given as
qn = cuNc = 5.7 × 36 = 205.2 kN/m2
Net bearing load = qn(B′ × 1) = 205.2 × 3.5 × 1 = 718.2 kN
Therefore, factor of safety against heave Fh = 718.2 = 3.33

216
When the hard material is present at 2.5 m, the factor of safety is obtained by taking B′ = D;
then,
1 ⎡ 5.7 cu ⎤⎥ 1 ⎡ 5.7 × 36 ⎤
Fh = ⎢ = ⎢ ⎥ = 7.13
H ⎢⎣ γ − cu / D ⎥⎦ 8 ⎢⎣ 18 − 36 / 2.5 ⎥⎦
Example 12.9 A 11.2 m thick layer of stiff saturated clay is underlain by a 2.3 m thick layer
of sand. The saturated clay has a saturated density of 1940 kg/m3 and the sand as 1825 kg/m3.
The sand layer is at a artesian pressure head of 6.2 m. Find the maximum depth of cut that
can be made in the clay.
Solution
1940 × 9.81
γsat of clay = 1,940 kg/m3 = = 19.03 kN/m3.
1000
Let the depth of cut be H, at that point the bottom of excavation will heave.
The stability of a point A, as the interface of both the layers is considered.
From Fig. 12.33 σA = (11.2 − H)γsat
uA = 6.2 γω.
For heave to occur σ A′ should be zero.
H
Clay 11.2 m
Sand 2.3 m
Fig. 12.33
M12_PURU1773_01_SE_C12.indd 462 2/12/2013 5:19:02 PM

Therefore, σA − uA = (11.2 − H) 19.03 − 6.2 × 9.81 = 0

6.2× 9.81
i.e., 11.2 − H =
19.03
6.2× 9.81
∴ H = 112 − = 8.00 m.
19.03
Maximum cut can be made is 8.00 m.
POINTS TO REMEMBER
12.1 Gravity-retaining walls provide slope and soil retention by their weight, which may
consist of masonry, concrete, concrete in combination with soil weight, or the weight
of earth mass alone. Cantilever wall is a type of gravity wall which is economical as
the backfill is designed to provide the most of the required dead weight.
12.2 Cantilever and gravity walls are both liable to rotational and translational movements,
and hence, Rankine’s and Coulomb’s theories may be used for the calculation of
lateral pressure.
12.3 Retaining walls have to satisfy the following stability requirements: (i) safety against
overturning, (ii) safety against sliding, (iii) safety against bearing capacity failures,
and (iv) safety against overall stability.
12.4 Backfill materials for retaining structures should have high long-term strength, free
drainage and impact, and less lateral pressure. Poorly graded to well-graded sands
and gravels form excellent backfill material.
12.5 Sheet pile walls are flexible structures compared to gravity-type retaining walls and
are widely used for both small and large water front structures. Sheet pile walls
are made out of wood, precast, concrete, or steel. The two types are cantilever and
anchored sheet pile walls.
12.6 A sheet pile wall may fail in any one of the following ways: (i) forward movement of
the base due to inadequate passive resistance in front of the wall, (ii) failure by bend-
ing, and (iii) failure of anchors. Depending on the type of failure, the earth pressure
distribution varies, and it does not follow the conventional distribution adopted in
rigid walls.
12.7 Wales, tie rods, and anchorages are provided to keep a sheet pile in the required
position for the expected lifetime. Wales are longitudinal members of a rolled channel
section usually provided back to back along the sheet pile length. A cable or a steel
rod, threaded to allow vertical alignment and tension adjustments, acts as a tie rod.
Anchors may be of tie-back or deadman type.
12.8 Sheeting for braced excavations basically consists of a sheet piling to support the
sides of the excavation, with stability being maintained by means of strut across the
excavation.
12.9 Failure of the soil of a braced excavation may occur due to deformation changes with
depth or heave of the bottom of a cut or due to upward seepage of water.
M12_PURU1773_01_SE_C12.indd 463 2/12/2013 5:19:02 PM

QUESTIONS
Objective Questions
12.1 State whether the following are true or false.
1. Structures that are restrained from yielding should be designed to resist at-rest
lateral pressures.
2. Stability analysis of retaining walls based on Rankine’s theory results in unconser-
vative wall design.
3. For free-standing retaining walls, active or passive pressures can develop only by
translation of the wall.
4. The free earth support method considers the lowest section of sheeting to be fixed
in the earth.
5. The spacing of the rods in anchored bulkheads depends on the forces taken by
each rod.
12.2 Rowe’s method for reducing the moment in anchored sheet piling basically depends
on the
(a) Depth of fixity of wall at the base
(b) Modulus of rigidity of the wall material
(c) Relative flexibility of the piling
(d) Area of cross-section of the pile
12.3 For the design of braced excavation, the earth pressure distribution is based on
(a) Rankine’s hydrostatic distribution
(b) Coulomb’s distribution in the classical form
(c) Apparent pressure envelopes based on field studies
12.4 Failure of braced excavation in clay due to bottom heave may be avoided by
(a) Reducing the flexibility of the wall system
(b) Increasing the time for installation of struts or anchors
(c) Loading the ground surface with some surcharge
(d) Increasing the γH/c value to be >8
12.5 The qualities required for a material to cause minimum earth pressure with the least
movement are
(a) Free draining, rigid, and light in weight
(b) Rigid, free draining, and with high angle of internal friction
(c) Free draining, light in weight, and with low angle of internal friction
(d) Free draining, loose, and light in weight
12.6 It is a general practice to provide the face of a cantilever retaining wall with a small
batter to compensate for the
(a) Forward tilting
(b) Lateral sliding
(c) Overturning
(d) Forward sliding
M12_PURU1773_01_SE_C12.indd 464 2/12/2013 5:19:02 PM

12.7 The criterion for the design of a gravity-retaining wall is

(1) Safe against sliding
(2) Safe against overturning
(3) Safe against tensile stress
(4) Safe against bearing capacity failure
(a) 1, 2, and 4 are correct
(b) All are correct
(c) 1, 3, and 4 are correct
(d) 2, 3, and 4 are correct
12.8 In the design of a cantilever sheet pile wall, the calculated depth is increased arbi-
trarily by 20% to allow for
(a) The development of passive resistance
(b) The reduction of active thrust
(c) Sufficient grip length
(d) Erosion
12.9 Identify the incorrect statement. A sheet pile wall may fail in any one of the follow-
ing ways:
(a) Forward movement of the base
(b) Failure by bending
(c) Failure by shear
(d) Failure of anchors
12.10 Horizontal timber plants placed by hand as the excavation proceeds are referred to as
(a) Wales
(b) Compression members
(c) Lagging
(d) Sheeting
12.11 In a sheet pile wall, supporting and penetrating clay, how is the thrust likely to alter
when the clay swells or consolidates?
12.12 Explain why the hydrostatic-type linear earth pressure distribution is not valid in a
strutted excavation?
12.13 What are the possible signs of distress of masonry retaining walls? Suggest a few
remedial measures.
12.14 Explain why weep holes are provided in retaining walls.
12.15 Explain why only granular materials are preferred for the backfill of a retaining wall.
12.16 What is the necessity to remove weak natural soil behind a bulkhead prior to place-
ment of a granular fill? Explain.
12.17 How will you design a deadman-type anchor. What factors would you consider in
deciding the location of a deadman?
M12_PURU1773_01_SE_C12.indd 465 2/12/2013 5:19:02 PM

12.18 What do you understand by repeated yielding? How can you counter this effect on a
wall?
12.19 How will you decide the location of a railway line on top of a cohesive backfill of a
rough-surfaced vertical retaining wall?
EXERCISE PROBLEMS
12.1 Check the stability of the concrete retaining wall shown in Fig. 12.34. The backfill
material is a mixture of sand and gravel with the following properties: γ = 19.6 kN/
m3 and φ = 33°. The tangent of the coefficient of friction between the concrete and the
soil is 0.48. The unit weight of concrete is 2.5 kN/m3. The retaining wall is placed on
a very dense gravelly bed with an allowable soil pressure of 380 kN/m2.
1m 15 kN/m2
7m
0.5 m 0.5 m
2m
1m
6m
Fig. 12.34
12.2 Estimate the minimum and maximum pressures under the base of a cantilever retaining
wall shown in Fig. 12.35. Also, check the stability against overturning and sliding. The
properties of the backfill material are γ = 18.2 kN/m3 and φ = 38°. The friction angle at
the base of the wall is given as 27°, and the unit weight of concrete is 23.5 kN/m3.
12.3 Determine the minimum safe width of a gravity-retaining wall, supporting 5 m of a gran-
ular fill having a dry unit weight of 18.5 kN/m3 and an angle of friction 32°. The pressure
surface of the retaining wall has a batter of 1:6. The backfill is sloped with an angle of incli-
nation of 15°. The base of the wall is located at a depth of 2 m from the ground surface.
The properties of the foundation soil are c = 15 kN/m2, φ = 25°, and γ = 19.0 kN/m3.
12.4 For the sheet pile wall system shown in Fig. 12.36, determine the depth of penetration
considering the sheet pile as a cantilever type. What will be the percentage of reduc-
tion in the depth if tie rods are placed at 1.5 m from the top and 3 m from the centres?
12.5 It is intended to design a cantilever sheet pile wall to support a varved clay as detailed
in Fig. 12.37. Compute the depth required considering a factor of safety of 2.5 against
M12_PURU1773_01_SE_C12.indd 466 2/12/2013 5:19:03 PM

9m
0.5 m 4.5 m
0.7 m
6m
Fig. 12.35
1m
c=0
φ = 32°
5m γ = 17 kN/m3
γ ′ = 12.2 kN/m3
cu = 45 kN/m2
Dd= ? φu = 0⬚
γ ′ = 12.0 kN/m3
FS = 3.5
Fig. 12.36
cu = 10 kN/m , φ u = 5°
2m 2
3
γ =18.2 kN/m
cu = 20 kN/m , φ u = 0°
2
1.5 m γ =18.4 kN/m
3
cu = 22 kN/m , φ u = 2°
3
1.6 m γ sat =19.0 kN/m
3
3.7 m cu = 20 kN/m3, φ u = 5°
1.4 m γ =18.9 kN/m3
sat
0.7 m cu = 25 kN/m3, φ u = 0°
3
γ sat = 19.2 kN/m
Dd = ? cu = 80 kN/m2
Fig. 12.37
M12_PURU1773_01_SE_C12.indd 467 2/12/2013 5:19:04 PM

passive resistance. What will be the change in depth of embedment if the top two
soft layers are replaced with sand with the following properties: φ = 30° and γ =
18 kN/m3? Consider the same factor of safety.
12.6 A sheet pile wall is driven 7 m into an estuarine clay which has the following proper-
ties: cu = 18 kN/m2, φu = 0°, and γsat = 20 kN/m3. The original groundwater is located
at 1.5 m from the ground surface. Excavation has been carried out on one side of the
wall up to a depth of 4 m. Check the adequacy of the depth of penetration of the piling
below the bottom of the excavation, to give a factor of safety of 2.0 with respect to
passive resistance.
12.7 Compute the embedment depth for a 6 m high cantilever pile supporting 4 m high
water above the dredge line. The soil of the backfill and that below the dredge line are
the same, having the following properties: γsat = 22 kN/m3 and φ = 30°. It is decided
at a later date to convert the sheet pile into a closed sheet pile by providing a tie rod at
1.5 m from the top. Determine the revised design depth of embedment (with a safety
factor of 1.4) and the force on the tie rod.
12.8 For a shipping channel, an anchored sheet pile is used to support a fill. The height
of the sheet pile above the bottom of the channel is 10 m, and it supports a 8 m
head of water in the channel. The backfill soil and the soil beneath the channel are
both granular and have an average bulk and submerged unit weights of 18.6 and
12.8 kN/m3, respectively, and an average angle of friction of 32°. The anchor rod
is positioned at 1.5 m from the top of the backfill and 2.5 m centre to centre. Using
the free earth support method, find the depth of embedment and the force on
the anchor rod. The design depth may be taken as 30% more than the theoretical
depth.
12.9 For the anchored sheet pile system in granular soil shown in Fig. 12.38, determine the
depth of embedment and the force on the rod. Tie rods are placed at 3.5 m centre to
centre horizontally. A safety factor of 2.0 is applied to the passive resistance.
1m Ta=?
c=0
2m φ = 25°
3
6m γ = 17.5 kN/m
c=0
3m φ = 38°
3
γ =18.6 kN/m
c=0
Dd = ? φ = 40°
3
γ ′ = 12.1 kN/m
Fig. 12.38
M12_PURU1773_01_SE_C12.indd 468 2/12/2013 5:19:04 PM

12.10 An anchored sheet pile wall supports 5 m of fully saturated soil having the following
relevant properties: c = 5 kN/m2, φ = 32°, and γsat = 20 kN/m3. The groundwater is
0.5 m below the top of the wall. Horizontal anchors are installed at depths of 1.2 and
2.5 m from the centre. Use the free earth support method and determine the mini-
mum safe driving depth, adopting a factor of safety of 1.50. Also, estimate the force
transmitted by each anchor rod.
12.11 In a river bank protection scheme, an anchored sheet pile wall is driven to support
sand up to a depth of 4.5 m. Anchor rods are provided at 1.0 m below the top and at
3 m centre to centre. The sand has a friction angle of 32°. The surface of the retained
material is to be horizontal and level with the top of the wall. During heavy rains the
water level rises to a level of 0.5 m below the top of the wall. Neglecting cohesion
and friction on the surface of the piles, use the fixed earth support method to find the
design depth of the pile. It is required to provide a factor of safety of 1.5 against the
depth of penetration. Determine the force and diameter of the anchor rod if the tensile
strength of the material of the rod is 90 × 103 kN/m2.
12.12 An anchored sheet pile wall is constructed by driving a line of piling as shown in
Fig. 12.39. The rods are spaced at 2.5 m centre to centre, 1.5 m below the surface of the
backfill. The water level in front of the wall and the water table behind the wall are
both 3 m below the surface of the backfill. Determine the design depth considering
a factor of safety of 2.0 with respect to passive resistance. Also, find the force on the
anchor rod. Adopt the fixed earth support method.
1.5 m Fa = ?
1.5 m c ′= 0 γ =16 kN/m3
1m φ′ = 35° γ sat = 19.2 kN/m3

7m
c ′ = 17.2 kN/m
2 φ ′ = 26°
3m γ
sat
= 20.5 kN/m3
c ′= 0 φ ′ = 43°
γsat = 21.2 kN/m3

Dd = ?
Fig. 12.39
M12_PURU1773_01_SE_C12.indd 469 2/12/2013 5:19:05 PM

12.13 A 3.2 m wide and 6.5 m deep cut is proposed to be made in a moist deposit of sand
with shear strength parameters c = 0 and φ = 30°. Find the total load on the timber
sheeting if γ = 19.5 kN/m3.
12.14 The sides of an excavation 5 m deep in stiff clay are to be supported temporarily by
timber. The struts are placed at 1, 2.5, and 4 m below the top. Assuming a suitable
pressure distribution, estimate the load that each strut can carry per metre run of
excavation. The relevant properties of the soil are γsat = 21 kN/m3, unconfined com-
pressive strength qu = 200 kN/m2, and φ = 0°.
12.15 A strutted excavation 1.5 m wide is executed in a saturated plastic clay with a unit
weight of 18 kN/m3. The bottom of the excavation yields when the height reaches
10 m. Estimate the approximate shear strength of the soil that prevailed during failure.
12.16 A strutted excavation 4 m × 8 m in plan is to be taken up for the installation of a
machinery. The depth of the excavation is 5 m in a saturated stiff clay which has an
undrained strength of 40 kN/m2 and a unit weight of 19 kN/m3. There is another
supporting machinery to be placed on the surface of the ground in line with the ver-
tical face of the excavation. This machinery will be inducing an overall surcharge of
15 kN/m2. Estimate the factor of safety against base failure.
M12_PURU1773_01_SE_C12.indd 470 2/12/2013 5:19:05 PM

13
Stability of Slopes
CHAPTER HIGHLIGHTS
Causes of slope failures – Short- and long-term failures – Types of land slides
and slope movements – Factor of safety – Infinite and finite slopes – Analysis
of infinite slopes – Analysis of finite slopes: planar and circular failure surfaces:
φu = 0 analysis, friction circle method, Fellenius method of slices, Bishop’s
simplified method – Taylor’s stability chart – Location of critical circle
13.1 INTRODUCTION
Landslides are the downward and outward movements of slope materials because of
exhaustion of required shear strength. The slope materials may be composed of natural
rock and soil, artificial fills, or combinations thereof. Potential landslides in natural
slopes may be identified either by aerial photographs or by ground reconnaissance.
Slides also occur in man-made structures such as embankments and earth dams. Suffi-
cient care has to be taken to choose the correct construction material and to adopt a
suitable construction procedure to avoid sliding of the slope during or after construc-
tion. Further, the stability of foundations and earth-retaining walls against ground
break or rupture of soil is also important. One of the causes of ground break is insuffi-
cient depth of the embedment of the foundation or the retaining wall combined with
low shear strength. In principle, the analysis consists of determining the factor of the
slopes against shear failure so as to ascertain the stability of natural slopes, cuts, embank-
ments, earth dams, and ground break.
13.2 CAUSES OF SLOPE FAILURES

The causes of failure of slopes may be external or internal. External causes are those
which produce an increase in the stress at unaltered shearing resistance of the material.
They include steepening of the slope, deposition of material along the edge of slopes,
and earthquake forces.
M13_PURU1773_01_SE_C13.indd 471 2/12/2013 7:55:59 AM

Internal causes are those that lead to a slide without any change in surface conditions
which involve unaltered shearing stresses in the slope material. Some of these conditions are
the decrease in shearing resistance brought about by excess pore water pressure, leaching
of salts, softening, breakage of cementation bonds, and ion exchange. Intermediate between
landslides due to external and internal causes are those due to rapid draw-down, to sur-
face erosion, and to spontaneous liquefaction. Terzaghi (1950) reviewed the processes which
cause landslides by several modes of action of agents and represented them in a lucid form,
as shown in Fig. 13.1. As an example, some of the activities which may provoke or improve
a landslide are shown in Fig. 13.2.
Spontaneous
unlined canal
liquefaction
from a new
Seepage
3
Beginning of erosion of
construction operation
Factor of safety
Gradual softening of 2
stiff fissured clay
1
Heavy rainfall
Slope failure Exceptionally rapid draw-down
during construction
0 10 20 30
Time (years)
Fig. 13.1 Variations in the factor of safety of different slopes of recent origin (Source: Terzaghi, 1950)
Excavation of head
Excavation at toe–
removes resistance Removes part of
driving force
k
oc
dr
Be
k
oc
dr
Be
(a) Toe excavation – provokes slide (b) Head excavation – increases stability
Cracks
Existing landslide
Drainage blocked – drainage intercepts
shearing resistance water headed for
reduced cracks and fissures Seepage
shearing
resistance
ck
Seepage decreased dro
Be
(c) Drainage blocked – provokes slide (d) Drainage improved – increases stability
Fig. 13.2 Activities that decrease or increase the probabilities of slides (Source: Woods, 1950)
M13_PURU1773_01_SE_C13.indd 472 2/12/2013 7:56:00 AM

Stability of Slopes 473
13.3 SHORT- AND LONG-TERM FAILURES

Stability of natural slopes and cuts may be studied under two conditions, namely, short- and
long-term conditions. The short-term instability is due to non-availability of sufficient time
for the dissipation of pore water pressure. The long-term condition is one in which the pore
water pressure gradually adjusts itself in the long run and shows values corresponding to a
certain groundwater condition.
In the stability analysis of slopes, one may adopt effective or total stress analysis depend-
ing on the field situation. In the effective stress analysis, the proportion of the shear strength
mobilized (actual stress) for limiting equilibrium is expressed (for a factor of safety of F) as
⎛ c′ ⎞ ⎛ tan φ ′ ⎞⎟
τ = ⎜⎜ ⎟⎟⎟ + (σn − uw )⎜⎜⎜ ⎟⎟ (13.1)
⎜⎝ F ⎠ ⎝ F ⎟⎠
Two factors that should be known for the application of the above equation are σn and
uw. Generally, a suitable stress distribution is assumed, and an appropriate value of the pore
water pressure, uw, has to be used, which depends on the class of stability problem (Bishop
and Henkel, 1962): (i) Class A problems, where the pore water pressure is an independent
variable and the value of uw is obtained from the groundwater level if there is no flow net if
a state of steady seepage exists and (ii) Class B problems, where the magnitude of pore water
pressure depends on a change in stress.
In the total stress analysis, the proportion of the shear strength mobilized for the zero
condition is expressed as
cu
τ= (13.2)
F
In natural and earth dam slopes (during steady seepage condition), pore water pressure
is controlled by the prevailing groundwater conditions, and hence, they fall under Class A
problems. On the other hand, in cuts and free-standing excavations in clay, pore water pres-
sure changes because of stress release due to excavation, and hence, they fall under Class B
problems.
13.4 TYPES OF LANDSLIDES AND SLOPE MOVEMENTS

A systematic classification of slides in clay and other mass movements was proposed by
Skempton and Hutchinson (1969). This includes five basic types and six complex forms of
movements (Fig. 13.3).
13.4.1 Basic Types of Landslides

Falls. The removal of earth support causes bulging at the toe and tension crack at the top.
The development of cracks induces additional stresses on the separating mass and leads to
an ultimate failure. Clay falls occur in steep slopes and are typical short-term failures. Such
failures are found mostly in over-consolidated fissured clays.
Rotational Slides (Slips, Slumps). These types of slides are common in fairly uniform
clays or shales. The curved surface of failure, being concave upwards, imparts a backward
tilt to the slipping mass, resulting in sinking at the rear and heaving at the toe. Such slides
are deep-seated, and the failure surfaces may be circular or non-circular.
M13_PURU1773_01_SE_C13.indd 473 2/12/2013 7:56:00 AM

Falls Circular Shallow Non-circular

Rotation slides (slips and slumps)
Block slide Slab slide

Competent substratum - Translational slide
Compound slides
Lobate
Lobate or elongate Sheet
Lobate
Earth flow Mud flow Solifluction sheet and lobate
(a) Some basic types of mass movements on clay slopes
Rotational Translational
Successive
slips Multiple retrogressive slips
Slump earth flow Slices in colluvium
Lateral spreading Bottleneck slides
(b) Multiple and complex landslides
Fig. 13.3 Types of mass movements on clay slopes (Source: Skempton and Hutchinson, 1969)
Compound Slides. The surface of failure is predetermined by the presence of heterogene-

ity within the slope material. Heterogeneity usually consists of a weak soil layer or a struc-
tural feature or a boundary between two materials, for example, clay and rock or weathered
and unweathered material. Such heterogeneity prevents simple rotational slides but intro-
duces a translational element in the movement in combination with or without rotational
slide. Compound slides usually occur in soils with heterogeneity at moderate depth.
Translational Slides. These are planar and most commonly occur in a mantle of weath-
ered material, the heterogeneity being at a shallow depth. Moreover, such slides occur as
block or slab slides. Block slides are found in marls and sandstones, whereas slab slides are
a type of translational failure in more weathered clay slopes.
Flows. These are mass movements which may be of either earth flow or mud flow. While
earth flows are slow movements of softened weathered debris, mud flows are glacier-like in
form and are often well developed below the bar in fissured clays.
13.4.2 Multiple and Complex Slides

A slide may include several types of basic movements within its various parts or at different
stages in its development. These are referred to as multiple and complex slides.
M13_PURU1773_01_SE_C13.indd 474 2/12/2013 7:56:01 AM

Successive Slips. Successive rotational slips consist of an assembly of individual shallow

rotational slips. Moreover, they are common in over-consolidated fissured clays at later
stages of the free degradation process of the slopes.
Multi Retrogressive Slips. Multiple slides develop from single failures and are predomi-
nantly rotational but sometimes translational. The cause for more numerous individual
retrogressive failures may be less cohesion of the sliding mass. These slides occur more fre-
quently in actively eroding slopes of fairly high relief in which a thick stratum of over-
consolidated fissured clay or clay–shale is overlain by a rock of considerable strength.
Slump Earth Flows. These are a common type of mass movement intermediate between
rotational slides and mud flows. They develop typically in rotational slides of considerable
displacement, where the toe of the slipping mass is much broken by over-riding, which in
the presence of water softens and forms a mud flow.
Slide in Colluvium. Colluvium develops typically in the accumulation zones below freely
degrading cliffs (Hutchinson, 1967). The sliding material is usually so shifted and weathered
that individual slipped masses are no longer distinguishable. Other types of slides in collu-
vium involve the renewal of movements in debris, which is associated with individual old
slides.
Spreading Failures. These are a particular type of retrogressive translational slides. The
initial rapid movement reduces considerably and stops within a few minutes because of the
gentle slopes involved.
Quick Clay Slides. Such slides generally begin with an initial rotational slip in the bank
of a stream incised into quick clay deposits. The slipping mass is in part re-moulded to the
consistency of a liquid, which runs out of the cavity carrying flakes of the stiff, weathered
crust. In general, quick clays may fail in one of the above-mentioned ways.
13.4.3 Rates of Land Movement

Terzaghi (1950) gave a qualitative description of movements typically associated with a
landslide. Excluding mud flow, four types of movements may be recognized (Skempton and
Hutchinson, 1969; Terzaghi, 1950): (i) creep, (ii) pre-failure movements, (iii) movements dur-
ing slide, and (iv) post-failure movements.
Creep. Invariably, all slopes are subject to creep, although at an imperceptible rate. Ter-
zaghi (1950) distinguished between seasonal or mantle creep and continuous or mass creep
(Fig. 13.4). Seasonal creep is confined within the zone of seasonal changes of moisture and
temperature; at least part of the horizontal component of the ground movement is produced
by thermal expansion and contraction, swelling and shrinkage, freezing and thawing, and
other seasonal processes. Mantle creep may range from less than 1 to a few millimetres per
year. But in moderate climates, significant movements may extend to a depth of as much as
25 mm (Terzaghi and Peck, 1967). The mass or continuous creep moves at a fairly constant
rate but at a depth below the material subject to mantle creep. The load at which creep begins
is much smaller than the shear stress at failure. Similar behaviour was reported by Bishop
(1966) using laboratory-drained long-term creep tests on clays.
Pre-failure Movements. Pre-failure movements form a basis for the prediction of fail-
ure. The line Oa shows the movement which preceded the slide. The distance OD1 depends
primarily on the thickness of the zone within which the state of stress approaches the state
M13_PURU1773_01_SE_C13.indd 475 2/12/2013 7:56:01 AM

Slide
Factor of
safety 2
1
0
Slide-producing agent Time
starts to act
Downhill displacement
a
D1
c
b
Fig. 13.4 Landslide movement (Source: Terzaghi, 1950)
of failure and on the type of clay. Skempton and Hutchinson (1969) confirmed this from
several field examples. Some techniques of measuring pre-failure movements in slopes are
discussed by Terzaghi and Peck (1967).
Movements During Slide. During the first phase of the slide, the sliding mass advances
at an accelerated rate, as shown by the upper part of the curve ab. The maximum velocity of
the movement depends on the average slope angle of the surface of sliding, the resistance
available, and the nature of stratification. For a clay with a perfectly plastic stress–strain
curve after failure, the downslope movement is slow and attains a stable position with a fac-
tor of safety 1. In clays, such as over-consolidated fissured clays, which show a pronounced
peak, the slide accelerates and is carried past the stable position by its own momentum,
coming to rest with a factor of safety higher than 1 on the residual strength. But quick clays
experience the fastest movement, and the decrease in shearing resistance may be of the order
of 90%. In contrast, slides in more or less homogeneous masses of residual soil, or clay with
low sensitivity, seldom attain a velocity of more than 0.3 m/min.
Post-failure Movement. After the descent (Fig. 13.4, Point b), the movement passes into a
slow creep unless the slide has radically altered the physical properties of the sliding mass.
In the majority of clays, the shear strength on the slip surface after failure may be at or very
close to the residual strength. Heavily over-consolidated clays experience post-failure move-
ments, and the speed of movement ranges from 0 to 6 m/year. Slides in normally consoli-
dated or quick clays generally exhibit no post-failure movements. Steps in the trend of the
line bc in Fig. 13.4 correspond to creep with seasonal effects.
13.5 FACTOR OF SAFETY

In any stability analysis, some measure of the degree of safety has to be provided. Such a
measure of safety may be a factor like a limiting stress or strain or a comparative ratio of
resistance. Working stresses in any earth structure are much less than the shear strength of
the soil so as to ensure the safety of the structure. The working stress is the actual stress at a
point or along a continuous surface and may be defined as developed or mobilized strength. In
slope stability problems, shear strength is the governing factor for stability; hence, the mobi-
lized or developed shear strength (τ) is also important. If this mobilized strength is less than
M13_PURU1773_01_SE_C13.indd 476 2/12/2013 7:56:01 AM

the available strength (τf) of the soil, then the slope is said to be stable. Thus, the factor of
safety may be defined, in a form most convenient and acceptable to practical engineers, as
the ratio of the shearing resistance available along a slip surface to the total mobilized shear-
ing resistance; that is,
τ
F= f
τ
In other words, F measures the factor by which the shear strength will have to be reduced
(τ = τf / F) to bring the structure to a state of imminent collapse.
If in the mobilization process, both cohesion and friction contribute in equal proportion,
then the factor of safety is referred to as the factor of safety with respect to strength; that is,
Available shear strength
Fs =
Mobilized shear strength
c ′ + σn′ tan φ ′
Fs = (13.3)
τ
or
c ′ σn′ tan φ ′
τ= +
Fs Fs
This may be written in a more general form as

c ′ σn′ tan φ ′
τ= + (13.4)
Fc Fφ
where Fc = c′/cm is the average factor of safety for the cohesional component of strength and
Fφ = tan φ ′ / tan φm is the average factor of safety for the frictional component of strength,
where cm and φm are the mobilized cohesion and friction, respectively. Thus, Eq. 13.3 is the
case for which Fc = Fφ = Fs.
In a non-cohesive soil, c = 0; hence,
τ f = σn′ tan φ ′
Hence, Eq. 13.4 reduces to
σn′ tan φ ′
τ= (13.5)
Fφ
where Fφ is the factor of safety with respect to friction.
If for a condition Fφ is unity (i.e., full friction has mobilized) or zero (φu = 0° condition),
then the ratio of the actual cohesion to the cohesion required for stability is defined as the
factor of safety with respect to cohesion (Fc).
The cohesion required for stability is directly proportional to the height of the slope.
Hence, the factor of safety with respect to cohesion (when Fφ is unity) is sometimes referred
to as the factor of safety with respect to height (FH). This is nothing but the ratio of the critical
height Hc to the actual height Ha, the critical height being the maximum height at which the
slope will be stable. For this case, Fc = FH; hence,
M13_PURU1773_01_SE_C13.indd 477 2/12/2013 7:56:02 AM

c′
FH = (13.6)
cm
and
c′
τ= + σn′ tan φ ′ (13.7)
FH
The safety factors defined above are simply standards of comparison and have no physi-
cal meaning beyond that given by their respective definitions.
13.6 BASIC CONCEPTS OF SLOPE STABILITY ANALYSIS

All sloping surfaces are subjected to shearing stresses on nearly all the internal surfaces. The
shear strength available should overcome the same at all points. If the shear stresses are
more and if these points are adjacent and continuous, then a surface of rupture is to be
formed at the verge of failure.
Thus, the stability analysis of slopes is based on two aspects, viz.,
1. finding the most severely stressed internal surface and the associated shearing stress
(mobilized shear strength) along the surface and
2. finding the shear strength along the above surface.
Finding the most severely stressed surface is possible by adopting a rigorous theory of
elasticity or plasticity approach. In a routine analysis, such a rigorous approach is not war-
ranted and is of only academic interest.
Thus, the problem is treated as a two-dimensional one which theoretically demands a
long length of slope normal to the section. This situation exists in a majority of the cases. A
reasonable shape of the failure surface can be assumed for the required ground condition,
and the stability of the soil above such a surface is analysed using the limit equilibrium or
limit analysis method.
All the methods of analysis are based on the following assumptions:
1. The shear stress (mobilized shear strength) along the assumed surface is the same at all
points (this may not be true in practice since the point that first failed would have experi-
enced large deformation and hence less mobilized shear strength).
2. Coulomb shear strength relationship is applicable (however, the correct shear strength
parameters should be used depending on the field condition).
3. The seepage and water pressure are uniform and known all along the surface.
4. Depending on the method of analysis, an assumption regarding the distribution of
stresses has to be made to make the problem a determinate one.
13.7 INFINITE AND FINITE SLOPES

The term infinite slope is given to any slope of great extent with uniform soil conditions at any
given depth below the surface. This implies that the soil stratum is not necessarily homoge-
neous with depth but the strata of different soils are parallel to the surface of the slope.
M13_PURU1773_01_SE_C13.indd 478 2/12/2013 7:56:03 AM

Nature never provides such an idealized condition, but from a practical standpoint, such a
simplification is enough. The usual plane of failure for such slopes is planar parallel to the
surface and along a weak layer. Generally, a typical column is taken as representative of the
soil mass, and the forces causing the flow are analysed (refer Section 13.8).
The term finite slope is given to any slope of finite extent (i.e., with limited height), e.g.,
slopes of embankments, dams, cuts, canals, etc. While analysing, the entire mass of soil
above a slip surface is considered and analysed along with the forces causing the flow.
The stability of infinite and finite slopes is related to earth pressure problems. A small
movement along the slope makes the upper portion of the slope to stretch. This movement
is sufficient to bring in the active state and causes tension cracks (Fig. 13.5). The lower por-
tion resists the movement and evidently attains the passive state. As the shear strength on
the slope surface of the upper portion fully mobilizes, the lower portion is no longer in a
position to support the weight of the material above it, and the passive state is fully reached,
resulting in the failure of the material.
Earth structures always have their lengths parallel to the bases of the slope, much greater
than their width or height. It is feasible to find the extent of soil mass parallel to the base.
Although some resistance is available at the ends of the slide, it is not easy to evaluate the
same. Hence, only the resistance available at the lower boundary of the slip is considered,
45°+f/2
Active 2c
z0 =
zone g√Ka
Passive zone H
Compression
Expansion
Stress
distribution
45°+f/2
(a) Infinite slope
Tension
crack
2c
z0 =
g√Ka
Full expansion
Partial
Passive zone expansion
Compression Active zone
(b) Finite slope
Fig. 13.5 Earth pressures acting in slopes (Source: Hunt, 1986)
M13_PURU1773_01_SE_C13.indd 479 2/12/2013 7:56:03 AM

ignoring the end effects. This allows the analysis to be treated as a plane strain problem. The
analysis is made adopting either a limit equilibrium technique or a limit analysis technique.
The limit equilibrium method is used commonly in stability analysis. This method does
not consider the stress–strain relationship of the soil but concentrates only on the equilib-
rium and yield conditions. Some researchers in geotechnical engineering have approached
the problem by using the limit analysis method (e.g., Chen, 1969; Ramiah et al., 1972, etc.)
The methods explained in the subsequent sections are based on limit equilibrium analysis.
13.8 ANALYSIS OF INFINITE SLOPES

Natural infinite slopes are of heterogeneous materials and are quite often subjected to seep-
age forces. It is extremely complicated to fit in a method to suit these requirements. For
convenience, the slopes may be delineated as cohesive or non-cohesive soil with or without
seepage. The engineer has to exercise his judgement and adopt a particular method where a
specific condition may fit.
13.8.1 Infinite Dry or Moist Cohesive Slope

Consider an element of soil of width b with unit thickness normal to the cross-section
(Fig. 13.6). The soil is assumed to be homogeneous, cohesive, and without seepage, and
the slip plane is parallel to slope. The earth pressures F1 and F2 are assumed to be equal.
Resolving the forces perpendicular and parallel to the slip surface, the normal and shear
stresses are obtained as in Eqs. 13.8 and 13.9. Thus,
N γ bH cos i
σn′ = =
b / cos i b / cos i
That is,
σn′ = γ H cos 2 i (13.8)
i
b
W H
F2
F1
T W = g Hb = R
N N = W cos i
i T = W sin i
R F1 = F2
Fig. 13.6 Forces on an element of an infinite dry or moist cohesive slope
M13_PURU1773_01_SE_C13.indd 480 2/12/2013 7:56:04 AM

and
T γ bH sin i
τ= =
b / cos i b / cos i
or
τ = γ H sin i cos i (13.9)
Equating Eq. 13.9 to mobilized shear strength,

τ = cm + σn′ tan φm
cm + γ H cos 2 i tan φm = γ H sin i cos i (13.10)
Rearranging Eq. 13.10, an expression for critical depth H = Hc for clay stratum is given as
cm ⎡ sec 2 i ⎤
Hc = ⎢ ⎥ (13.11)
γ ⎢ tan i − tan φ ⎥
⎢⎣ m ⎥⎦
The strength envelope for a cohesive soil is represented by the line ABC (Fig. 13.7), and
line ODC is the line parallel to the slope. The shear strength corresponding to the normal
stress OF is BF, which is larger than the mobilized shear strength on the slip represented by
FD. Hence, under such stress conditions, no sliding occurs. But sliding would occur when
the normal stress is OE, and under this condition there is an increase in shear stress and
complete mobilization has taken place. The depth, H, at which the shear stress on the slip
plane equals the shear strength of the soil is referred to as the critical depth, Hc . Any depth
greater than this will not be stable, and sliding would occur. The factor of safety F can be
represented as
τf
F=
τ
c′ tan φ ′
F= + (13.12)
γ H sin i cos i tan i
Slope
f¢
C
B
Mohr's envelope
A D
c¢
i
O F E sn
Fig. 13.7 Limiting slope condition
M13_PURU1773_01_SE_C13.indd 481 2/12/2013 7:56:05 AM

13.8.2 Infinite Cohesive Slopes with Seepage

Let us consider a condition in which the water table is at the surface of the slope and seepage
takes place (Fig. 13.8).
The pore water pressure at a depth H is given as γ w H cos 2 i . Thus, the normal and shear
stresses are given by Eqs. 13.13 and 13.14, respectively.
σn′ = (γ − γ w )H cos 2 i
or
σ n′ = γ ′H cos 2 i (13.13)
and
τ = γ H sin i cos i (13.14)
But the mobilized shear strength is
τ = cm + σn′ tan φm
or
τ = cm + γ ′H cos 2 i tan φm (13.15)
Equating Eqs. 13.14 and 13.15
cm + γ ′H cos 2 i tan φm = γ H sin i cos i (13.16)
Rearranging,
cm sec 2 i
Hc = (13.17)
γ tan i − γ ′ tan φm
and
c′ ⎛ γ ′ ⎞ tan φ ′
F= + ⎜⎜ ⎟⎟⎟ (13.18)
γ H sin i cos i ⎜⎝ γ ⎟⎠ tan i
w
Flo
t
ne
b
i g H cos i
g H cos i sin i
g H cos2 i
Fig. 13.8 Forces on an element of an infinite cohesive slopes with seepage
M13_PURU1773_01_SE_C13.indd 482 2/12/2013 7:56:07 AM

13.8.3 Infinite Non-cohesive Slopes with Seepage

Consider the same Fig. 13.8 with cm = 0. Thus, the critical slope ic for this condition is obtained
by letting cm be equal to zero in Eq. 13.16; that is,
⎛γ′⎞
ic = tan−1 ⎜⎜ ⎟⎟⎟ tan φm (13.19)
⎜⎝ γ ⎟⎠
and
τf
F=
τ
or
γ ′H cos 2 i tan φ ′
F=
γ H sin i cos i
or
⎛ γ ′ ⎞⎛ tan φ ′ ⎞⎟
F = ⎜⎜ ⎟⎟⎟⎜⎜⎜ ⎟ (13.20)
⎜⎝ γ ⎟⎠⎜⎝ tan i ⎟⎟⎠
13.8.4 Infinite Dry or Moist Non-cohesive Slope

Considering Fig. 13.8 and letting cm be equal to zero in Eq. 13.10, we have
γ H sin i cos i = γ H cos 2 i tan φm
or
tan i = tan φm
i = φm (13.21)
So the maximum angle that could be maintained by a dry slope in a cohesionless soil is the
angle of shearing resistance of the soil.
The factor of safety is
γ H cos 2 i tan φ ′
F=
γ H sin i cos i
or
tan φ ′
F= (13.22)
tan i
13.9 ANALYSIS OF FINITE SLOPES

13.9.1 Planar Failure Surface
It is not uncommon to find a plane surface in a soil deposit or embankment with a specific
plane of weakness. Excavation in a stratified deposit quite often leads to a planar failure
M13_PURU1773_01_SE_C13.indd 483 2/12/2013 7:56:09 AM

along a plane parallel to the strata. In embankment dams with sloping cores, planes of weak-
ness within the bank consist of two or three planar surfaces.
Culmann, in 1866, considered a simple failure mechanism of a slope of homogeneous soil
with the plane failure surface passing through the toe of the slope.
Figure 13.9 shows a typical slope with a plane failure surface. The weight of the wedge is
given as
W = 12 hLγ (13.23)
An expression for b can obtained from geometry as
h H
AB = =
sin (i − θ ) sin i
Thus,
H sin(i − θ )
h= (13.24)
sin i
sin (i − θ)
= 12 Lγ H (13.25)
sin i
The force due to shear strength along plane AC is
S = c ′L + W cos θ tan φ ′
The weight component parallel to the plane AC is W sin θ. Thus, the factor of safety is
c ′L + W cos θ tan φ ′
F=
W sin θ
That is,
c ′ + 12 γ H[sin(i − θ ) sin i] cos θ tan φ ′
F= (13.26)
1
2 γ H[sin (i − θ)/sin i] sin θ
C b
B
i
h
W L
cm
H fm
P
L
f -line
q
A i
Fig. 13.9 Culmann’s slip plane
M13_PURU1773_01_SE_C13.indd 484 2/12/2013 7:56:10 AM

Now, referring to the force polygon (Fig. 13.10) and using Fig. 13.10, the sine rule is
cm L W
=
sin(θ − φm ) cos φm
Substituting for W, we have

1 Lγ H sin(i − θ)
cm L
=2
sin(θ − φm ) sin i cos φm
or
cm ⎡ sin (θ − φm ) sin(i − θ) ⎤
= 12 ⎢ ⎥ (13.27)
γH ⎢ sin i cos φ ⎥
⎣ m ⎦
where cm/γ, known as the stability number. Thus for failure to occur the stability number
has to be at a maximum. Thus, differentiating Eq. 13.27 with respect to θ, making φm = φ′ and
equating it to zero, we get
cos (θ − φ ′) sin(i − φ)− sin(θ − φ ′) cos(i − θ) = 0
from which we have

sin(θ − φ ′) sin(i − θ )
=
cos(θ − φ ′) cos (i − θ)
or
tan(θ − φ ′) = tan(i − θ)
Thus,
(θ − φ ′) = (i − θ )
Representing θ = θf,
θf = 12 (i + φ ′) (13.28)
c mL
90° – q
90°+fm
P W
q – fm
Fig. 13.10 Forces acting on sliding mass
M13_PURU1773_01_SE_C13.indd 485 2/12/2013 7:56:12 AM

which is the expression for the angle of inclination of the critical slip plane. This method is
suitable for very steep slopes.
This factor of safety is not comparable with the one defined previously.
13.9.2 Circular Failure Surfaces

The planar surface discussed above is not the one usually associated with most slope failures.
But the actual failure surfaces are curved. The mode of rupture in clay masses is reported to
be deep-seated with rotational movements over a curved rupture surface. It is reported that
the rupture mass slides down a sliding surface in a definite pattern resembling that of a
cycloid. Generally, the failure surfaces have arcs somewhat flatter at the ends and sharper at
the centre. Due to large variations in the soil properties and slope characteristics, the failure
surface with a general shape could be the best (discussed in the next section)
Based on the studies of failure of the quay wall in the harbour of Gothenburg, Sweden,
in 1916, the circular rupture surface was first proposed by Petterson (1955). Further, field
investigation by the Swedish Geotechnical Commission justified circular arcs as close
approximations of actual slip surfaces in homogeneous and isotropic soil conditions. But
some significant deviations may occur if discontinuities exist in the soil.
The methods described under this section consider the circular arc as the shape of the
failure surface. In addition to the assumptions made in the limit equilibrium analysis, it is
further assumed that the mass of soil above the rupture surface moves as a single rigid mass
and the movement is similar to a rigid body motion.
The centre of rotation for a slip circle lies somewhere above the slope. For a given slope,
a large number of potential slip circles exist with varying radii and different centres. Some
circles may pass through the toe of the slope, while others may be deep-seated and cut the
ground surface in front of the toe. A number of slip circles are chosen and safety factors
calculated adopting a method of analysis. The slip circle giving the lowest factor of safety is
referred to as the critical slip circle, along which failure is most likely to occur.
The φu = 0 Analysis. This is a total stress analysis which may be applied to the case of
a newly constructed slope or a cut in a fully saturated condition. As this method was first
adopted in Sweden, it is also referred to as the Swedish circular arc method.
X
Rotation
centre
r
B
.
t
AB = L
W AB = L
t
Fig. 13.11 The φu = 0 analysis
M13_PURU1773_01_SE_C13.indd 486 2/12/2013 7:56:12 AM

A trial slip circle with radius r is shown in Fig. 13.11. The disturbing force is the weight
(W) of the segment of soil within the arc (taking the full weight, both above the water level
and that submerged below). This force causes an instability due to the moment of the weight
(W); that is,
Disturbing moment = W x
where x is the moment arm.
This force produces the resisting moment, which is the strength along the surface and is given as
Resisting moment = cu Lr
The factor of safety is given as
Resisting moment
Fc = (13.29)
Disturbing moment
cu Lr
Fc =
W x
Alternatively, let cm be the mobilized shearing strength of the soil along the slip surface nec-
essary for equilibrium; then,

W x = cm Lr
or
Wx
cm =
Lr
Therefore,
Available cohesion c
Fc = = u (13.30)
or Mobilized cohesion cm

cu Lr
Fc =
W x
Both definitions (Eqs. 13.29 & 13.30) give to the formula as the centre of rotation is the
same for the assumed slip surface.
If the minimum factor of safety is less than unity, the slope is considered unstable. The
minimum factor of safety should be generally equal to or greater than 1.5.
In cohesive soils, due to stretching of the upper portion of the slope, tension cracks are
formed, and the development of the slip circle is terminated at the tension crack depth (Fig.
13.12). The depth of the tension crack is given as z0 = 2cu/γ, where γ is the unit weight of
the soil. No shear strength mobilization is possible along this length; instead, if the crack is
filled with water, the disturbing moment due to water pressure has to be taken into account.
If Pw (= 12 γ w z02 ) is the force due to water pressure, then the disturbing moment is Pw y; thus,

c L r
Fc = u AG (13.31)
W x + Pw y
M13_PURU1773_01_SE_C13.indd 487 2/12/2013 7:56:14 AM

x
Moment
centre
Tension
crack y Hydrostatic
pressure
r B
r 2cu
z0 = g
Pw
D gw z0
A
t W
Fig. 13.12 Effect of tension crack on φu = 0 analysis
r sin fm
Radius DR R
Tension
= r sin fm Tension
crack
Friction or crack B
B
f -circle
Pw
r D r W D
dl Cm
cm dl + sn tan fm
sn dl
A fm A
Fig. 13.13 The friction circle method
This method can be extended for multi-layered soils (under undrained condition) and for
submerged slopes.
Friction Circle Method. In this method, which is based on total stress analysis, both cohe-
sion and the angle of internal friction are taken into account. A circular failure arc from a trial
centre is shown in Fig. 13.13.
Consider an element of length dl on the trial slip circle ADB. The reaction φR on the
elemental length is directed against the direction of motion of the sliding wedge and inclined
at an angle φm to the normal, at the point of ΔR. Thus, ΔR is tangent to a concentric circle of
radius r sin φm. This smaller circle of radius r sin φm is known as the friction circle or φ-circle.
The forces acting on the element dl are
1. the shearing force due to cohesion, cm dl;
2. the shearing force due to friction, σn tan φm dl; and
3. the normal force, σn dl.
The reaction ΔR is the vector sum of the forces σn tan φm dl and σn dl. Consider the sum
of all the forces cm dl along the arc AD. Resolve the forces parallel and perpendicular to
chord AD. The sum of forces parallel to chord AD is given as
Cm = cm L
M13_PURU1773_01_SE_C13.indd 488 2/12/2013 7:56:15 AM

where L is the chord length of AD and the sum of components normal to AD is zero. Take the
moment of the forces cm dl about the centre, and equating it to the moment due to resultant
force Cm, we have

Cm r1 = cm rL

where L is the arc length of AD, or

cm Lr1 = cm rL
or

L
r1 = (13.32)
L
That is, the resultant mobilized cohesive force Cm acts at a distance r1 and parallel to the
chord. The equilibrium of the wedge is analysed by considering the following four vectors
(Fig. 13.14): the weight W, a resultant cohesive force Cm, the reaction R, and the force due to
hydrostatic pressure of water in the tension crack Pw.
The weight vector is equal to the area of the wedge times the unit weight of the soil. It
acts vertically downwards through the centroid of the wedge. This vector can be drawn to
a suitable scale. The force Pw = 12 γ w z02 . This acts horizontally at a height of 32 z0 from the top
ground surface.
Let Q be the resultant of W and Pw. Thus, the direction and magnitude of Q are known.
The direction of Cm is known, but the magnitude and/or direction of R should be fixed to
draw the force polygon.
If Fφ is assumed, the friction circle for φm equal to tan–1(tan φ/Fφ) can be drawn. Hence,
the direction of R is fixed if we assume that the resultant reaction also makes a tangent with
the φ-circle. This is not strictly true, and the resultant R actually makes a tangent with a fric-
tion circle with a slightly larger radius (say, Kr sin φ). The error involved in the assumption
is only 20% for deep-seated circles. The value of K may be read out from Fig. 13.15 for a
particular central angle.
With the knowledge of the directions of the forces Cm and R, the force polygon is com-
pleted and the required Cm is measured (Fig. 13.14). The factor of safety with respect to
cohesion, Fc = C / Cm = cL / Cm, is then computed with the assumed, Fφ. The value obtained
for Fc is compared with the assumed Fφ . If Fc ≠ Fφ , the analysis is repeated until Fc = Fφ.
Cm
R Q W
Pw
Fig. 13.14 Force polygon for the φ-circle method
M13_PURU1773_01_SE_C13.indd 489 2/12/2013 7:56:17 AM

Coefficient K
1.16 b°
1.08
Central
angle
1.0
0 40 80 120
Central angle, b °
Fig. 13.15 The coefficient K of the φ-circle assumption
Alternatively, a series of Fc values for assumed Fφ values are obtained, and the same are plot-
ted versus assumed Fφ values. A 45° line drawn from the origin intersects the curve at a point
whose projection on both the axes gives the value F = Fc = Fφ (Fig. 13.16). This value of F is
nothing but the factor of safety with respect to strength, Fs.
The friction circle method is limited to homogeneous soils and a total stress analysis. This
can be extended for problems considering effective stress also; however, the method of slices
(discussed in the next section) is more adaptable for such problems.
Taylor’s Stability Chart. Taylor (1937, 1948) proposed stability coefficients for the analysis
of homogeneous slopes in terms of total stress based on the friction circle method. Neglect-
ing tension cracks, consider two slopes of different heights with similar slope and failure
surfaces (Fig. 13.17). For such geometrically similar failure surfaces, the force diagrams are
similar. This shows that the ratio Cm /W is a constant.
But Cm = cL / Fc and W = (Area)× γ .
The factor L and area are functions of height of slope H. So
cH
Cm ∝ and W ∝ Η 2 γ
Fc
Hence,
Cm cH / Fc
=
W H 2γ
Ff = Fc Fs = Fc = Ff
Ff
45∞ Fc = Ff
Fc
Fig. 13.16 Fc–Fφ plot to find Fs
M13_PURU1773_01_SE_C13.indd 490 2/12/2013 7:56:18 AM

R2
W2 C m2
R1 Cm2
W2
R2
C m1
W1 W1
Cm1
R1
Fig. 13.17 Concept diagrams for Taylor’s stability numbers
or
c c
= m = Sn (13.33)
Fc γ H γ H
This coefficient Sn, which is non-dimensional and depends only on the geometry of the
embankment, is referred to as Taylor’s stability number.
Values of Sn and slope angle i are related for different values of φm and the depth factor
D, as shown in Fig. 13.18a and b. This chart can be used to find safety factors with respect to
cohesion, friction, or strength.
As Taylor’s stability numbers were determined from total stress analysis, the use of these
charts for effective stress conditions may lead to a serious error. These charts are applicable
at the end of construction and under short-term stability conditions. The curves are often uti-
lized to determine the safe inclination for a given height or the maximum or critical height
for a given inclination.
Thus, for the Fφ = 1 condition, if Hc is the critical height for the given slope and soil prop-
erties and Ha is the actual height, then the factor of safety with respect to height may be
calculated as
FH = Hc / Ha
The critical height, Hc, of a slope in c–φ soils is expressed as
c
Hc = Ns
γ
where Ns is a stability factor depending on φ and i. Keeping in view the non-possibility of a
base failure unless φ < 3°, a chart (Fig. 13.19) is available (based on Taylor’s data) to determine
the critical height (Terzaghi and Peck, 1967). Evidently, all the points on the curve corres-
pond to failure along the circles. The stability factor Ns is the reciprocal of Taylor’s stability
number for a particular case of Fc = Fφ = 1.0.
M13_PURU1773_01_SE_C13.indd 491 2/12/2013 7:56:19 AM

0.35 0.19
i = 53°
0.18
°
0.30
45
0.17
°
30
°
.5
22
0.25 0.16
°
Stability number, cm/gH
15
Stability number, cm/γH
0.15
0°
m
=
f
0.20 fm =0, D = ∞ 0.14
5°
5°
°
10
7.
2 °
15 0.13 n dH
0.15 25 0° H DH
°
0.12
0.10 0.11
H DH
0.10
0.05
0.09
1 2 3 4
0 Depth factor D
0 10 20 30 40 50 60 70 80 90 (b) Chart of stability numbers for the case of
Slope angle zero friction angle and limited depth
(a) Chart of stability numbers
Fig. 13.18 Stability charts (Source: Taylor, 1937, 1948)
12
°
g Hc
25
11
c
f=
°
20
°
15
Values of stability factor NS =
10
f=
f=
°
10
5°
=
f=
9
f
8
7
6
0° NS = 5.52
f= 5.3
5
4
3.85
3
90 80 70 60 50 40 30 20 10 0
Values of slope angle (degrees)
Fig. 13.19 Relation between slope angle and stability factor (Source: Terzaghi and Peck, 1967)
M13_PURU1773_01_SE_C13.indd 492 2/12/2013 7:56:20 AM

Location of Critical Circle for Total Stress Analysis. Fellenius, in 1936, proposed a
simple method of finding the centre P of a critical toe circle for a homogeneous slope with
φ = 0° condition. This point P is located with the help of directional angles α1 and α2 as
given in Table 13.1 (Fig. 13.20). Jumikis (1962) extended this for c–φ soils and gave a method
of locating the locus on which the probable centre of a toe circle may lie. P is a point on the
straight line PQ, the locus of the centre of critical slip circles. The point Q has its coordinates
H downwards from the toe and 4.5H horizontally away.
After obtaining the line PQ, trial centres are taken on PQ, and the factor of safety
corresponding to each centre is calculated. These factors of safety are plotted as shown in
Fig. 13.21. The point on the extended line PQ corresponding to the lowest factor of safety is
thus the critical centre. This method is applicable only to homogeneous soils and provides
an approximate location of the critical centre for use in an iterative method.
Taylor’s analysis also provides the data necessary to locate the critical centre of the circle
for relatively steep slopes (refer Taylor, 1948).
Table 13.1 Location angles for critical circle (based on

φ = 0 analysis)
Slope H:V Slope angle i (°) Directional angles (°)
α1 α2
0.58:1 60 29 40
1:1 45 28 37
1.5:1 30.8 26 35
2:1 26.6 25 35
3:1 18.4 25 35
5:1 11.3 25 35
Source: Fellenius (1936).
2b a2
a1
a2
i
Fig. 13.20 Location of critical centre for the φ = 0 case (Source: Fellenius, 1936)
M13_PURU1773_01_SE_C13.indd 493 2/12/2013 7:56:20 AM

F
P (F c
c)
m in
Curve of Fc
Trial centres
Slip circle
Critical centre O
D corresponding to
for f = 0° critical centre for f > 0°
Critical a
centre for f > 0° a1
H
Locus of centre of
critical slip circle
Slip circle (passing through toe)
H corresponding to critical
centre for f > 0°
Q
R 4.5H
Fig. 13.21 Location of critical centre for the φ > 0° case
Method of Slices – Fellenius Method. The method of slices is a more generalized analysis
suitable for different soils and pore water pressure conditions. This method is quite often
referred to as effective stress analysis. This method was pioneered by Swedish engineers and
more particularly by Fellenius (1936) and Petterson (1955).
The soil profile inside the assumed slip circle is divided into a convenient number of
vertical strips or slices, as shown in Fig. 13.22. The base of each slice is assumed to be a
planar surface, and other dimensions of a slice are shown in Fig. 13.22b.
The factor of safety is defined with respect to strength. Mutual support between slices
comes by way of inter-slice forces.
r sin a
O
Xn + 1
db
D Xn En + 1 dh
r C En
dW
a
dT
a dN′
dN dU
B dl
A
(a) (b)
Fig. 13.22 The method of slices
M13_PURU1773_01_SE_C13.indd 494 2/12/2013 7:56:21 AM

Considering a unit dimension normal to the section, the forces acting on a slice will be as
follows:
dW Weight of each slice including any external boundary forces
dN Total normal force at the base of slice = σn dl
dU Force due to water pressure at the base of slice = uw dl
dN′ Effective normal force at the base of slice = dN − dU = σ′n dl
dT Shear force induced along the base = dW sin α = τ dl
α Angle of inclination of base of slice
Also, En and En + 1 are the inter-slice normal forces on the nth and (n + 1)th faces, and X n
and Xn + 1 are the inter-slice shear forces on the nth and (n + 1)th faces. An assumption for
(En − En + 1) and (Xn − X n + 1) has to be made to remove the statical indeterminacy of the
problem.
Considering the moment of forces dT and dW about the centre of rotation,
∑ dT r = ∑ dW r sin α (13.34)
or
∑ τ dl r = ∑ dW r sin α
but
τf
τ=
Fs
Therefore,
τ f dl
∑ = ∑ dW sin α
Fs
or
∑ τ f dl
Fs =
∑ dW sin α
∑(c ′ dl + σ n′ dl tan φ ′)
Fs =
∑ dW sin α
Therefore,
∑(c ′ dl + dN ′ tan φ ′)
Fs = (13.35)
∑ dW sin α
A proper estimation of dN′ in each slice will yield the factor of safety Fs for a given failure arc.
In the Fellenius method, the inter-slice forces are assumed to be equal and opposite, i.e.,
(En – En + 1) = 0 and (Xn – Xn + 1) = 0. An estimation of dN′ can be obtained by resolving the
forces normal to the base; that is,
dN ′ = dW cos α − uw dl (13.36)
M13_PURU1773_01_SE_C13.indd 495 2/12/2013 7:56:23 AM

Hence, the factor of safety in terms of effective stress is given as

∑[c ′ dl + (dW cos α − uw dl) tan φ ′]
Fs = (13.37)
∑ dW sin α
From the α values for each slice, dW sin α and dW cos α are determined. A minimum
factor of safety is obtained by choosing different slip surfaces. This method in general gives
conservative values with an error of about 5% to 20% in comparison with more exact meth-
ods. For φu = 0, the factor of safety reduces to the same Eq. 13.31 for the φ = 0 analysis.
Method of Slices – Bishop’s Simplified Method. In the Fellenius approach, the omission
of side forces violates the equilibrium requirements with respect to translation. Bishop (1955)
suggested a method considering all the equilibrium equations. In the exact method, both the
inter-slice forces were considered along with the moment equilibrium. He presented a sim-
plified form of the exact method by assuming (Xn – Xn + 1) = 0 but En ≠ En + 1.
Resolving the forces parallel to the base of the slice,
1
dT = (c ′ dl + dN ′ tan φ ′)
Fs
Resolving the forces in the vertical direction,
c ′dl dN ′
dW = dN ′ cos α + uw dl cos α + sin α + tan φ ′ sin α
Fs Fs
Therefore,
⎛ c ′ dl ⎞ ⎛ tan φ ′ sin α ⎞⎟
dN ′ = ⎜⎜⎜dW − sin α − uw dl cos α⎟⎟⎟/⎜⎜⎜cos α + ⎟⎟ (13.38)
⎜⎝ Fs ⎟⎠ ⎝⎜ Fs ⎟⎠
Substituting for dl = db sec α and substituting for dN′ from Eq. 13.38 in Eq. 13.35, we have
1 ⎡ sec α ⎤
Fs = ∑ ⎢{c ′ db + (dW − uw db) tan φ ′} ⎥
∑ dW sin α ⎢⎣ 1 + tan φ ′ tan α / Fs ⎥⎦
or
Fs =
∑ [{c ′ db + (dW − uw db)tan φ ′}1/ mα ] (13.39)
∑ dW sin α
where
⎡ 1+ tan φ ′ tan α ⎤
mα = cos α ⎢ ⎥ (13.40)
⎢ Fs ⎥
⎣ ⎦
The pore pressure can be taken as a function of the overburden pressure at any point by
means of a non-dimensional pore pressure ratio,
uw
ru = (13.41)
γ dh
M13_PURU1773_01_SE_C13.indd 496 2/12/2013 7:56:24 AM

Therefore, for any slice,

uw
ru =
dW / dh
Hence, Eq. 13.39 can be rewritten as
Fs =
∑ [{c ′ db + dW (1 − ru )tan φ ′}1/ mα ] (13.42)
∑ dW sin α
As the factor of safety appears on both sides of Eq. 13.42, an iterative procedure has to be
adopted to arrive at its value. For manual use of the equation, the value of mα may be read
from Fig. 13.23 for an assumed Fs value, and a new Fs value can be obtained. Similarly, for a
different Fs value on the RHS, the corresponding Fs value on the LHS can be found. From a
plot of (Fs)RHS and (Fs)LHS, the value of Fs can be determined. This method does not satisfy
the force equilibrium condition fully, and the error involved in Fs is insignificant. The simpli-
fied method errs on the conservative side (about 3%) with reference to the exact method, and
the two methods may not lead to the same critical circle. Because of the repetitive nature of
the calculation, the method is more suitable for solution by a computer.
Effective Stress Stability Charts. Bishop and Morgenstern (1960) have presented stability
coefficients similar to Taylor’s coefficients in principle. These coefficients are based on effec-
tive stress and the pore pressure ratios. A simple expression for factor of safety, in terms of
two stability coefficients, has been suggested; that is,
Fs = m − nru (13.43)
where m and n are the stability coefficients and ru is the pore pressure ratio. The coefficients
m and n, and in turn Fs , depend on (i) the slope angle, i; (ii) the angle of shearing resistance,
φ′; (iii) the depth factor, D; and (iv) the non-dimensional parameter, c′/γ H. Charts are
1.6
Note:g is + when slope of failure arc
is in the same quadrant as ground slope 1.0
1.4
0.8
1.2 0.6
Values of ma
tan f′ 0.4
1.0 Fs
0.2
8 0.6 .4 0.2 0
0.8
0
tan f′
0
Fs
0.6
0.
0
1.
0.4
-40° -30° -20° -10° 0° 10° 20° 30° 40° 50° 60°
Values of ma
Fig. 13.23 Graph for determination of mα (Source: Lambe and Whitman, 1978)
M13_PURU1773_01_SE_C13.indd 497 2/12/2013 7:56:25 AM

available for three depth factors, viz., D = 1.0, 1.25, and 1.5 (refer Bishop and Morgenstern,
1960). It has been reported that the factor of safety is not very sensitive to changes in the
value of the depth factor.
Cousins (1978, 1980) developed stability charts considering homogeneous soil with
constant pore pressure ratio, effective shear strength parameters, and tension crack. They
are based on the friction circle method. Cousins’ stability number is the reciprocal of Taylor’s
stability number, but in terms of effective cohesion,
γ HFs
NF = (13.44)
c′
Cousins defined another term, λ cφ , to group the soil properties and slope height, where
γ H tan φ ′
λ cφ = (13.45)
c′
Charts are provided for different slope angles and pore pressure ratios, with and without
tension cracks and water in tension cracks (refer Cousins, 1978, 1980).
Cousins also provided separate charts to locate the coordinates X and Y for the centre
of the critical slip circle for different pore pressure ratios and c–φ values (refer Cousins,
1978, 1980).
It has been shown by Cousins (1980) that a tension crack tends to reduce the factor of
safety by 8% to 10%, and the presence of water in the tension crack further reduces the factor
of safety by 10%.
13.9.3 Non-circular Failure Surfaces

For simple idealized problems, the assumption of a circular failure surface is sufficiently
accurate. However, there are many practical cases where the slip surface departs from the
simple circular shape. These conditions may arise in homogeneous dams* (Bennett, 1951)
with (i) a foundation of infinite depth, (ii) rigid boundary planes of maximum or zero shear,
and (iii) a relatively stronger or weaker layer (Fig. 13.24a). Morgenstern and Price (1965)
have shown the conditions for non-circular failure surfaces which may prevail in
non-homogeneous dams when (i) a soft layer is present in the foundation, (ii) different types
of soil or rock are used in the dam cross section with varying strength and pore pressure
condition, and (iii) drainage blankets are used to facilitate the dissipation of pore pressures
(Fig. 13.24b). Similarly, there may be different field situations which may demand on analysis
using a non-circular shape for the slip surface.
For the first time, Cooling and Golder (1942) analysed an earth dam failure (for φ = 0°
condition) in which the failure surfaces were composed of two circular arcs tangential at
their point of contact and two circular arcs joined by a horizontal straight line through the
centre of a weak layer. Nonveiller (1953), based on model studies on non-homogeneous earth
dams, suggested that the stability analysis should consider a cylindrical sliding surface in the
core and straight sliding surface in the retaining body, with the factor of safety being defined
*Discussed in Chapter 20.
M13_PURU1773_01_SE_C13.indd 498 2/12/2013 7:56:26 AM

General General slip

slip surface surface
Relatively Relatively
Rough rigid boundary Frictionless rigid stronger weaker layer
plane of maximum boundary plane of layer
shear zero shear
(i) Foundation of infinite depth (ii) Effect of extreme discontinuities (iii) Effect of moderate discontinuities
(a) Homogeneous dam and foundation (Source: Bennett, 1951)
Weak foundation Cohesive core Clay core

stratum Clay core Clay shoulder
General slip
General slip
surface Drainage
surface
Granular blanket
Rock fill
fill
(i) Effect of weak foundation (ii) Effect of types of bank material (iii) Effect of drainage blanket
(b) Non-homogeneous dam and foundation (Source: Morgenstern and Price, 1965)
Fig. 13.24 Practical cases for non-circular failure surfaces
as the ratio of the passive pressure of the retaining body necessary for the maintenance of
equilibrium to the available passive pressure.
Janbu (1954) was the first to present a stability analysis with a general shape for the slip
surface, adopting the requirement of the sum of horizontal forces to be equal to zero as the
stability criterion in finding the factor of safety for a given surface. Janbu’s solution may
be applied safely to elongated shallow slip surfaces, but it errs when applied to deep slip
surfaces.
Nonveiller’s (1965) method is an extension of Bishop’s exact method but with a general
shape for the slip surface and an arbitrary point as the moment centre. This method needs a
justifiable X n and En distribution.
Morgenstern and Price (1965, 1967) presented a method of slice analysis with a general
shape for the slip surface. Two different equations have been formed, one satisfying the
no-rotation condition of the slice about its mid-point and the other satisfying the Cou-
lomb–Mohr failure criterion for effective stress. The solutions of these equations have to
be obtained by suitable inter-slice force distributions. It is mandatory to use computers to
obtain the solution.
In general, a stability analysis problem can be made determinate only when the unknown
normal stress is explored by suitable assumptions. Except for the Fellenius (1936) method, all
methods were based on the equilibrium equation and thereby placed the burden of indeter-
minacy on the internal forces. In these methods (based on the method of slices with circular
M13_PURU1773_01_SE_C13.indd 499 2/12/2013 7:56:26 AM

or non-circular slip surfaces), different distributive assumptions for the internal forces have
been made. Such distributions in no way consider the actual or approximately real normal
stress distribution. Taylor (1937) and Brown and King (1966) stressed the necessity for the
assumption of a normal stress distribution rather than an internal force distribution. A solu-
tion of Kotter’s (1903) equation with a suitable shear strength law as a function of normal
stress may yield a justifiable normal stress distribution. Such an equation was proposed by
Brinch Hansen (1953), but its application in stability analysis resulted in complex equations
(Purushothama Raj, 1967). The author adopted a polynomial of the form
σn = (dl)p (V1 + V2 dl + V3 dl2 )2 + V4 (13.46)
where dl is the elemental length on the slip surface and p a distribution factor. The unknown
constants V1, V2, V3, and V4 and the factor of safety Fs are determined from three equilibrium
equations and two boundary conditions.
For the limited slopes analysed, the method yielded very close values with Bishop’s
method for circular surfaces and Morgenstern and Price’s method for non-circular surfaces.
The method is very simple in operation and has been shown to fit well with finite element
analysis (Narain et al., 1971).
13.10 SELECTION OF SHEAR STRENGTH PARAMETERS AND STABILITY

ANALYSIS
Construction of earth structures involves stability requirements in four cases: (i) during con-
struction, (ii) at the end of construction, (iii) during the working stage (intermediate time),
and (iv) under long-term condition. During and end of construction fall under the same
category. Table 13.2 gives the four cases and the corresponding stability analysis.
Table 13.2 Selection of strength parameter in slope stability
Situation Preferred method Comment
1. End of construction with saturat- φu = 0° analysis c′−φ′ analysis permits

ed soil; construction period short check during construction
compared to consolidation time using actual pore pressures
2. Long-term stability c′−φ′ analysis with pore-
pressures given by equilibrium
groundwater conditions
3. End of construction with Either method: cu, φu from c′−φ′ analysis permits
partially saturated soil; unconsolidated undrained tests check during construction
construction period short or c′, φ′ plus estimated pore using actual pore pressures
compared to consolidation time pressures
4. Stability at intermediate times c′−φ′ analysis with estimated Actual pore pressures must
pore pressures be checked in field
Source: Lambe and Whitman (1979).
M13_PURU1773_01_SE_C13.indd 500 2/12/2013 7:56:26 AM

13.11 SLOPE PROTECTION MEASURES

Slopes that are susceptible to sliding should be protected so that the area will be safe. Slopes
which have failed recently are likely to fail under long-term condition.
Slopes have been protected by adopting some successful techniques. In general, the
corrective or protective measures involve (i) reducing the mass or loading which contributes
to sliding (ii) improving the shearing strength along the anticipated zone of failure, and
(iii) providing certain materials which will provide resistance to movement.
The protective measure to be adopted depends on different field conditions, viz., the type
of soil in the slope, the volume or depth of the soil involving in sliding, the groundwater
conditions, assessment of the complete area which may require stabilization, the space avail-
able to undertake corrective measures, topographical conditions prevailing in the area, and
the possible changes that could occur due to the vibratory measure undertaken.
Some of the protective measures which could be adopted are given in Fig. 13.25
(McCarthy, 1982). Figure 13.25a shows different techniques for reducing the weight of the
moving mass. When a base failure is anticipated, a term may be provided near the toe
(Fig. 13.25b). If a zone near the toe is susceptible to erosion, a protective rock-fill blanket
followed by a riprap can be provided (Fig. 13.25c). Shearing resistance of the soil is reduced
due to high groundwater and excess pore-water pressure. This could be avoided by lower-
ing the groundwater or intercepting the surface water. Figure 13.25d shows such a situation.
Driven piles are sometimes used to keep the moving part intact with the original ground
(Fig. 13.25e). Sometimes driven piles, sheet piling, and construction of retaining wall help
by providing lateral support and increasing the resistance of slopes to sliding (Fig. 13.25f).
If it is intended to construct a building in the vicinity of a slope, the procedure given in
Fig. 13.25g may be adopted.
M13_PURU1773_01_SE_C13.indd 501 2/12/2013 7:56:27 AM

Removed
soil Removed soil
Original
slope
Revised Modified
slope slope Benched
slope
(a) Slopes flattened or benched
Gravel–rock
fill Critical
slip circle
Earth
berm
Zone susceptible
to erosion (wave
action, etc.) if no protection
(b) Berm provided at toe (c) Protection against erosion provided at toe
Interceptor ditch for

diverting surface
flow
Collector drains
(perforated pipe in
gravel-filter envelope). Instal driven piles
Collected water can be close to the slope
Lowered first and back piles
discharged below the
water table last, to reduce the
toe by utilizing man-
holes connected to effects of driving
transverse drains on the slope's stability
(d) Lowering of groundwater table (e) Use of driven or cast-in-place piles
to reduce pore pressures in the slope
Building
General location
Soil added/removed for piling
if wall is utilized Building Basement
Sub-basement
Shallow foundations Deep foundations

or basement for or basement for
buildings below toe buildings on
(f) Retaining wall or sheet piling or cylinder piles top of slope
provided to increase resistance to sliding (g) Plan for building design to aid slope stability
Fig. 13.25 Methods to improve and protect slope stability (Source: McCarthy, 1982)
M13_PURU1773_01_SE_C13.indd 502 2/12/2013 7:56:27 AM

WORKED EXAMPLES
Example 13.1 An infinitely long slope having an inclination of 26° in an area is underlain
by firm cohesive soil (G = 2.72 and e = 0.50). There is a thin, weak layer of soil 6 m below and
parallel to the slope surface (c = 25 kN/m2, φ′ = 16°). Compute the factor of safety when the
slope is dry. If groundwater flow could occur parallel to the slope on the ground surface,
what factor of safety would result?
Solution
When the slope is dry, the factor of safety can be obtained from Eq. 13.12; that is,
c′ tan φ ′
F= +
γd H sin i cos i tan i
Here,
Gγ w 2.72× 9.807
γd = = = 17.8 kN / m 3
1+ e 1 + 0.5
Substituting
i = 26°, c ′ = 25 kN / m 2 , φ ′ = 16°, γd = 17.8 kN / m 3 , and H = 6 m
25 tan 16°
F= + = 1.18
17.8 × 6 × sin 26° cos 26° tan 26°
When there is seepage of water, the factor of safety can be obtained from Eq. 13.18; that is,
c′ ⎛ γ ′ ⎞ tan φ ′
F= + ⎜⎜ ⎟⎟⎟
γ H sin i cos i ⎜⎝ γ ⎟⎠ tan i
Here,
G+e 2.72 + 0.5
γ= γw = × 9.807 = 21.05 kN / m 3
1+ e 1 + 0.5
or
γ ′ = 21.05 − 9.807 = 11.24 kN / m 3
Hence,
25 ⎛ 11.24 ⎞⎟ tan 16°
F= + ⎜⎜⎜ ⎟ = 0.816
21.05× 6 × sin 26° cos 26° ⎝ 21.05 ⎟⎠ tan 26°
Example 13.2 A finite slope has an inclination of 48° with a horizontal ground surface.
The height of the slope is 15 m, and the details of the soil are c = 26 kPa, φ = 18°, and γ =
17.2 kN/m3. Compute the factor of safety assuming a plane rupture surface. Adopt
Culmann’s method.
M13_PURU1773_01_SE_C13.indd 503 2/12/2013 7:56:29 AM

Solution
Here, i = 0 as the ground is horizontal (Fig. 13.9). The angle of the inclination of the critical
slip surface is given as
θf = 12 (i + φ) = 12 ( 48° + 18°) = 33°
H 15
L= = = 27.54 m
sin θf sin 33°
H sin(i − θf ) 15 sin( 48°− 33°)
b= = = 5.22 m
sin i sin 48°
W = 12 bLγ = 12 × 5.22× 27.54 ×17.2 = 1236.3 kN
S = c ′L + W cos θf tan φ ′ = 26 × 27.54 + 1236.3 × cos 33° tan 18°

S = 1052.93 kN
T = W sin θf = 1236.3 sin 33° = 673.34 kN
S 1052.93
F= = = 1.56
T 673.34
Example 13.3 A 12 m deep cut with 1:1 slope is made in a layered clay deposit with the
following details:
Depth (m) Soil Cohesion (kPa)

0–5 Very soft clay 10
5–8 Medium stiff clay 50
8–15 Stiff clay 100
15 Rock –
Assume the average unit weight of the three layers to be 18 kN/m3. Compute the factor of
safety against sliding corresponding to the rotation centre shown in Fig. 13.26.
Rotation
5m
centre
7m 75° 16°
9°
Very soft clay
m
8 5m
.6
7 c1 = 10 kPa
21
5 6 3m Medium stiff clay

c2 = 50 kPa
3 4 7m
1 2
Stiff clay
c3 = 100 kPa
Rock
Fig. 13.26
M13_PURU1773_01_SE_C13.indd 504 2/12/2013 7:56:31 AM

Solution
As a hard surface is available near the toe of the slope, a base failure should be anticipated.
The slip surface is drawn tangential to the rock base.
In order to calculate the overall moment produced by the sliding mass, the mass of soil
above the slip surface is divided into slices and the moments of individual slices is taken
about the rotation centre.
Slice No. Weight of slice Lever arm (m) Moment

(Area×1×unit wt. of soil) (kN) (kN-m)
1 63.45 8.2 –520.3

2 303.04 5.0 –1515.2
3 627.65 1.0 –627.7
4 896.36 3.0 2689.1
5 966.97 7.0 6768.8
6 833.60 11.0 9169.6
7 621.76 15.0 9326.5
8 234.39 18.4 4312.7
Driving moment = 32266.7 – 2663.2 = 29603.5 kN-m

Resisting moment = r(c1l1 + c2l2 + c3l3)
Since,
π
l1 = rθ1 = 21.6 ×16°× = 6.03 m
180°
π
l2 = rθ2 = 21.6 × 9°× = 3.39 m
180°
π
l3 = rθ3 = 21.6 ×75°× = 28.27 m
180°
Therefore, the resisting moment = 21.6 (10 × 6.03 + 50 × 3.39 + 100 × 28.27) = 66026.88 kN m
Resisting moment
F=
Driving moment
66026.88
F= = 2.23
29603.5
Example 13.4 The bank of a canal is 9.4 m in height and has a face inclination of 30°. The
material is homogeneous silty clay of unit weight 20 kN/m3, cohesion 30 kPa, and angle of
shearing resistance 20°. For the trial slip circle shown in Fig. 13.27, find the factor of safety
with respect to cohesion by using the friction circle method, if Fφ = 1.50.
M13_PURU1773_01_SE_C13.indd 505 2/12/2013 7:56:32 AM

3.07m R
r = 13 m
r1 = 14.42 m 9.4 m
4m
W R 200 kN
Scale
Cm
Fig. 13.27
Solution
⎛ tan φ ′ ⎞⎟ ⎛ ⎞
φm = tan−1 ⎜⎜⎜ ⎟⎟ = tan−1 ⎜⎜ tan 20° ⎟⎟ = 13.64°
⎜⎝ Fφ ⎟⎠ ⎟ ⎝ 1.50 ⎟⎠
⎜
Radius of friction circle r0 = r sin φm = 13 sin 13.64° = 3.07 m

Area of sliding mass = 2
3
× 18 × 3.8 = 45.6 m2
Weight of driving mass = 45.6 × 1 × 20 = 912 kN
Central angle = 88°
Therefore,
π
L = 13 × 88°× = 19.97 m
180°

L 19.97
r1 = = 13 × = 14.42 m
L 18
Cm is read from the force polygon as 125 kN. Therefore,
Cm 125
cm = = = 6.94 kPa
L ×1 18 ×1
Therefore,
c 30
Fc = = = 4.32
cm 6.94
Example 13.5 A 60° sloping embankment has a height of 6.5 m. The embankment soil
possesses the following properties: γ = 18 kN/m3, φ = 28°, and c = 20 kPa. Determine the
factor of safety with respect to strength. Use Taylor’s chart.
M13_PURU1773_01_SE_C13.indd 506 2/12/2013 7:56:33 AM

Solution
Mobilized friction
⎛ tan φ ⎞⎟
φm = tan−1 ⎜⎜⎜ ⎟⎟
⎜⎝ Fφ ⎟⎟⎠
Assuming Fφ = 1.6,
⎛ tan 28° ⎞⎟
φm = tan−1 ⎜⎜ ⎟ = 18.4°
⎜⎝ 1.6 ⎟⎠
From Fig. 13.18a, for i = 60° and φm = 18.4°, the value of cm/γH is obtained by interpolation
as 0.1007. Therefore,
cm = 0.1007 ×18 × 6.5 = 11.78 kPa
and
c 20
Fc = = = 1.698
cm 11.78
Now take Fφ = 1.65, then
⎛ tan 28° ⎞⎟
φm = tan−1 ⎜⎜ ⎟ = 17.86°
⎜⎝ 1.65 ⎟⎠
Again, after interpolation, cm/γH for i = 60° and φm = 17.86° is obtained as

cm
= 0.1018
γH
Therefore,
cm = 0.1018 ×18 × 6.5 = 11.91 kPa
and
20
Fc = = 1.68
11.9
cm
Fφ = 1.67 , φm = 17.66°, and = 0.1022
γH
Trying again with cm = 0.1022 × 18 6.5 = 11.96 kPa,
20
Fc = = 1.672
11.96
Fc = Fφ = 1.671
Therefore, the factor of safety with respect to strength Fs = 1.671.
Example 13.6 It is proposed to construct a 10 m high highway embankment with the

following soil properties: c = 18.8 kN/m2, γ = 17 kN/m3, and φ = 10°. What is the inclination
required for the embankment if the design Fc = 1.5 and Fφ = 1.0 ?
M13_PURU1773_01_SE_C13.indd 507 2/12/2013 7:56:35 AM

Solution c c
Stability number, Sn = m =
γ H Fc γ H
Substituting the values, we have
18.8
Sn = = 0.0737
1.5×17 ×10
As Fφ = 1.0, the mobilized friction angle φm = φ = 10°.

From Fig. 13.18a, for Sn = 0.0737 and φm = 10°, the slope angle is read as 29°. Therefore,
the required inclination of the embankment, i = 29°.
POINTS TO REMEMBER
13.1 Causes of failure of slopes may be external or internal. External causes are those
which produce an increase in the shearing stresses at unaltered shearing resistance
of the material. Internal causes are those which lead to a slide without any change in
surface conditions which involve unaltered shearing stresses in the slope material.
13.2 The short-term instability of a slope is due to non-availability of sufficient time for the
dissipation of pore water pressure. The long-term condition of a slope is one in which
the pore water pressure gradually adjusts itself in the long run and shows values
corresponding to a certain groundwater condition.
13.3 Types of land slides may be falls, rotational slides, compound slides, translational
slides, flows, and multiple and complex slides. Rates of landslides are recognized as
creep, pre-failure movements, movements during slide, and post-failure movements.
13.4 Factor of safety of a slope is defined as the ratio of shearing strength available along a
slip surface to the total mobilized shearing strength. Factor of safety is also defined in
certain cases with respect to cohesion, friction, or height of a slope.
13.5 Any slope of great extent with uniform soil conditions at any given depth below the
surface is termed as infinite slope. Any slope of finite extent, i.e., with limited height,
is termed as finite slope. All natural slopes are infinite slopes: slopes of embankments,
dams, cuts, canals, etc., are finite slopes.
13.6 Slip surfaces are generally curved, deep-seated, somewhat flatter at the ends, and
sharper at the centre. Analysis with general shape as the slip surface is cumbersome.
For all practical purposes, most of the analyses use a circular slip surface.
13.7 φu = 0 analysis is a total stress analysis which may be applied to the case of a newly
constructed slope or a cut in a fully saturated condition.
13.8 Friction circle method (assumes a circular slip surface) is based on total stress analysis;
both cohesion and the angle of internal friction are considered with friction completely
mobilized. Then, the factor of safety is defined with respect to cohesion.
13.9 Fellenius’ method of slices (assumes a circular slip surface) is a more generalized
analysis suitable for different soils and pore water pressure conditions and based on
effective stresses.
M13_PURU1773_01_SE_C13.indd 508 2/12/2013 7:56:36 AM

13.10 Bishop’s method of slices (assumes a circular slip surface) considers all equilibrium
conditions, including side forces on slices.
13.11 Taylor’s stability chart for homogeneous soil is based on the friction circle method
and total stresses and provides a factor of safety for a given slope angle and mobilized
friction angle.
QUESTIONS
Objective Questions
1. The maximum possible slope angle in a granular soil is equal to the friction angle
of the soil.
2. Gravitational forces tend to cause instability in natural slopes.
3. The term infinite slopes is given to earth masses of varying inclinations and
non-uniform soil conditions of unlimited extent.
4. Tension cracks do not significantly affect the safety factor of a slope.
5. The most critical circle is the one along which failure is most likely.
13.2 Total stress method of stability analysis may be applied to find the factor of safety in
the case of a newly cut slope in
(a) Fissured over-consolidated saturated clay
(b) Non-fissured over-consolidated saturated clay
(c) Normally consolidated saturated clay
(d) Partially saturated expansive clay
13.3 Total stress method of stability analysis may be applied to find the factor of safety of
an embankment dam under end-of-construction condition
(a) Method of slices with φ > 0° condition
(b) φu = 0° analysis
(c) Friction circle method (ignoring the effect of tension cracks)
(d) Friction circle method with tension crack
13.4 In stability analysis, mobilized shear strength is referred to as
(a) Maximum shear stress
(b) Applied shear stress
(c) Developed cohesion only
(d) Developed friction only
13.5 Bishop’s simplified method of slices satisfies
(a) All the statical equilibrium conditions
(b) Only the vertical force equilibrium condition
(c) Only the moment equilibrium condition
(d) All the conditions except the horizontal force equilibrium conditions
13.6 Inclination of a clay slope
(a) Can be greater than the angle of shearing resistance
(b) Cannot be greater than the angle of shearing resistance
M13_PURU1773_01_SE_C13.indd 509 2/12/2013 7:56:36 AM

(c) Cannot be greater than the angle of repose

(d) Cannot be greater than 45°
13.7 Identify the incorrect statement. The following factors cause instability of slopes
because of increased stresses:
(a) Removal of part of the slope by excavation
(b) Shock caused by earthquake or blasting
(c) Water pressure in cracks
(d) Swelling of clays by adsorption of water
13.8 The effective stress method of stability analysis is used
(1) For analysing the long-term stability of slopes
(2) For analysing dense, moderately compressible soil material
(3) For analysing the stability of compressible soils where some drainage of water
takes place when a load is applied
(a) 1, 2, and 3 are correct (b) 1 and 2 are correct
(c) 2 and 3 are correct (d) 3 and 1 are correct
13.9 A base failure is likely to occur when
(a) φu > 0° and β < 53° (b) φu = 0° and β > 53°
(c) φu = 0° and β < 53° (d) φu > 0° and β > 53°
where β is the slope angle and φu is the undrained friction angle
13.10 In order to use Taylor’s stability chart for sudden draw-down condition, the weighted
friction angle, φw, should be equal to
(a) (γ′/γw)φu (b) (γw/γ′)φu (c) (γ′/γsat)φu (d) (γw/γ′)φu
13.11 Derive an equation for the factor of safety of an infinite slope in a cohesionless soil,
assuming that seepage is
1. emerging from the slope at an angle α, which is less than the slope angle i, or
2. flowing parallel to the slope at a certain depth from the surface.
13.12 Explain the various causes of the failure of earth slopes.
13.13 Explain why a high factor of safety of 2.5 to 3 for shallow foundations and a low factor
of safety of 1.1 to 1.5 for stability of slopes are adopted.
13.14 It is often stated that refinements in stability analysis by using different methods
is generally not as significant as the correct use of the shear parameters of the soil.
Discuss the validity of this statement.
13.15 Discuss the different investigations needed to effect corrective measures in a landslide
area.
13.16 Distinguish between the total and effective stress approaches of stability analysis.
Indicate the advantages and shortcomings of the total stress approach.
13.17 Explain the various types of failures of finite slopes indicating the situations in which
they are likely to occur.
M13_PURU1773_01_SE_C13.indd 510 2/12/2013 7:56:36 AM

EXERCISE PROBLEMS
13.1 A 21° infinite slope consists of an uniform 5 m thick layer of sandy clay. At 5 m depth,
a shale ledge runs parallel to the surface. A laboratory investigation on the sandy clay
revealed the following properties: c = 20 kPa, φ = 15°, γ = 18 kN/m3. Compute the
factor of safety against sliding on the shale and ledge if (i) no water exists at the top of
the shale and (ii) the water level is at the surface of the slope.
13.2 A sub-surface investigation on a 12° natural slope revealed the presence of bedding
planes dipping toward the slope at an angle of 40°. A 60° cut slope is to be excavated to
a depth of 8 m as shown in Fig. 13.28. Estimate the factor of safety of the slope. The shear
strength parameters of the soil in the bedding plane are, c = 15 kN/m2 and φ = 28°. The
average unit weight of the soil, on the bedding plane and above, is 18.5 kN/m3.
12°
Bedding plane
8m
40°
60°
Fig. 13.28
13.3 A 45° cut was made in a clayey silt soil with c′ = 12 kPa, φ = 30°, and γ = 19.5 kN/m3.
A sub-surface exploration revealed the presence of a thin soft clay with c = 13 kPa and
φ = 0°, at a depth of 18 m from the ground surface. Estimate the factor of safety of the
slope against sliding along the composite slip surface, as shown in Fig. 13.29.
18 m
A
Clayey silt
15 m
45° 6m D
3m
B C
Soft clay 6m
Fig. 13.29
M13_PURU1773_01_SE_C13.indd 511 2/12/2013 7:56:37 AM

13.4 A cutting in clayey soil is shown in Fig. 13.30. The undrained shear strength param-
eters are cu = 48 kN/m2 and φu = 0°. The unit weight of the soil is 20 kN/m3. Compute
the factor of safety against the slip surface shown when (i) no tension crack is formed,
(ii) a tension crack exists with no water in the crack, and (iii) the tension crack is com-
pletely filled with water.
4m
r
6m
r z0
10 m
40°
Fig. 13.30
13.5 A 15 m deep 45° cut is excavated in a soil profile as follows:
Depth Soil Shear strength parameters Unit weight

2
c (kN/m ) φ (°) (kN/m3)
0–2.5 Medium stiff clay 52 0 17.2
2.5–8.6 Stiff clay 60 0 18.5
8.6–18.2 Very stiff (firm) clay 73 0 19.1
18.2 Shale (rock) – – –
1. Compute the factor of safety with respect to base failure assuming the centre of failure
circle to be above the mid-point of the slope. Also, verify by various centres and radii.
2. Check the results, using Taylor’s stability chart for an average cohesion.
13.6 Determine the factor of safety for the trial as shown in Fig. 13.31, using the friction
circle method. The soil parameters are γ = 16 kN/m3, c = 15 kPa, and φ = 28°.
4m
r
14 m
1 8m
1.5
Fig. 13.31
M13_PURU1773_01_SE_C13.indd 512 2/12/2013 7:56:38 AM

13.7 An embankment is made of soil having a cohesion of 50 kPa, an angle of internal fric-
tion of 22°, and a unit weight of 19 kN/m3. Locate the centre of rotation (for φ = 0)
by the Fellenius method and determine the factor of safety along a slip circle passing
through the toe. Use the friction circle method.
13.8 A 15 m high clay embankment with a 45° slope has the following parameters: c =
22 kPa, φ = 0°, and γ = 18.2 kN/m3. What will be the factor of safety of this slope if a
rock stratum exists 15 m beneath the toe elevation?
13.9 An excavation has to be made with an inclination of 35° in a soil with c′ = 28 kPa, φ′ =
26°, and γ = 18 kN/m3. What is the maximum height to which the excavation can be
made if Fc = 1.25?
13.10 A canal is excavated to a depth of 5 m below the ground level through a soil stratum
having the shear strength τ = c + σn tan 15°, c = 16 kN/m2, void ratio e0 = 0.72, and
specific gravity G = 2.70. The bank of the canal has a slope of 1:1. Compute the factor
of safety of the slope with respect to cohesion when the canal runs full. If it is sud-
denly and completely drawn down, what will be the change in the factor of safety?
13.11 A 10 m deep silty clay cut has an inclination of 45° and the following soil parameters:
cu = 30 kPa, φu = 10°, and γ = 18 kN/m3. Estimate the critical height of the slope in
this soil.
13.12 A proposed cutting in a c – φ soil will be 15 m deep with a slope of 1V:2.5H. The soil
has an average unit weight of 18.6 kN/m3 and an average pore pressure ratio ru of
0.45. The shear strength parameters of the soil under different conditions are
cu = 85 kN / m 2 , φu = 0°
c ′ = 12 kN / m 2 , φ ′ = 26°
Estimate the factor of safety against (i) immediate shear failure and (ii) long-term
shear failure.
13.13 For the soil slope and trial slip surface shown in Fig. 13.32, estimate the factor of safety
adopting Bishop’s simplified method. A preliminary approximate calculation for the
slip surface, based on the Fellenius method, gave a factor of safety of 2.
5m
Rotation centre
5.4 m
Soil 1
c ′=15 kN/m2
9m
f′=18°
g =17.5 kN/m3 15 m
Soil 2
c′ = 0 45°
f′ = 34°
g = 19.2 kN/m3 Slip surface
Fig. 13.32
M13_PURU1773_01_SE_C13.indd 513 2/12/2013 7:56:39 AM

M13_PURU1773_01_SE_C13.indd 514 2/12/2013 7:56:39 AM
14
Bearing Capacity of Soils
CHAPTER HIGHLIGHTS
Modes of failure – Bearing capacity theories: Terzaghi’s bearing capacity
theory – Effect of soil compressibility – Effect of water table – Foundation
pressures – Special loading and ground conditions: eccentric load, inclined
load, stratified soils, partially saturated soils and desiccated soils – Other
bearing capacity theories: Modified bearing capacity formulae, Skempton’s
bearing capacity theory, Meyerhof’s bearing capacity theory, Brinch
Hansen’s bearing capacity theory – Bearing capacity from building codes –
Permissible settlements – Allowable bearing pressure – Bearing capacity from
field tests – Bearing capacity from building codes – Factors affecting bearing
capacity
14.1 INTRODUCTION
A foundation is that part of the structure which is in direct contact with the ground and
transmits the load of the structure to the ground. It includes the soil or rock of the earth’s
crust or any special part of the structure which serves to transmit the loads into the soil
or rock. The main purpose of the transmissions of load can be satisfied by a particular
type of foundation that takes into account the properties of the supporting soil. A foun-
dation functions properly only if the supporting soil performs properly. Consequently,
the structural support is actually being provided by a soil–foundation system. This
combination of soil and foundation (now referred to as soil–structure interaction) can-
not be separated. Although engineers are aware of this relationship, it is common prac-
tice to consider the structure to be sound and to attribute the failure of the foundation to
the failure of the supporting soil.
Foundations may be grouped as shallow or deep foundation depending on the depth
of installation of foundation.
M14_PURU1773_01_SE_C14.indd 515 2/12/2013 7:54:46 AM

14.2 BEARING CAPACITY

The bearing capacity of a soil is the maximum intensity of loading which the soil can carry
without being detrimental to the normal functioning of a foundation.
14.2.1 Bearing Capacity Criteria

The bearing capacity of a soil is based on the stability requirement of a foundation.
The two criteria on which the bearing capacity of a soil depends are shear strength and
settlement.
1. The shear strength criterion is that the shear failure of the foundation or bearing capacity
failure should not occur.
2. The settlement criterion is that the foundation shall not settle more than the safe or
tolerable magnitude of settlement such that the anticipated settlement due to the applied
pressure on the soil should not be detrimental to the stability of the foundation.
These two criteria are independent and have to be dealt with separately. The bearing
capacity value to be decided for the design requirement of a foundation is the smaller of
these two values based on the above two criteria. This smaller value of bearing capacity is
referred to as allowable soil pressure (dealt in detail in Section 14.10).
14.2.2 Factors Affecting Bearing Capacity

Keeping in view the above two criteria, the following factors directly or indirectly affect the
bearing capacity of a soil.
1. Type of soil, i.e., homogeneous, layered, expansive, etc., and its physical and engineering
properties
2. Initial stress condition of the soil due to pre-history and due to the existing structure in
and around the proposed foundation
3. Location of groundwater in the soil and its fluctuations with time
4. Type of foundation, i.e., shallow or deep, and other factors such as shape, size, and rigid-
ity condition of the foundation
5. Depth and location of foundation
6. Allowable settlement of the foundation which shall not be detrimental to the functioning
of the foundation
7. Natural calamities such as earth quake, floods, heavy wind, etc., of the region where the
structure has to be located
14.3 MODES OF SHEAR FAILURE

The load–deformation relationship of a soil is not unique as it is in more homogeneous
materials, like steel. Nevertheless, it can be generalized to a certain extent even in soils.
A stratum subjected to loading through a footing will depict a reasonable elastic relationship
up to a certain percentage of the ultimate strength. This phase of deformation is attributed
to the densification of the stratum. Further increase in load causes a rapid increase in
deformation. This increased rate of yielding is due to a combination of decrease in void ratio
M14_PURU1773_01_SE_C14.indd 516 2/12/2013 7:54:46 AM

Bearing Capacity of Soils 517
and the lateral displacement. Subsequent loading leads to excessive deformation and
ultimate shear failure of the soil stratum. This pressure which has caused a shear failure of
the supporting soil is usually referred to as the ultimate bearing capacity of the foundation.
Figure 14.1 typifies a load–settlement relationship for the case of a footing on a hypothetical
stratum. Different types of soils with varied conditions show wide variation in load–
settlement relationships.
Three principal modes of shear failure have been identified, based on the model tests of
strip footings on sand (Vesic, 1973).
14.3.1 General Shear Failure

General shear failure, usually associated with dense or stiff soils of relatively low compress-
ibility, is said to occur when a continuously well-defined slip surface develops on one or both
sides of the footing and extends from the edge of the footing to the soil surface. As the pressure
is increased towards the ultimate value (qf) the state of plastic equilibrium is attained, initially
in the soil around the edges of the footing, which then gradually spreads downwards and
outwards. Then, plastic equilibrium develops throughout the soil above the failure surface.
The failure is reflected by the heave of the ground surface on both sides of the footing. The
load–settlement curve is linear up to a substantial percentage of the ultimate load but thereaf-
ter shows a rapid yielding till the load intensity approaches the ultimate value (Fig. 14.2a).
Load
Settlement
Theoretical failure
cracking
Local
Distortion Rapid downward

movement shear
failure
Fig. 14.1 Load–settlement curve (Source: Vesic, 1973)
Q QM
Q
d Strain Stress
controlled controlled
(a) General shear

Q QM
Q
Q (b) Local shear Qu Qu

Q
d Test at
greater depth
Surface
(c) Punching shear test
Fig. 14.2 Modes of bearing failure (Source: Vesic, 1973)
M14_PURU1773_01_SE_C14.indd 517 2/12/2013 7:54:46 AM

14.3.2 Local Shear Failure

In case of local shear failure, usually associated with medium dense or medium stiff soils,
the slip surface extends from the edges of the footing to a certain length (approximately up
to the boundary of Rankine’s passive state) but does not reach the ground surface. The soil
undergoes a significant compression beneath the footing and only partial development of
plastic equilibrium takes place. The heave of the ground is comparatively less and no tilting
of foundation is expected (Fig. 14.2b).
The load–settlement curve displays a lesser degree of linearity. Because of high compress-
ibility, a large settlement is characterized in the load–settlement curve, and no well-defined
ultimate load is observed.
14.3.3 Punching Shear Failure

Punching shear failure, usually associated with loose or soft soils, is said to occur when there
is compression beneath the footing accompanied by shearing in the vertical direction around
the edges of footing. There is little horizontal strain and no apparent heave of soil around the
footing (Fig. 14.2.c).
The load–settlement curve shows a relatively large settlement and the ultimate load is not
well-defined. Although deformations are considerable, sudden collapse or tilting failures
are not common. Punching shear failure is also possible in soils, if the foundation is located
at a considerable depth or as a result of the compression of an underlying soft layer.
Vesic (1973) showed the dependence of the mode of failure on the compressibility of the
soil and the depth of the foundation relative to its breadth. Figure 14.3 depicts the relation-
ship for mode of failure of foundations on sands as proposed by Vesic (1973).
In the expression for B* = 2BL /(B + L) , L is always greater than B for square and circular
footings.
Table 14.1 illustrates the types of shear failure of footings.
0
Punching Local General
shear shear shear
1
of foundation Dr/B
Relative depth
4 Dr
B
5
0 20 40 60 80 100
Density index of sand, Dr %
B∗ = B for a square or circular footing
B∗ = 2BL/(B + L) for rectangular footing
Fig. 14.3 Density index versus relative depth of foundation (Source: Vesic, 1973)
M14_PURU1773_01_SE_C14.indd 518 2/12/2013 7:54:47 AM

Table 14.1a Conditions for Modes of Failure (Kaviraj, 1988)
Sl. no. Foundation condition Types of shear failure
1. Footings on the surface or at shallow depths in very dense sand. General shear failure
2. Footings on saturated and normally consolidated clay under General shear failure
undrained loading.
3. Footings at deeper depth in dense sand. Punching shear failure
4. Footings on the surface or at shallow depths in loose sand. Punching shear failure
5. Footings on very dense sand loaded by transient dynamic Punching shear failure
load.
6. Footings on very dense sand underlain by loose sand or soft Punching shear failure
clay.
7. Footings on saturated and normally consolidated clay under Punching/local shear failure
drained loading.
14.4 TERZAGHI’S BEARING CAPACITY THEORY

Most theories which are in wide use have emerged from Prandtl’s theory of plastic equilib-
rium. Prandtl’s theory considers the deformation or penetration effects of hard objects on
soft materials. Based on this basic principle, the bearing capacity problem is considered as a
rigid footing penetrating into a soft homogeneous material. The implied assumptions in
adopting this theory are as follows: (i) the soil is isotropic and homogeneous, (ii) the soil is
weightless, and (iii) the footing is long with a smooth base.
Figure 14.4 shows a generalized failure mechanism for a strip footing. This is the basic
mechanism suggested by Prandtl. The condition of the footing (smooth or rough) and var-
ied boundary wedge angles (α and β) have been adopted by researchers. Prandtl stud-
ied the effect of a long, narrow metal tool bearing against a smooth metal mass with α =
45° + φ/2 and β = 45° – φ/2 and the curved portion of the slip surface is assumed as
logarithmic spiral. Based on model studies, most researchers (e.g., Ko and Davidson, 1973)
agreed that the curve fits to a logarithmic spiral for φ > 0˚ and a circle for φ = 0˚.
B
qc a
f a b
1 b b
3 a
2 d Straight line
e 2b
Logarithmic spiral
180–2a
Zone I – abd – Zone of active state
Zone II – ade – Zone of plastic state
Zone lII – aef – Zone of passive state
Fig. 14.4 Generalized failure mechanism
M14_PURU1773_01_SE_C14.indd 519 2/12/2013 7:54:48 AM

Zone I is assumed to remain intact and at plastic state. The load is transmitted through
this soil-wedge of Zone I. In Zone II, plastic flow develops with the formation of slip planes,
as shown by broken lines in Fig. 14.4. Zone III is at passive state with plane slip surface.
The penetrating wedge (Zone I) pushes aside Zones II and III, and the shearing residence
mobilizes along the logarithmic spiral and straight line segment. Based on this premise, the
ultimate bearing capacity for a surface footing*
qnf = c cot φ [exp(π tan φ) tan 2 (45°+φ/2) − 1] (14.1)
For the general case, it is necessary to consider the overburden pressure as surcharge q,
otherwise when c = 0, the bearing capacity of the weightless soil would be zero. Ressner
(1924) extended Prandtl’s work by including the condition that the bearing area is located
below the surface of the soil and the overburden is represented by a surcharge equation for
ultimate bearing capacity given as
qf = c cot φ[exp(π tan φ) tan 2 (45°+φ/2) − 1]+q[exp(π tan φ) tan 2 (45°+φ/

/2)] (14.2)
In order to consider the effect of the self-weight of the soil, an additional term must be
added.
Terzaghi (1943) applied the developments of Prandtl–Ressner to soil foundation prob-
lems. He identified a foundation as shallow if the depth Df of the foundation is less than or
equal to the width B of the foundation. The assumed failure mechanism is shown in Fig. 14.5.
Terzaghi assumed a strip footing with a rough base placed at the depth Df on a homo-
geneous and isotropic soil medium. In the analysis, the shearing resistance of the soil
above the base (ab and a′b′ in Fig. 14.5) of the footing is not considered, but the effect of
soil-weight above the base is considered by superimposing an equivalent surcharge inten-
sity q = γDf. The development of the failure surface in the soil is governed by the general
shear failure.
The soil immediately beneath the foundation forms a wedge (Zone I) which moves down-
wards (Fig. 14.5). The movement of the wedge forces the soil aside and produces two zones
of shear (Zones II and III) consisting of radial shear zone (Zone II) immediately adjacent to
B
Qf
b b′
qr q = gDr
Dr
E D
a 45°–f/2 45°–f/2 C f I f C
45°–f/2 45°–f/2 a′
III III
II f II
PP PP
A
Fig. 14.5 Terzaghi’s failure mechanism
*This is defined as the ultimate beating capacity less overburden pressure, γDf, i.e., qf − γDf, where Df
is the depth of footing.
M14_PURU1773_01_SE_C14.indd 520 2/12/2013 7:54:48 AM

the wedge and linear shear (Zone III) beyond the radial. Here, Zone I is considered to be
at Rankine active state, Zone II under radial shear, and Zone III at Rankine’s passive state.
This situation can be compared with the passive case of a retaining wall. The penetrat-
ing wedge is in equilibrium when the downward load is resisted by forces on the inclined
faces of the wedge. Cohesion and the resultant passive pressures contribute to the resistance
along the inclined faces. For equilibrium in the vertical direction, at the verge of failure,
∑V = 0, thus,
qf B = 2Pp + ADc sin φ
substituting AD = B / 2 cos φ ,
qf B = 2Pp + Bc tan φ (14.3)
The value of Pp has been represented as the vector sum of three components, viz., (i) cohe-
sion, (ii) surcharge, and (iii) weight of the soil. Terzaghi assumed the method of superposi-
tion to be valid and presented the unit ultimate bearing capacity
qf = cNc + qN q + 12 γ BN γ (14.4)
where Nc, Nq, and Nγ are non-dimensional bearing capacity factors and functions only for
the angle of shearing resistance φ,
Nc = cot φ [N q − 1] (14.5)
exp[2(3π/4 − φ / 2) tan φ]
Nq = (14.6)
2 cos 2 [π/4+(φ / 2)]
⎛ Kp ⎞⎟
tan φ ⎜⎜⎜
1
Nγ = − 1⎟⎟ (14.7)
2 ⎜⎝ cos φ
2 ⎟⎠
where Kp is the coefficient of passive pressure from Zones II and III.

The bearing capacity factors are given in Table 14.1.
Terzaghi (1943) stated that the value of the wedge angle α may lie between φ and
45° + φ/2. Purushothama Raj et al. (1972) using limit theorem (based on kinematical consid-
erations) analysed the bearing capacity of shallow foundation varying the boundary wedge
angles. The individual minimum values of Nc, Nq, and Nγ for critical wedge angles of α
and β based on the analysis is presented in Table 14.2. As it was generally observed that
Terzaghi’s values are high, the same behaviour has been highlighted by the limit analysis
approach also. The critical wedge angles α and β in Table 14.2 clearly show that β angles
were almost equal to 45° − φ/2, but α angles were found to vary significantly; however,
their effect on bearing capacity factors was found to be negligible.
M14_PURU1773_01_SE_C14.indd 521 2/12/2013 7:54:50 AM

Table 14.1b Bearing capacity factors from Terzaghi’s theory and the Bureau of
Indian Standards
φ (°) Nc Nq Ng
Terzaghi IS Terzaghi IS Terzaghi IS
0 5.70 5.14 1.00 1.00 0.00 0.00

5 7.30 6.14 1.60 1.57 0.50 0.45
10 9.60 8.35 2.70 2.47 1.20 1.22
15 12.90 10.90 4.40 3.94 2.50 2.65
20 17.70 14.83 7.40 6.40 5.00 5.39
25 25.10 20.72 12.70 10.66 9.70 10.88
30 37.20 30.14 22.50 18.40 19.70 22.40
35 57.80 46.12 41.40 33.30 42.40 48.03
40 95.70 75.31 81.30 64.20 100.40 109.41
45 172.30 138.88 173.30 134.88 297.50 271.76
50 347.50 266.89 415.10 319.07 1153.20 762.89
Table 14.2 Critical wedge angles and bearing capacity factors
φ (°) Nc Nq Ng
α (°) β (°) Value α (°) β (°) Value α (°) β (°) Value
40 60.50 25.00 74.10 60.50 25.00 63.05 54.00 25.00 138.90

30 51.50 30.00 29.30 51.40 30.00 17.72 46.00 25.00 24.23
20 45.25 35.00 14.19 43.25 30.00 5.97 38.50 30.00 4.86
10 38.00 40.00 7.89 40.00 33.00 2.20 40.00 37.00 0.67
5 35.00 40.00 6.07 27.50 42.50 1.35 – – 0
Source: Purushothama Raj et al. (1972).
Other notable contributions are made by Meyerhof (1951), Hansen (1970), and Vesic
(1973). Meyerhof (1951) considered the effects of shearing resistance within the soil above
foundation level, the shape and roughness of foundation. Hansen (1970) proposed a more
generalized equation with shape and depth of foundation and the inclination of the load.
Vesic (1973) reviewed different theories and showed that Meyerhof’s and Hanzen’s theories
give almost same Nc and Nq values. Although Nγ value of Meyerhof’s has been in use, Vesic
(1973) suggested that Nγ is best represented by Eq. 14.8.
N γ = 2( N q + 1) tan φ (14.8)
Based on these facts, Indian Standards recommended Vesic’s values of bearing capacity
factors, as given in Table 14.1. Plotted values are shown in Fig. 14.6.
Terzaghi’s expression is valid for simplified conditions but can be modified, as discussed
below, to adopt to different field conditions.
M14_PURU1773_01_SE_C14.indd 522 2/12/2013 7:54:50 AM

1,000
800
600
400
200 Ng
Bearing capacity factors

100
80
60
40
20 Nq
Nc
10
8
6
4
2
1
0 10 20 30 40 50
Friction angle, f
Fig. 14.6 Bearing capacity factors
Effect of Soil Compressibility. The general expression considers a relatively

incompressible soil. This expression has to be modified in order to apply it to materi-
als which undergo large vertical compression. Such a soil condition is identified as local
shear failure and Terzaghi suggested modified shear strength parameters, cl = (2/3)c and
tan φl = (2/3) tan φ. (The conventional notations for reduced shear parameters are c′ and φ′,
but they are referred to here as cl and φl to differentiate them from effective shear strength
parameters.)
Accordingly, the ultimate bearing capacity for local shear failure is
qf′ = cl Nc + qN q′ + 12 γ BN γ′ (14.9)
For obtaining values of Nc′ , N q′ , and N γ′ , φl is calculated as φl = tan−1 (0.67 tan φ). Then,
Nc, Nq, and Nγ (IS recommended values) are read from Table 14.1 corresponding to the value
of φl instead of φ, which are values of Nc′ , N q′ , and N γ′ , respectively (IS: 6403, 1981).
Vesic (1973) suggested that in sands the effect of relative density may be combined with
the reduction factor ( 23 + D r − 43 D r2 ) for the range 0 < Dr < 67%. But from a practical point of
view foundations will never be laid on loose sand without proper densification.
Effect of Water Table. The general equation is based on the assumption that the water
table is located well below the foundation. Some modifications are necessary depending
on the location of the water table. In the general equation, there are two terms which are
affected by water table movement: (i) the soil-weight component, (½)γBNγ and (ii) the
surcharge component, γDfNq.
Let us consider three locations of water table.
Case I: When the water table is well below the foundation, i.e., dw ≥ B. For this case, no
correction is needed for both the components (Fig. 14.7).
M14_PURU1773_01_SE_C14.indd 523 2/12/2013 7:54:52 AM

Qf ′ = 0.5 m
Rw
d ′w
= 0
Df
′
dw
III Df
Rw = 0.5 m Df
d′w
= 1
Df d w /B = 0
dw B
B R w′ = 1.0 m
II
dw /B = 1
I
Rw = 1.0
Fig. 14.7 Water table locations
Case II: When the water table is anywhere from the base of the footing to a level well
below the foundation, i.e., 0 ≤ dw ≤ B. In this case, only the soil-weight component is
affected. This aspect can be considered by substituting an equivalent unit weight γe in
place of γ, i.e.,
dw γ + (B − dw )γ ′
γe =
B
or
dw
γe = γ ′ + ( γ − γ ′) (14.10)
B
The surcharge component is not affected.
Case III: When the water table is anywhere between the ground surface and the base of the
′ ≤ Df . In this case, both the components are affected. For the surcharge
footing, i.e., 0 ≤ dw
component, the required substitution is
′ γ + (Df − dw
q = dw ′ )γ ′ (14.11)
For the soil-weight component, the required substitution is γ′ in place of γ in the term
1
2
γ BN γ .
Teng (1962) suggested water table correction factors, assuming the submerged unit weight
of soil as 50% of the bulk unit weight of soil. Considering Case III, when the water table
′ / Df = 0 and γ = γ′, and at the base of the footing, dw
is at the ground surface, dw ′ / Df = 1
and γ = γsat. This suggests a correction factor to have a value of 0.5 at d′w/Df = 0 and 1.0 at
d′w/Df = 1, and such a factor may be Rw ′ = (1/ 2)(1 + dw
′ / Df ) . For any intermediate point, a
linear interpolation is made.
M14_PURU1773_01_SE_C14.indd 524 2/12/2013 7:54:54 AM

Considering Case II for dw/B = 0, the correction factor should have a value of 0.5 when
water table is at the base and for dw/B = 1.0, the correction factor should be 1.0, such a
factor may be Rw = (1/ 2)(1 + dw / B) . For any intermediate point, a linear interpolation is
made.
Thus, the general expression can be written with modification for water table as
′ qN q + 12 Rw γ BN γ
qf = cNc + Rw (14.12)
The variations of Rw and Rw′ are also shown in Fig. 14.7. For Case I, both Rw and Rw
′ have a
′ = 1.0 and Rw = 0.50.
value of 1. When the water table is at the base of the footing, Rw
14.5 FOUNDATION PRESSURES

The total foundation pressure on the soil due to the weight of the structure is called the
gross foundation pressure (qg). The gross foundation pressure at the time of failure is
nothing but the ultimate bearing capacity of the soil (qf). The net foundation pressure (qn)
is the foundation pressure in excess of the pressure caused by the surrounding soil. So the
net ultimate bearing capacity (qnf) is the net foundation pressure at the time of failure.
Thus,
qn = qg − γ Df (14.13a)
and
qnf = qf − γ Df (14.13b)
If F is the factor of safety with respect to shear failure, then

qnf
F= (14.14)
qn
or
qnf
qn = (14.15)
F
Usually, a factor of safety of 3.0 is adopted against shear failure, and hence qn may be
called the net bearing pressure qns such that qns = qnf/F. This leads to another definition of a
term called gross safe pressure (qs). That is,
qs = qns + q
q
qs = nf + q
F (14.16)
1⎡ 1 ⎤
qs = ⎢ cNc + q( N q − 1) + γ BN γ ⎥ + q
F ⎣⎢ 2 ⎥⎦
M14_PURU1773_01_SE_C14.indd 525 2/12/2013 7:54:55 AM

When the value of φ is relatively high there is no appreciable difference between the
values of F defined in terms of net and gross pressures. The safe pressure defined above is
based on shear failure only and need not be minimum.*
14.6 SPECIAL LOADING AND GROUND CONDITIONS

The bearing capacity expressions given in the previous section deal with an idealized condi-
tion of loading and soil property. But in practice loading may be eccentric and inclined, and
soil conditions may not be homogeneous but layered with varying shear parameters. Such
special cases are briefly dealt with below.
14.6.1 Foundations with Eccentric Loading

A footing subjected to a concentric loading, with a moment or a load applied off the centre,
causes the loading to be eccentric.
The analysis of an eccentrically loaded footing involves the evaluation of contact pressure
beneath the footing and the ultimate bearing capacity. Let us consider the eccentricity in one
direction only.
Eccentricity ey is given as (Fig. 14.8)
M
ey = (14.17)
Q
The maximum and minimum pressures are given as
Q ⎛⎜ 6 e y ⎞⎟
qmax = ⎜⎜1 + ⎟⎟ (14.18)
BL ⎜⎝ B ⎟⎠
and
Q ⎛⎜ 6 e y ⎞⎟
qmin = ⎜1 − ⎟⎟
BL ⎜⎜⎝ B ⎟⎠
B′ = B – 2ey
B ey
Fig. 14.8 One-directional eccentricity
*A minimum pressure is obtained after considering the settlement aspect also.
M14_PURU1773_01_SE_C14.indd 526 2/12/2013 7:54:56 AM

L′ = L–2ex
ex
B ′ = B – 2ey
B ey
Fig. 14.9 Two-directional eccentricity
For ey = B/6, qmin is zero, and for any condition where ey > B/6, tension will develop.
As soil cannot take tension, this has to be avoided; otherwise, there will be separation
between the foundation and the soil.
If the eccentricity is in both the directions, ex and ey (Fig. 14.9) then the pressure is
given as
Q ⎛⎜ 6e 6 e y ⎞⎟
q= ⎜⎜1 ± x ± ⎟⎟ (14.19)
BL ⎜⎝ L B ⎟⎠
Here again, a negative value of q indicates tension between the soil and the bottom of
footing. The footing has to be sufficiently weighted down by surcharge loads so as to rely on
a proper bonding between the soil and the footing.
The concept of useful width was introduced by Meyerhof (1953) for the determination of
the ultimate bearing capacity of eccentrically loaded footing and is also adopted in IS: 6403
(1981). The effective footing dimensions are
Effective length L′ = L − 2ex
Effective width B′ = B − 2ey
Effective footing area = A′ = B′ × L′
By this concept, the area of the footing which is symmetrical about the load is taken as
useful. The other portion is assumed to be excess. It is evident that the bearing capacity will
decrease with increase in eccentricity. Meyerhof (1953) suggested a reduction factor to obtain
the ultimate bearing capacity, as determined in the conventional way, considering the load
is acting at the centroid of the footing. This bearing pressure is reduced by a reduction factor
Re. Thus, the reduced bearing pressure (qf)e is
(qf )e = qf Re (14.20)
where Re = 1 − 2(ey/B) for the cohesive soil and Re = 1 − (ey/B)1/2 for non-cohesive soils (for
the range 0 < ey/B < 0.3). The reduction factor can also be read from Fig. 14.10.
M14_PURU1773_01_SE_C14.indd 527 2/12/2013 7:54:57 AM

1.0
Cohesive soil
Reduction factor, Ri
0.8
Granular soil
0.6
0.4
0.2
0 0.1 0.2 0.3 0.4 0.5

Eccentricity ratio ey /B
Fig. 14.10 Bearing capacity of eccentrically loaded footing (Source: AREA, 1958)
14.6.2 Foundation Subjected to Inclined Load

The conventional method of analysis of footing subjected to inclined load is to resolve the
force vertically and horizontally. The vertical component is used to determine the relevant
bearing capacity and the horizontal force is ignored. However, the stability of footing against
the horizontal force is analysed and a suitable factor of safety is adopted.
Meyerhof (1953) analysed the condition of inclined loads and presented a chart to find the
reduction factor, Ri. According to his approach, the load is assumed vertical and the ultimate
bearing capacity is determined. Then, it is corrected by a correction factor obtained from the chart
(Fig. 14.11). Thus, the ultimate bearing capacity for a footing subjected to inclined load (qf)i is
(qf )i = qf Ri (14.21)
where Ri is the reduction factor.

Jambu (1957) extended Terzaghi’s theory for inclined loads with due consideration of
horizontal force and introduced an additional factor Nh as
Qu + N h Qb 1
= cN q + qN q + γBN γ (14.22)
A 2
f Q
Q
20°
25°
30°
35°
40°
45°
10°
15°
0°
5°
300 Dt
200 Nr B Qh
100
50 (Area = A)
Nc Nq
20 Qv + Nh Qh
= Ncc + NqgDr + 1 NgB
10 2
A
5 Nq Nh Qh cannot exceed Qv tan f
Nr
2 c = Cohesion
1
0 01 02 03 04 05 06 07 08 09 f = Angle of internal friction
Fig. 14.11 Bearing capacity charts for inclined loads (Source: Jambu, 1957)
M14_PURU1773_01_SE_C14.indd 528 2/12/2013 7:54:58 AM

Figure 14.11 shows the bearing capacity factors to use in the above equation. Modified
bearing capacity formula (Eqs. 14.29 and 14.30) can also be used.
14.6.3 Foundations on Sloping Surface

Meyerhof (1957) proposed a method for determining the bearing capacity of footings on
sloping ground surfaces. The bearing capacity equation for strip footing is given as
qf = cNcq + 12 γBN γ q (14.23)
The bearing capacity factors Ncq and Nγq depend on the slope of the ground and the
relative position of the ground, in addition to the angle of shearing resistance of the soil.
Footing should not be placed on unsafe slopes. Before construction, the stability of the
slope has to be checked. The construction of footing should not provoke a slide and the
aspect has to be analysed. Further, if the slope material is under slow creep, the construction
of footing on such slopes has to be avoided.
14.6.4 Foundations on Stratified Soil

Footings are sometimes constructed on soils where the rupture surface may not lie entirely
on the first layer but may extend to the second layer. Further, depending on the strength of
the soil in each layer, the rupture surface tends to increase in a weaker material and decrease
in a stronger material.
A solution for the case of a footing at the surface of a two-layered saturated clay under
undrained condition was first made by Button (1953). For φ = 0° condition, the curved sector
becomes circular, and Button used a circular surface. The analysis was made for a two-layer
system with cohesions c1 in the top layer and c2 in the bottom layer. The ultimate bearing
capacity for surface footing is given as
qf = c1 Nc (14.24)
where Nc is the bearing capacity factor depending on c2/c1 ratio and the ratio of the thickness
of the top layer to the width of footing. Figure 14.12 (Button, 1953) shows the variation of Nc
with d/B and c2/c1. It is observed that when the upper layer is harder than the lower, the
bearing capacity increases with the thickness of the top layer, and when the upper layer is
softer, the bearing capacity decreases as its thickness increases. In this approach c1 and c2 are
isotropic within their respective layers. This chart has been included in IS: 6403 (1981).
Button’s solution was extended by Siva Reddy and Srinivasan (1967) for anisotropic soils,
defined by a coefficient of anisotropy as
K = qv / qh (14.25)
where qv is the vertical shear strength and qh the horizontal shear strength.
Charts for various K values are available.
Different model analyses for two-layer systems were also attempted by Yamaguchi (1963),
Meyerhof (1974), and Hanna and Meyerhof (1980). In these model analyses, some curves for
estimating the capacity of sand overlying clay were presented.
M14_PURU1773_01_SE_C14.indd 529 2/12/2013 7:54:59 AM

0.2
10
0
0.4
=
B
d/
8
0.6
0.8
Nc 6 1.0
b Top layer
0
2.
4
1.5 c1
1.0 B d
0
0.5 =
2
B
c2
d/
Bottom
layer
0 0.4 0.8 1.2 1.6 2.0

c1/c2
Fig. 14.12 Bearing capacity factors for layered cohesive soil deposits (Source: IS: 6403, 1981)
A detailed bearing capacity for a more general case with both c and φ was presented by
Purushothama Raj et al. (1974) based on the upper bound limit theorem. Bearing capacity
charts for varying cohesion with constant angle resistance were provided by them.
Purushothama Raj et al. (1975) also extended Button’s solution for concentrically loaded
footing to eccentrically loaded footing on an isotropic two-layer cohesive soil system based
on upper bound limit theorem. Bearing capacity charts for different depth–width ratios
(d/B), cohesion ratios (c2/c1), and eccentricity–width ratios (ey/B) for conventional and tri-
angular foundation–soil contact conditions were presented.
14.6.5 Foundations on Partially Saturated Soil

Siva Reddy and Mogalaiah (1976) have presented a solution to the problem of ultimate bear-
ing capacity of a strip footing on a partially saturated soil in the approach. Bishop’s (1955)
concept for effective stress of partially saturated soils and Skempton’s (1954) pore pressure
parameters have been used along with an initial negative pore water pressure. The solution
is obtained by adopting the method of characteristics. Numerical results are given in the
form of bearing capacity factors Nc, Nq, and Nγ, for different values of pore pressure param-
eters A and B, coefficient of earth pressure at rest K0, ratio of initial negative pore water pres-
sure (uw)0, and cohesion c.
It is inferred from their studies that the degree of saturation and initial negative pore
water pressure have significant influence on the bearing capacity of partially saturated soils.
14.6.6 Foundations on Desiccated Soil

Desiccated cohesive soils show decrease in undrained cohesion with depth. Further, this
decrease stabilizes at depths of about 3 to 4 m from the ground level. For this situation, a linear
decrease of cohesion with depth may be considered. The gradient λ and cohesion at the
M14_PURU1773_01_SE_C14.indd 530 2/12/2013 7:54:59 AM

Table 14.3 Data for determining ultimate

net bearing capacity
8λB/qnf qnf /c1
0.0 5.7
0.2 5.0
0.3 4.5
0.6 4.0
0.7 3.6
1.0 3.2
Source: IS: 6403 (1981).
ground surface c1 can be obtained from borehole data. A set of values relating 8λB/qnf and
qnf/c1 are presented in Table 14.3. For a given footing width B, by trial and error qnf can be
estimated by matching qnf/c1 and 8λB/qnf (IS: 6403, 1981), or a plot can be made between
(8λB/c1) = (8λB/qnf × qnf/c1), and (qnf/c1) using the values from Table 14.3, and with the
knowledge of (8λB/c1), (qnf/c1) can be read from the plot.
Hence, from this value qnf is obtained.
14.6.7 Foundations on Rock

Sound rocks have strengths extremely higher than the pressure transferred by footings.
Unfavourable rock conditions, heterogeneity, and overstressing of foundations may cause
large differential settlement leading to failure. Particularly, porous limestones, volcanic
rocks, some shales, and highly fractured rocks need special consideration.
In principle, Terzaghi’s bearing capacity equation can be used provided strength param-
eters are obtained from triaxial shear tests on rocks. However, in rocks settlement criteria is
more critical than shear failure.
The Bureau of Indian Standards (IS: 12070, 1987) suggests different methods for evaluat-
ing the bearing capacity of shallow foundation on rocks. They are explained below.
Evaluation Based on Classification. For preliminary design, the net safe bearing pres-
sure based on classification (Table 14.4) has been suggested with a caution that it has to be
Table 14.4 Net safe bearing pressure based on classification
Material qns (kN/m2)
Massive crystalline bed rock, including granite, diorite, gneiss, trap rock 10,000
Foliated rocks such as schist or slate in sound condition 4,000
Bedded limestone in sound condition 4,000
Sedimentary rock, including hard shales and sandstone 2,500
Soft or broken bed rock (excluding shale) and soft limestone 1,000
Soft shale 400
Source: IS: 12070 (1987).
M14_PURU1773_01_SE_C14.indd 531 2/12/2013 7:55:00 AM

Table 14.5 Net safe bearing pressure based on RMR
Classification no. I II III IV V
Description of rock Very good Good Fair Poor Very poor

RMR 100–81 80–61 60–41 40–21 20–0
qns (kN/m2) 6000–4480 4400–2880 2880–1510 1450–900–580 550–450–400
Source: IS: 12070 (1987).
checked before the final design. Further, net bearing pressure is also recommended based
on rock mass rating (RMR) for different rocks (Table 14.5). The RMR values obtained up to a
depth equal to width of the foundation should be used. The recommended values consider
a limiting settlement of 12 mm.
Evaluation Based on Core Strength. For a rock mass with favourable bedding planes
(i.e., rock surface parallel to the base of the foundation) and the walls of discontinuities
closed, the safe bearing pressure is given as
qs = q0 N j (14.26)
where q0 is the average uniaxial compressive strength of rock cores and Nj the empirical coef-
ficient depending on the spacing of discontinuities (Table 14.6) given as
3 + s / Bf
Nj =
10 1 + 300 δ / s
where δ is the thickness of discontinuities (cm), s the spacing of discontinuities (cm), and Bf
the footing width (cm).
The above relationship is valid for a rock mass with spacing greater than 0.3 m, opening
of discontinuities less than 10 mm, and a foundation width greater than 0.3 m.
Evaluation Based on Plate Load Test. It is recommended that plate load test be con-
ducted on poor rocks with safe bearing pressure less than 1,000 kN/m2. From the plate load
testing, the settlement of plate is computed from the formula as follows:
Bp
1. For massive or sound rock, Sp = Si (14.27)
Bf
Table 14.6 Value of Nj
Spacing of discontinuities (cm) Nj
300 0.40
100–300 0.25
30–100 0.10
Source: IS: 12070 (1987).
M14_PURU1773_01_SE_C14.indd 532 2/12/2013 7:55:00 AM

⎡ B ⎛ B + 0.3 ⎞⎤ 2
p⎜ ⎟⎟⎥
2. For laminated or poor rocks, Sp = Si ⎢⎢ ⎜⎜ f ⎟⎟⎥ (14.28)
B ⎜
⎢⎣ f ⎝ p
B + 0. 3 ⎟⎠⎥
⎦
where Sp is the settlement of plate (mm), Si the settlement of footing (mm), Bp the width of
plate (m), and Bf the width of footing (m).
From the pressure–settlement curve, the safe bearing pressure is read for the calculated
settlement of plate. It is further recommended to conduct at least three tests with different
sizes of plates to check the results.
14.7 OTHER BEARING CAPACITY THEORIES

14.7.1 Modified Bearing Capacity Formulae (IS: 6403, 1981)
Because of mathematical complexity, no theoretical analysis is available for the shape of foot-
ing like square, rectangular, and circular. Empirical corrections based on model tests are
recommended by Terzaghi (1943) and are subsequently simplified.
Terzaghi’s expression (Eq. 14.4) is valid for Df/B ≤ 1. But in practice the depth factor has
also to be considered. Further, the load may be inclined instead of vertical. Thus, modified
ultimate net bearing capacity formulae taking into account the shape of the footing, depth of
embedment, inclination of loading, and effect of water table has been recommended by the
Bureau of Indian Standards (IS: 6403, 1981) as follows:
General shear failure:
qnf = cNc sc dc ic + q( N q − 1)sq dq iq + 12 γ BN γ sγ dγ iγ Rw (14.29)
Local shear failure:

qnf = cl Nc′ sc dc ic + q( N q′ − 1)sq dq iq + 12 γ BN γ′ sγ dγ iγ Rw (14.30)
where sc, sq, and sγ are shape factor corrections (values are given in Table 14.7), dc, dq, and dγ
are depth factor corrections (to be applied only when the backfilling is done with proper
compaction) calculated as follows:
dc = 1 + 0.2Df / B Nφ
dq = dγ = 1 for φ < 10°
dq = dγ = 1 + 0.1Df / B Nφ for φ > 10°
2
Nφ = tan ( 45° + φ / 2)
and ic, iq, and iγ are load inclination factor corrections given as
2
⎛ α⎞
ic = iq = ⎜⎜⎜1 − ⎟⎟⎟
⎝ 90 ⎠
⎛ α ⎞2
iγ = ⎜⎜1 − ⎟⎟⎟
⎜⎝ φ ⎟⎠
M14_PURU1773_01_SE_C14.indd 533 2/12/2013 7:55:02 AM

Table 14.7 Shape factor corrections
Shape of base of Shape factor corrections

footing
sc sq sg
Continuous strip 1.00 1.00 1.00

Rectangle 1 + 0.2B/L 1 + 0.2B/L 1 – 0.4B/L
(width B, length L)
Square (side B) 1.30 1.20 0.80
Circle (diameter B) 1.30 1.20 0.60
Source: IS: 6403 (1981).
where α is the inclination of the load to the vertical in degrees.

Rw = 1, if the water table remains permanently at or below a depth (Df + B) beneath the
ground level.
Rw = 0.5, if the water table is permanently located at a depth of Df likely to rise above the
base of footing.
Rw = 12 [1 + (dw / B)] , if the water table is likely to be permanently located at a depth Df <
dw < (Df + B).
14.7.2 Skempton’s Bearing Capacity Theory

Skempton (1951) proposed a simple expression for the ultimate bearing capacity of satu-
rated clay under undrained condition (φ = 0), for a rectangular footing of length L and width
B, and is given as
qf = cu Nc + γ Df (14.31)
Values of Nc may be obtained from Fig. 14.13 or from the following expression:
10 Circle or square
B/L =1
9
Intermediate values
8 by interpolation
Nc
7
Strip B/L = 1
6
4
0 1 2 3 4 5
z/B
Fig. 14.13 Skempton’s values of Nc when φu = 0° (Source: Skempton, 1951)
M14_PURU1773_01_SE_C14.indd 534 2/12/2013 7:55:02 AM

B⎞ ⎡ ⎛ D ⎞ ⎤
1/ 2
⎛
Nc = 5.14 ⎜⎜⎜1 + 0.2 ⎟⎟⎟ ⎢⎢1 + ⎜⎜⎜0.053 f ⎟⎟⎟ ⎥⎥ (14.32)
⎝ L ⎠ ⎢⎣ ⎝ B ⎠ ⎥⎦
The limiting maximum values (Df/B) are
when B/L = 0, i.e., for strip footing, Nc ⬎ 7
when B/L = 1, i.e., for circular or square footing, Nc ⬎ 9
14.7.3 Meyerhof’s Bearing Capacity Theory

Meyerhof (1963) presented a more comprehensive solution to solve the bearing capacity
problem. This theory takes into account depths, shape, and inclination factors together in the
bearing capacity equation. This theory is applicable for foundation of all shapes and loading
conditions, except foundations on built up slopes.
The solution proposed by Meyerhof (1963) is given as
qd = cNc dc sc ic + γ Ddq sq iq + 12 γ BN γ dγ sγ iγ (14.33)
where c is the unit cohesion, Nc, Nq, Nγ are the bearing capacity factors for a strip foundation,
dc, dq, dγ the depth factors, sc, sq, sγ the shape factors, ic, iq, iγ the inclination factors for the
load inclined at an angle α (degrees) to the vertical, γ the effective unit weight of soil above
base level of foundation, γ the effective unit weight of soil below foundation base, and D the
depth of foundation.
The depth, shape, and inclination factors are given in Table 14.8.
Table 14.8 Depth, shape and inclination factors
Factors Equation For

⎛D⎞
Depth dc = 1 + 0.2 Nφ ⎜⎜⎜ ⎟⎟⎟ any φ
⎝B⎠
⎛D⎞
dq = dγ = 1 + 0.1 Nφ ⎜⎜⎜ ⎟⎟⎟ φ>0
⎝B⎠
dγ = dq = 1 φ=0
⎛B⎞
Shape sc = 1 + 0.2 Nφ ⎜⎜⎜ ⎟⎟⎟ any φ
⎝L⎠
⎛B⎞
sq = sγ = 1 + 0.1Nφ ⎜⎜⎜ ⎟⎟⎟ φ>0
⎝L⎠
sq = sγ = 1 φ=0
2
Inclination ⎛ α° ⎞
ic = iq = ⎜⎜1 − ⎟⎟⎟ any φ
⎜⎝ 90 ⎠
2
⎛ α° ⎞
iγ = ⎜⎜1 − ⎟⎟⎟ φ>0
⎜⎝ φ° ⎟⎠
iγ = 0 φ=0
Source: Meyerhof (1963).
M14_PURU1773_01_SE_C14.indd 535 2/12/2013 7:55:04 AM

In Table 14.8, B = width, L = length of foundation, and Nφ = tan(45°+φ/2).

The theoretical bearing capacity factors for a shallow horizontal strip foundation are
N q = eπ tan φ Nφ ⎪⎫⎪
⎬ (14.34a)
Nc = ( N q − 1) cot φ⎪⎪
⎪⎭
N γ = ( N q − 1) tan(1.4φ) (14.34b)
Equation 14.34a is the same as per Prandtl (1920), whereas Eq. 14.34b is proposed by
Meyerhof (1961).
1,000
Strip (D < B)
Square (D < B)
Pile (D /B > 4 – 10)
100
Bearing capacity factors - Nc , Nq , Ng
N′c
Nc
Nq
10
N′q
Ng
1
0° 10° 20° 30° 40°
Angle of internal friction, f
Fig. 14.14 Bearing capacity factors for spread and pile foundations (Source: Meyerhof, 1963).
M14_PURU1773_01_SE_C14.indd 536 2/12/2013 7:55:05 AM

Figure 14.14 gives the N factors for strip and square foundations. For rectangular
foundations, the N factors have to be interpolated. The N′ factors given in the same figure
applies to point bearing capacity of pile foundation. At φ = 0°, Nc = 5.14 for strip foundation
and Nc = 6.2 for square foundation.
Depth Factors. The simple bearing capacity factors in Eq. 14.34 do not take into account
the resistance of the soil above the foundation level which increases the bearing capacity. If
the soil above the foundation level is quite compact, the effect of this on the bearing capacity
may be considered by means of depth factors given in Table 14.8. The increase of bearing
capacity can be estimated from depth factors by which the individual bearing capacity
factors have to be multiplied. As the depth of the foundation increases, the depth factors
increase at a decreasing rate and approach a maximum value which can be used for an
estimate of the point resistance of piles.
Shape Factors. The bearing capacity factors given in Eq. 14.34 is for a strip foundation.
The bearing capacity factors for rectangular foundations can be obtained by multiplying the
individual N factors in Eq. 14.34 of the corresponding shape factors given in Table 14.8.
Eccentric Loading. If the foundations are subjected to eccentric loads, vertical or inclined,
the effective width B′ of the foundation has to be used in Eq. 14.34.
14.7.4 Brinch Hansen’s Bearing Capacity Theory

Brinch Hansen’s bearing capacity theory is in a way an extension of Meyerhof’s work. In
addition to the factors considered by Meyerhof, the foundation base tilt and foundations on
slopes are included in Hansen’s (1970) equation.
Hansen’s equation can be put in a simple form as follows:
qd = cNc Ac + γ DN q Aq + 12 γ BN γ Aγ (14.35)
where
Ac = dc sc ic bc gc
Aq = dq s q iq bc g q
Aγ = dγ sγ iγ bγ g γ
γ is the effective unit weight of soil above the base, γ the effective unit weight of soil below
the base, dc, dq, dγ are depth factors, sc, sq, sγ are shape factors, ic, iq, iγ are load inclination
factors, bc, bq, bγ are base inclination factors, and gc, gq, gγ are ground surface inclination
factors. The other factors are the same as given earlier. The equations for the various factors
are given below:
1. N-factor
N q = eπ tan φ Nφ
Nc = ( N q − 1) cot φ Meyerhof’s factors (14.36)
N γ = 1.5( N q − 1) tan φ
The bearing capacity factors are given in Fig. 14.15a.
M14_PURU1773_01_SE_C14.indd 537 2/12/2013 7:55:06 AM

(a) 250 Qv
(b)
Qa
200
160
120
100
80
60 Qh
Bearing capacity factors - Nc , Nq , Ng
40
b
30 Qv
(c) D
20
16 Qh
Nc
12
10 a°
8
6
(d)
Nq Q
4 b
3 Ng
1
0 5 10 15 20 25 30 35 40 45
Angle of shearing resistance f, degrees
Fig. 14.15 Bearing capacity factors and the definition of Qv, Qh, α0 , and β. (a) Bearing capacity
factors, Nc, Nq, Nγ. (b) Definition of Qv and Qh. (c) Definition of α° and β. (d) Definition
of ground inclination factor (Source: Hansen, 1970)
2. Depth factors
Equation Limiting value of D/B Limiting value of φ
dc = 1 + 0.4 tan−1 D/B >1 >0

dc = 1 + 0.4D/B ≤1 >0
dc = 0.4D/B ≤1 φ=0
dc = 0.4 tan−1 D/B ≥1 φ=0
dq = 1 + 2 tan φ(1 − sin φ)D/B 1
dq = 1 + 2 tan φ(1 − sin φ) × tan−1(D/B) >1
dγ = 1 for all φ
3. Shape factors
⎛ N q B ⎞⎟
sc = ⎜⎜⎜1 + ⎟⎟ for φ >0
⎝⎜ Nc L ⎟⎠
M14_PURU1773_01_SE_C14.indd 538 2/12/2013 7:55:07 AM

B
sc = 0.2 for φ =0
L
sc =1 for strip foundation
B
sq = 1 + tan φ
L
B
sγ = 1 − 0.4
L
4. Load inclination factors for horizontal base
1 − iq
ic = iq − for φ > 0°
Nq − 1
Qh
ic = 0.5 − 0.5 1 − for φ=0°
Af ca
⎛ 0.5Qh ⎞⎟5
iq = ⎜⎜⎜1 − ⎟
⎜⎝ Qv + Af ca cot φ ⎟⎟⎠
⎛ 0.7Qh ⎞⎟5
iγ = ⎜⎜⎜1 − ⎟
⎜⎝ Qv + Af ca cot φ ⎟⎟⎠
The definitions of Qv and Qh are given in Fig. 14.15b.

5. Base inclination factors
⎛ α° ⎞⎟
bc = ⎜⎜1 − for φ > 0°
⎜⎝ 147° ⎟⎟⎠
α°
bc = for φ > 0°
147°
bq = e−2α tan φ
bγ = e−2.7 α tan φ
The definitions of α° and β are shown in Fig. 14.15c.

6. Ground surface inclination factor
⎛ β ° ⎞⎟
gc = ⎜⎜1 − for φ > 0°
⎜⎝ 147° ⎟⎟⎠
⎛ β ° ⎞⎟
gc = ⎜⎜ for φ > 0°
⎜⎝ 147° ⎟⎟⎠
g q = g γ = (1 − 0.5 tan β )2
The definition of ground inclination is shown in Fig. 14.15d.

Wherever footings are subjected to eccentric loadings, the effective width B′ has to be used
in Eq. 14.35 for evaluating the shape factors,
B′ = B − 2e
M14_PURU1773_01_SE_C14.indd 539 2/12/2013 7:55:10 AM

If the eccentricity is in two directions for a rectangular foundation of width B and length
L, then the effective widths in each direction are
B ′ = B − 2e x
L ′ = L − 2e y
Af′ = B ′L ′
where ex and ey are the eccentricities in the B and L directions, respectively, and Af′ the effective
area. If qd is found out from Eq. 14.35 for an effective width of B′, then the ultimate load is
Qd = Af′ × qd (14.37)
For footings on a slope, g factors are used to reduce the bearing capacity, however, these factors
should be used cautiously as there is little experimental data available to confirm this factor.
14.8 BEARING CAPACITY OF SOILS FROM BUILDING CODE

Building codes are prepared in a traditional way based on the vast database on soils of
different locations. These codes give a list of soil types and their safe bearing capacity. It is
assumed that soil can sustain the pressure with respect to shear failure and without appre-
ciable settlement.
As discussed in the previous section that bearing capacity of a soil depends on various
factors, the bearing capacity given in a building code should be taken as a guiding value and
not as one exact value (IS: 1904, 1986, revised).
The Bureau of Indian Standards has given presumptive bearing capacity values which are
presented in Table 14.9. Following are the limitations of the bearing capacity values given in
building codes:
1. The code values do not consider the effect of shape, size, depth of foundation, and the
base condition (rough or smooth) of a foundation.
2. The code values do not consider the effect of water table and its fluctuation.
3. The code values assume that the soil is homogeneous in all directions.
4. The effect of settlement is not taken into account in the code values.
5. The code values are more simplified based on experience and have no theoretical base.
6. The code values used are to be updated.
Table 14.9 Safe bearing capacitya
S. no. Type of rock or soil Safe bearing Remarks

capacity,
kN/m2 (t/m2)
I. ROCKS
1. Rocks without lamination and defects, 3,240 (330)
e.g., granite, trap, diorite
2. Laminated rocks, e.g., sandstone and 1,620 (165)
limestone, in sound condition
Table 14.9 Contd.
M14_PURU1773_01_SE_C14.indd 540 2/12/2013 7:55:11 AM

Table 14.9 Contd.

capacity,
kN/m2 (t/m2)
3. Residual deposits of shattered and broken 880 (90)

bed rock and hard shale, cemented material
4. Soft rock 440 (45)
II. COHESIONLESS SOILS
5. Gravel, sand and gravel, compact and 440 (45) See note b
offering high resistance to penetration
when excavated by tools
6. Coarse sand, compact and dry 440 (45) Dry means that the
groundwater level
is at a depth not less
than width of the
foundation below the
base of the foundation
7. Medium sand, compact and dry 245 (25)
8. Fine sand, silt (dry lumps easily 150 (15)
pulverized by fingers)
9. Loose gravel or sand–gravel mixture; 245 (25) See note b Susceptible
loose coarse to medium sand, dry to long-term
10. Fine sand, loose and dry consolidation 100 (10) Consolidation
settlement settlement
III. COHESIVE SOILS
11. Soft shale, hard or stiff clay, dry 440 (45) Susceptible to long-
term consolidation
settlement
12. Medium clay, readily indented with a thumb nail 245 (25)
13. Moist clay, and sand–clay mixture 150 (15)
which can be indented
with strong thumb pressure
14. Soft clay indented with moderate 100 (10)
thumb pressure
15. Very soft clay which can be penetrated 50 (5)
easily with the thumb
16. Black cotton soil or other shrinkable See note c
or expansive clay To be determined after
in dry condition (50% saturation) investigation
IV. PEAT
17. Peat See notes c and d To
be determined after
investigation
Table 14.9 Contd.
M14_PURU1773_01_SE_C14.indd 541 2/12/2013 7:55:11 AM

Table 14.9 Contd.

capacity,
kN/m2 (t/m2)
V. MADE-UP GROUND
18. Fills or made-up ground See notes b and d
To be determined
after investigation
Source: IS: 1904 (1986, revised).

a
Values listed in the table are from shear consideration only.
b
Values are very much rough for the following reasons:
(i) Effect of characteristics of foundations (i.e., effect of depth, width, shape, roughness, etc.) has not
been considered.
(ii) Effect of range of soil properties (i.e., angle of internal friction, cohesion, water table, density, etc.)
has not been considered.
(iii) Effect of eccentricity and inclination of loads has not been considered.
c
For non-cohesive soils, the values listed in the table shall be reduced by 50% if the water table is above or
near the base of the footing.
d
Compactness or looseness of non-cohesive soils may be determined by driving a cone of 65 mm dia and
60° apex angle by a hammer of 65 kg falling from 75 cm. If the corrected number of blows (N) for 30 cm
penetration is less than 10, the soil is called loose; if N lies between 10 and 30, it is medium; if more than
30, the soil is called dense.
14.9 PERMISSIBLE SETTLEMENTS

Settlement may be classified as uniform (or total) settlement, tilt, and non-uniform settle-
ment. Structures on rigid foundations undergo uniform settlement (Fig. 14.16a). When the
entire structure rotates, the structure is said to be under uniform tilt (Fig. 14.16b). If founda-
tions of different elements of a structure undergo varied settlements, the foundation is said
to be under non-uniform settlement (or differential settlement).
Foundations may settle uniformly due to (IS: 1904, 1986) (i) elastic, consolidation and second-
ary compression of soil, (ii) groundwater lowering, (iii) swelling and shrinkage of expan-
sive soils caused by seasonal variations, (iv) surface erosion, creep, or landslides and effects
of vegetation, and (v) mining subsidence, underground erosion by streams or floods, and
adjacent excavation. Tilt of a structure generally occurs due to eccentric loading or sudden
subsidence of a corner of a rigid foundation.
l l
d d1
d2
d1 d2
d
d = d2 – d1 d = d2 – d1
Angular distortion = d Angular distortion = d
l l
(a) Uniform settlement (b) Tilt (c) Non-uniform settlement
Fig. 14.16 Types of settlement (Source: Lambe and Whitman, 1979)
M14_PURU1773_01_SE_C14.indd 542 2/12/2013 7:55:12 AM

Table 10.10 Permissible maximum, differential settlements, and tilt (angular distortion) for shallow foundation in soils
Isolated foundation Raft foundations

Sl. Type of structure Sand and hard clay Plastic clay Sand and hard clay Plastic clay
no.
M14_PURU1773_01_SE_C14.indd 543
Bearing Capacity of Soils
Maximum
settlement
Differential
settlement
Angular
distortion
Maximum
settlement
Differential
settlement
Angular
distortion
Maximum
settlement
Differential
settlement
Angular
distortion
Maximum
settlement
Differential
settlement
Angular
distortion
(mm) (mm) (mm) (mm) (mm) (mm) (mm) (mm) (mm) (mm) (mm) (mm)
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14)
(i) For steel structures 50 0.0033L 1/300 50 0.0033L 1/300 75 0.0033L 1/300 100 0.0033L 1/300
(ii) For reinforced 50 0.0015L 1/666 75 0.0015L 1/666 75 0.0021L 1/500 100 0.002L 1/500
concrete structures
(iii) For multi-storeyed
buildings
(a) Reinforced concrete 60 0.002L 1/500 75 0.002L 1/500 75 0.0025L 1/400 125 0.0033L 1/300
or steel framed
buildings with
panel walls
(b) For loading
bearing walls
1. L/H = 2a 60 0.0002L 1/5,000 60 0.0002L 1/5,000 } Not likely to be encountered
2. L/H = 7a 60 0.0004L 1/2,500 60 0.0004L 1/2,500
(iv) For water towers and 50 0.0015L 1/666 75 0.0015L 1/666 100 0.0025L 1/400 125 0.0025L 1/400
silos
Source: IS: 1904 (1986).

The values given in the table may be taken only as a guide and the permissible total settlement/differential settlement and tilt (angular distortion)
in each case should be decided as per requirements of the designer.
L denotes the length of deflected part of wall/raft of centre-to-centre distance between columns.
H denotes the height of wall from foundation footing.
a
For intermediate ratios of L/H, the values can be interpolated.
543
2/12/2013 7:55:12 AM
Non-uniform settlement can result from (IS: 1904, 1986) (i) non-homogeneous subsoil
condition, (ii) non-uniform pressure distribution on soil due to unequal loading, (iii) vari-
ation of water regime at the construction site, (iv) overstressing of adjacent site due to heavy
structures and interference of pressure distribution, (v) unequal expansion of the soil due to
excavation, (vi) non-uniform development of extrusion settlements, and (vii) non-uniform
structural disruptions or volume changes due to freezing and thawing, shrinkage and
swelling, etc.
Generally, the amount of uniform settlement is not a critical factor, but it is only a ques-
tion of convenience. In practice, the settlement is often non-uniform and is of concern in the
design of a foundation. Estimation of uniform settlement is much more simpler than that of
differential settlement. On important jobs, it is essential to investigate and identify stronger
and weaker subsoils and accordingly estimate the movements. On jobs of less importance, an
empirical relationship between total and differential settlements is enough (e.g., 75% of total
settlement may be taken as differential settlement).
The total and differential settlements should not exceed the permissible values. The per-
missible values of settlement for different types of structures are given in Table 14.10 (IS:
1904, 1986). The permissible differential settlement is obtained by taking the difference of
maximum and minimum settlements. Tilt is computed by dividing the differential settle-
ment by the distance between the points of related maximum and minimum settlements.
14.10 ALLOWABLE BEARING PRESSURE

A loaded foundation settles in direct proportion with increase in load. At higher load levels,
the rate of increase of settlement is extremely large and the foundation is said to have broken
into the ground or to have experienced a bearing capacity failure. It is evident that distinc-
tion between excessive settlement and failure by breaking into the ground is, in many
instances, quite arbitrary.
qB < qs qB > qs
qa = qB qa = qs Soil pressure
qB qf
SB < Ss
SB
Ss
SB
SB > Ss
Settlement
Fig. 14.17 Load–settlement curve of a foundation
M14_PURU1773_01_SE_C14.indd 544 2/12/2013 7:55:12 AM

Thus, every foundation has to satisfy two independent conditions. The first condition is
that there should be adequate factor of safety against shear failure of foundation. Second,
the settlement of the structure should not be great enough to damage the structure. Out of
these two conditions, whichever gives a lower value of load intensity is referred to as the
allowable soil pressure.
Let us consider the load–settlement curve of a foundation (Fig. 14.17). Let qf be the ultim-
ate bearing capacity, F the factor of safety, qs = qf/F the safe soil pressure with respect to
shear failure, Ss the settlement corresponding to qs, SB the permissible settlement of the foun-
dation, qB the intensity of pressure corresponding to SB, and qa the allowable soil pressure.
Now we can identify two conditions:
1. If the settlement Ss corresponding to safe soil pressure is less than the permissible settle-
ment SB, the pressure qB corresponding to SB is greater than qs. That is, the settlement
criterion is satisfied but the shear failure criterion is violated. Hence, the allowable soil
pressure is governed by the lesser pressure qs. That is,
SB > Ss
Therefore,
qa = qs (since qB > qs)
2. If the settlement Ss is greater than SB, the settlement criterion will be violated when qs is
adopted. Hence, the allowable soil pressure is governed by the lesser pressure qB. That is,
SB < Ss
Therefore,
qa = qB (since qB < qs)
In general, the allowable soil pressures in sands, gravelly sands, and silty sands are
governed only by the settlement considerations, except in narrow footings on loose sand.
In many situations, the permissible settlement is reached at a pressure for which the factor of
safety against shear failure is greater than 3.0. Settlement in sands occurs rapidly and about
80% to 90% of settlement takes place during construction.
The allowable soil pressure for clays, silty clays, and sandy clays is generally determined
con-sidering a factor of safety of 3.0 with respect to shear failure. However, in certain cases
the settlement criterion may predominate, for example, in normally consolidated clays. For
homogeneous clays with less permeability, the factor of safety has to be checked immedi-
ately after construction, adopting the undrained shear strength. But in case of fissured clays,
the permeability will be very high and the undrained shear strength condition may be very
much on the conservative side.
The most important soils intermediate between sand and clay are silt and loess which
have different characters. Loose silts behave worse than soft clays and are unsuitable for
supporting footings. Medium or dense silts are those which have characteristics of a rock
flour, or which have a certain plasticity. The allowable pressure on silts of the rock flour type
may be computed roughly by adopting the procedure for sand, and that on plastic silts by
the methods used for clay. No general rules can be established for silts in determining soil
pressure.
M14_PURU1773_01_SE_C14.indd 545 2/12/2013 7:55:12 AM

14.11 ESTIMATION OF BEARING CAPACITY FROM FIELD TESTS

Over the years, various attempts were made for obtaining bearing capacity values by more
direct approaches instead of using the more rigorous theoretical approaches. These methods
are, in general, empirical in nature and warrants sufficient judgement from the user. There
are three in situ tests used to estimate the bearing capacity of soils, viz., standard penetration
test, cone penetration test, and plate load test.
14.11.1 Bearing Capacity Based on Standard Penetration Test

On the basis of the results of standard penetration tests (as explained in Chapter 11), Terzaghi
and Peck in 1948 proposed correlations in the form of curves for estimating allowable soil
pressure for footings on sand. The correlations represented in Fig. 14.18 allowable to situa-
tions in which the water table is at least 2B below the foundation level, where B is the width
of the footing. The N values represented in Fig. 14.18 are the corrected values. If the water
table is at or close to foundation level and the depth–width ratio of foundation is small,
either the settlement is doubled or the allowable soil pressure is reduced by 50% for the same
permissible settlement of 25 mm. If the depth–width ratio is close to unity, the values need
to be reduced by only one-third.
Teng (1962) provided Eq. 14.38, which closely approximates the curves presented by
Terzaghi and Peck (Fig. 14.18)
⎛ 0.305B + 1⎞⎟
qna = 34.5( N − 3)⎜⎜⎜ 2
⎟ (kN/m ) (14.38)
⎝ 0.7 B ⎟⎠
700
Max. settlement 25mm
Allowable bearing pressure, kN/m2
600
Standard penetration resistance
500 50
400 40
300 30
200 20
100 10
5
0
0 1 2 3 4 5 6
Width of footing, m
Fig. 14.18 Allowable soil pressure based on N values (Source: Terzaghi and Peck, 1967)
M14_PURU1773_01_SE_C14.indd 546 2/12/2013 7:55:13 AM

where B is given in metres. Meyerhof (1956) suggested a correlation as given in Eqs. 14.39
and 14.40
qna = 12 N (kN/m 2 ) for B ≤ 1.22 m (14.39)
⎛ 0.305B + 1⎞⎟2
qna = 8 N ⎜⎜⎜ 2
⎟ (kN/m ) for B > 1.22 m (14.40)
⎝ 0.305B ⎟⎠
where qna is the net allowable soil pressure for a permissible settlement of 25 mm.
The correlations given in the chart generally give conservative values. These values were
so intended that the largest footing should not settle more than 25 mm even if it were situated
on the most compressible pocket of sand. In general, the soil pressure for any settlement is
S′
′ =
qna qna
Sa
′ is the net allowable soil pressure (in kPa) corresponding to the settlement S′ (in
where qna
mm) and Sa = 25 mm.
Peck et al. (1974) revised the Terzaghi and Peck curves, accounting for water table loca-
tion, based on research and observational data as
′ = Cw (0.41)NS ′
qna (14.41)
where Cw is the water table correction factor (0.5 < Cw < 1.0) and
d′
Cw = 0.5 + 0.5 w
df + B
Terzaghi and Peck curves and the correlations for qna are primarily intended for non-
cohesive soils like sand and gravel and may be used for silts with judgements.
The Bureau of Indian Standards (IS: 6403, 1981) recommends to find the angle of shearing
resistance φ from the corrected N values and to compute Nq and Nγ (from Table 14.1) to in
turn evaluate the net ultimate bearing capacity.
14.11.2 Bearing Capacity Based on Cone Penetration Test

Meyerhof (1956, 1965) has suggested formulae for allowable soil pressures based on cone
penetration test values restricting the settlement not to exceed 25 mm. His formulae are
based on Terzaghi and Peck’s curves for spread or strip footings on dry sands as
qcs
qa = (kPa) and B ≤ 1.2 m (14.42)
30
and
qcs ⎛ B + 0.3 ⎞⎟
qa = ⎜⎜ ⎟ (kPa) and B > 1.2 m (14.43)
50 ⎜⎝ B ⎟⎠
M14_PURU1773_01_SE_C14.indd 547 2/12/2013 7:55:14 AM

These formulae are based on the approximate rule that the N-value is one-quarter of the
static cone resistance.
The value of qcs obtained using the above equations should be reduced by 50% if the sand
within the stressed zone is submerged. Meyerhof further suggests that the values have to be
doubled for raft or pier foundation.
Schmertmann (1975) gave a method of calculating the allowable soil pressure indirectly
from the cone penetration test. He related Nγ with qcs as
qcs
Nγ = (kPa) (14.44)
80
With Nγ, one may work back to compute φ and then in turn Nq. Any standard theory may
be used to calculate the bearing capacity with Nq and Nγ values in silts and sands. Thorn-
burn (1971) has presented a correlation between qcs and N for particle range from 0.006 to 6
mm (Fig. 14.19). From a knowledge of qcs and the average particle size, we can find N and
hence the net allowable soil pressure. No standard correlation is yet available for clays using
cone penetration test.
Standarad penetration resistance N
Static cone resistance, kN/m2
10
8
6
4
2
0
Ratio of
0 0.002 0.006 0.02 0.06 0.2 0.6 2.0 6.0 20 60 200
Particle size, mm
Fig. 14.19 Relationship between static cone resistance and standard penetration resistance
(Source: Tomlinson, 1986)
0.2500
0
0.1875
qnf
0.5
qcs 0.1250
Dr
=1
0.0625 B
0 100 200 300 400

B, cm
Fig. 14.20 Chart for static cone test
M14_PURU1773_01_SE_C14.indd 548 2/12/2013 7:55:15 AM

For cohesionless soils, the Indian Standards code (IS: 6403, 1981) provides a chart relating
qnf/qcs and of footing B (Fig. 14.20). Knowing the depth and width of footing, the value of
qnf/qcs is obtained and hence the net ultimate bearing capacity.
14.11.3 Bearing Capacity Based on Plate Load Test

The object of a plate load test is to obtain a load settlement curve of a soil at a particular
depth. Such a curve is needed to estimate the ultimate bearing capacity, allowable soil pres-
sure, and the settlement of footings.
Wooden joists of Sand bags

suitable size at 300 mm Wooden planks
Wooden joists of suitable size

M S plate 15 cm f loading column (with
About 100 cm plum bob arrangement)
Clamp Dial gauge
Angle iron stakes

As required
Wooden guide joists
Test plate
As required
(a) Gravity loading platform
Ball and socket Loaded platform

Arrangement Head room for person to
Jack sit and observe dial gauge
Dial gauge As required

Dial gauge
Fixture Pit, strutted if necessary
Test plate As required
or block
(b) Reaction loading platform
Load truss
Spikes
Test pit
(c) Loading truss
Fig. 14.21 Typical set-up for loading (Source: IS: 1888, 1982)
M14_PURU1773_01_SE_C14.indd 549 2/12/2013 7:55:16 AM

In order to load the soil to the required level a suitable loading device is needed. Three
types of loading devices (Fig. 14.21) are recommended (IS: 1888, 1982), viz., gravity loading,
reaction loading, and loading truss. The load is measured using a pressure gauge, proving
ring or a cell. Settlements are recorded using dial gauges. Circular or square plates of size 300
to 750 mm are used. A Plate size of 300 mm is used for dense soils and 450 mm for loose soils.
Site for load test is located based on exploratory borings. A test pit of width equal to five
times the size of plate is made at the proposed foundation level.
After placing the plate over a thin layer of sand at the bottom of the pit a seating pressure
(70 g/cm2) is applied. The load is applied in equal load increments with each increment not
exceeding one-fifth of the estimated ultimate bearing capacity or 100 kN/m2, whichever is
less. The settlement is observed for each load increment.
A load–settlement curve in arithmetic scale is plotted. In dense or stiff soils, the failure is
well-defined (Fig. 14.22), whereas in loose or soft soils the failure is not pronounced. In such
cases, a plot of load and settlement, both being taken in logarithmic scales, gives two straight
lines the intersection of which is taken as the yield value of soil.
The safe bearing pressure is calculated from the ultimate bearing capacity after allowing
a certain factor of safety. In case of sandy soils, the plate settlement (Sp) corresponding to
safe soil pressure from the graph is found. The footing settlement (St) is computed from Eq.
14.32b. If the settlement St is less than the permissible settlement, then the safe soil pressure
computed above is the allowable soil pressure. Otherwise St is made equal to permissible
settlement and plate settlement Sp is computed back from Eq. 14.32b. Then, the soil pressure
corresponding to this Sp in the load–settlement curve is the allowable soil pressure.
The plate load test is adequate for light or less important structures under normal condi-
tions. However, in the case of unusual soil types and for all heavy and special structures
the plate load test results have to be supplemented with additional laboratory tests or field
tests. In order to arrive at a reasonable value for settlement and allowable soil pressure, tests
at different depths and with different sizes of plates have to be done. This is expensive and
time-consuming. However, plate load tests are best suited in weak-jointed rocks or soils
containing large gravel or boulders.
Housel (1929) suggested an entirely different procedure, but based on the results from
plate load test, for determining the load bearing capacity of shallow foundations considering
Ultimate bering capacity

Load per unit area
(D) Dense cohesionless soil

Settlement
(C) Partially cohesive soil
(A) Loose to medium dense

cohesionless soil
(B) Cohesive soil
Fig. 14.22 Typical load–settlement curves (Source: IS: 1888, 1982)
M14_PURU1773_01_SE_C14.indd 550 2/12/2013 7:55:16 AM

settlement criterion. That is, it is required to find the dimensions of a foundation that has to
carry a load Q with a tolerable settlement SB.
To obtain the relevant parameters, two plate load tests are conducted with different sizes
of plates. From the load–settlement curves for equal settlement, the loads Q1 and Q2 are
obtained. Then, the load is related to the area, perimeter, bearing pressure, and perimeter
shear. That is,
Q1 = A1 m + P1 n (14.45)
Q2 = A2 m + P2 n (14.46)
where A1 and A2 are the areas of plates No. 1 and No. 2, respectively, P1 and P2 are the perim-
eters of plates No. 1 and No. 2, respectively, m the constant corresponding to bearing pres-
sure, and n the constant corresponding to perimeter shear.
The constants m and n are obtained by solving the above equations. Then, for a given load
Q of the foundation, the area (A) and perimeter (P), and hence the dimensions are obtained
from
Q = Am + Pn
WORKED EXAMPLES
Example 14.1 In a mass-housing complex scheme over a vast area, two types of soils were
encountered. One of which is a partially saturated silty clay with cu = 5.8 kN/m2, φu = 25°,
and γ = 18.5 kN/m3 and extends over most of the area. The other, predominantly clay hav-
ing cu = 55 kN/m2 spreads to a lesser extent. The water table is at a greater depth. As per the
design, strip footings of the building have to be placed at 1 m depth. Compute the width of
the footing required in each type of soil if the load intensity is 150 kN/m run. Adopt a factor
of safety of 2.5 in both the soils, and only shear failure need to be considered. For φ = 25°,
take Nc = 20.7, Nq = 10.7, Nγ = 10.8.
If there is a possibility of the water table rising to the ground surface, what should be the
change in the width of footing in both areas. The submerged unit weight of the silty clay is
11.2 kN/m3.
Solution
For partially saturated silty clay, the net safe bearing pressure
1⎡ 1 ⎤
qns = ⎢ cNc + q( N q − 1) + γ BN γ ⎥
⎢
F⎣ 2 ⎥⎦
or 150 1 ⎡ 1 ⎤
= ⎢ 5.8 × 20.7 + 18.5×1(10.7 − 1) + ×18.5× B×10.8⎥
⎢
B×1 2.5 ⎣ 2 ⎥⎦
or 150
= 119.8 + 39.96B
B
M14_PURU1773_01_SE_C14.indd 551 2/12/2013 7:55:18 AM

39.96B2 + 119.8B − 150 = 0

or
B2 + 3.03B − 3.75 = 0
Therefore,
−3.03 ± (3.03)2 + 4 ×1× 3.75 −3.03 ± 4.92

B= =
2 ×1 2
Therefore, width of footing on partially saturated silty clay = 0.95 m.
For the clay soil with φ = 0°, the values for bearing capacity factors are Nc = 5.7, Nq = 1,
and Nγ = 0. Hence, the net bearing pressure reduces to
1
qns = [cNc ]
F
or
150 1
= [55× 5.7 ]
B×1 2.5
or
150 × 2.5
B= = 1.2 m
55× 5.7
Therefore, the width of footing on clay = 1.2 m.

Because of submergence the unit weight of soil will be reduced to the submerged unit
weight. Hence, the terms containing γ should be replaced by γ′. Thus, footings only on silty
clay will be affected. The footings on clay are independent of the unit weight. Hence, for
footings on silty clay
1⎡ 1 ⎤
qns = ⎢ cNc + γ ′Df ( N q − 1) + γ ′BN γ ⎥
F ⎢⎣ 2 ⎥⎦
or
150 1 ⎡ 1 ⎤
= ⎢ 5.8 × 20.7 + 11.21(10.7 − 1) + ×11.2× B×10.8⎥
B×1 2.5 ⎢⎣ 2 ⎥⎦
or
150
= 91.48 + 24.19B
B
or
B2 + 3.78B − 6.2 = 0
M14_PURU1773_01_SE_C14.indd 552 2/12/2013 7:55:19 AM

or
−3.78 ± (3.78)2 + 4 ×1× 6.2 −3.78 ± 6.25
B= = = 1.24
2 ×1 2
Hence a width of 1.24 m has to be provided for footings on silty clay.
Example 14.2 A 1 m wide long footing is located at a depth of 1.5 m from the ground sur-
′ = 30 kN/m 2
face. The supporting soil is compressible and has shear strength parameters, ccu
3
and φcu′ = 25° . The total unit weight of the soil, γ = 18.3 kN/m . The water table is at a
greater depth. Compute the safe load that can be carried by the long footing per metre length
of the wall. Adopt a factor of safety of 3.0.
Solution
As the soil is compressible, the reduced shear strength parameters and bearing capacity
factors corresponding to the local shear condition are used.
Therefore,
′ = 32 × 30 = 20 kN/m 2
cl = 32 ccu
and
tan φl = 2
3
′
tan φcu
or
φl = tan−1 [ 2
3
tan 25°] = 17.3°
For φl = 17.3°, the bearing capacity factors are taken as

Nc′ = 13.91
N q′ = 5.17
N γ′ = 4.02
Based on Eq. 14.20,
1⎡
qs = cl Nc′ + q( N q′ − 1) + 12 γ BN γ′ ⎤⎥ + q
F ⎣⎢ ⎦
qs = 31 [30 ×13.91 + 18.3 ×(5.17 − 1) + 12 ×18.3 ×1× 4.02] + 18.3 ×1.5
or
qs = 31 [429.45] + 27.45 = 170.6 kN/m 2
Therefore,
Qs = qs × B = 170.6 × 1 = 170.6 kN
Safe load that can be carried by the wall = 170.6 kN/m.
M14_PURU1773_01_SE_C14.indd 553 2/12/2013 7:55:21 AM

Example 14.3 The construction of a strip footing is undertaken during a summer period
and the water table is observed at 2.5 m from the ground surface (Fig. 14.23). During monsoon
the water table rises to the ground surface. The relevant soil parameters are γ = 19.2 kN/m3
and φ = 32°. Determine the gross safe bearing capacity in both the cases for a factor of safety
of 2.5. Use Teng’s water table correction factors.
Solution
For the summer condition,
1⎡
qns = ′ q( N q − 1) + 12 Rw γ BN γ ⎤⎥ + Rw
Rw ′ γ Df
F ⎢⎣ ⎦
For φ = 32°, the bearing capacity factors are Nq = 23.2 and Nγ = 30.2.
Rw is obtained from the expression for dw = 0.5 m.
That is,
⎛ d ⎞ 1⎛ 0.5 ⎞⎟
Rw = 12 ⎜⎜1 + w ⎟⎟⎟ = ⎜1 +
2⎜ ⎟ = 0.58
⎜⎝ B⎠ ⎜⎝ 3 ⎟⎠
and
′ = 1.0
Rw
Hence,
1 ⎡
qns = 1×19.2× 2(23.2 − 1) + 12 × 0.58 ×19.2× 3 × 30.2⎤⎦ + 1×19.2× 2
2.5 ⎣
= 542.8 + 38.4 = 581.2 kN/m 2
For the monsoon condition,
′ = 0.50 and Rw = 0.5

Rw
Hence,
1 ⎡
qs = 0.5×19.2× 2(23.2 − 1) + 12 × 0.5×19.2× 3 × 30.2⎤⎦ + 0.5×19.2× 2
2.5 ⎣
= 344.45 + 19.2 = 363.7 kN/m 2
3m
qns Monsoon
condition
2.0 m
0.5 m Summer
condition
Fig. 14.23
M14_PURU1773_01_SE_C14.indd 554 2/12/2013 7:55:23 AM

Example 14.4 In a warehouse building, two unequally loaded columns are combined by a
rectangular combined footing. It is proposed to place the footings at a depth of 1.5 m on a
saturated clay with the following soil properties: cu = 72 kN/m2, φu = 0°, γ = 17.8 kN/m3.
The loads on the columns are 720 and 1,170 kN, with a spacing of 5 m, and the centre of the
720 kN column is placed at a distance of 0.8 m from the property line (Fig. 14.24). Neglecting
the weight of the footing, estimate the dimension of the footing. Adopt a factor of safety of 3.
Solution
The general expression (Eq. 14.13) for net bearing capacity can be written as
qnf = sc cu Nc + sq q( N q − 1) + 12 sγ γ BN γ
For a rectangular footing,

⎛ B⎞
sc = ⎜⎜⎜1 + 0.2 ⎟⎟⎟
⎝ L⎠
⎛ B⎞
sγ = ⎜⎜⎜1 − 0.4 ⎟⎟⎟
⎝ L⎠
and sq = sc .
Further, for φu = 0°, adopting Terzaghi’s values Nc = 5.7, Nq = 1, and Nγ = 0,
⎛ B⎞
qnf = ⎜⎜⎜1 + 0.2 ⎟⎟⎟× 5.7 cu
⎝ L⎠
⎛ B⎞
= ⎜⎜⎜1 + 0.2 ⎟⎟⎟× 5.7 ×72
Assume L/B = 4, then ⎝ L⎠
⎛ 0.2 ⎞⎟
qnf = ⎜⎜⎜1 + ⎟× 5.7 ×72 = 430.92 kN/m
2
⎝ 4.0 ⎟⎠
q 430.92
qns = nf = = 143.64 kN/m 2
F 3
1,170 kN 720 kN
5m 0.8
m
Property line
Fig. 14.24
M14_PURU1773_01_SE_C14.indd 555 2/12/2013 7:55:24 AM

The width B and length L can be obtained by satisfying ∑V = 0 and the centre of gravity
requirement.
Equating the upward force with the downward force, or qns × footing area = column loads,
143.64 × B × L = 720 + 1170
1890
B× L = = 13.16 m 2
143.64
The centres of gravity of the footing pressure and the loading should be in one line. Tak-
ing moment about the property line
720 × 0.8 + 1170 × 5.8 = 1890 × x
7362
x=
= 3.9 m
1890
Therefore, L/2 = 3.9 m or L = 7.8 m. That is,
13.16 L
B= = 1.69 m and = 4.6
7.8 B
L and B are modified such that L/B = 4.
Hence, L = 7.25 m and B = 1.82 m.
Example 14.5 A circular concrete pier of 3 m diameter carries a gross load of 3,500 kN.
The supporting soil is a clayey sand having the following properties: c = 5 kN/m2, φ = 30°,
and γ = 18.5 kN/m3. Find the depth at which the pier is to be located such that a factor of
safety of 3.0 is assured. The bearing capacity factors for φ = 30° are Nc = 30.1, Nq = 18.4,
and Nγ = 22.4.
Solution
The gross safe bearing pressure is given as
1⎡
qs = 1.3 Nc + q( N q − 1) + 0.6 γ BN γ ⎤⎥ + q
F ⎢⎣ ⎦
3500
= 31 [1.3 × 5× 30.1 + 18.5× Df (18.4 − 1) + 0.6 ×18.5× 3.0 × 22.4] + 18.5Df
π×(3.0)2
4
or
495.2 = 313.86 + 107.3Df + 18.5Df
or
495.2 − 313.86
Df = = 1.44 m
107.3 + 18.5
Thus, depth of the location of pier = 1.44 m.
M14_PURU1773_01_SE_C14.indd 556 2/12/2013 7:55:25 AM

Example 14.6 The weight of a heavy machinery is 7,600 kN and the base dimensions are
5.5 m × 3.5 m. The machinery has to be installed on a stiff clay soil with a cohesion of
150 kN/m2, at a depth of 0.8 m below the ground surface. The total unit weight of the soil is
19.2 kN/m3. Determine the size of the foundation required if the minimum factor of safety
is 3.0. Assume the load to be rapidly applied so that undrained condition prevails (φ = 0).
Neglect the weight of the foundation.
Solution
Since the loading is made rapidly and the stratum is clay, Skempton’s bearing capacity equa-
tion (Eq. 14.15) may be used. Thus,
qf = Cu Nc + γ Df
Provide an all-round clearance of 0.25 m then the length and width of footing may be
taken as 6 and 4 m, respectively. For the condition, Nc may be computed from Eq. 14.16.
Thus,
B ⎞ ⎡ ⎛ 0.053Df ⎞⎟ ⎤⎥
1/ 2
⎛
Nc = 5.14 ⎜⎜⎜1 + 0.2 ⎟⎟⎟ ⎢⎢1 + ⎜⎜⎜ ⎟⎟ ⎥
⎝ L ⎠ ⎢⎣ ⎝ B ⎠ ⎥⎦
4⎞⎡ ⎛ 0.8 ⎞ ⎤
1/ 2
⎛
= 5.14 ⎜⎜⎜1 + 0.2× ⎟⎟⎟ ⎢⎢1 + ⎜⎜⎜0.053 × ⎟⎟⎟ ⎥⎥
⎝ 6 ⎠ ⎢⎣ ⎝ 4 ⎠ ⎥⎦
= 5.4 ×1.25 = 6.43

Therefore,
cu Nc 150 × 6.43
qns = = = 321.5 kN/m 2
F 3
Actual pressure transferred by the machine is 7600 /(6 × 4) = 316.67 kN/m 2
Hence, the recommended dimension of the foundation is 6 m × 4 m.
Example 14.7 An eccentrically loaded rectangular footing of size 2.5 m × 3.5 m is placed at
a depth of 1 m on a stiff saturated clay. The eccentricity is 0.2 m in each direction. The footing
is loaded rapidly and the soil properties are c = 105 kN/m2 and γ = 17.8 kN/m3. Compute
the safe net allowable bearing load on the footing if the factor of safety is 3.0 and the settle-
ment is negligible.
Solution
Based on the useful width concept, the width and length of the footing are given as
L ′ = L − 2e x = 3.5 − 2× 0.2 = 3.1 m
B ′ = B − 2e y = 2.5 − 2× 0.2 = 2.1 m
The net safe soil pressure for strip footing is given by Eq. 14.19 as
qnf 1
qns = = ⎡⎢ cu Nc + q( N q − 1) + 12 γ BN γ ⎤⎥
F F⎣ ⎦
Modifying the above equation with shape factor correction and taking Nc = 5.7, Nq = 1.0,
and Nγ = 0, we have
M14_PURU1773_01_SE_C14.indd 557 2/12/2013 7:55:27 AM

1 ⎡⎢⎛⎜ B′ ⎞ ⎤
qns = ⎜1 + 0.2 ⎟⎟⎟ cu Nc ⎥
F ⎢⎣⎜⎝ L′ ⎠ ⎥
⎦
⎡⎛ 2.1⎞ ⎤
= 31 ⎢⎜⎜⎜1 + 0.2× ⎟⎟⎟×105× 5.7 ⎥ = 226.5 kN/m 2
⎢⎣⎝ 3.1⎠ ⎥⎦
As the settlement is negligible, the net safe soil pressure with respect to shear strength is
the net allowable soil pressure also.
Therefore, the net allowable load,
Qna = (qns )(useful area) = 226.5× 2.1× 3.1 = 1474.5 kN
Example 14.8 The corrected blow count from standard penetration test in a medium sand,
observed at an average depth of 2.5 m was 22 blows per 305 mm. Laboratory tests conducted
on the sample revealed the following physical properties: c = 0, φ = 30°, and γ = 18.5 kN/m3.
The water table was located at 4.5 m from the ground surface. It is planned to place a 2 m
wide square footing at a depth of 2 m. Estimate the allowable gross bearing pressure for the
soil if the factor of safety against shear failure is 2.5 and the limiting total settlement is 25 mm.
Solution
For limiting settlement, the net bearing pressure is given by Eq. 14.41 as
′ = Cw (0.41)NS ′
qna
Here,
dw′ 4.5
Cw = 0.5 + 0.5 = 0.5 + 0.5× = 1.063
Df + B 2+2
Therefore,
′ = 1.063 × 0.41× 22× 25 = 239.7 kN/m 2

qna
The net bearing pressure based on the shear condition is given as

1⎡
qns = 1.2q( N q − 1) + 0.8 ×γ BN γ ⎤⎥
F ⎣⎢ ⎦
For φ = 30°, Nq = 18.4, and Nγ = 22.4,

1 2
qns = [1.2×18.5× 2 (18.4 − 1) + 0.8 ×18.5× 2× 22.4 ] = 574.2 kN/m
2.5
The lower of the two net bearing pressures is taken as the net allowable bearing pressure.
Therefore, net allowable bearing pressure = 239.7 kN/m2
Gross bearing pressure = 239.7 + 18.5 × 2 = 276.7 kN/m2
Example 14.9 Two plate load tests with square plates were performed on a soil deposit. For
a 30 mm settlement, the following loads were obtained.
M14_PURU1773_01_SE_C14.indd 558 2/12/2013 7:55:28 AM

Width of square plate (mm) Load (kN)
300 38.2
600 118.5
Determine the width of a square footing which would carry a net load of 1,500 kN for a lim-
iting settlement of 30 mm.
Solution
Plate No. 1: 38.2 = (0.3)2m + 4 × 0.3n (a)
Plate No. 2: 118.5 = (0.6)2m + 4 × 0.6n (b)
Solving by elimination, we get,
(a) × 0.36 → 0.032m + 0.43n = 13.75
(b) × 0.09 → 0.032m + 0.22n = 10.67
Therefore,
13.75 − 10.67
n= = 14.37 kN/m
0.43 − 0.22
and
13.75 − 0.43 ×14.67
m= = 232.6 kN/m 2
0.032
For the foundation,

Am + Pn = Q
Therefore,
B2 × 232.6 + 4B ×14.67 = 1500
or
B2 + 0.25B − 0.65 = 0
or
1/ 2
2 −0.25 ± [(0.25)2 + 4 ×10.65] −0.25 ± 1.63
B = = = 0.69 m
2 2
Therefore, the width of the square foundation = 0.69 m ≈ 0.70 m.
Example 14.10 Plate load test data are given below. Plot the load–settlement curve and find
the ultimate bearing capacity.
Width of plate = 300 mm
Least count of dial gauge = 0.01 mm
M14_PURU1773_01_SE_C14.indd 559 2/12/2013 7:55:29 AM

Load intensity (kN/m2) Dial gauge reading

A B C
0 0 0 0
55 186 192 192
110 362 365 353
165 766 758 756
220 1,886 1,889 1,865
280 4,810 4,806 4,784
335 14,006 14,010 13,984
Solution
Settlements are calculated by multiplying average values of dial gauge readings and the
least count of dial gauges. The load–settlement curve is plotted as shown in Fig. 14.25. The
ultimate bearing capacity is read from graph as 242 kN/m2.
Load per unit area, kN/m2

60 120 180 240 300 360
0
10
20
30
40
50
60
Settlement, mm
70
80
90
100
110
120
130
140
150
Fig. 14.25
M14_PURU1773_01_SE_C14.indd 560 2/12/2013 7:55:29 AM

POINTS TO REMEMBER
14.1 General shear failure, usually associated with dense stiff soils of relatively low
compress-ibility, is said to occur when a continuously well-defined slip surface devel-
ops on both sides of a footing and extends from the edge of the footing to the soil surface.
14.2 In case of local shear failure, usually associated with medium dense or medium stiff
soils, the slip surfaces extend from the edges of the footings to a certain length only
and do not reach the ground surface.
14.3 Punching shear failure, usually associated with loose or soft soils, is said to occur
when there is compression beneath the footing accompanied by shearing in the ver-
tical direction around the edges of the footing.
14.4 Prandtl’s theory considers the deformation or penetration effects of hard objects on
soft materials which is considered as the basic principle adopted in different bearing
capacity theories.
14.5 Nc, Nq, and Nγ are the bearing capacity factors which depend on angle of internal
friction only.
14.6 Effect of soil compressibility is taken into account by considering the failure as a local
shear failure and the corresponding bearing capacity factors, Nc′ , N q′ , and N γ′ , for the
reduced friction angle φl = tan−1[2/3(tan φ)]. The cohesion is also reduced as cl = 2/3c.
14.7 Effect of water table is accounted by considering submerged unit weight in place of
total unit weight depending on the location of water table.
14.8 Effect of different shapes of foundation are taken in to account by appropriate shape
factors sc, sq, and sγ .
14.9 The total pressure on the soil due to the weight of the structure is called the gross
foundation pressure. The net foundation pressure is the foundation pressure in
excess of the pressure caused by the surrounding soil. Thus, the net ultimate bearing
capacity is the net foundation pressure at the time of failure.
14.10 Settlements may be classified as uniform (or total) settlement, tilt, and non-uniform
settlement. The total and differential settlements should not exceed the permissi-
ble values which are denoted as permissible settlements. Permissible settlements
depend on the type of structure and type of soil.
14.11 Allowable soil pressure is one which gives the lowest value based on the two condi-
tions, viz., adequate factor of safety against shear failure and settlement should be
less than permissible settlement.
QUESTIONS
Objective Questions
(1) The bearing capacity factors for a clayey soil will depend on cohesion, shape, and
size of the footing.
(2) Greater the width of foundation, greater is the settlement for the same pressure
intensity.
M14_PURU1773_01_SE_C14.indd 561 2/12/2013 7:55:30 AM

(3) The safe bearing capacity of a surface strip footing on a saturated clay is
approximately equal to the unconfined compressive strength.
(4) The correction factors to account for the effect of shapes are based on sound theo-
retical analysis.
14.2 The total settlement of a soil layer under any given loading is
(a) Proportional to the thickness of the layer
(b) Proportional to the square of the thickness of the layer
(c) Dependent on the length of the drainage path
(d) Dependent on factors other than the above
14.3 The ultimate bearing capacity of a footing on strip footing is reduced by 50% when the
position of water table is at
(a) The base of the footing
(b) The ground surface
(c) A depth equal to 1.5 times the depth of foundation
(d) A depth equal to 0.5 times the depth of foundation
14.4 Two strip surface footings of equal lengths are placed on dry sand and the width of
footing A is equal to half the width of footing B? Then the ratio of the load carrying
capacities of A and B (i.e., qA/qB) is
(a) 1/2 (b) 1/4 (c) 2 (d) 1
14.5 Bearing capacity of a footing consists of the following components:
(1) The cohesion and friction of a weightless material carrying no surcharge
(2) The friction of a weightless material upon addition of a surcharge on the ground
surface
(3) The friction of a material possessing weight and carrying no surcharge
(a) 1, 2, and 3 are correct
(b) 1 and 2 are correct
(c) 2 and 3 are correct
(d) 3 and 1 are correct
14.6 Identify the incorrect statement. Bearing capacity of a footing on sand depends on
(a) Depth of footing
(b) Width of footing
(c) Position of water table
(d) Undrained shear strength
14.7 Plate load test results reflect only the character of the soil located within a depth
______ the width of the bearing plate. Choose the correct statement.
(a) Equal to
(b) Less than twice
(c) Equal to 2.5 times
(d) More than twice
14.8 The technique of reducing the net load on a soil by excavating soil up to a certain
depth is called
(a) Load relief
(b) Buoyancy method
M14_PURU1773_01_SE_C14.indd 562 2/12/2013 7:55:30 AM

(c) Flotation
(d) Load reduction
14.9 Identify the incorrect statement. Meyerhof’s bearing capacity equation includes
(a) Shape factor
(b) Width factor
(c) Depth factor
(d) Inclination factor
14.10 What factors determine whether a foundation type is shallow or deep?
14.11 Explain why bearing capacity equations for clay usually employ the undrained shear
strength.
14.12 How will you proportion footings for equal settlements?
EXERCISE PROBLEMS
14.1 On a cohesive friction soil, a square foundation of 2 m × 2 m is founded at 1 m depth.

The soil has the following properties: ccu = 15.5 kPa, φcu = 28°, and γ = 18.2 kN/m3.
Determine (i) the net ultimate load, (ii) the gross ultimate load, (iii) the net safe load,
and (iv) gross safe load on the footing if the factor of safety with respect to shear fail-
ure is 3.0. Assume that general shear failure occurs in the soil.
14.2 A strip footing of width 3.5 m is to be placed at a depth of 0.5 m below the ground
surface on a compressible sandy silt having a bulk unit weight of 18.7 kN/m3. The
shear strength parameters of the soil are c = 5 kN/m2 and φ = 22°. Determine the net
ultimate bearing pressure and the net safe load if the factor of safety against shear
failure is 4.0.
14.3 The width of a square footing of an existing building is 2 m and is located at 1 m below
the ground surface. It is proposed to add one additional floor which would make the
total load on the column 1,800 kN. Check whether the existing footing is adequate if it
is intended to maintain a factor of safety of 3.0. The soil at the location has the follow-
ing properties: c = 20 kPa, φ = 35°, and γ = 18.5 kN/m3.
14.4 Estimate the factor of safety of a 2 m square footing, located at a depth of 1.5 m and
subjected to a 1,500 kN vertical load. A horizontal load of 300 kN is also applied at the
base of the footing. The soil is dry sand with φ′ = 33° and γ = 18.2 kN/m3.
14.5 A long bridge pier, 3 m wide, carries a load of 1,640 kN per liner metre of its length. It
is founded 5 m below the ground level on a soil whose angle of friction is 15° and unit
cohesion 38.4 kN/m2. The unit weight of the soil is 15.7 kN/m3. Check for the safety
of the pier if the factor of safety is 3.0.
14.6 A square footing of width 2.5 m is positioned on a medium dense sand at a depth of
2 m from the ground surface. The sand has a void ratio e = 0.72, specific gravity of
soil solids G = 2.65, and the angle of shearing resistance φ = 35°. Adopting a factor of
safety of 2.5, find the safe load on the footing for the following water table positions:
M14_PURU1773_01_SE_C14.indd 563 2/12/2013 7:55:30 AM

(i) at 5 m from the ground surface, (ii) at 1.5 m from the base of the footing, and (iii) at
1.2 m from the ground surface.
14.7 The depth of a wall footing to be constructed on a saturated clay is 1 m. The soil
parameters are cu = 65 kPa, φu = 0°, and γ = 17.5 kN/m3. The wall imposes a load of
170 kN/m of wall length. Estimate the width of the footing to be provided so as to
have a factor of safety of 3.0.
14.8 The load on a reinforced cement concrete column is 1,000 kN. The supporting soil is
a dry, dense sand with the angle of friction of 41° and a unit weight of 18.2 kN/m3.
Find the size of the square footing for the following conditions considering a factor of
safety of 3.0:
(a) if it is placed on the ground surface
(b) if it is placed at 1.5 m below the ground surface
(c) if water table rises to the ground surface for the case (b) above, the saturated unit
weight is 21.3 kN/m3.
14.9 A square column transfers a load of 1,650 kN on a c–φ soil and rests on a soil which
weighs 19 kN/m3 and has shear strength parameters as c = 10 kN/m2 and φ = 36°.
Considering a factor of safety of 2.5, find the size of the footing if it is placed at the
ground surface. Examine whether it would be cheaper to lower the footing if the col-
umn is 450 mm2 and the footing is 500 mm thick than to place it at the ground surface.
The cost of con-crete and the cost of excavation for hard soil at a site are Rs. 4,600/m3
and Rs. 60/m3, respectively.
14.10 Calculate the minimum depth of footing required below ground level in a clay stra-
tum if the footing is to be safe
(i) For a continuous wall footing with a contact pressure of 65 kN/m2 and width 1.6 m.
(ii) For a square footing with a contact pressure of 65 kN/m2 and side width 1.6 m.
The undrained shear strength parameters are cu = 25 kN/m2, φu = 0°, and γ =
16 kN/m3. Adopt a factor of safety of 3.0. Discuss the effect of footing shape on
the depth of footing.
14.11 During a sub-surface exploration programme, two cohesive layers are encountered.
One forms the top layer of finite thickness 3 m, which is stiff clay and the bottom
one is soft and showed undrained shear strengths of 135 and 50 kN/m2, and the
respective unit weights are 17.2 and 16.7 kN/m3. It is intended to design a foundation
1.5 m from the ground surface. Compute the gross load for the foundation with a
factor of safety of 2.5. If the layered system is assumed as homogeneous and isotropic
with average values of cohesion and unit weights of both the layers, what is the per-
centage error involved?
14.12 Determine the safe load which can be imposed normal to the base of a strip footing
which is 1.2 m wide and has its base inclined at 12° from the horizontal. One corner
of the footing is located at 1.2 m from the ground surface. The footing rests on a satu-
rated cohesive soil with a cohesion of 75 kN/m2 and unit weight 18.2 kN/m3.
14.13 A load bearing wall of an industrial building is to be located close to the edge of a
slope as shown in Fig. 14.26. The shear strength parameters of the soil are cu = 45 kN/
m2, φu = 0°, and the unit weight of soil γ = 18.2 kN/m3. Suggest a suitable width of
strip footing for the given condition.
M14_PURU1773_01_SE_C14.indd 564 2/12/2013 7:55:30 AM

3m
Q = 100 kN/m
1.5 m
B
10 m
Saturated clay
c = 45 kN/m2
fu = 0°
45°
g = 18.2 kN/m3
Fig. 14.26
14.14 A standard penetration test conducted at 2 m depth and a subsequent laboratory test
revealed the soil at a location as medium dense sand with a blow count of 29 blows
and a moist unit weight of 18 kN/m3. It is planned to design a square footing on this
sand to carry a load of 3,500 kN. As per the design requirement, the footing has to be
designed for settlement criterion and the maximum total settlement should be lim-
ited to 25 mm. The water table is at 6 m from the ground surface. Find the size of the
footing.
14.15 A square footing of 4 m width and 0.8 m thickness is supported by a sand having an
average N-value of 30. The top of the footing is 1 m below the ground surface, and the
water table is 1.2 m below the base of the footing. Determine the maximum load that
the footing can carry if the settlement is not to exceed 15 mm.
14.16 The results of a plate load test conducted on a 300 mm square plate at a depth of 1 m
on a dry sand is given below.
Until applied pressure (kN/m2) Settlement (mm)
50 3
100 5
150 9.8
200 13.0
250 19.0
275 22.0
300 28.0
325 39.0
350 65.0
Determine (i) the ultimate bearing pressure, (ii) the safe bearing pressure if the factor of
safety is 3.0, (iii) the size of a square footing to be placed at the same depth and to carry
M14_PURU1773_01_SE_C14.indd 565 2/12/2013 7:55:31 AM

a load of 2,500 kN, considering the safe bearing pressure obtained in (ii), and (iv) the
settlement of the footing.
14.17 In a plate load test using a 305 mm square plate on a sandy soil under a pressure
of 150 kN/m2, a settlement of 8 mm was recorded: (i) estimate the settlement of a
600 mm square plate at the same contact pressure and (ii) what should be the size
of a square footing if the settlement is to be restricted to 25 mm?
14.18 Develop an allowable bearing pressure chart for square footing on sand. The aver-
age corrected N-value from 2 to 10 m is 25, and the groundwater was encountered at
a 15 m depth during sub-surface exploration. The depth of footing is 2 m. The other
properties of the soil are φ′ = 35°, γ = 18.5 kN/m3. Adopt a safety factor of 2.5 against
shear failure and a limiting total settlement of 25 mm.
M14_PURU1773_01_SE_C14.indd 566 2/12/2013 7:55:31 AM

15
Shallow Foundations
CHAPTER HIGHLIGHTS
Design criteria – Types of shallow foundations – Selection of type of founda-
tion – Location and depth of foundation – Settlement of shallow foundation
– Design considerations for a shallow foundation – Proportioning of combined
footing – Mat foundation
15.1 INTRODUCTION
Structural foundations may be grouped under two broad categories – shallow founda-
tions and deep foundations. This classification indicates the depth of foundation instal-
lation. A shallow foundation is one which is placed on a firm soil near the ground, and
beneath the lowest part of the superstructure. A deep foundation is one which is placed
on a soil that is not firm, and which is considerably below the lowest part of the super-
structure. There is no exact definition which distinguishes one from the other.
15.2 DESIGN CRITERIA

While considering a shallow foundation for a given loading system, the foundation
must meet certain design requirements. The three basic requirements are as follows:
1. Foundation placement, which involves the location and depth of foundation, requires
a careful investigation of the past usage of the site and detailed information of the
sub-surface stratum. The foundation placement should be such that any future influ-
ence should not affect its performance adversely.
2. Safety against bearing capacity is a requirement that involves suitable proportioning
of footing to avoid a catastrophic collapse of the soil beneath the foundation. This
M15_PURU1773_01_SE_C15.indd 567 2/12/2013 8:48:08 AM

occurs if the shear strength of the soil is inadequate to support the applied load. This
requirement makes it essential to have a complete knowledge of the geotechnical proper-
ties of the soils and rocks involved.
3. Tolerable foundation settlement involves keeping a check on the excessive settlement of a
structure. Excessive settlement is caused due to the distortion of the soil mass as a result of
the applied shear stresses and due to the consolidation of the supporting soil. This again
requires a complete knowledge of the geotechnical properties of the soil to assess the
anticipated settlement of the structure and the time required for completion of the same.
15.3 TYPES OF SHALLOW FOUNDATIONS

Shallow foundations are subdivided into a number of types according to their size, shape,
and general configuration. They are described below.
15.3.1 Spread Footings

These footings are the most common of all types of footings involving minimum cost and
complexity of construction (Fig. 15.1a). It necessarily provides the function of distributing
the column load to a value compatible with the strength and deformation characteristics of
the soil or rock on which the foundation is placed. These types of footings are also known as
pad footings, isolated footings, and square or rectangular footings (for an L/B ratio less than 5).
15.3.2 Combined Footings

These footings are formed by combining two or more columns (even with unequal loadings)
into one footing. This arrangement averages out and provides a more or less uniform load
distribution in the supporting soil or rock and, thus, prevents differential settlement. These
footings are usually rectangular in shape but may be modified to a trapezoidal one to accom-
modate unequal column loadings (or columns close to property lines) and provided with a
strap to accommodate wide column spacings or column close to property lines (Fig. 15.1b).
15.3.3 Continuous Footings

These footings carry closely spaced columns or a continuous wall so that the load intensity
is low and uniform on the supporting soil or rock (Fig. 15.1c). In such footings, the load per
unit length is considered accordingly. The load intensity is given in terms of force per unit
length of the footing. These footings are also referred to as strip footings or wall footings (for
an L/B ratio greater than 5).
15.3.4 Mat Foundations or Footings

These are characterized by the feature that columns frame into the footing in two directions.
Any number of columns can be accommodated, and the number can be as low as four
(Fig. 15.1d). These are recommended for poor foundation soils and when the total area of
footings exceeds 50% of the total plinth area.
M15_PURU1773_01_SE_C15.indd 568 2/12/2013 8:48:08 AM

Shallow Foundations 569
Column Q Q Column
Column Q
Elevations
(a) Spread footings Q2
Q1
Q1 Q2 Q1 Q2
Elevations
Property line
Plans
Rectangular Trapezoidal
Q Q Strap
(b) Combined footings
Wall Elevations
Section-XX Section-XX X
Plans
X
(c) Continuous footings X
Elevations
Flat plate Plans

Beam and slab
(d) Mat foundations
Fig. 15.1 Types of shallow foundations
M15_PURU1773_01_SE_C15.indd 569 2/12/2013 8:48:08 AM

15.3.5 Floating Foundations

The total load of a structure may cause a pressure that is more than the safe bearing
capacity of the soil or an excessive settlement. In such cases, instead of changing the foun-
dation size, the structure may be placed at a greater depth. By this means, the weight of
the excavated soil reduces the total load and only a net load (total load – weight of exca-
vated soil) is transferred to the soil. This technique of reducing the net load by more
excavation is called floatation and the foundation is called a floating foundation. The
technique where the load of a structure is partially adjusted by the relief of load due to
excavation is called partial floatation and if fully adjusted it is full floatation. These tech-
niques are suitable for light structures on soft or loose soils and for heavy structures
constructed over a limited area.
15.4 SELECTION OF THE TYPE OF FOUNDATION

The selection of a foundation suitable for the type of structure to be constructed or for a
given size depends on several factors. The most important factors are
1. the type of structure, its intended function, and the load it is expected to carry;
2. the cost of the sub-structure, including the treatment of the foundation of soil or rock
if any.
The choice of the foundation should be such that it will be stable under all adverse condi-
tions and for the particular type of structure under all loading conditions; at the same time,
it should involve less expenditure.
The loads to be considered for a given structure may include dead loads, live loads, wind
loads, impact loads, lateral pressure, etc.
The sub-surface conditions should be favourable for the given structure, failing
which it should be treated to meet the requirements of the superstructure and its load-
ing condition.
As several factors contribute to the choice, the engineer concerned has to look into vari-
ous acceptable solutions.
The following are the general steps to be followed by the concerned engineer in choosing
the type of foundation:
1. Collect the necessary data about the type of structure and the loads anticipated to be car-
ried by the structure.
2. Get adequate information about the subsoil condition through a suitable soil investigation.
3. Explore the possibility of constructing a different foundation keeping in mind the basic
design criteria for a foundation. During this exercise, all unsuitable types may be elimi-
nated in the preliminary choice.
4. Select one or two types of foundations based on the preliminary studies, which may be a
shallow or deep foundation, and carry out more detailed studies regarding the stability
of the foundation and superstructure.
5. Work out cost estimates of the one or two chosen foundations.
6. Finally, decide on three types of foundations to satisfy all the requirements.
M15_PURU1773_01_SE_C15.indd 570 2/12/2013 8:48:08 AM

15.5 LOCATION AND DEPTH OF THE FOUNDATION

The foundation must be located properly keeping in view both the horizontal and the verti-
cal orientations, such that it is not affected by outside influences, apart from the general
design criteria of bearing capacity and excessive settlement. Outside influences may include
groundwater, volume changes, underground defects, adjacent structure, etc.
Thus, the depth and location of a foundation depend on the following data:
1. volume change of soil

2. adjacent structures
3. groundwater
4. underground defects
15.5.1 Volume Change of Soil

In general, the foundation must be placed below the zone of volume change. Highly plastic
clays (called expansive soils) shrink significantly upon drying and swell significantly upon
wetting. This volume change is greatest near the surface and decreases with increase in
depth. So, during weather changes, a certain depth of the soil undergoes a volume change.
This zone is called the zone of volume change.
Similarly, in areas where the air temperature falls below the freezing point, the mois-
ture near the soil surface may freeze. The frozen moisture may melt due to increase in
temperature. As the soil moisture freezes and melts, it alternately expands and contracts
leading to volume changes of the soil.
Repeated expansion and shrinkage due to soil type or temperature change may cause the
foundation to lift and drop. Such a sequence cannot be acceptable for the stability of a structure.
The IS code (IS: 1904, 1986) recommends that a foundation should be located at a mini-
mum depth of 50 cm below the natural ground surface.
For expansive soils (black cotton soils of India) the zone of volume change varies from
1.5 to 3.5 m.
In order to avoid the above expansion and shrinkage, it is advisable to place the footing
below the zone of volume change.
15.5.2 Adjacent Structures

The horizontal location of a footing is often affected by adjacent structures and property
lines. The construction of a new structure may damage the existing adjacent structure by
vibration and shock due to blasting, caving in due to nearby excavation, lowering of the
water table or increasing the stress.
The Indian Standards code (IS: 1904, 1986) recommends the following for footings placed
adjacent to a sloping ground or when the bases of footings are at different levels.
When the ground surface slopes downwards adjacent to a footing, the sloping surface
should not encroach upon a frustum of bearing material under the footing, as shown in
Fig. 15.2(a) and (b) for granular soils and clayey soils, respectively.
In order to avoid damage to an existing structure, the following norms may be followed
(as shown in Fig.15.2c):
M15_PURU1773_01_SE_C15.indd 571 2/12/2013 8:48:08 AM

G.S.
G.S.
Upper 1
Upper 1 footing 2
footing 2
Lower footing Lower footing

Slope of joining Slope of joining
line not steeper line not steeper
than two horizontal than two horizontal
to one vertical to one vertical
(a) (b)
S
G.S.
Old footing
30° B1
45°
S is larger of B1 and B2
B2
New footing on average soil
New footing on poor soil

B2
(c)
Fig. 15.2 Footings at different levels for (a) granular soil, (b) clay soil, and (c) footings for old and
new structures (Source: IS: 1904, 1980)
1. The footing should be placed at least at a distance S from the edge of the existing footing
where S is the width of the larger footing.
2. The line from the edge of the new footing to the edge of the existing footing should make
an angle of 45° or less.
3. When a new footing is placed lower than an old footing, the excavation for the founda-
tion must be carefully done with a suitable bracing system so as to prevent damage to the
existing structure.
Special care must be taken in placing a footing at or near a property line, so as to avoid
encroachment of the footing into the adjacent property.
M15_PURU1773_01_SE_C15.indd 572 2/12/2013 8:48:09 AM

15.5.3 Groundwater
For several reasons, the presence of groundwater within the soil immediately around a
footing is not desirable. The following points need consideration:
1. Construction of a footing below the groundwater level is difficult and expensive because
the area must be dried prior to construction.
2. Existence of groundwater around a footing may reduce the bearing capacity of the soil,
particularly in sands.
3. Excess groundwater around a footing may cause hydrostatic uplift problems.
4. In areas of sub-zero temperatures, frost action may predominate.
5. Existence of groundwater below a floor may add to waterproofing problems.
For the above reasons, as far as possible, footings should be placed above the groundwater
level.
15.5.4 Underground Defects

The presence of underground defects may also affect the location of a footing. The under-
ground defects may be faults, caves, mines, and man-made discontinuities, such as sewer
lines, underground cables, and utilities. Construction of structures on or near tectonic faults
should be avoided. Further, foundations should not be placed directly on caves mines or on
man-made discontinuities.
15.6 CAUSES OF SETTLEMENT

Foundation settlement may occur due to the following reasons:
1. Elastic compression of the foundation and the underlying soil, also called immediate
settlement, may be one cause. It is computed by idealizing the soil as an elastic material
(dealt with in Section 15.7.1 in detail).
2. Plastic or inelastic compression of the underlying soil, called time-dependent settlement
or settlement due to consolidation (both primary and secondary), which was dealt with
in Chapter 8, could be another cause.
3. Groundwater lowering is another major cause for settlement to occur. Repeated raising
and lowering of groundwater, particularly in granular soils, tends to reduce the void vol-
ume and causes settlement of the ground surface. Prolonged lowering of water table may
cause settlement in fine-grained soils. Pumping of water or draining of water without
proper filter material may also cause settlement.
4. Vibrations caused by pile driving, machinery, blasting, etc. may cause settlement, particu-
larly in granular soils.
5. Other causes of settlement include volume change of soil, ground movement and excava-
tion for adjacent structures, mining subsidence, etc.
15.7 SETTLEMENT OF SHALLOW FOUNDATIONS

Analytical methods are available for computing the settlement of shallow foundations under
a symmetrical static vertical load only. Settlement due to other causes such as deterioration
M15_PURU1773_01_SE_C15.indd 573 2/12/2013 8:48:09 AM

of the foundation, mine subsidence, and catastrophic settlement are not dealt with. The
methods of estimating immediate settlement are discussed below.
15.7.1 Immediate Settlement

All highly permeable soils, including all non-cohesive soils, undergo immediate settlement.
Immediate settlement also occurs in fine-grained soils. It is computationally convenient to
idealize the soil as an elastic material, and the results from the mathematical theory of elas-
ticity can be applied with full confidence to compute the settlement (Lambe and Whitman,
1979). It is generally accepted that immediate settlement predominates in non-cohesive soils
and the estimation of settlement based on elastic theory is quite appropriate. In saturated
clays, Leonards (1962) attributed the immediate settlement to shear strains caused by shear
stresses. Further, as the shear stresses are small, the immediate settlement may be computed
assuming that the soil mass behaves like an elastic solid (Leonards, 1962). Thus, immediate
settlement, in metres, is calculated from
1 − v2
Si = qB If (15.1)
Eu
where q is the intensity of contact pressure (kN/m2), B the least lateral dimension of the
loaded area (m), Eu the undrained modulus of elasticity (kN/m2), and If the influence factor,
which depends on rigidity and the shape of the loaded area (Table 15.1).
Equation 15.1 is basically given for a surface flexible loaded area. The same can be used
for rigid footings by modifying the influence factor. As per IS: 8009 – Part 1 (1976), the total
settlement of a rigid footing is taken to be 0.8 times the settlement at the centre of the flexible
foundation. Hence, the influence factors for a rigid foundation are also given in Table 15.1.
Equation 15.1 is further based on the assumption that the elastic medium is a semi-
infinite mass. For a compressible stratum of finite thickness (Ht), Steinbrenner’s influence
factor for settlement at the corner of the loaded area is recommended by Indian Standards
(IS: 8009 – Part 1, 1976) (Fig. 15.3).
Table 15.1 Influence factors for vertical displacement due to

elastic compression
Shape Flexible Rigid
Centre Corner Average
Circle 1.00 0.64 0.85 0.80
Rectangle
1.0 1.12 0.56 0.95 0.90
1.5 1.36 0.68 1.20 1.09
2.0 1.53 0.77 1.31 1.22
5.0 2.10 1.05 1.83 1.68
10.0 2.52 1.26 2.25 2.02
100.0 3.38 1.69 2.96 2.70
Source: IS: 8009 – Part 1 (1976).
M15_PURU1773_01_SE_C15.indd 574 2/12/2013 8:48:09 AM

2
L=
Ht
B
5
B
Depth factor
4
L=
10
B
6 L=∝
L = 1 B
L=
B 2
8 B
0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
Values of lf
Fig. 15.3 Steinbrenner’s chart for the influence factor (Source: IS: 8009 – Part 1, 1976)
The above discussion has centred on foundations located at the ground surface.
For foundations located at a certain depth, a depth factor correction has been suggested
(IS: 8009 – Part 1, 1976). The depth factor can be read from Fig. 15.4 for different L/B ratios.
Hence,
0 L/B = 100
GL
25
0.2
D
0.4
D B×L 9
LB
0.6
1
0.8
1.0
0.8
1
LB 0.6
D 9
0.4
0.2 25
100
0
0.5 0.6 0.7 0.8 0.9 1.0
Depth factor
Fig. 15.4 Fox’s correction curves for settlements of flexible rectangular footings (Source: IS: 8009 –
Part 1, 1976)
M15_PURU1773_01_SE_C15.indd 575 2/12/2013 8:48:10 AM

Corrected settlement = Si × depth factor (15.2)

If the compressible layer is of a thickness less than twice the breadth, it is observed that
the settlement is overestimated. Janbu et al. (1956) considered the depth as well as the finite
compressible layer and derived an expression to compute the average immediate settlement
under a flexible foundation as
μ0 μ1 qB (1 − v 2 )
Si = (15.3)
Eu
The values of μ0 and μ1 are given in Fig. 15.5. In the case of a thin layer (of thickness
Ht) existing immediately below the foundation (Fig. 15.6), we obtain a value of μ1(t) cor-
responding to the thickness Ht and similarly find a value of μ1(b) corresponding to the
thickness Hb. Then, the immediate settlement due to the thin layer is computed by taking
μ1 = μ1( b) − μ1( t ) .
Improved relationship for estimating the immediate settlement of shallow foundation
has been presented by Mayne and Poulos (1999). This improved relationship considers the
following:
1. the rigidity of the foundation,
2. the depth of embedment of the foundation,
3. the increase in the modulus of elasticity of the soil with depth, and
4. the location of rigid layer at limited depth.
Figure 15.4a shows the improved relationship for the immediate settlement. Hence, Be is
an equivalent diameter of a rectangular foundation, which is as follows:
4BL .
Be =
π
Be
q
Df
df Ef E0 Es
Compressible soil
Es , μs Es = E0 + kz
h
Rigid layer Depth Z
Fig. 15.4a Improved relationship for immediate settlement
M15_PURU1773_01_SE_C15.indd 576 2/12/2013 8:48:11 AM

In the above equation, B and L are the width and length of the foundation. For a circular
foundation, Be is equal to the diameter B of the foundation.
Hence Ef is the modulus of elasticity of the foundation, df is the thickness of the founda-
tion, and Df is the depth of the foundation below the ground surface. A rigid layer is located
at a depth h below the bottom of the foundation.
Then, modulus of elasticity of the compressible soil, Es, is given as:
Es = Eo + kz,
where, Eo is the modulus of elasticity at the base of the foundation.

z = depth.
k = slope.
Now, the improved expression for the immediate settlement is given as follows:
⎛ 1 − v 2 ⎞⎟
(Si )i = q Be ⎜⎜⎜ s ⎟
⎟I I I ,
⎜⎝ Eo ⎟⎠ G F E
where, IG = influence factor for the variation of Es with the depth, which is a function of
Eo, k, Be, and h.
IF = foundation rigidity correction factor.
IE = foundation embedded correction factor.
Figure 15.4b shows the variation of IG with β = Eo/k Be and h/Be.
The foundation rigidity correction factor IF is given as:
1.0
>30
10.0
5.0
2.0
0.8
1.0
0.6
IG
0.5
0.4
h/Be = 0.2
0.2
0
0.01 2 4 6 8 0.1 1 10 100
E0
b=
kBe
Fig. 15.4b Variation of IG with β (Source: Mayne and Poulos, 1999)
M15_PURU1773_01_SE_C15.indd 577 2/12/2013 8:48:11 AM

1.0
0.95
0.9
0.85
IF
Ef 3
0.8
KF =
( E0 +
Be
2
k
) ( ) 2t
Be
0.75 = Flexibility factor
0.7
0.001 2 4 6 8 0.01 0.1 1.0 10.0 100.0
Kf
Fig. 15.4c Variation of the rigidity correction factor IF with the flexibility factor Kf (Source:
Mayne and Poulos, 1999)
π 1
IF = +
4 ⎛ ⎞⎟
⎜⎜ ⎟⎟⎛ 2d ⎞3
⎜ Ef ⎟⎟⎜⎜ f ⎟⎟
4.6 + 10 ⎜⎜ ⎟⎟⎜ ⎟⎟
⎜⎜ Be
⎜⎝ Eo + k ⎟⎟⎝ Be ⎠
2 ⎠
Figure 15.4c presents the IF as a function of kf that is given as:

⎛ ⎞⎟
⎜⎜ ⎟⎟⎛ 2d ⎞3
⎜ Ef ⎟⎟⎜⎜ f ⎟⎟
kF = ⎜⎜ ⎟⎟⎜⎜ ⎟⎟
⎜⎜ Be
⎜⎝ Eo + k ⎟⎟⎟⎝ Be ⎠
2 ⎠
Similarly, the embedded correction factor IE is given as:
1
IE = 1 −
⎛B ⎞
3.5exp (1.22vs − 0.4)⎜⎜⎜ e + 1.6⎟⎟⎟
⎝ Df ⎟⎠
Figure 15.4d shows the variation of IE with Df /Be.
M15_PURU1773_01_SE_C15.indd 578 2/12/2013 8:48:12 AM

1.0
0.95
0.9
us = 0.5
0.85 0.4
IE
0.3
0.2
0.8
0.1
0.75 0
0.7
0 5 10 15 20
Df
Be
Fig. 15.4d Variation of the embedded correction factor IE (Source: Mayne and poulos, 1999)
15.7.2 Consolidation Settlement

As discussed in Chapter 8, settlements due to primary and secondary compression are
important in fine-grained soils. Evaluation of settlement based on secondary compression
has not yet been standardized. The methods of computing primary compression (already
discussed) need certain modifications depending on the field situations. The usual field
situation is that the clay layer is sandwiched between cohesionless soil layers or between a
cohesionless soil layer at the top and rock at the bottom. These are conventional situations
for which settlement equations, discussed in Chapter 8, can be used.
One major limitation of the estimation of consolidation settlement (based on oedometer
results) is that the laboratory condition is one-dimensional whereas the actual field condi-
tion is different. Under one-dimensional conditions, the lateral strain is zero and the increase
in pore water pressure is equal to the increase in total stress, i.e., A = 1.0. The field condition
required for A = 1 is that the extent of the loaded area should be large compared with the
thickness of the layer.
Under such conditions, the lateral strain is not zero and the total settlement comprises
1. immediate settlement for the undrained condition and
2. consolidation settlement due to drainage.
Skempton and Bjerrum (1957) have considered the thickness of the compressible layer,
width of the loaded area, and pore pressure parameter A and suggested a settlement
coefficient μc to obtain the corrected consolidation settlement (Sc′).
M15_PURU1773_01_SE_C15.indd 579 2/12/2013 8:48:12 AM

1.0
0.9
μ0 0.8
200
0.7
L/B 1.2 510 20 50
100
0.6
0.5
1 0.2 0.5 1 2 5 10 25 50 100 1000
D/B Log scale
3.0
q 100
2.5 D 50
L/B = ∝
H B 20
2.0 L-Length
μ1μ0
qB 10
Si (1–n 2)
μ1 1.5 E(u) 5
1.0 2
Square
1
0.5
Circle
0.0
0.1 0.2 0.5 1 2 5 10 25 50 100 1000
Log scale
Fig. 15.5 Coefficients for immediate settlement under a flexible foundation (Source: Janbu et al.,
1956)
Uniform-load = q
Ht B
Hb
Fig. 15.6 Thin layer below foundation
Sc′ = μc Sc
where μc is a factor related to the pore pressure parameter A and the ratio Ht/B.
This settlement coefficient chart (Fig. 15.7) suggested by Skempton and Bjerrum has been
adopted in the Indian Standards (IS: 8009 – Part 1, 1976). In the absence of data for the
parameter A, μc values from Table 15.2 may be used.
15.7.3 Evaluation of Settlement from Field Tests

There is much difficulty in the sampling of cohesionless soils and no standard test procedure
is available for finding the compressibility characteristics of cohesionless soils. In such soils,
the settlement may be estimated based on the data from in situ bearing tests such as the static
cone penetration test, standard penetration test, and plate bearing test.
From a static cone penetration test (IS: 4968 – Part 3, 1976), a curve showing the depth
and static cone penetration resistance (qcs) is drawn (Fig. 15.8). The entire profile is divided
M15_PURU1773_01_SE_C15.indd 580 2/12/2013 8:48:13 AM

1.2
Hc
Values on curves are
B
Settlement coefficient μm
1.0
0.25
0.8
0.25 B
1.0
0.5
0.6 4
0.5 Clay
Layer
Hc
0.4 1.0 Very

4 Over Normally sensitive
consolidated consolidated clay
0.2
0 0.2 0.4 0.6 0.8 1.0 1.2
Pore pressure coefficient, A
Fig. 15.7 Settlement coefficients for circular and strip footings (Source: IS: 8009 – Part 1, 1976)
Table 15.2 Values of μc
Type of clay μc
Very sensitive clays (soft alluvial, estuarine, and marine clays) 1.0–1.2
Normally consolidated clays 0.7–1.0
Over-consolidated clays 0.5–0.7
Heavily over-consolidated clays 0.2–0.5
into strata or layers with constant qcs for each layer. The settlement for each layer due to the
foundation load is calculated from Eq. 15.4. Then, the settlements corresponding to all the
layers are added to get the total settlement (S)
Ht ⎡ p + Δp ⎤
S cs = 2.303 log10 ⎢ 0 ⎥ (15.4)
C ⎢ p ⎥
⎣ 0 ⎦
where
3 qcs
C=
2 p0
Ht is the thickness of each layer and p0 the initial effective pressure at mid-height of the layer.
Hence, S = ∑Scs.
Indian Standards (IS: 8009 – Part 1, 1976) also provides a chart for dry cohesionless soils
relating settlement of a footing of width B under unit intensity of pressure for different N
values (Fig. 15.9). For a given pressure on the footings, the settlement is proportional to
the intensity of pressure. If the water table is located at a shallow depth from beneath the
footing base the correction factor Rw is found and the settlement obtained from Fig. 15.9 is
multiplied by Rw.
M15_PURU1773_01_SE_C15.indd 581 2/12/2013 8:48:14 AM

Cone resistance
C kd
Layer l
Depth below ground level

Layer Il
Layer IIl
Layer IV
Actual cone reading

Average cone resistance
in each layer
Fig. 15.8 Static cone penetration resistance diagram (Source: IS: 8009 – Part 1, 1976)
Settlement (metre per unit pressure) (1 kg/cm2)
N=5
10–1
N = 10
N = 15
N = 20
N = 25
10–2 N = 30
N = 40
N = 50
N = 60
10–3
0 1 2 3 4 5 6
Width 'B' of footing m
Fig. 15.9 Settlement per unit pressure from standard penetration resistance (Source: IS: 8009 – Part
1, 1976)
The settlement can also be estimated from the plate load test (IS: 8009 – Part 1, 1976)
data. The test details are explained in Chapter 14. The plate load test (IS: 1888, 1982) is con-
ducted at the required depth and the settlement (St) of the proposed foundation is found
from Eq. 15.5.
⎡ Bf (Bp + 0.3) ⎤ 2
St = Sp ⎢⎢ ⎥
⎥ (15.5)
⎢⎣ Bp (Bf + 0.3) ⎥⎦
The water table correction factor can also be applied, if necessary, as explained in the
previous paragraph.
M15_PURU1773_01_SE_C15.indd 582 2/12/2013 8:48:15 AM

15.7.4 Reliability of Settlement Computations

In general, the allowable bearing pressure in cohesionless soils is governed by settlement
criteria rather than shear. Further, in such soils, the immediate settlement predominates with
some creep effects. In sands, most of the settlement occurs during the initial stage of con-
struction of the building due to the combination of construction vibrations and loads. The
settlement predictions for cohesionless soils are based on conservative methods (Bowles,
1984). But there are many factors which affect the settlement of footings in sand, viz., relative
density, amount of fines and gradation of sand, the size of the loaded area, the position of the
water table, in situ stresses, and capillarity. A comprehensive critical review of the evaluation
of settlement in sand was presented by Ramasamy (1984) and was modified by him later
(Ramasamy, 1986). This modified method considers, among other routine factors, the effect
of past loading history, non-linear load settlement behaviour of footings, and the effect of
fines in sand. Further, the effects of embedment (e.g., Kaniraj, 1977; Ramasamy et al., 1982),
capillary zone (Ramasamy et al., 1986), bearing area and plan dimension (Kaniraj, 1977), and
stiffness of footing (Arora and Varadarajan, 1984) on settlement in sand have been reported
in the literature.
In other soils, depending on the percentage clay fraction and organic matter, all the three
components of settlement may be present. Consolidation settlement may predominate in
inorganic silts and clays and in highly organic soils it could be secondary compression.
Consolidation settlement can be reasonably predicted provided test results from undis-
turbed samples are available (Mac Donald and Skempton, 1955; Skempton, 1955). The effect
of depth of embedment change in bearing area and plan dimensions on the settlement of
normally consolidated clay has been studied by Kaniraj and Ranganathan (1977). It has been
shown that there is a decrease in settlement with increase in depth of embedment, bearing
area, and length of footing.
15.8 DESIGN STEPS FOR A SHALLOW FOUNDATION

Generally, a footing may have to carry a load-bearing wall or a single column or more than
one column to support the structure. Columns generally carry different loads depending on
the location and based on the type of structure.
Each column has to carry a different type of load, and the major areas are dead and
live loads. Further, the entire amount of live load is not borne by the column for the entire
life period. Hence, it is generally considered a service load which will be applied on the
column during the entire lifespan, which is taken as dead load plus 50% of the live load
for ordinary buildings. A large percentage of live load should be used in warehouses and
other storage floors. The design of column footings based on service loads is generally
adequate.
Footings may be designed by adopting the following procedure (Teng, 1962):
1. Calculate the loads acting on the footing.
2. Obtain a soil profile or soil profiles showing the soil stratification at the site.
3. Establish the maximum water level.
4. Obtain the pertinent field and laboratory measurements and test results.
5. Determine the depth and location of the footing.
M15_PURU1773_01_SE_C15.indd 583 2/12/2013 8:48:15 AM

6. Determine the bearing capacity of the supporting stratum.

7. Proportion the footing sizes.
8. Check the footing contact pressure.
9. Check the stability of the footing against sliding, overturning, and uplift pressures.
10. Estimate the total and differential settlements.
11. Design the footing structure.
12. Assess the need for foundation drains, waterproofing or damp proofing.
15.9 PROPORTIONATING FOOTING SIZE

Footing sizes are basically designed for safe bearing and then checked for permissible total
and differential settlements. The size is modified if the permissible settlements are not satis-
fied. While proportionating, the following procedure may be followed (Teng, 1962):
1. Calculate the load on the structure:
Ll + d = live load + dead load
for the column which has the largest live load to dead load ratio.
2. Calculate the service load for the same column:
Ls = dead load + C (live load)
where C is ½ for ordinary buildings and ¾ for warehouses and storage floors.
3. Decide the storage floor’s safe bearing pressure:
Qa = safe bearing pressure from theories or field tests
4. Compute the design pressure:
Qd = design pressure
for all footings except the one with the lowest live load to dead load ratio.
qd = Lc / A
where Ls is the service load. That is,
service load
proportioned area of footing =
qd
5. Compute the area of footing supporting the column with the lowest live load to dead load
ratio:
A = (Ll+d )/ qs
6. Decide the length and width of footing and check for permissible settlement and alter
width if needed.
M15_PURU1773_01_SE_C15.indd 584 2/12/2013 8:48:15 AM

15.10 DESIGN OF COMBINED FOOTINGS

As discussed earlier, a combined footing is used when equal or unequal columns are posi-
tioned so closely that individual footings are not practicable. These footings are usually
rectangular in shape. In order to accommodate unequal column loads or columns close to
property lines, the rectangular shape is modified to a trapezoidal shape. Sometimes, a strap
is provided to combine two columns which have a wider spacing or if one of them is close
to a property line.
The conventional methods of design of combined footings are based on the following
considerations:
1. The footing is infinitely rigid and does not have any bearing on the pressure distribution.
2. The soil pressure beneath the footing is linearly distributed (or is distributed on a plane
surface) such that the centre of the soil pressure coincides with the line of action of the
resultant force of all the loads.
15.10.1 Rectangular Combined Footing

The following steps are adopted for the design of a rectangular combined footing (Fig. 15.10):
1. Find the total column loads which are to be positioned in the footing and determine the
line of action of the resultant.
2. Obtain the soil pressure distribution (stress per unit length of the footing).
3. Find the width, B, of the footing.
4. Draw the shear force diagrams along the length of the footing.
5. Draw the bending moment diagram along the length of the footing.
6. Design the footing as a continuous beam to withstand the shear and bending moment.
7. Design the footings for transverse bending also in the same manner as for spread
footings.
Then, the structural design details for the footing are worked out.
15.10.2 Trapezoidal Combined Footing

As discussed earlier, a trapezoidal combined footing is preferred (i) when two unequally
loaded columns are encountered and (ii) when the property line is quite close to the exterior
column. In the first case, the higher loaded column will have a larger width. The soil pres-
sure distribution is linear or uniformly varying (not uniform) as shown in Fig. 15.11.
The width of the trapezoid is determined from the following equations:
L
(B1 + B2 )qa = Pe + Pi
2
where qa is the allowable soil pressure. That is,

2(Pe + Pi )
B1 + B2 =
Lqa
M15_PURU1773_01_SE_C15.indd 585 2/12/2013 8:48:16 AM

C. G. of base
Property line
Plan
Pe R Pi
Section and loading
+
– –
S.F. diagram
+
+
B.M. diagram
Fig. 15.10 Rectangular combined footing
By taking moments about the property line or left edge and simplifying,
2B1 + B2 3⎡ 2Pi L ⎤⎥
= ⎢ e1 +
B1 + B2 ⎢
L⎣ Pe + Pi ⎥⎦
M15_PURU1773_01_SE_C15.indd 586 2/12/2013 8:48:16 AM

C.G. of base
(area A)
B2 B1
L'
Plan Pi
R Pe > Pi
Pe
e1
q1
q2
Section and loading
+
– –
S.F. diagram
+ +
B.M. diagram
Fig. 15.11 Trapezoidal combined footing
B1 and B2 may be obtained from the above two equations. That is,
2 A ⎛⎜ 3 x ⎞ 2A
B1 = ⎜⎜⎝ − 1⎟⎟⎟⎠ and B2 = − B1
L L L
M15_PURU1773_01_SE_C15.indd 587 2/12/2013 8:48:17 AM

The pressure distribution values q1 and q2 are calculated as

q1 = B1 qa and q2 = B2 qa
The other design procedure is the same as that for the rectangular combined footing.
15.10.3 Combined Strap Footing

A combined strap footing is designed based on the following assumptions:
1. The strap beam transfers the column loads onto the soil with equal and uniform soil pres-
sure under both the footings, and it acts as an infinitely stiff beam.
2. No load is transferred through the bottom of the strap, and it acts as a pure flexural member.
As a matter of fact, the bottom of the strap is made free standing without touching the soil.
The design procedure is as follows:
1. To start with, a trial value of e (as shown in Fig. 15.12) is assumed.
2. Reactions R1 and R2 are computed based on the principles of statics as
⎛ e ⎞⎟ Q1 e
R1 = Q1 ⎜⎜⎜1 + ⎟ and R2 = Q2 −
⎝ LR ⎟⎠ LR
where R is the distance between R1 and R2.
P1 P2
B1
B2
(a) Cantilever principle of strap footing
P1 P2
q q
R1 R2
e R
(b) Load–reaction details
Fig. 15.12 Strap footing
M15_PURU1773_01_SE_C15.indd 588 2/12/2013 8:48:17 AM

3. With a knowledge of the safe soil pressure, the tentative footing areas are computed.
4. Knowing the footing areas, e is calculated.
5. Steps 1 to 4 are repeated till the e value is identical to the final one.
6. The footings are designed as a simple spread footing subjected to uniform soil pressure.
7. Shear and bending moments are computed and the strap beam is designed.
Based on the above design procedure, the centre of gravity of the two footing areas will
coincide with the resultant of the column loads Q1 and Q2.
15.11 MAT FOUNDATION

A mat or raft foundation is a large footing, usually supporting several columns in two or
more rows. The choice between strap footing, combined footing, and raft foundation
depends on the soil type and the relative cost. In a well-bearing ground, strap footing may
be more economical than combined footing, whereas in a soil of low bearing capacity a large
strap footing may be less preferred due to the cost.
As a general rule, mat foundations are used where the soil has low bearing capacity. Since
the bearing capacity increases with increase in width and depth of the foundation, the mat
foundation gains a two-fold advantage.
The mat foundation, in general, is a flat concrete slab having uniform thickness through-
out the entire area. This is adopted where the column spacing is fairly small and uniform
and the column loads relatively small. However, for large column loads, the slab has to be
thicker.
15.11.1 Types of Mat Foundations

Some of the common types of mat foundations are shown in Fig. 15.13 (Das, 1984):
1. flat plate with a mat of uniform thickness (Fig. 15.13a);
2. flat plate thickened under columns (Fig. 15.13b);
3. beams and slab, wherein the beams run both ways and the columns are located at the
intersection of the beams (Fig. 15.13c);
4. slab with basement walls as a part of the mat, with walls providing stiffness for the mat
(Fig. 15.13d).
Mats may be placed directly on the soil or may be supported on piles. Piles are preferred
particularly when the height of groundwater is large.
15.11.2 Bearing Capacity of Mat Foundations

The gross ultimate bearing capacity (qug) of a mat foundation can be determined in the same
way as for a shallow foundation. The term B to be used in the general equation is the small-
est dimension of the mat. The net ultimate bearing capacity (qun) is
qun = qug − q
where q is the surcharge.
M15_PURU1773_01_SE_C15.indd 589 2/12/2013 8:48:17 AM

Section Section
at A – A at A – A
A A A A
Plan Plan
(a) (b)
Section
Section at A – A
at A – A
A A
A A
Plan Plan
(c) (d)
Fig. 15.13 Types of mat foundations (Source: Das, 1984)
M15_PURU1773_01_SE_C15.indd 590 2/12/2013 8:48:18 AM

A suitable factor of safety, varying from 1.75 to 3, is used. For mats on clay, the factor of
safety should not be less than 3 under dead load and maximum live load. Also, for mats on
sand, again, the factor of safety should not be less than 3. On no account should the factor
of safety be less than 1.75. In general, under normal working conditions, the factor of safety
against bearing capacity failure of mats on sand is very large. On granular deposits, the net
ultimate bearing capacity may be found based on Standard Penetration Resistance numbers.
Generally, the settlement permissible is about double that for spread footing, for the rea-
son that the depth of the zone of influence is likely to be much larger than that of the spread
footing. For this reason, the loose soil pockets under a mat may be evenly distributed, result-
ing in a smaller differential settlement. Generally, the permissible total settlement is 50 mm
and differential settlement is 19 mm.
15.11.3 Design Methods

The design of mat foundations may be done using any one of the following three methods:
conventional (rigid) method, simplified elastic (flexible) method, and truly elastic founda-
tion method.
The conventional method of design assumes that (i) the mat is infinitely rigid; i.e., the
deflection of the mat does not influence the pressure distribution and (ii) the soil pressure is
a plane surface such that the centroid of the soil pressure coincides with the line of action of
the resultant force of all the loads acting on the foundation.
The simplified flexible method assumes that the soil behaves like an infinite number of
springs where none of the springs is affected by the others. The elastic constant of the springs
is equal to the coefficient of sub-grade reaction, which is defined as the unit pressure required
to produce a unit settlement. A number of procedures have been developed for the analysis
of beams based on the simplified elastic foundation concept.
In the truly elastic foundation method, the soil is assumed to be a truly elastic mate-
rial obeying Hooke’s law in all directions. This method is very rarely used because of its
complexity.
The conventional method of design is similar to that of the design of combined footings.
WORKED EXAMPLES
Example 15.1 The circular foundation of a ground-level oil tank of 20 m diameter, trans-
mits to the soil a uniform contact pressure of 250 kN/m2 at a 3 m depth. Determine the
immediate settlement under the centre of the foundation. The properties of the soil are as
follows: Eu = 60 MN/m2, v = 0.45, and γ = 22 kN/m3.
Solution
As the foundation is for an oil tank, consider it to be a flexible one. From Table 15.1 for circu-
lar flexible footings, the value of If at the centre is 1.0
qB(1 − v 2 )
Si = If
Eu
M15_PURU1773_01_SE_C15.indd 591 2/12/2013 8:48:18 AM

Net contact pressure q = 250 – (22 × 3) = 184 kN/m2. Therefore,

184 × 20(1 − 0.452 )×1×10 3
Si = = 48.9 mm
60 ×10 3
Example 15.2 A rectangular footing of dimensions 4 m × 2 m is founded at a depth of 2 m

on a soil with Eu = 48 MN/m2 and ν = 0.50. A rigid layer is laid under the soil at a depth
of 10 m from the ground surface. The foundation transmits a uniform contact pressure of
200 kN/m2. Estimate the average immediate settlement likely to occur.
Solution
Refer to Fig. 15.5
D 2 L 4
= =1 and = =2
B 2 B 2
For
D L
=1 and = 2, μ0 = 0.78
B B
H 10
and = =5
B 2
For H/B = 5 and L/B = 2.0, μ1 = 0.84. Therefore,
qB
Si = μ0 μ1 (1 − υ 2 )
Eu
0.78 × 0.84 × 200 × 2(1 − 0.52 )10 3
= = 4.1 mm
48 ×10 3
Example 15.3 A flexible rectangular foundation of dimensions 3 m × 1.5 m is placed at a

depth of 2 m saturated in a clayey soil of infinite depth. The undrained modulus of clay,
Eu = 45 MN/m2, ν = 0.42, and the unit weight of clay γ = 19.8 kN/m3. The foundation trans-
mits uniform contact pressure of 230 kN/m2. Determine the average immediate settlement
expected to occur under the foundation. If a hard stratum exists below the clay stratum at a
depth of 6 m, what will be the change in the settlement?
Solution
From Eq. 15.1 the immediate settlement for flexible footing on an saturated clay of infinite
depth is given as
1− ν 2
Si = qB If
Eu
(1 − 0.422 )
= 230 ×1.5× 3
×1.20 ×10 3 = 7.6 mm
45×10
From Eq. 15.3, the immediate settlement for flexible footing on saturated clay underlain by
a hard stratum is given as
qB(1 − ν 2 )
Si = μ0 μ1
Eu
M15_PURU1773_01_SE_C15.indd 592 2/12/2013 8:48:19 AM

From Eq. 15.3 for L/B = 2 and D/B = 2/1.5 = 1.33, μ0 = 0.85. For L/B = 2 and H/B = 6/1.5 = 4,
μ1 = 0.80. Therefore,
1.5(1 − 0.422 )
Si = 0.85× 0.80 × 230 × = 4.03 mm
45×10 3
Example 15.4 Settlement of a square footing of dimensions 1.2 m × 1.2 m carrying a load
of 220 kN/m2 is 30 mm. What would be the settlement of a footing measuring 3 m × 3 m
carrying a load of 160 kN/m2 ? The subsoil conditions are identical in both the footings.
Solution
Consider the immediate settlement equation
⎛ 1 − ν 2 ⎞⎟
Si = qB ⎜⎜⎜ ⎟I
⎜⎝ Eu ⎟⎟⎠ f
For a given soil,

Si = qBI f X
⎛ 1 − ν 2 ⎞⎟
where X = ⎜⎜⎜ ⎟
⎜⎝ Eu ⎟⎟⎠
For the 1.2 m × 1.2 m footing, 30 = 220 × 1.2 × 0.62(X).
⎛ 30 ⎞⎟
Therefore, X = ⎜⎜⎜ ⎟
⎝ 220 ×1.2× 0.62 ⎟⎠
For the 3 m × 3 m footing,
⎛ 30 ⎞⎟
Si = 160 × 3 × 0.84 ⎜⎜⎜ ⎟ = 73.9 mm
⎝ 220 ×1.2× 0.62 ⎟⎠
Example 15.5 A combined footing has to be proportioned for the two columns detailed
below.
Column loads Column C1 Column C2

Dead load 550 kN 700 kN
Live load 400 kN 800 kN
The distance between the columns is 5.0 m. The footing should not be beyond 0.5 m from
the face of the column.
The soil pressures to be considered are:
For dead load (DL) + reduced live load (LL) = 160 kN/m2
For dead load + live load = 230 kN/m2
M15_PURU1773_01_SE_C15.indd 593 2/12/2013 8:48:20 AM

Solution
Total column loads Column C1 (kN) Column C2 (kN) Total (kN)

DL + reduced LL 550 + 200 = 750 700 + 400 = 1100 1860
DL + LL 550 + 400 = 950 700 + 800 = 1500 2450
1. Uniform soil pressure under DL + 50% LL condition.

(1100 × 5.0)
Let x= = 2.94 m
1860
Therefore,
Length L = 2(2.94 + 0.50) = 6.88 m say 7.00 m
1860
Width B = = 1.66 m say 1.70 m
160 ×7.00
2. Uniform soil pressure under DL + LL condition
Let y be the distance of a resultant from column C1
1500 × 5.0
y= = 3.06 m
2450
From column C1 to the centre of gravity of footing is the eccentricity, e = 3.06 − 3.00 =
0.06 m
2450 ⎛⎜ 6 × 0.06 ⎞⎟ 2 2
qmax = ⎜⎜⎝1 + ⎟⎟ = 216.5 kN/m < 230 kN/m
7.0 ×1.70 7.0 ⎠
2450 ⎛⎜ 6 × 0.06 ⎞⎟ 2
qmin = ⎜1 − ⎟ = 195.3 kN/m
7.0 ×1.70 ⎜⎝ 7.0 ⎟⎠
The size of the footing is 7.0 × 1.70 mm.
Example 15.6 Proportion a trapezoidal combined footing for two columns 300 mm × 300 mm
carrying column loads of 800 kN and 1200 kN if the spacing between the columns is 4.5 m. Take
the allowable soil pressure as 250 kN/m2 and the length of the footing as 5 m.
Solution
Let A and B be columns of loads 800 kN and 1,200 kN, respectively.
Length of the footing L = 5 m
Let the projection of footing beyond the column face be 0.5 m
Then, distance from centre to centre of the column, L′ = 5 – 2 × 0.5 = 4 m.
Total area required
800 + 1200
A= = 8 m2
250
Let the distance of column loads from column A be

1200 × 4
x1 = = 2.4 m
2000
M15_PURU1773_01_SE_C15.indd 594 2/12/2013 8:48:21 AM

Let x be the distance of the resultant column from the left edge = 2.4 + 0.5 = 2.9 m.
Let B1 be the longer width and B2 be the shorter width. B1 near to load of 1,200 kN and B2
near to 800 kN. Therefore,
2 A ⎛⎜ 3 x ⎞ 2× 8 ⎛ 3 × 2.9 ⎞⎟
B1 = ⎜ − 1⎟⎟⎟⎠ = ⎜⎜ − 1⎟⎟ = 1.73 m
L ⎜⎝ L 5 ⎜⎝ 5 ⎠
2A 2× 8
B2 = − B1 = − 1.73 = 3.2 − 1.73 = 1.47 m
L 5
Total area provided is

1.73 + 1.47
×5 = 8 m2
2
The trapezoidal footing will have widths of 1.73 m and 1.47 m with a length of 5 m and
with the centre to centre of columns as 4 m.
POINTS TO REMEMBER
15.1 Shallow foundations are those placed on a firm soil near the ground and beneath the
lowest part of the superstructure.
15.2 A shallow foundation for a given loading system must meet three design require-
ments, regarding (i) foundation placement, (ii) safety against bearing capacity failure,
and (iii) safety against permissible settlement.
15.3 The types of shallow foundations are spread footings, combined footings, continuous
footings, and mat foundations.
15.4 Settlement of a shallow foundation could be of two types: immediate settlement and
consolidation settlement. Immediate settlement is computed by idealizing the soil as
an elastic material and using the results from the mathematical theory of elasticity.
Consolidation settlement is due to drainage.
15.5 Combined footings are used for combining two or more columns into one footing.
If the loading on the columns is not very much different, a rectangular footing may be
used. If the difference in column loads is more, a trapezoidal footing is preferred.
15.6 The pressure transmitted from the base of a foundation to the soil is termed the contact
pressure. This depends on the rigidity of the foundation structure and the nature of
the soil.
QUESTIONS
Objective Questions
15.1 Ultimate settlement of footings on cohesive soils is best estimated from the data from
(a) Plate load test
(b) Consolidation test
M15_PURU1773_01_SE_C15.indd 595 2/12/2013 8:48:22 AM

(c) Cone penetration test

(d) Standard penetration test
15.2 The permissible settlement of a shallow foundation on rock for evaluation of safe
bearing pressure from a plate load test is taken as
(a) 12 mm (b) 25 mm
(c) 40 mm (d) 50 mm
15.3 Identify the incorrect statement. Non-uniform settlement can result from
(a) Non-uniform bearing stress
(b) Non-homogeneous subsoil conditions
(c) Non-uniform stress acting upon a homogeneous soil
(d) Variation of water required at the construction site
15.4 Identify the incorrect statement. Settlement of a structure can be important for the
following reasons:
(a) Imminent rupture of the structure
(b) Appearance of the structure
(c) Utility of the structure
(d) Damage to the structure
15.5 The dependence of the settlement of a footing in sand on the width of the footing is
(a) Directly proportional
(b) Indirectly proportional
(c) Logarithmically proportional
15.6 The major problem with settlement analysis is
(1) obtaining reliable values of the elastic parameters;
(2) obtaining a reliable stress profile from the applied load.
Of these statements
(a) 1 alone is correct
(b) 2 alone is correct
(d) None of them is correct
15.7 The vertical displacement of a rigid loaded area under a pressure q is taken to be
______ times the displacement at the centre of the flexible area. Choose the correct
factor from those listed below:
(a) 1.4 (b) 1.2
(c) 0.80 (d) 0.60
15.8 The influence factor for vertical displacement at the centre of a rectangular area varies
from
(a) 1.52 to 2.10
(b) 1.12 to 1.52
(c) 1.12 to 1.00
(d) 1.00 to 0.76
15.9 The influence factor for vertical displacement at the centre of a flexible circular area is
(a) 1.0 (b) 1.5 (c) 2.0 (d) 0.80
M15_PURU1773_01_SE_C15.indd 596 2/12/2013 8:48:22 AM

15.10 Indicate the circumstances under which combined footings are adopted.
15.11 What precautions are to be taken while locating a footing (i) on a slope and (ii) adjacent
to an existing structure?
15.12 What are the different types of settlements which are to be considered in the design of
a shallow foundation?
EXERCISE PROBLEMS
15.1 Determine the dimensions of a rectangular combined footing for the data given below:
Column A (kN) Column B (kN)

Dead load 480 680
Live load 350 450
Average allowable soil pressure is 250 kN/m2. Distance between centre to centre of
column is 4.6 m. The projection beyond column A should not exceed 0.4 m.
15.2 It is decided to provide a strap footing for two columns A and B as detailed below:
Column loads: load on A = 1,500 kN; load on B = 1,450 kN.
Size of column = 0.5 m.
Centre to centre of columns = 5.8 m.
Allowable soil pressure = 370 kN/m2.
Determine the size of the footing for columns A and B.
15.3 Column loads on columns A and B are 1,920 kN and 1,500 kN, respectively. Column
B is a boundary column. Proportion a trapezoidal footing. The allowable soil pressure
is 200 kPa.
M15_PURU1773_01_SE_C15.indd 597 2/12/2013 8:48:22 AM

M15_PURU1773_01_SE_C15.indd 598 2/12/2013 8:48:22 AM
16
Pile Foundations
CHAPTER HIGHLIGHTS
Classification of piles: material composition, installation methods, ground
effects, function as foundation – Pile-driving equipment – Bearing capacity of
single pile: statistical methods, pile-driving formulae, wave equation, based on
SPT values, pile load test – Negative skin friction – Under reamed piles – Pile
groups: group capacity, group in filled ground – Group settlement – Pile cap
16.1 INTRODUCTION
The design and construction of deep foundations for transferring the weight of the
superstructure through soft or weak soils, to deep load-bearing strata is a challenging
job for a civil engineer. Piles, piers, and caissons are the most common types of deep
foundations. The mechanism for deriving support from the soil or rock below and adja-
cent to the foundations is similar for any system. Each system differs in its method of
construction. Piles are slender structural members normally installed by driving with
hammer or by vibrating, and occasionally by auguering. Pre-drilling or other proce-
dures may be necessary to permit penetration to the desired depth.
16.2 CLASSIFICATION OF PILES

Piles can be classified according to their material composition, installation method,
ground effect, and their function as a foundation.
16.2.1 Material Composition

Based on material composition, the piles may be further classified as timber, steel,
concrete, or composite piles. Timber piles are the oldest types of piles made from tree
M16_PURU1773_01_SE_C16.indd 599 2/12/2013 5:15:09 PM

Weld
Butt diameter
300-500 mm
Pile may be treated
with wood preservation
Splicing by welding
Cross-section
Tip diameter
150–250 mm Splicing by welding
Cross- Splicing by
section riveting
(a) Timber pile (b) Steel pile
300–600 f m
Cased
or
uncased
D concrete
Circular
300–600 mm
Timber
Steel
pipe
concrete
2D filled
Square
cross-section
(c) Concrete pile (d) Composite pile
Fig. 16.1 Types of piles
trunks that have had their branches carefully trimmed off. The maximum length of a pile is
about 20 m. The timber should be straight, sound and without any defects (Fig. 16.1a). Timber
piles are installed by driving. Overdriving of timber piles may result in splitting, crushing,
and/or shearing of piles. Timber piles have a long life if prevented from alternative wetting
and drying. A plain timber pile permanently below water will not decay. The life of timber
piles may be increased by treating them with preservatives. Preservative protection treat-
ment is required to protect timber piles from marine borers (if for a marine environment), or
from wood-infesting insects, such as termites, or from decay (wet rot) if the pile is embedded
in soil above water table. Timber piles find extensive use for compaction of soils, for support-
ing structures, and for protecting water-front structures. As per Indian Standards (IS: 2911,
Part 2, 1980), piles are classified as Class A or Class B depending on the use. Piles used for
railway and highway bridges, trestles, docks, and wharves are categorized as Class A. These
piles shall have butt diameter or the sides of square not less than 30 cm. Piles used for founda-
tion work and other temporary works are categorized as Class B. Such piles may have diam-
eters less than 30 cm. For compaction piles, usually a 10 cm diameter is recommended.
M16_PURU1773_01_SE_C16.indd 600 2/12/2013 5:15:10 PM

Pile Foundations 601
Steel piles generally used are either pipe piles or rolled steel H-section piles. Pipe piles
are normally filled with concrete. Steel piles are spliced by welding or riveting. The steel
piles may be provided with driving points or shoes – for driving through dense materials.
Because of the high strength of steel, these piles withstand driving pressures. Strength, rela-
tive ease of splicing, and economy are some of the advantages of steel piles (Fig. 16.1b). Steel
piles are affected by corrosive agents such as salt, acid, moisture, and oxygen. To account for
corrosion, an additional thickness of steel is generally recommended. Protection from corro-
sion is effected by applying epoxy coatings on the pile surface before driving, or by concrete
encasement in most corrosive zones of the steel pile.
Concrete piles are cast to specified lengths and shapes of circular, square, or octagonal
cross-sections with reinforcement. The reinforcement is provided to enable the pile to resist
the bending moment developed during lifting and transportation (Fig. 16.1c).
Piles can also be pre-stressed using high tensile steel cables. The cables are pre-tensioned
up to about 1,300 MN/m2 before pouring concrete around them. After curing, the cables are
cut, thus producing a compressive force on the pile section. Concrete piles are frequently
used in marine environment. These piles are limited to a length of 25 m and the diameter is
generally less than 0.5 m. Concrete piles may be of precast or in-place type. The installation
process of cast-in-place piles are discussed in the next section.
For cast in situ concrete piles the reinforcing cage depends on the installation condition, the
nature of the subsoil, and the nature of load transmission is 0.4% of the sectional area. The mini-
mum cover for main reinforcement should not be less than 50 mm. The lateral reinforcement
imparts adequate rigidity. The minimum diameter and spacing of links or spirals are 6 and
150 mm, respectively (IS: 2911, Sections 1 and 3, 1979). The area of main reinforcement for pre-
cast piles shall not be less than the following percentages of the cross-sectional area of the piles:
1. Pile length < 30 times the least width: 1.25%
2. Pile length 30 to 40 times the least width: 1.5%
3. Pile length > 40 times the least width: 2%
The lateral reinforcement resists the driving stresses induced in the piles and should be in
the form of hoops or links of diameter not less than 6 mm. The volume of lateral reinforce-
ment shall not be less than the following:
1. At each end of the pile for a distance of about three times the least width – not less than
0.6% of the gross volume of piles and
2. In the body of the pile – not less than 0.2% of the gross volume of piles.
Close spacing is provided near the ends and the maximum spacing over a length of three
times the least width of the pile (IS: 2911 – Part 1/Section 3). The cover of concrete is not less
than 40 mm, and in places where corrosion is anticipated, the cover should be at least 50 mm.
Materials and method of manufacture for cement concrete are based on IS: 456. Consistency
of concrete for cast in situ piles shall be suitable to the method of installation of piles. The mini-
mum grade of concrete to be used is M15. The grade of concrete for driven piles is as follows:
1. For hard driving (where driving stress is more than 1,000 kN/m2) Grade not less than M20
2. For easy driving (where driving stress is less than 1,000 kN/m2) Grade not less than M15
Clean water, free from acids and other impurities, shall be used in the manufacture of
concrete (IS: 2911 – Part 1, 1979).
M16_PURU1773_01_SE_C16.indd 601 2/12/2013 5:15:10 PM

Upper and lower portions of composite piles are made of different materials. Composite
piles may be of steel and concrete or timber and concrete. Steel and concrete piles consist
of a lower portion of steel and an upper portion of cast-in-place concrete (Fig. 16.1d). Tim-
ber and concrete piles usually have their lower portion as timber and their upper portion
above water table as concrete. It is extremely difficult to give proper connection between two
different materials. Hence, these types of piles are not widely used.
16.2.2 Installation Methods

Based on installation techniques, piles are classified as driven piles and cast in situ piles. Driven
piles may be concrete, steel, or timber. Concrete piles are classified as driven precast concrete
piles, driven cast in situ concrete piles, and bored cast in situ concrete piles (IS: 2911 – Part 1, 1979).
Driven precast pile is the one constructed in concrete (reinforced or pre-stressed) in a casting
yard and subsequently driven in the ground when it has attained sufficient strength. Driven cast
in situ pile is formed within the ground by driving a casing (with bottom closed) of uniform
diameter, permanent or temporary, and subsequently filling in the hole so formed with plain or
reinforced concrete. When the casing is left permanently, it is termed as cased pile and when the
casing is taken out it is referred to as uncased pile. Bored cast in situ pile is formed within the
ground by excavation or boring a pile within it, with or without the use of a temporary casing
and subsequently filling it with plain or reinforced concrete. Cased and uncased piles hold good
here too. There is another type of bored cast in situ pile in which the compaction of surrounding
ground and freshly filled concrete in pile bore is simultaneously achieved by suitable method,
and such a pile is referred to as bored compaction pile.
Driven piles are installed by hammer impact or by a vibrating machine. The installation
of any type of driven pile causes displacement and disturbance of the soil around the pile.
However, in case of pipe piles without shoe and in H-piles, the displacement is negligible.
Loose granular soils are densified because of driving, thereby increasing the frictional resis-
tance. On the other hand, driving causes re-moulding in cohesive soils, and temporarily
reduces the soil shear strength along the pile surface. The reduced strength is regained with
time due to thixotropic effect.
For cast in situ piles after making the pre-excavated hole, required reinforcement is placed
and concrete is poured around. These piles tend to relieve the lateral earth pressure, and
hence, reduce the strength along the shaft. Such piles are designed taking into account only
the end bearing. However, with time the resistance along the shaft improves substantially.
Thus, driven piles show higher resistance than cast in situ piles immediately after placement.
Figure 16.2 shows some types of cast in situ piles.
16.2.3 Ground Effects

Piles are sometimes employed to compact soils and such piles are referred to as displacement
or compaction piles. These piles displace a substantial volume of soil during installation
(Fig. 16.3a). In granular soils, there is a tendency for compaction, whereas in clays, heaving
of the ground surface often results. The ratio of heave volume to pile volume is about 50%
for clays and 30% for silty clays. Driven piles installed in pre-drilled holes are also called
non-displacement piles. In foundations for bridge abutments and piers, to safeguard the foun-
dation from damage due to scour, piles are used (Fig. 16.3b). Piles are also used to prevent
the movement of earth slopes and earthen banks by driving piles through the moving mass
into the firm ground (Fig. 16.3c).
M16_PURU1773_01_SE_C16.indd 602 2/12/2013 5:15:10 PM

(a) Step taper (b) Pipe (c) Raymond (d) Uncased (e) Bulb type
Fig. 16.2 Some types of cast in situ piles
16.2.4 Function as Foundation

Piles function as foundation under different conditions and they are referred to accordingly.
When the topsoil is soft or too weak to support the superstructure, piles are employed to
transmit the load to the underlying bedrock; such piles are called end-bearing piles or point-
bearing piles. If the bedrock is not existing at a reasonable depth below the ground surface,
the load is transferred through friction along the pile shaft. The frictional resistance devel-
oped at the soil–pile interface is utilized to support the structural load; such piles are called
friction piles (Fig. 16.4a–c).
Original
Bridge pier bed level
Zone of
erosion
Scoured
bed level
(b) Sefeguard scour damage

Original surface
Soft material
(a) As soil compactor Piles

Firm ground
(c) Soil stabilizer
Fig. 16.3 Piles for improving ground condition
M16_PURU1773_01_SE_C16.indd 603 2/12/2013 5:15:11 PM

Friction
Soft soil Soft soil carrying
material
Friction
Hard stratum carrying
material
(a) Point bearing pile (c) Friction pile

(b) Friction cum
bearing pile
Tie rod
Deadman
Uplift
pressure
Uplift piles
(d) Uplift piles (e) Batter piles (f) Batter pile as anchor pile
Sheet pile Dolphin
Fender pile
(g) Other functions of some piles
Fig. 16.4 Classification of piles based on function
Transmission towers, offshore platforms, and basement mats are subjected to uplift forces
and piles are used to resist the uplift forces, which are called uplift piles or tension piles. Some
of the water- and earth-retaining structures are subjected to horizontal and inclined forces.
Such forces are better resisted by providing piles in inclined position; such piles are referred
to as batter piles (Fig. 16.4d and e).
Flexible earth-retaining structures are tied at the top by anchor rods supported by a dead-
man. The deadman is in turn supported by piles. These piles function as anchorage against
horizontal pull from the sheet pile walls or other pulling forces; such piles are called anchor
M16_PURU1773_01_SE_C16.indd 604 2/12/2013 5:15:12 PM

piles. In order to protect water front structures against impact from ships or other floating
objects, fender piles and dolphins are used (Fig. 16.4f and g).
16.3 PILE-DRIVING EQUIPMENT

Pile-driving equipment mainly consists of pile frames, pile winches, and piling hammers.
Pile frames consist essentially of leaders which are a pair of steel members extending to full
height of the frame (Fig. 16.5). They guide the hammer and the pile during the process of driv-
ing. Leaders can be extended at the top by telescopic boom when long piles have to be drawn.
Pile frames are usually mounted on standard tracked crane base machines for mobility on land
sites or on framed bases for mounting on stagings for marine construction. Pile frames have
to remain in correct position throughout the driving of pile. Any possibility of a settlement of
frame should be avoided, so that the weight of frame is not transmitted to a partially driven pile.
Piling winches may have single, double, and triple drums which can raise hammer and
pile separately. Light winches have only one drum. Winches are provided with reversing
facilities, such that in addition to their main purpose of lifting the hammer and pile, they can
also be used for raking and rotating. Winches are powered by steam, diesel or petrol engines,
or electric motors. Pile winches are mounted on the base of pile frames.
Most piles are installed from the ground surface by means of hammers or vibratory
drivers. Piles can also be inserted by jetting or partial augering. General categories of pile
hammers include drop hammer, single-acting hammer, double-acting hammer, and diesel
hammers (Fig. 16.6).
Drop hammer is the simplest form of hammer used in conjunction with light frames and
for piles driven for pile load test. Drop hammers are solid masses of steel, 10 to 50 kN in
mass, fitted with a lifting eye and lugs for sliding in the leaders. The main disadvantages of
a drop hammer are uncontrolled drop of fall and slow rate of hammering. Drop hammers
have the advantage that they can be operated within a sound-proof box in places where
noise abatement procedures are necessary.
Single-acting hammers use steam or compressed air to raise the hammer ram. At the
ready-for-driving position, the steam is cut off and the cylinder falls freely on to the pile
helmet. The maximum height of a drop is usually about 1.4 m and the rate of blow is about
Backward rake 1 in 5
Forward rake 1 in 25
Tubular struts
9.0 m
Hand winch
Wide flange beam
leader
Three wheel base
3 m × 3.5 m
Fig. 16.5 Schematic diagram of light pile frame
M16_PURU1773_01_SE_C16.indd 605 2/12/2013 5:15:12 PM

Exhaust Exhaust
Ram
Cylinder and
Hammer intake
cushion Intake
Ram Ram
Pile cap
Pile cushion Hammer
Pile cushion
(a) Drop hammer (b) Single-acting (c) Double-acting

hammer hammer
Static weight
Ram
Oscillator
Hammer Pile
cushion Anvil Clamp
(d) Diesel hammer (e) Vibratory pile driver
Fig. 16.6 Pile-driving devices
60 strokes per minute. These hammers have to rely on the mass of ram. For effective driving,
the mass of hammer ram should be equal or greater than the mass of pile.
Double-acting hammers use steam or compressed air to raise the ram to the driving posi-
tion and also to accelerate the ram’s downward thrust. This increases the impact velocity of
the ram. The rate of driving varies from 300 blows per minute for light types to 100 blows
per minute for heavier types. The weight of the hammer ram is in the range of 0.9 to 23 kN.
The major advantage of the double-acting hammer over single-acting hammer is its tremen-
dous operating speed. Special maintenance is required for efficient functioning. These ham-
mers are used mainly for sheet pile driving.
Diesel pile hammers provide an efficient means of pile driving in favourable ground con-
ditions. These are self-contained and have self-activated units. They essentially consist of a
ram, an anvil block, and a fuel injection system. The ram is mechanically raised to the top of
the cylinder and released. A fuel mixture is injected into the cylinder and compressed by the
falling ram. The fuel is detonated and the resulting explosion imparts an additional impact
to the pile, which is already moving downwards under the effect of the hammer blow. The
explosion also raises the ram, keeping it ready for the next downstroke. Diesel hammers
work well under hard driving conditions. In soft or yielding soils, the downward movement
of the pile is large compared to the upward movement of the ram. This upward distance cov-
ered may not be sufficient to ignite the air–fuel system unless the ram is lifted up manually.
Vibratory methods of driving sheet piles or bearing piles are best suited to sandy or grav-
elly soils. Pile-driving vibrators consist of two counter-rotating masses, which produce a
M16_PURU1773_01_SE_C16.indd 606 2/12/2013 5:15:13 PM

75 mm diameter
M.S. pipe
Pile
50 mm diameter
nozzle
Fig. 16.7 Pile with central jet pipe
dynamic vertical force on the pile and move the pile downwards. Driving steel piles in loose
to medium dense sands is easy by this method but difficult in dense sands. Vibrators are
rarely used in stiff clays. Vibrators are also used for extracting piles, and in large-diameter
bored and cast in situ piling works, for sinking and extracting of pile casings.
Water jetting is a technique used to aid the penetration of pile into a sand or sandy gravel
stratum. In this technique, water is discharged at the pile point to wash and loosen the sand
and gravel. Jetting should be cut off at least 1 m above the required level and the pile is driven
for the balance length. It is sometimes a difficult problem to dispose of the large quantity of
water and sand flowing at ground level from around the piles (Fig. 16.7).
Each pile is provided with a cap or helmet at the top of the pile. The purpose of the helmet
is to hold the resilient dolly and packing, which are provided between the hammer and the
pile to prevent shattering of the pile head. A cushion may be used between the pile and the
cap. This has the effect of evening out the hammer impulses. Longer dollies or followers
are used when driving piles below the level of the bottom of the leaders.
16.4 BEARING CAPACITY OF SINGLE PILE

The bearing capacity of a single pile depends on the structural strength of the pile and the
supporting strength of the soil, and the smaller of the two controls the design load. In order
to satisfy the first criterion, the pile load is restricted so as to avoid damage from overdriv-
ing, to avoid overstressing of the pile under design load, and to avoid buckling failure of the
pile. Generally, based on the material and dimension of the pile, allowable stress in piles is
fixed by codes to account for the above-mentioned factors. However, the failure of a pile is
controlled by the second criterion, i.e., the supporting strength of the soil, unless the pile
material or the construction is below standard. Thus, further discussion is confined to the
supporting strength of the soil.
Load on a pile is partly carried by skin friction and partly by the resistance offered by
the soil at the base of the pile called the point resistance (Fig. 16.8). As observed in shallow
foundations, an increase in load (within 20% of the failure load) on a pile is followed by a
proportionate increase in settlement. But after a certain load level, the rate of increase of
settlement is far out of proportion with the rate of increase of load. Thus, the ultimate load
M16_PURU1773_01_SE_C16.indd 607 2/12/2013 5:15:13 PM

Qu
f f L
qp
Fig. 16.8 Ultimate load of single pile
or bearing capacity of a single pile indicates the load at which the settlement of the pile
increases continuously with no further increase in load.
Therefore, the ultimate bearing capacity of a pile Qu is given as
Qu = Qf + Qp (16.1)
where Qf is the resistance due to skin friction and Qp the resistance due to point bearing. That is,
Qu = fAs + qp Ap (16.2)
where As is the effective surface area of the pile in contact with soil along the embedded shaft
length, Ap the bearing area of pile tip, f the average unit skin friction or adhesion between
soil and pile surface, and qp the bearing pressure of soil at the tip.
Depending on the type of soil penetrated, Qf and Qp are evaluated. For simplicity, the anal-
ysis for piles in sand and clay are discussed below. However, the analysis can be extended
for layered soils by computing the skin friction for the appropriate layer.
16.4.1 Pile Capacity from Statical Methods for Driven Piles

For a driven pile in sand, the unit skin friction depends on the soil pressure acting normal to
the pile surface and the coefficient of friction between the soil and pile material (Fig. 16.9).
Thus, the unit skin friction acting at any depth in sand is
f = σh′ tan δ
f = Kσ v′ tan δ
where σh′ is the horizontal soil pressure acting at any depth z in a soil mass, σ v′ the effective
over-burden pressure acting at the same depth within a soil mass, K the lateral earth pres-
sure coefficient, and tan δ the coefficient of friction between soil and the pile surface.
Values of coefficient of friction depend on the type of soil and pile material. McCarthy
(1982) reports values varying from 0.20 to 0.45 depending on the roughness of pile surface.
Indian Standards (IS: 2911 – Part 1, 1979) recommends δ = φ.
The earth pressure is approximately equal to the passive condition at the top of the pile
and may be near to the at-rest-condition at the pile tip. Further, the method of installation
also influences the earth pressure coefficient. The following values of K (Table 16.1) are
recommended by Das (1990).
M16_PURU1773_01_SE_C16.indd 608 2/12/2013 5:15:14 PM

Qv
Direction of movement
under loading of Qu
z
s¢v
Soil Area sh= ks¢y
surrounding pile =1 d Area=1
Resultant of sh and f
\ f = sh tan d
Fig. 16.9 Soil–solid friction developing against pile shaft
Hence, the total skin friction acting along the embedded length of the pile is computed
considering an average effective vertical overburden pressure. Thus,
Qf = fAs = [K (σ v′ )a tan δ ]As (16.3)
where (σ v′ )a is the average effective vertical overburden pressure,

γL
(σ v′ )a =
2
where L is the length of the pile,

As = πdL
where d is the average diameter of the pile.

The end-bearing component of pile capacity can be determined by a method similar to
that of shallow foundations. This is given for deep foundations in non-cohesive soils as
Qp = 0.5γ dp N γ + σ v′ Nq (16.4)
where Nq is the bearing capacity factor for pile foundation (Fig. 16.10), Nγ the bearing capac-
ity factor for shallow foundations, and dp the pile tip diameter or width.
Table 16.1 Coefficient of earth pressure
Pile K
Bored or jetted piles K0 = 1− sin φ
Low-displacement-driven piles K0−1.4 K0
High-displacement-driven piles K0−1.4 K0
Source: Das (1990).
M16_PURU1773_01_SE_C16.indd 609 2/12/2013 5:15:15 PM

300
200
(also) Indian standards
150 IS : 2911
120
100
80
60
40
30
Nq
5
d= 0
20 L/ 2
15 d =
L/
70
10 =
8 /d
L
6
4
3
2
20∞ 25∞ 30∞ 35∞ 40∞ 45∞
Angle of shearing resistance, f¢
Fig. 16.10 Values of Nq for pile formulae (Source: Berezantzev et al., 1961)
In most of the driven piles, the first term of Eq. 16.4 is small compared to the second term
because of limited dp dimensions. Thus, for practical considerations, the point-bearing resis-
tance can be written as
Qp = (σ v′ Nq )Ap (16.5)
Various theoretical analyses for the point-bearing pressure have been attempted and
among them the Berezantzev et al. (1961) value for Nq have been in use for commonly
encountered soil conditions (Fig. 16.10). Nq curve as recommended by Indian Standards
(IS: 2911 – Part 1, 1979) is also presented in Fig. 16.10. For design purposes, σ v′ = γ L is
considered. Thus, the ultimate bearing capacity of driven piles in sand is given as
Qu = [K (σ v′ )a tan δ ]As + (σ v′ Nq )Ap (16.6)
In saturated clays, pile driving causes the re-moulding of soil in the vicinity of the pile
because of displacement and disturbance. But the strength is regained with time, and the
rate of strength gain depends on the consolidation and thixotropic characteristics of the clay.
Presently, there are several approaches available for obtaining the unit skin friction.
Two of the approaches adopt effective stresses in the analysis (Burland, 1973; Vijayvergiya
and Focht, 1972) and the other one uses an empirical adhesion factor. Burland’s method,
commonly referred to as the β-method, is widely used.
The common approach is to use undrained cohesion on the presumption that the dis-
sipation of pore water pressure takes much more time. But recent studies have shown that
dissipation of excess pore pressure takes place fairly quickly, and at the time of the final
loading of pile, the soil is almost at the drained condition. Thus, the unit skin friction based
on Burland’s approach is
f = β (σ v′ )a (16.7)
′ = drained friction angle of re-moulded clay.
′ and φre
where β = K tan φre
M16_PURU1773_01_SE_C16.indd 610 2/12/2013 5:15:17 PM

The value of K is conservatively taken as
′
K = K 0 = 1 − sin φre for normally consolidated clays (16.8a)
and
′ ) OCR
K = K0 OCR = (1 − sin φre for over-consolidated clays (16.8b)
Thus,
Qf = [β (σ v′ )a ]As (16.9)
Indian Standards (IS: 2911 – Part 1, 1979) recommends Eq. 16.1 taking into account the
adhesion factor
Qf = α cs As (16.10)
where α is the adhesion (or reduction) factor as given in Table 16.2 and cs the average
undrained cohesion along pile shaft.
The end bearing is related only to the undrained strength of clay. For φ = 0° and Nq = 1,
the contribution by end bearing is very small. So the point bearing resistance is expressed as
a function of cp and Nc, i.e.,
Qp = (cp Nc )Ap (16.11)
where cp is the average undrained cohesion at pile tip and Nc is 9.0 for intact clays (Skemp-
ton, 1951 and IS: 2911 – Part 1, 1979) and Nc is 6.75 for fissured clays (Skempton, 1951).
Thus, the ultimate bearing capacity of a pile in clay is
Qu = [β (σ v′ )a ]As + [cp Nc ]Ap (16.12a)

and
Qu = αcs As + cp Nc Ap (as per IS: 2911 − Part 1, 1979) (16.12b)
In both the cases, a factor of safety (F) of 2.5 is adopted to arrive at the allowable bearing
capacity, Qa, as
Qu
Qa = (16.13)
F
Table 16.2 Value of α
Consistency Value of α
Soft to very soft 1.0
Medium 0.7
Stiff 0.4
Stiff to hard 0.3
Source: IS: 2911 – Part1 (1979).
M16_PURU1773_01_SE_C16.indd 611 2/12/2013 5:15:18 PM

16.4.2 Pile Capacity from Statical Methods for Cast In Situ or Bored Piles
For such piles in relatively homogeneous soils, the ultimate capacity is due to end-bearing
and skin friction, as considered in driven piles. Because of the disturbance caused by
construction, there will be a possibility of the loss of strength at the bottom of the pile for a
depth of 2 m. Similarly, there is an equal possibility of disturbance and loss of strength occur-
ring in the surface zone of soil for a depth of about 2 m. Thus, for computing skin friction, a
depth of 4 m has to be omitted in the evaluation of skin friction. This is more important in
clayey soils.
The ultimate capacity of the cast in situ pile in sand is similar to that of driven piles in sand
(Eq. 16.6); that is,
Qu = [k(σ v′ )a tan δ ]As + (σ v′ Nq )Ap
Here, (σ v′ )a is the effective average vertical pressure considering the limits imposed for
effective depth L′ (i.e., L′ = L − 4) where L is in metres.
As no effective stress approach is made yet for bored piles, the ultimate pile capacity in
clay is obtained from Eq. 16.12b; that is,
Qu = α cs As + cp Nc Ap
Sometimes, bored piles may be provided with an enlarged base, and in such cases the
values of f for belled foundation has to be used. The value of Ap in such cases is evaluated
for the foundation base. The values may be adopted as discussed for driven piles.
16.4.3 Pile Capacity from Pile-Driving Formulae

In qualitative terms, it may be considered that a pile is capable of sustaining a greater load if
it exerts a greater resistance against driving. Based on this principle, many dynamic formu-
lae were suggested to obtain pile capacity considering the energy needed for driving the
piles. One of the earliest of these dynamic equations, commonly referred to as the Engineer-
ing News Record (ENR) formula, is derived on the basis of the work–energy theory.
The total driving energy caused by the hammer hitting the pile is equal to the weight of
the hammer times the height of drop or stroke. This energy is consumed by the work done
in penetrating the pile and by certain losses. This means that
Driving energy = (work of pile penetration) + (loss of energy)
or
E = RS + EL (16.14)
where E is the driving energy, R the pile resistance, S the pile penetration per blow, and EL
the loss of energy, including loss in impact, in driving cap, in pile, and in soil. If EL is assumed
to be proportionate to the pile resistance, it can be written as
EL = RC
where C is the empirical constant.
According to the ENR formula, the pile resistance is the ultimate load Qu. Then
E = Qu S + Qu C (16.15)
M16_PURU1773_01_SE_C16.indd 612 2/12/2013 5:15:19 PM

or
E
Qu = (16.16)
S+C
The average value obtained from the last few driving blows is taken as the pile penetra-
tion S (m). Recommended values of C are
C = 2.54 ×10−2 (m) for drop hammers

C = 0.254 ×10−2 (m) for steam hammers (16.17)
and
E = Wh
E = η HE
where W is the weight of ram or hammer (kN), h the height of free fall of the ram (m), η the
hammer efficiency, and HE the rated energy of single or double acting hammer (kN-m).
A factor of safety, F = 6.0, has been used to estimate the allowable pile capacity. The ENR
formula has been modified by Hiley and is given as
ηWhη b
Qu = (16.18)
S + C /2
where η is the hammer efficiency as given in Table 16.3, S the final set or penetration per
blow (m), C the sum of the temporary elastic compressions of the pile (m) = C1 + C2 + C3,
where C1 is the temporary compression of dolly and packing, C2 the temporary compression
of pile, C3 the temporary compression of ground, and ηb the efficiency of the blow, represent-
ing the ratio of energy after impact to the striking energy of ram
W + Per2
= , where W ≥ Per
W +P
W + Per2 W − Per2
= − , where W < Per
W +P W +P
where P is the weight of pile, anvil, helmet, and follower, if any (kN), and er the coefficient
of restitution of the materials under impact.
The factor ηb is given for the condition that pile is driven into the penetrable ground. If the
pile finds refusal in rock, 0.5P has to be adopted in place of P in the expression for ηb.
Table 16.3 Hammer efficiency
Hammer type Efficiency

Single- and double-acting hammers 0.70–0.85
Diesel hammer 0.80–0.90
Drop hammers 0.70–0.90
Source: Das (1990).
M16_PURU1773_01_SE_C16.indd 613 2/12/2013 5:15:20 PM

Table 16.4 Coefficient of restitution
Pile material Coefficient of restitution, er

Cast-iron hammer and concrete piles (without cap) 0.40–0.50
Wood cushion on steel piles 0.30–0.40
Wooden piles 0.25–0.30
Source: Das (1990).
The coefficient of restitution, er, of the material under impact are given in Table 16.4.
A factor of safety of 2 to 2.5 may be adopted to obtain the allowable load-bearing capacity
of a pile.
The dynamic formulae discussed above are based on the premise that the soil resistance
remains constant during and after the driving operation. In coarse-grained soils, this con-
dition is fulfilled because of high permeability. In fine-grained soils, water cannot escape
readily during driving, and thus, the excess water tends to reduce the frictional resistance
along the periphery of the pile. Further, the driving operation reduces the shear strength of
the surrounding soil. But because of thixotropic effect, the soil regains its strength rapidly.
Thus, dynamic formulae are not dependable for the determination of pile capacity in soils
containing more fines. Dynamic formulae give varying results even in coarse-grained soils.
The dynamic formulae can be only applied to small jobs in granular soils, localities of known
soil conditions, and to terminate the driving of a pile based on the data from a test pile.
16.4.4 Pile Capacity from Wave Equation

The major drawback with dynamic pile-driving formulae is that they do not stimulate soil–
pile interaction or the time-dependent nature of the problem. The application of driving data
to determine pile capacity requires knowledge of the effect that a hammer blow has on the pile
and supporting soil. The impact of a pile-driving hammer causes stress waves to be transmit-
ted through the length of the pile. These waves cause elastic compression and tension in the
pile and interact with the soil both along the surface of the pile and at the pile tip. Apart from
this, the pile-driving accessories, like hammer, anvil, cap block, pile cap, and cushion, influ-
ence the pile-driving behaviour. These variables were accounted for by Smith (1962) through
an analogy, termed the wave equation, of the pile behaviour and a mathematical model.
The model presented by Smith was based on the propagation of an elastic wave through a
long rod. The model considered the complete pile-driving operation, including pile-driving
accessories, soil–pile interaction and the time-dependent nature of the elastic pile deforma-
tion. The model has been given in the form of a partial differential equation. The actual and
the idealized systems are shown in Fig. 16.11. The springs simulate the axial resistance of the
pile and the spring constant, k relate to the elasticity of the pile. The spring damping R rep-
resents the frictional resistance of the soil surrounding the shaft of the pile and the soil resis-
tance at the pile tip. The various weight values W correspond to the weight of the incremental
sections of the pile, and the spring damping at the pile tip accounts for the point resistance.
The partial differential equation has been solved using computers and is now widely used
by civil engineers. Bowles (1984) has presented a more detailed discussion on this approach.
M16_PURU1773_01_SE_C16.indd 614 2/12/2013 5:15:20 PM

Ram W1
Cap block K1
Pile cap W2
Cushion K2
W3
Spring K3
v4
K4 Dash pot
W5
K5
W6
Pile K6
Side friction
W7
K7
W8
K8
W9
K9
W10
K10
Point resistance
Actual system Idealized system
Fig. 16.11 Pile representation for wave equation analysis (Source: Smith, 1962)
16.4.5 Pile Capacity Based on SPT Values on Non-Cohesive Soils

The empirical correlation of pile capacity and SPT value has been widely used. The correlation
between blow counts and pile capacity has not been accepted as a standard method. A care-
ful judgement has been exercised by the engineers while adopting the correlation.
Meyerhof (1976) has suggested a formula that may be used for a non-cohesive material,
particularly for sand deposits. It is
L
q = 40 N ≤ 400 N (kN / m 2 ) (16.19)
d
and
f = 2 N (kN / m 2 ) (16.20)
Hence, the pile capacity can be expressed as

Qu = fAs + qAp (16.21)
16.4.6 Pile Capacity from Pile Load Test

The pile load test is the most reliable method of determining the capacity of a pile. A test pile
is installed adopting the same proposed procedure. It may be loaded to near-failure condition
or up to the working load level. In the latter case, the pile shall form one of the permanent
piles of the foundation. A careful record has to be maintained during installation and during
the load test.
Three types of tests are conducted on piles, namely, vertical load test, lateral load test, and
pull-out test. The results from vertical load test is used to estimate the vertical load-carrying
capacity of the soil while the lateral and pull-out tests are used to estimate the lateral load-
carrying capacity and the frictional resistance of the pile. Only the vertical load test is dis-
cussed below and the reader may refer to IS: 2911 – Part 4 (1985) for other tests.
The test procedure consists of applying a static load on the pile top in convenient load
increments and recording the vertical deflections of the pile. Suitable reaction device is
M16_PURU1773_01_SE_C16.indd 615 2/12/2013 5:15:21 PM

Dead load Jack Jack
Settlement
gauge support
outside zone of
Reaction pile
influence
Test pile Reaction pile Test
pile
(a) Gravity loading (b) Reaction loading
Fig. 16.12 Pile load test arrangements
adopted (Fig. 16.12). The reaction may be obtained from (i) a kentledge placed on a platform
supported clear of the test pile with the centre of gravity of the kentledge passing through the
axis of the pile or (ii) anchor piles installed at a distance not less than three times the test pile
shaft diameter or 1.5 m, whichever is greater. The reaction for the test should be 25% more
than the proposed final test load. Measurement of pile movements is related to a fixed refer-
ence mark. Reference marks would be supported on objects located outside the soil zone.
The pile head is made level by chipping off to natural horizontal plane and finished
smooth and level with plaster of Paris. A bearing plate is placed before seating the hydraulic
jack. Datum bar is set on immovable supports beyond a distance of 1.5 m from the edge of
the pile. At least two dial gauges are fixed to the datum bar. A series of vertical downward
increment of loads with intensity of 20% of the safe load on the pile are applied. Settlement
readings are taken for each load increment till the rate of displacement is 0.003 mm/min.
The test is continued till the maximum load is 1.5 times the working load or the maximum
settlement of the test not exceeding 12 mm.
Pile load settlement curve is drawn and the safe load is obtained as the least of the follow-
ing (IS: 2911 – Part 4, 1985):
1. The straight portions of the curve are extended and the ultimate load is found. A factor of
safety is adopted and the safe load against shear failure is determined.
2. Two-thirds of the final load at which the total displacement attains a value of 12 mm.
3. Fifty percent of the load at which the total displacement equals 10% of the pile diameter
in case of uniform diameter piles or 7.5% of bulb diameter in case of under-reamed piles
(discussed in next section).
Typical test data and a load–settlement curve are given in Worked Example 16.5.
If soil conditions are uniform, a relatively low factor of safety of 1.5 to 2.0 may be enough.
But in non-uniform deposits, being reflected by varying load test results, a higher factor of
safety (about 3) is justified.
In addition to assessing the design load, the pile test may be used to establish the construc-
tion driving criteria. A comparison can be made with pile-driving records and the measured
M16_PURU1773_01_SE_C16.indd 616 2/12/2013 5:15:21 PM

ultimate load, with dynamic pile-driving formulae, or with the wave equation to establish
the driving specification. The skin friction and end-bearing components of the pile support
capacity can be separated by running a tension test or a cyclic-load test on the pile (refer IS:
2911 – Part 4, 1985).
16.4.7 Negative Skin Friction

Negative skin friction is downward drag acting on the piles due to relative movement
between the piles and the surrounding soil. This condition can develop where a soft or loose
soil stratum located anywhere above the pile tip is subjected to compressive loading. The
effect of negative skin friction is to increase the axial load in the pile and the pile settlement.
Negative skin friction can develop under different field conditions, such as the following
(Fig. 16.13):
1. A cohesive fill is placed over a non-cohesive soil layer and a pile is driven into such a
medium. The cohesive fill consolidates, and during the process of consolidation the fill
imparts a downward drag on the pile.
2. A non-cohesive fill is placed over a soft cohesive layer and a pile installed in such a
medium. Due to the overburden pressure, the cohesive fill consolidates and during the
process imparts a downward drag on the pile.
3. In a saturated soil, lowering of the groundwater increases the vertical effective stresses in
the soil medium. This induces settlement, and a pile driven in such a medium is subjected
to a downward drag force.
The negative skin friction is computed for cohesive and non-cohesive fills as follows:
Cohesive Fill Overlying Non-Cohesive Soil. Based on β-method (as discussed earlier for
piles in non-cohesive soils), the unit negative skin friction can be given as
f n = Kσ v′ tan δ (16.22)
Q Q
Lf fn fn Lf
Clay fill Sand fill
L L L Neutral plane
fn fn
Sand Clay
Fig. 16.13 Negative skin friction
M16_PURU1773_01_SE_C16.indd 617 2/12/2013 5:15:22 PM

where K = K 0 = 1 − sin Pre′ , σ v′ = γ fz = vertical effective stress (0 < z < Lf), δ = soil–pile
friction angle = 0.5 − 0.7 fre′ , and γf = unit weight of fill.
Hence, the total downward drag force, Qn, is given as
Lf
Qn = ∫ πd(K γ f tan δ )z dz
0
πdK γ f L2f tan δ

Qn =
2
where Lf is the depth of fill.

If the water table is at the ground surface γf = γ′ and if the fill is above the water table,
γf = γ.
Cohesive Soil Underlying Non-Cohesive Fill. Vesic (1977) has shown that negative skin
friction develops at the top portion of the pile in the compressing medium. That is, in the
depth range z = 0 to z = L1 (Fig. 16.13), and this depth is referred to as the neutral depth. The
neutral depth may be given as (Bowles, 1982)
(L − Lf ) ⎡ L − Lf γ f′Lf ⎤ 2γ f′Lf
L1 = ⎢ + ⎥− (16.24)
L1 ⎢⎣ 2 γ ′ ⎥⎦ γ′
where γ f′ and γ ′ are submerged unit weights of the fill and the underlying cohesive layer,
respectively.
Hence, the total drag force
L1 L1
Qn = ∫ πdf n dz = ∫ πdK (γ f′Lf + γ ′z) tan δ dz
. 0 0
L21πdK γ ′ tan δ
Qn = (πdK γ f′Lf tan δ )L1 + (16.25)
2
If the soil and the fill are above the water table, the submerged unit weights should be
replaced by moist unit weights.
16.5 UNDER-REAMED PILES

Under-reamed piles are of bored cast in situ and bored compaction concrete piles with
enlarged base. The enlarged base is termed a bulb or under-ream. An under-reamed pile
may have one, two, or more bulbs, accordingly, they are referred to as single-, double-, or
multi-under-reamed piles. The bulb provides adequate bearing or anchorage. Under-reamed
piles are used for a variety of field applications, e.g., to obtain adequate capacity for down-
ward, upward, and lateral loads and moments as in transmission tower foundation, to take
the foundation to deeper stratum in order to prevent the effect of seasonal changes as in
expansive soils, to reach firm strata and to take the foundations below scour level.
M16_PURU1773_01_SE_C16.indd 618 2/12/2013 5:15:24 PM

The under-reamed pile should satisfy all the design requirements for a conventional
pile. In addition to these, in deep deposits of expansive soils, the minimum length of piles
(irrespective of any other factors) should be 3 to 5 m below ground level. In poor bearing
stratum or in recently filled-up grounds, the pile should pass through such soils and be
seated in good bearing strata (IS: 2911 – Part 3, 1980).
The bulb diameter is taken two to three times the diameter of the shaft. For 30 cm diameter
piles, the recommended spacing between the bulbs should not exceed 1.5 times the diameter
of bulb and for greater diameter piles, the spacing is reduced to 1.25 times the stem diam-
eter. The position of the top most bulb should be at a minimum depth of two times the bulb
diameter, and for expansive soil, the minimum depth is 1.75 m from ground level. Further,
the minimum clearance below the underside of pile cap and the bulb should be a minimum
of 1.5 times the bulb diameter. Indian Standards (IS: 2911 – Part 3, 1980) recommends only
two bulbs. Typical details of single and double under-reamed piles are shown in Fig. 16.14.
The bearing capacity of a single under-reamed pile (Fig. 16.15a) may be given based on
static formula. Thus,
Qu = Qf + Qp
or
Qu = fAs + qb ( Ab − Ash ) + qsh Ash (16.26)
where As is the surface area of the embedded shaft of the pile above and below the bulb, Ab
the cross-sectional area of the bulb (= πdu2 / 4 ), Ash the cross-sectional area of the shaft
GL GL
Stirrups
expansive
2du min or
1750 for
soil
First bulb
Bring level
for making
first bulb
1.25 to
1.5 du
d.4
Second /last
Approx
bulb
Bucket length
φ1 φ1
φ1 φ2
+ 0.55
d d Bucket length
Cover 75 to 100 du
φ1 = 45° φ2 = 30° – 45° du = 2.5 d (normally)
Approx Approx
(a) Single under-reamed pile (b) Double under-reamed pile
Fig. 16.14 Typical details of under-reamed piles (Source: IS: 2911 – Part 3, 1980)
M16_PURU1773_01_SE_C16.indd 619 2/12/2013 5:15:25 PM

Qu Qu
f f f
d d
f
A A
f
f 1.25 du to 1.5 du
f f
A′ A′
qb
qb
qsh
du du
(a) Single under-reamed pile (b) Double under-reamed pile
Fig. 16.15 Details of forces acting on under-reamed piles
( = πd 2 / 4 ), f the unit skin friction on the shaft above and below the bulb, qb the bearing
pressure of the soil at the under-ream section, qsh the bearing pressure of the soil at the pile
base, du the diameter of under-ream, and d the diameter of the shaft.
The values of f, qsb, and qb are determined following the procedure suggested under stati-
cal methods.
In order to increase the pile capacity, more than one under-ream may be provided
(Fig. 16.15b) for double or multi under-reamed piles (with bulbs suitably spaced), the soil
between the bulbs tends to act as part of the pile. Thus, on the surfaces AA′ full soil resis-
tance mobilizes. Mohan et al. (1967, 1969) have confirmed this behaviour from model and
field tests. The bearing capacity of a double under-ream is given in Eq. 16.27:
Qu = fAs + f As + qb ( Ab − Ash ) + qsh Ash (16.27)
where f is the unit skin friction between soil to soil (along the cylindrical surface AA′) and
As the surface area of the cylinder bounded by the diameter of the bulb and the distance
between the centres of the extreme bulbs.
Indian Standards (IS: 2911 – Part 3, 1980) suggests separate equations for clay and sand.
However, Eqs. 16.26 and 16.27 can be effectively used depending on the type of soil and
water table position as dealt for conventional piles using static methods. The bearing capac-
ity of an under-reamed pile can also be determined by load test. Approximate safe loads on
under-reamed piles are also provided in the code (IS: 2911 – Part 3, 1980).
M16_PURU1773_01_SE_C16.indd 620 2/12/2013 5:15:26 PM

16.6 PILE GROUPS

In the previous section, we discussed the pile capacity of single piles. When piles are used
for foundation support, they are always used in a group. This is an important requirement
to ensure that the imposed structural load lies within the support area provided by the foun-
dation. Building codes never permit the use of less than three piles to support a major col-
umn, and less than two piles to support a foundation wall. Thus, the bearing capacity and
settlement of pile groups are the end results needed for the design of the foundation.
16.6.1 Pile Group Capacity

The pile group capacity is not necessarily the individual pile capacity multiplied by the
number of piles in the group. Soil disturbance caused by pile installation techniques and
overlap of stresses between adjacent piles may reduce the group capacity from the sum of
individual capacities. On the other hand, soil between individual piles might get densified
or increased with adhesion, and the group may tend to behave as an equivalent single large
pile, which may show higher group capacity.
Pile spacing (centre-to-centre) plays an important role in the group capacity of piles. Pile
spacing is considered from two aspects of installing the piles and the nature of the load transfer
to the soil. For end-bearing piles, the minimum spacing should be 2.5d, where d is the diameter.
If the piles are resting on solid rock, the spacing of 2d is recommended. For friction piles, the
spacing should be sufficient to avoid overlapping of stresses. In such cases, the recommended
spacing is 3d. In case of loose sand or fill, a spacing of 2d may be adopted (IS: 2911 – Part 1, 1979).
Ideally, the piles in a group should be spaced in such a way that the group capacity is not
less than the sum of individual pile capacity.
Vesic (1977) recommends an optimum spacing of 3 to 3.5d. Mohan (1981) suggests that the
Indian Standards should follow Vesic’s recommendations of 3 to 3.5d. As a general guide, 2d
spacing is the minimum requirement and 3 to 3. 5d is the preferred spacing.
In order to understand the behaviour of pile groups, it is necessary to identify two types
of them (Poulos and Davis, 1980), viz.,
1. A free-standing pile group, in which the pile cap is not in contact with the underlying soil.
2. A pile group in which the pile cap is in contact with the underlying soil (referred to as
piled foundation).
In both the cases, an efficiency factor ηg is defined as the ratio of the ultimate capacity of
the group (Qug) to the sum of the individual pile capacity. It is expressed as
Qug
ηg = ×100% (16.28)
npQu
where np is the number of piles in a group.
Free-Standing Groups. Several group efficiency formulae are in use relating pile spacing
and number of piles. Converse–Labarre formula (Eq. 16.29) is one such formula quite often
used by engineers
⎡ (n − 1)n2 + (n2 − 1)n1 ⎤
ηg = 1 − ⎢ 1 ⎥θ (16.29)
⎢ 90 n1 n2 ⎥
⎣ ⎦
where n1 is the number of rows, n2 the number of piles in a row, θ = arctan d/s (degrees), and
s the centre-to-centre spacing.
M16_PURU1773_01_SE_C16.indd 621 2/12/2013 5:15:26 PM

Chellis (1962) compared several formulae and found wide variation in ηg values. The most
widely used approach in clays is the one suggested by Terzaghi and Peck (1967). As per their
approach, the group capacity is analysed based on the following two conditions and the
lesser value is considered as the design load:
1. The sum of ultimate capacity of the individual piles in the group
2. The bearing capacity for block failure of the group (Fig. 16.16). That is,
Qus = npQu (16.30)
or
Qub = fAsg + qApg (16.31)
where Qus is the sum of individual pile capacity, Qub the ultimate load capacity of block,
and Asg the surface area of the group, that is,
Asg = 2(B1 + B2 )L
where
B1 = (n1 − 1)s + d
B2 = (n2 − 1)s + d
Apg = B1 × B2
For pile groups in clay

f = αcu
q = cu Nc
and Nc is Skempton’s bearing capacity factor. Whitaker (1957) confirmed the existence of
such failures from model studies in clays. Poulos and Davis (1980) observed that the values
obtained from Eqs. 16.30 and 16.31 are not smooth transitions but abrupt and suggested the
following empirical equation for clays:
Qug
s s
B1 = (n1–1)s + d
L s B1 B (n –1)s + d
2= 2
s
A A
s s B2
(a) Sectional elevation (b) Sectional plan – AA
Fig. 16.16 Pile group
M16_PURU1773_01_SE_C16.indd 622 2/12/2013 5:15:28 PM

1 1 1
2
= 2 2+ 2 (16.32)
Qug npQu Qub
Equation 16.32 may be re-expressed as
2 2
1 npQu
2
= 1+ 2 (16.33)
ηg Qub
Comparative study of De Mello (1969) on efficiency of groups in clays showed that higher
efficiency factors occur for
1. piles with smaller length–diameter ratios,
2. larger spacings, and
3. smaller number of piles in the group.
For spacing commonly used in practice, the efficiency factor is of the order of 0.70 to 0.85
in clays.
Only limited information is available for free-standing pile groups in sand. It has been
established that the group efficiency in sands may often be greater than one (Poulos and
Davis, 1980). But practical considerations limit the spacing to approximately 3d or the spac-
ing corresponding to ηg = 1 (i.e., Qug = nQu).
Piled Foundations. In piled foundations, the pile group has the cap cast on or beneath
the surface of the soil. For such cases in clays, Poulos and Davis (1980) suggested to take the
lower value of ultimate load-bearing capacity from the following:
1. The ultimate load capacity based on block failure (Eq.16.31) plus the ultimate load capac-
ity of that portion of the cap outside the perimeter of the block.
2. The individual pile capacity of cap and piles. That is,
Qug = np (α cu As + ApCp Nc ) + Ncc Cc (B1 × B2 − npπ d 2 / 4) (16.34)
where cp is the undrained cohesion at the level of pile tip, cc the undrained cohesion
beneath pile cap, Ncc the bearing capacity factor for rectangular cap = 5.14(1 + 0.2B1/B2)
for B1 > B2 (Skempton, 1951), and Nc the bearing capacity factor (Skempton’s values).
The first value applies to close pile spacing while the second for wider spacing (when
individual action of piles occur). Model studies on piled foundations by Whitaker (1957)
showed good agreement between the model test results and the predicted efficiency from
the block-failure criterion.
In sands, it has been reported (Vesic, 1969) that the pile cap contributes significantly to the
group capacity. For practical purposes, the contribution of the cap can be taken as equivalent
to the bearing capacity of a strip footing with a half-width equal to the distance from the
edge of the cap to the outside of the pile (Poulos and Davis, 1980).
16.6.2 Pile Group in Filled Ground

In pile groups installed in a fill which is in the process of consolidation under its own weight
or under the weight of the surface load, the weight of the whole mass of soil entrapped
within the periphery of the group is transferred to the piles. For the first case (Fig. 16.17a),
M16_PURU1773_01_SE_C16.indd 623 2/12/2013 5:15:28 PM

the additional load causing the drag down is due to the entire compressible soil located
within the periphery. Thus, the total load on the pile group is
Qug = (working load) + B1 × B2 × γ f′Lf′ (16.35)
where γ f′ is the submerged unit weight of fill and Lf′ the depth of fill over which the move-
ment is sufficient to cause a drag down.
For the second case (Fig. 16.17b), the total load on the pile group at the level of bearing
stratum is
Qug = (working load) + B1 × B2 × γ f′Lf′ + B1 × B2 × γ n Ls′ (16.36)
where γ f′ is the submerged unit weight of soft stratum and Lf′ the depth of soft stratum over
which movement is sufficient to cause a drag down.
However, the total load on the pile group will not exceed the ultimate skin friction on
piles from fill and soft clay, that is,
Qug > (working load) + f ′As′ + f ′As′′
where As′ is the sum of surface areas of piles embedded in fill, As′′ the sum of surface areas of
piles embedded in soft clay, f ′ the skin friction between fill and piles, and f″ the skin friction
between soft clay and piles.
Group settlement is computed considering both the working load and the load trans-
ferred by the fill. Thus, there is no increase in the weight of soft stratum which should cause
additional loading on the bearing stratum (Fig. 16.17b). Again, only the working load and
the weight of the fill are causing settlement of the bearing stratum. The above approach
assumes that the fill has been recently placed and has not had time to cause appreciable
consolidation of the underlying strata.
The above-mentioned negative skin-friction problem will not occur if the piles are taken
through the fill on to an incompressible stratum, such as bedrock or very compact sand and
gravel.
Qug Qug
Filling L′f
Soft
L′s compressible
L′s
L layer
Bearing
stratum
B1 × B2 B1 × B2
(a) (b)
Fig. 16.17 Pile groups in filled ground
M16_PURU1773_01_SE_C16.indd 624 2/12/2013 5:15:30 PM

16.6.3 Pile Group Settlement

The settlement observed at the pile cap level is due to compression within the pile from the
loading plus the settlement occurring in the soil supporting the pile. The total settlement
may be attributed to four separate causes: (i) the axial deformation of the pile, (ii) the defor-
mation of the soil at the pile–soil interface, (iii) the compression of the soil between the piles,
and (iv) the compression of the stratum below the tips of the pile. The settlement due to (i)
and (ii) are very small and generally neglected, and settlement due to (iii) is difficult to
evaluate. Hence, only the fourth factor is considered and expressed as the total settlement of
a pile group.
A widely used procedure assumes that the pile group acts as a single large deep foun-
dation, such as a pier or a raft. For pile groups that are essentially end bearing, the load
is assumed to be applied at the pile tips as a uniform load over the area of the group. The
stress distribution below the tips can be evaluated using the approximate 30° or 2:1 stress
distribution (Fig. 16.18). For friction piles in clay, the load is assumed to be applied load at a
depth of about two-thirds the length of the piles. The 30° or 2:1 distribution is adopted from
that level.
For pile groups in sand and gravel, Meyerhof (1976) suggested an expression for the elas-
tic settlement S (mm) as
0.92q B1l
S= (16.37)
N′
2
where q = Qg /(B1 × B2 ) (kN / m ), N′ is the corrected SPT value within the seat of settlement
(≈ B1 deep below the tip of the pile), l the influence factor (= 1 − L / B, B1 ≥ 0.5 ), Qg the allow-
able load on the pile group (kN), and L the length of the embedment of pile (m).
Qug Qug
2/3L
Dense Soft clay
bearing stratum L
30°
30°
Compressible
layer
Firm layer Firm layer
Fig. 16.18 Stress distribution for settlement analysis
M16_PURU1773_01_SE_C16.indd 625 2/12/2013 5:15:31 PM

16.7 UPLIFT RESISTANCE OF PILES

Under certain field conditions, such as tall chimneys, transmission towers, or jetty structures,
foundation of structures are subjected to large overturning moments resulting in uplift
forces. If foundations are on piles, they have to resist the upward forces. Methods of calculat-
ing the adhesion to resist uplift are the same as those used for bearing piles.
16.7.1 Uplift of Single Pile

For a uniform diameter in clay, the ultimate uplift resistance, Puu, is given as
Puu = ca As + Wp
where Wp is the weight of pile and ca the average adhesion along pile shaft.
For piles of uniform diameter in sand, the ultimate uplift capacity may be calculated as
the sum of the shaft resistance plus the weight of the pile. Only a few data are available
about the skin friction for upward loading.
For piles of uniform diameter in c–φ soils, Meyerhof and Adams (1968) suggest the fol-
lowing formulae for ultimate uplift load capacity.
1. Shallow depths (L < db)
π
Puu = πcdb L + s γ db L2 K u tan φ + W
2
2. Great depths (L > H)
π
Puu = πcdb H + s γ db (2L − H )KK u tan φ + W
2
where s is the shape factor ( = 1 + mL/db) with a max value of 1 + mH/db, Ku the earth
pressure coefficient (approximately 0.90 to 0.95 for φ values between 25° and 40°), m the
coefficient depending on φ, H the limiting height of failure surface above base, and W the
weight of soil (buoyant or total) and pile in cylinder above base.
The upper limit of the uplift capacity is the sum of the net bearing capacity of the base, the
side adhesion of the shaft, and the weight of the pile, that is,
π 2
(Puu )max = ′ Nq ) + As fs + W
(db − d 2 )(cNc + σ vb
4
′ the
where Nc, Nq are bearing-capacity factors, fs the ultimate shaft shear resistance, and σ vb
effective vertical stress at the level of pile base.
Meyerhof and Adams (1968) suggest that the values of Nc and Nq for downward load
can be used in this case also. Values of H/db, m, and s for various values of φ are given in
Table 16.5.
M16_PURU1773_01_SE_C16.indd 626 2/12/2013 5:15:32 PM

Table 16.5 Factors for uplift analysis
φ° 20 25 30 35 40 45 48
H/db 2.5 3.0 4.0 5.0 7.0 9.0 11.0
m 0.05 0.10 0.15 0.25 0.35 0.50 0.60
s 1.12 1.30 1.60 2.25 3.45 5.50 7.60
Source: Meyerhof and Adams (1968).
16.7.2 Uplift of Pile Group

Uplift of a group can be calculated as the lesser of the following (Meyerhof and Adams, 1968):
1. The sum of the uplift of the individual footings/piles.
2. The uplift load of an equivalent pier foundation consisting of the footings/piles and the
enclosed soil mass.
Based on the uplift testing of footings in clays, Meyerhof and Adams (1968) have reported
that the uplift efficiency increases with the spacing of footing or bases and as the depth of
embedment decreases, but the efficiency decreases as the number of footings or bases in the
group increases. These results are reported to be true for free-standing groups of piles in
clays with downward loading (Whitaker, 1957).
In sands, for a given sand density, the uplift efficiency of the groups has been reported to
increase roughly with the spacing of the footings or shafts and to increase when the depth
of embedment becomes smaller (Meyerhof and Adams, 1968). Further, the uplift efficiency
is reported to decrease as the number of footings or shafts in the group increases and as the
sand density increases.
16.8 LATERAL RESISTANCE OF PILES

16.8.1 Causes of Lateral Forces
Sometimes, piles have to face lateral forces and moments; such a situation may arise in the
following cases:
1. Quay walls and harbour structures where lateral forces are caused due to wave action and
by the impact of ships during bettering.
2. Offshore structures subjected to wind and wave actions.
3. Earth-retaining structures supported by piles.
4. Transmission tower-foundation structures, that are to face a high wind force.
5. Lock-structure in dockyards.
6. Pile-supported structures constructed in earthquake areas.
Ultimate lateral resistance and deflection are the two main aspects of laterally loaded
pile that are of interest. These problems are solved by simplified methods involving the
M16_PURU1773_01_SE_C16.indd 627 2/12/2013 5:15:32 PM

concept of coefficient of sub-grade reaction. The ratio between the lateral pressure, P and the
deflection, y produced at a particular point is called the coefficient of sub-grade reaction, a
modulus of sub-grade reaction, or simply as the soil modulus, k (= p/y).
16.8.2 Short and Long Piles

A vertical pile resists a lateral load by mobilization of passive pressure by the soil surrounding
the pile. The distribution of the reaction of the soil depends on the following factors:
1. stiffness of the pile,
2. stiffness of the soil, and
3. fixity of the ends of the pile.
Thus the mode of failure, lateral resistance, and the deflection of a laterally loaded pile
depend on the length of the pile and the surrounding soil. Accordingly, laterally loaded
piles can be classified into two major categories, viz., short or rigid piles and long or elastic
piles.
This behaviour of a pile is governed by the stiffness (EI) of the pile and the compressibility
of the soil, the latter being a function of the soil modulus, K.
Pile stiffness and soil modulus are expressed as stiffness factors, which is explained as
follows:
1. For a stiff preconsolidated clay with a constant soil modulus with depth, stiffness factor,
⎛ EI ⎞1 / 4
R = ⎜⎜⎜ ⎟⎟⎟ .
⎝ KB ⎠
2. For cohesionless soils and most of the normally consolidated clays with soil modulus
⎛ EI ⎞1 / 5
assumed to increase linearly with depth, then stiffness factor, T = ⎜⎜⎜ ⎟⎟⎟ .
⎜⎝ n ⎟⎠h
In the above equation, where E = modulus of elasticity of the pile material, I = moment of
inertia of the pile section, and nh = coefficient of modulus variation (force/L3).
A laterally loaded pile is delineated as short or long based on the non-dimensional factor
L/R or L/T as illustrated in Table 16.6.
References may be made to Kaviraj (1988) and Das (2002) for design charts for lateral
resistance and deflection.
Table 16.6 Criteria for short and long piles
Pile type Soil modulus

Constant Linearly increasing
Short rigid L/R ≤ 2 L/T ≤ 2

Long L/R ≥3.5 L/T ≥ 4
Source: Tomlinson, 1977.
M16_PURU1773_01_SE_C16.indd 628 2/12/2013 5:15:32 PM

16.9 INCLINED LOADING OF VERTICAL PILES

The ultimate load-carrying capacity of a vertical pile subjected to inclined loading is a func-
tion of both the lateral resistance and the vertical load capacity of the pile.
When there is only a small deviation from vertical loading the failure will occur essentially by
axial slip or by bearing failure of the tip for downward loading. If the inclination of the applied
load is large, lateral failure will occur due to load being applied perpendicular to the pile axis.
The above tow modes of failure will occur under the following conditions:
1. Axial failure is bound to occur when the ultimate lateral capacity exceeds the horizontal
component of the ultimate inclined load, i.e., when
Hu > Qu sin δ
or
Hu > Pu tan δ
where Qu is the ultimate inclined-load capacity of pile, Hu the ultimate lateral-capacity of pile,
Pu the ultimate axial-load capacity of pile, and δ the angle of inclination of load from vertical.
2. Lateral failure is bound to occur when the ultimate lateral capacity is less than the hori-
zontal component of the ultimate inclined load, i.e., when
Hu < Pu tan δ
For cohesive soils, it is generally assumed that the ultimate axial capacity of the pile is
independent of the lateral component of the load and that the lateral load capacity is inde-
pendent of the axial component of load. Then the inclined load capacity, Qu, can be calcu-
lated as the lesser of the two values:
For axial failure
Qu = Pu sec δ
For lateral failure
Qu = Hu cosec δ
16.10 PILE CAP

The axial and lateral load carrying capacity of a pile group is significantly affected by a pile
cap because of the resistance offered by the surrounding soil. This aspect is generally ignored
in view of the potential danger of loss of soil support due to scour. However, in situations
where the pile cap is fully embedded, it can contribute significantly to the group capacity.
The pile group stiffness is increased by 5% to 15% when the pile cap rests in contact with the
soil and also the settlement is reduced by 5% to 15% (Mohan, 1981).
Pile caps are almost invariably made of reinforced concrete. For designing pile caps, the
load from the column at top of the cap and the reaction from the pile at the bottom of the cap
are assumed to be distributed at 45° till the mid-depth of the cap. Based on this concept, the
maximum bending moment and shear force are calculated at the critical sections (IS: 2911 –
Part 1/Section 1). Other recommendations of the code are as follows:
1. The depth of pile cap should be sufficient enough to provide necessary anchorage of the
column and pile reinforcement.
M16_PURU1773_01_SE_C16.indd 629 2/12/2013 5:15:33 PM

2. The pile cap should be rigid enough to distribute the imposed load on the piles in
a group equitably.
3. The overhang of the pile cap should normally be 100 to 150 mm.
4. The pile should project 50 mm into the cap concrete.
5. The cap is cast over a levelling course of thickness of 75 mm.
6. In the case of large cap, where differential settlement may be imposed between piles
under the same cap, due consideration for the consequential movement should be given.
WORKED EXAMPLES
Example 16.1 For a waterfront structure, round concrete piles are planned for use as founda-
tion. The soil at the site is a medium-dense to dense sand having φ = 39°, γ = 19.7 kN/m3,
γ′ = 10.5 kN/m3. The water table is located at a depth of 3.0 m from the ground surface. As
per the design requirement, driven piles of 300 mm diameter and 8 m length have to be
used. Determine the design load considering a factor of safety of 2.0. Assume K = 3.0, tan
δ = 0.45.
Solution
Here, static formula can be used. Thus,
Qu = Qf + Qp
f = Kσ v′ tan δ
Because of the presence of water table, the effective overburden pressure should be sepa-
rately calculated.
Hence, resistance due to skin friction
Qf = fAs = (σ v′ )K tan δ As
= [(σ v′ )1 L1 + (σ v′ )2 L2 ] tan δπd
⎡ 0 + 3 ×19.7 ⎛ 3 ×19.7 + 5×10.5 ⎞⎟ ⎤
=⎢ × 3 + ⎜⎜⎜3 ×19.7 + ⎟⎟× 5⎥ ×(3.0 × 0.45×π× 0.30)
⎢⎣ 2 ⎝ 2 ⎠ ⎥⎦
= (88.65 + 426.75)×(3 × 0.45×π× 0.30) = 655.8 kN
Point bearing resistance
0.3 2
Qp = qAp = (σ v′ Nq )Ap = (3 ×19.7 + 5×10.5)Nq ×π×
4
From Fig. 16.10, for φ = 39°, Nq = 108. Therefore,
0.3 2
Qp = (3 ×19.7 + 5×10.5)×108 ×π× = 852.0 kN
4
Qu = 655.8 + 852 = 1507.8 kN
Qu 1507.8
Qa = = = 754 kN
F 2.0
M16_PURU1773_01_SE_C16.indd 630 2/12/2013 5:15:34 PM

Example 16.2 In a two-layered cohesive soil, bored piles of 400 mm are installed. The top
layer has a thickness of 5 m and the bottom one is of considerable depth. The shear strength
of the top clay layer is 45 kN/m2 and that of the bottom is 100 kN/m2. Determine the length
of the bored pile required to carry a safe load of 380 kN, allowing a factor of safety of 2.0.
Solution
Consider an adhesion factor, α = 0.50.
Let L1 and L2 be the depths of embedment of pile in top and bottom layers, respectively. Then,
Q
Qa = u
F
Qu = Qa × F = 380 × 2 = 760 kN
Also,
πd2
Qa = α (Cu )1 π dL1 + α (Cu )2 π dL2 + (Cp )2 Nc
4
Therefore,
0.4 2
760 = 0.5× 45×π× 0.4 × 5 + 0.5×100 ×π× 0.4 L2 + 100 × 9×π
4
or
760 = 254.5 + 62.8 × L2
or
760 − 254.5
L2 = =8m
62.8
Therefore, the length of the pipe is as given below:
L1 + L2 = 5 + 8 = 13 m
Example 16.3 A 12 m long 300 mm2 square pre-cast concrete pile is driven into a sand stra-
tum by a single-acting steam hammer. The weight of the CI hammer ram is 14 kN and the
stroke is 750 mm. The pile showed a driving resistance of 5 blows/25 mm penetration. Esti-
mate the ultimate bearing capacity of the pile based on the Hiley formula. Take C = 0.00508 m.
Solution
For a single-acting steam hammer, the hammer efficiency can be taken as 80% (from Table 16.3).
The coefficient of restitution for CI hammer and pile can be taken as 0.45 (from Table 16.4).
Weight of pile, Wp = 0.3 × 0.3 ×12× 23.5 = 25.38 kN
25 1
Penetration per blow = × = 0.005 m
5 1000
Per = 25.38 × 0.45 = 11.4 kN
That is, W > Per. Therefore,
ηWh W + er2 p
Qu = ×
S + (C / 2) W + p
0.80 ×14 × 0.75 14 + 0.452 × 25.38
= ×
0.005 + 0.00254 14 + 25.38
M16_PURU1773_01_SE_C16.indd 631 2/12/2013 5:15:36 PM

or
Qu = 541.5 kN
Example 16.4 A 6 m thick layer of medium dense sand overlies a deep dense gravel. Series of
standard penetration tests were undertaken and the sand stratum showed an average N-value
of 21. From the tests on gravel, the N-value at the interface has been interpolated as 42. A round
pile of 250 mm diameter is to be driven down through the sand gravel to give adequate end
bearing. Taking a factor of safety of 3, determine the allowable load that the pile can carry.
Solution
As suggested by Meyerhof (1976), the ultimate bearing capacity is given as (Eq. 16.21)
Qu = fAs + qAp
The lower value of q obtained from Eq. (a) or (b) is to be taken. Now,
L
q = 40 N kN / m 2 (a)
or d
q = 400 N kN / m 2 (b)
From (a)
6.0
q = 40 × 42
0.25
= 40 , 320 kN / m 2
From (b)
q = 400 × 42 = 16 , 800 kN/m 2
Therefore,
π(0.25)2
Qu = 2× 21×π× 0.25× 6 + 16, 800 ×
4
= 197.8 + 824.3
= 1022.1 kN
Allowable load,
1022.1
Qa = = 340.7 kN
3
Example 16.5 A pile load test is made on a 300 mm diameter test pile and the following
data are obtained.
Load (kN) Settlement (mm)
0 0.00
300 1.25
600 3.75
900 7.50
1,200 13.75
1,500 23.75
1,800 36.75
M16_PURU1773_01_SE_C16.indd 632 2/12/2013 5:15:37 PM

Determine the design load on the pile considering the settlement and shear criteria.
Adopt a factor of safety of 2.0.
Solution
The load–settlement curve is plotted as shown in Fig. 16.19 and the tangents are drawn from
the two straight portions. The ultimate load is read from the graph as
Qu = 1050 kN
Based on shear failure, the allowable load
1050
(Qa )1 = = 350 kN
3
Based on the settlement criterion, the safe load is least of the following (IS: 2911 – Part4, 1985):
1. 2/3 of the final load corresponding to 12 mm settlement
(Qa )2 = 32 ×1125 = 750 kN
2. 50% of the final load corresponding to 10% of the pile diameter (= 0.10 × 300 = 30 mm)
(Qa )3 = 12 ×1650 = 825 kN
Based on settlement, the allowable load = 750 kN.

Hence, the design load is least of the values obtained from shear and settlement criteria.
Therefore, design load
Qa = 350 kN
Load on pile, kN
300 600 900 1,200 1,500 1,800 2,100
0
8
Settlement, mm
16
24
32
40
Fig. 16.19
M16_PURU1773_01_SE_C16.indd 633 2/12/2013 5:15:38 PM

Example 16.6 A single under-reamed pile is to be installed in a hard clay deposit.

The undrained strength obtained from a series of vane tests at different depths have shown
a linear relationship, cu = 50 + 8D in kN/m2, where D is the depth in metres. The diameters
of the pile shaft and the bulb are, respectively, 1 and 3 m. The centre of the under-ream is
located at 16 m from the ground surface. Determine the allowable load on the pile to ensure
an overall factor of safety of 2.0. Neglect the resistance offered by the shaft beneath the bulb.
Solution
At D = 16 m, cu = 178 kN/m2.
Therefore, average cohesion on shaft is
50 + 178
= 114 kN / m 2
2
From Table 16.2, α = 0.3. Thus,
Q u = f As + q Ap
πd u2
= α(c u )1 πd L + (c u )2 N c
4
32
= 0.3× 114 ×π× 1× 16 + 178 × 9 ×π×
4
= 13 , 043 kN
Q 13 , 043
Qa = u = = 6 , 522 kN
F 2
Example 16.7 A 4 × 3 pile group has the following details:

Diameter of each pile, d = 350 mm
Centre-to-centre spacing of pile = 1,050 mm
Capacity of a single pile = 400 kN
Determine the efficiency of the free-standing pile group.
Solution
The efficiency of a pile group can be obtained using the Converse–Labarre formula. So,
n1 = 4 , n2 = 3
d = 350 mm , s = 1050 mm
Therefore,
⎛ 350 ⎞⎟
θ = tan−1 (d / s) = tan−1 ⎜⎜⎜ ⎟ = 18.4°
⎝ 1050 ⎟⎠
From Eq. 16.29,
⎧⎪ ⎡ ( 4 − 1)3 + (3 − 1)4 ⎤ ⎫⎪
ηg = ⎪⎨1 − ⎢ ⎥ 18.4⎪⎬×100
⎪⎪⎩ ⎢⎣ 90 × 4 × 3 ⎥⎦ ⎪⎪⎭
= (1 − 0.29)×100 = 71%
Qug
=
npQu
M16_PURU1773_01_SE_C16.indd 634 2/12/2013 5:15:39 PM

Therefore,
Qug = ηg × npQu = 0.71×12× 400 = 3408 kN
Example 16.8 A group of concrete piles is square in plan and consists of 9 piles each 12 m
long and 500 mm diameter. The piles are bored piles and installed at a spacing of 3 d in a
deep clay deposit having an unconfined compressive strength of 62.4 kN/m2.
At the tip of the pile and below, the undrained shear strength cu = 45 kN/m2. The average
unit weight of the soil and concrete are 19.2 and 22.5 kN/m3, respectively. Estimate the total
ultimate load of the pile group.
Solution
Since the piles are bored piles, the difference between the weight of concrete and the exca-
vated earth has to be considered.
The general equation for a single pile is written as
Qu + (weight of pile – weight of excavated soil) = Qf + Qp

Qu = fAs + qAp − (weight of pile − weight of excavated soil)
Assuming the clay as medium stiff, α = 0.50. Thus,
⎛ 62 ⎞ 1 ⎡ π×(0.5)2 ⎤
Qu = 0.5×⎜⎜⎜ ⎟⎟⎟ π× 0.50 ×12 + 9× 45×π×(0.5)2 × ⎢⎢− ×12(23.5 − 19.2)⎥⎥
⎝2⎠ 4⎣ 4 ⎦
Qu = 292.2 + 79.5 − 10.1 = 361.6 kN
Considering the individual pile capacity,

npQu = 9× 361.6 = 325.4 kN
Considering block failure of the group,

Qug = fAsg + qApg
Now,
B1 = B2 = (n1 − 1)s + d = (3 − 1)× 3 d + d = 7 d = 3.5 m
For a block failure, the skin friction development is predominantly between soil to soil,
hence the adhesion factor can be taken as 1.0.
Qug = 2α(cu )1 [B1 + B2 ]L + (cu )2 Nc [B1 × B2 ]

= 2×1×(62 / 2)[3.5 + 3.5]×12 + 45× 9[3.5× 3.5]
= 5208 + 2835 = 8043 kN
Here,
npQu < Qug
Therefore, group capacity ≈3,254 kN.
M16_PURU1773_01_SE_C16.indd 635 2/12/2013 5:15:40 PM

POINTS TO REMEMBER
16.1 Pile is a type of deep foundation used for transferring the weight of the superstructure
through soft of weak soils, to deep load-bearing strata. They are slender structural members
normally installed by driving, by hammer or by vibrating, and occasionally by auguering.
16.2 Piles are classified based on material composition, installation methods, ground
effects, and function as foundation.
16.3 Pile-driving equipment consist mainly of pile frames, pile winches, and pile hammers.
16.4 Bearing capacity of a single pile depends on the structural strength of the pile and the
supporting strength of the soil, and the smaller of the two controls the design load.
16.5 Bearing capacity of a single pile may be found theoretically by statistical methods,
pile-driving formulae, and wave equation.
16.6 Bearing capacity of a single pile may be found from field tests, viz., pile load test and
SPT values.
16.7 Pile-driving formula is based on the qualitative principle that a pile is capable of sus-
taining a greater load if it exerts a greater resistance against driving.
16.8 Commonly used dynamic formula is the Engineering News Record (ENR) formula
which is derived on the basis of the work–energy theory.
16.9 Pile capacity from wave equation is a model based on the propagation of an elas-
tic wave through a long rod. The model considers the complete pile-driving opera-
tion, including pile-driving accessories, soil–pile interaction and the time-dependent
nature of the elastic pile deformation.
16.10 Pile load test is the most reliable method of determining the capacity of a pile. The test
necessitates a careful record to be maintained during installation and during load test.
16.11 Negative skin friction is a downward drag acting on the pile due to relative move-
ment between the piles and the surrounding soil. The effect of negative skin friction
is to increase the axial load in the pile and the pile settlement.
16.12 Piles are always used in a group. Bearing capacity and settlement of pile groups are
needed for the design of a deep foundation.
QUESTIONS
Objective Questions
1. The term composite pile is applied to a cross-section with more than one material.
2. Closely spaced piles embedded in clay often behaves like a group acting as a single
large unit.
3. Dynamic formulae are generally used to establish pile-driving criteria for pile
installation.
4. The development of skin friction along the shaft of a pile requires no shear strain in
the soil adjacent to the pile.
5. In friction piles, the load is transmitted to the soil through the adhesion or frictional
resistance along the shaft of the piles.
M16_PURU1773_01_SE_C16.indd 636 2/12/2013 5:15:40 PM

16.2 The negative skin friction or down-drag of a pile is a phenomenon which occurs
when
(a) A compressible organic soil is found at the pile tip
(b) A soil layer surrounding a portion of the pile shaft settles more than the pile
(c) The groundwater table suddenly rises from the tip of the pile to ground surface
(d) A long pile is driven next to the short pile under consideration
16.3 Load tests on piles in clays are not carried out immediately but sufficient time is
allowed for the soil to
(a) Come in contact with the pile surface
(b) Develop skin friction
(c) Regain its thixotropic strength
(d) Re-distribute the initial weight of pile
16.4 The action of driven piles in sands increases the relative density and thus the effi-
ciency of a pile group may be
(a) Equal to 100% (b) Greater than 100%
(c) Approximately equal to 100% (d) Well below 100%
16.5 The point bearing and average skin frictional resistance of bored piles in cohesionless
soil are ______ that for driven piles.
(a) Less than (b) Greater than
(c) Equal to (d) About 80% of
16.6 Indicate the incorrect statement. High group efficiency of pile can be obtained in a
clayey stratum for
(a) Piles having smaller length-to-diameter ratio
(b) Larger spacing
(c) Small number of piles in a group
(d) Closer spacing
16.7 Assertion A: Longer pile groups should settle more than smaller groups for the same
pile loads.
Reason R: This is due to the overlapping effect of stresses below the pile point from the
additional piles.
Choose the correct statement.
16.8 Consider the following statements:
1. The pile cap is in contact with the ground.
2. The piles are vertical.
3. Load is applied at the centre of the pile group.
4. The pile group is symmetrical and the cap is very thick.
The assumption that each pile in a group carries equal load may be correct when the
criteria in statements
(a) 1 and 2 are met (b) 1, 2, and 3 are met
(c) 1, 3, and 4 are met (d) All are met
M16_PURU1773_01_SE_C16.indd 637 2/12/2013 5:15:40 PM

16.9 Settlement of a pile group predominantly depends on the

(a) Axial deformation of the pile
(b) Deformation of the soil at the pile–soil interface
(c) Compression of the soil between the piles
(d) Compression of the stratum below the tips of the pile
16.10 For settlement evaluation of friction pile groups in clay, the load is assumed to be
applied as a uniform load at a depth of about ______ the length of the piles. Choose
the appropriate statement
(a) One-thirds (b) Two-thirds
(c) One-fourths (d) One-half
16.11 Does the choice of a pile hammer have any relevance to the type of pile? If so, give reasons.
16.12 While driving large number of displacement piles for a foundation, how would you pro-
ceed: centre to out, outside to centre, or progressively from one side to the other? Explain.
16.13 In a pile group, what are the geometrical properties that are to be considered in bring-
ing out a proper spacing of piles to ensure that they carry equal load?
16.14 List the circumstances under which a pile foundation becomes necessary.
16.15 What are the factors to be considered in the selection of pile hammer?
16.16 What type of piles would you recommend for the following types of soil and site con-
ditions:
1. The subsoil is a weak underlain by a hard rock
2. For a multi-storeyed building in the central part of a city surrounded by existing
buildings
3. For a harbour structure
16.17 Why do deep foundation units have typically long slender members?
EXERCISE PROBLEMS
16.1 A tapered 10 m long wooden pile with a 200 m diameter at the tip and a 350 mm diam-
eter at the butt is driven into a dry sand φ = 28°, and dry unit weight, γ = 17.2 kN/m3.
Compute the ultimate bearing capacity of the pile.
16.2 A single test pile of 0.25 m × 0.25 m square cross-section is driven through a stratum
up to a depth of 10 m. It is observed that the undrained cohesive strength is varying
from 12.4 kN/m2 at its surface to 65.3 kN/m2 at the depth of 10 m. Determine the safe
load the pile can carry if the factor of safety is 2.5.
16.3 A pre-stressed concrete pile of 350 mm diameter is to be driven into a layered deposit
of intact clay. The deposit consists of a normally consolidated clay up to a depth of
8 m from the ground surface, with cu = 30 kPa, φre ′ = 15ο, and γ = 16 kN/m3 followed
by 7 m of over-consolidated clay having cu = 100 kPa, γ = 19.6 kN/m3, and an over-
consolidation ratio of three. If the piles are driven to a depth of 12 m, estimate the total
skin friction expected to develop.
M16_PURU1773_01_SE_C16.indd 638 2/12/2013 5:15:41 PM

16.4 A drilled, dry straight pile foundation is to be constructed in a layered clay stratum.
The boring log and subsequent laboratory tests furnished the following details:
0–8 m depth cu = 45 kPa γ = 16.6 kN/m3
8–15 m depth cu = 80 kPa γ = 18.2 kN/m3
16.5 A sub-surface investigation at a site revealed the soil profile as shown in Fig. 16.20.
A clock tower is to be constructed on the site over a pile foundation. As per the design
requirement, piles of 300 mm diameter are to be driven to a depth of 14 m. Estimate
the ultimate pile capacity of one single pile.
g = 18.5 kN/m3 Sand

1.5 m 2.5 m
N = 18
g = 18.1 kN/m3 Medium

6.5 m cu = 35 kPa stiff
clay
g = 19.6 kN/m3 Sand
N = 33
g = 18.6 kN/m3 Stiff clay

8m cu = 65 kPa
Fig. 16.20
16.6 Compute the safe bearing of a 500 mm diameter and 12.5 m long concrete pile driven
into a granular medium by a hammer with a 30 kN ram and 900 mm stroke. The set is
observed to be 12 mm per blow and the constant C = 25 mm. The hammer efficiency
is 85% and coefficient of restitution is 0.50. Allow a factor of safety of 3.
16.7 In a multi-storeyed building site, soil boring revealed the presence of fine to coarse
sand The standard penetration test information obtained from the boring log is as
follows:
Depth (m) N value

0.9 12
2.0 14
3.0 18
4.5 20
6.0 25
7.5 27
Compute the length and diameter of driven single pile to take a design load of 450 kN,
considering a factor of safety of 2.0.
M16_PURU1773_01_SE_C16.indd 639 2/12/2013 5:15:41 PM

16.8 On a project, a pile load of 600 kN was desired. A pile load test was conducted and the
results are given below.
Load (kN) Penetration (mm)

0 0.0
300 4.5
550 5.5
800 9.5
1,050 16.0
1,300 21.3
1,550 30.1
Determine the adequacy of the test pile. Also, mention the criterion adopted to specify
the adequacy.
16.9 A two-bulb under-reamed pile is in a medium stiff clay having an average undrained
cohesion of 78 kN/m2 obtained from a triaxial test. The length of the shaft from the
ground surface up to the centre of the first bulb is 8 m, and the centre-to-centre of bulb
is 1.2 times the diameter of the bulb. The diameter of the shaft is 800 mm and that of
the bulb is 2,000 mm. Determine the allowable load on the pile, allowing a factor of
safety 2.0.
16.10 An eight-pile group consists of 300 mm diameter circular piles with centre-to-centre
spacing of 900 mm both ways. The piles are driven to a depth of 20 m into a clay which
has an unconfined compressive strength of 20 kPa and a unit weight of 17.5 kN/m3.
Also, the clay stratum is very deep. Determine the allowable load for the group, con-
sidering a factor of safety of 3.
16.11 It is proposed to transfer the total load of 3,000 kN of a structure through 10 m long
bored piles in a deep deposit of clay having an average undrained shear strength
of 90 kN/m2. The design diameter of the pile is 400 mm. Estimate the number of
piles required, considering a factor of safety of 2.5. Also, suggest the arrangement
of piles.
16.12 A nine-pile group is placed in a square pattern with centre-to-centre spacing of 0.9 m.
The diameter and length of the pile are 300 mm and 11.5 m, respectively. The piles
are driven into a sand deposit having f = 32° and g = 18.5 kN/m3. The sand deposit
extends to a depth of 12.5 m followed by a 5 m depth of clay having e0 = 0.85 and
Cc = 0.32. Determine the pile group efficiency and the settlement of the group if the
pile group carries a safe load of 4,500 kN. Allow a factor of safety of 2.0.
16.13 Forty RCC piles of 300 mm diameter are needed to carry a structural load of 1,650 kN
from a column on a deep soft clay deposit. The undrained cohesion of the deposit
is 22 kPa and the unit weight is 18.2 kN/m2. Estimate the length and spacing of the
piles such that the group has an efficiency of 100%. The adhesion factor can be taken
as 0.75 and the factor of safety against shear is 2.5.
M16_PURU1773_01_SE_C16.indd 640 2/12/2013 5:15:41 PM

17
Drilled Piers and Caisson
Foundations
CHAPTER HIGHLIGHTS
Drilled piers: uses and types of drilled piers, bearing capacity and settlement
of drilled piers, construction methods – Caissons: types of caissons, bearing
capacity and settlement of caissons – Construction procedures – Well founda-
tions: types of wells, components of well foundations, design of wells, stability
analysis, construction of well foundations – Shifts and tilts in well sinking
17.1 INTRODUCTION
Drilled piers and caissons provide a solid massive foundation for heavy loads and high
horizontal thrusts. Drilled piers are structural members of relatively large-diameter mas-
sive struts constructed and placed in a pre-excavated hole. They are referred to variously by
civil engineers as bored piles, large-diameter piles, foundation piers, sub-piers, and drilled
caissons. The shafts can be enlarged at the base, resulting in belled or under-reamed piers.
Caissons or well foundations are structural boxes or chambers that are sunk in place through
the ground or water by systematically excavating below the bottom of the unit, which
thereby descends to the final depth. These have a cross-sectional area and hence provide
high bearing capacity, which is much larger than what may be offered by a cluster of piles.
17.2 DRILLED PIERS

17.2.1 Uses
Piers have distinct advantages compared to other types of deep foundations:
1. In areas where pile penetration is difficult, piers can be provided.
2. Vibration and heave of soil are not caused as in installation of a driven pile. This is a
decisive factor when the adjacent structure is on spread footings or short piles.
M17_PURU1773_01_SE_C17.indd 641 2/12/2013 5:13:20 PM

3. Equipment used in the construction of drilled piers produces less noise and, hence, is
quite suitable for areas near hospitals and similar institutions.
4. There is a possibility of inspection and physical testing of the soil or rock conditions at the
bottom of the pier.
5. In the construction of piers, there is no displacement of volume of soil, and the problems
of shifting and lifting are eliminated.
6. Drilled piers generally require light construction equipment.
7. They can resist high lateral stresses.
17.2.2 Types of Drilled Piers

The common type of drilled pier is the straight type (Fig. 17.1); such shafts are taken through
the upper soil layers, and the end is placed on a firm soil layer or bed rock. In some instances,
the shafts are provided with casing.
Sometimes a broad base called a bell is provided at the bottom of straight shafts; such
piers are referred to as belled piers. The bell may be shaped like a dome or it may be angled.
Under-reaming equipment is used to make the bells.
The third type is the extended straight shaft or socketed piers, in which the straight shafts
are extended into the underlying rock layer.
17.2.3 Bearing Capacity of Drilled Piers

A drilled pier derives its supporting power from both skin friction and bottom bearing as in
a pile. Generally, the skin friction developing along the shaft is less compared to the end
bearing capacity. Since drilled piers are placed invariably on a hard stratum, which is less
compressible, the maximum percentage of load is taken by the hard stratum only. The sur-
face area of a drilled pier is less compared to the one available in a pile group. Thus, in many
Soft soil Soft soil Soft soil
Stiff clay or Stiff clay or

dense sand dense sand
Rock or firm soil Rock
(a) Straight-shafted pier (b) Belled pier (c) Pier socketed into rock
Fig. 17.1 Types of drilled piers
M17_PURU1773_01_SE_C17.indd 642 2/12/2013 5:13:20 PM

Drilled Piers and Caisson Foundations 643
instances, the drilled piers are designed as a compression member subjected to a load on top
and an equal reaction at the bottom, neglecting the skin friction.
The ultimate load-bearing capacity of drilled piers can be computed as for piles as
Qu = Qf + Qp
Piers in Cohesive Soils. The skin resistance for the shaft in cohesive soils is computed as
Qf = α cAs (17.1)
where As is the surface area of the shaft, α is the empirical adhesion factor ( 0.35 to 0.40), and
the load-bearing capacity for a circular base is
Qp = Ap (cNc + q ′N q + 0.3γ Db N γ ) (17.2)
where Nc , N q , and N γ are the bearing capacity factors for a deep foundation and q′ is the
vertical effective stress at the level of the bottom of the pier. Also, Db is the diameter of the
base and Ap is the area of the base = πDb2 /4 . The last term is generally considered only for
short piles and is neglected in other cases. Thus,
Qp = Ap (cNc + q ′N q ) (17.3)
The net ultimate bearing capacity (Qp)n at the base can be approximated as
(Qp )n = Ap (cNc + q ′N q − q ′)
(Qp )n = Ap [cNc + q ′( N q −1)] (17.4)
It is customary to omit the surcharge term, for φ = 0°, N q = 1; then,
(Qp )n = cu Nc Ap (17.5)
Skempton’s value of N q = 9 for L/Db ≥ 4 is widely used, where L is the length of the pier.
Thus,
(Qp )n = 9 cu Ap (17.6)
Piers in Non-cohesive Soils. For piers in non-cohesive soils, Berezantzev (1965) proposed
the net allowable load (Qp)a at the base limiting relative settlement (ΔH/Db) to be about 0.20.
Thus,
(Qp )a = [γ Db (Bk )]Ap (17.7)
Figure 17.2 presents values of Bk for several selected L/Db ratios. For piers in non-
cohesive soils, Eq. 17.4 may be written in terms of net base bearing pressure as
(Qp )n = Ap q ′( Nq −1) (17.8)
As the piers are invariably drilled and placed, compared to piles which are driven, a
lower bound value of N q , as suggested by Vesic (1963), may be adopted (Fig. 17.3).
M17_PURU1773_01_SE_C17.indd 643 2/12/2013 5:13:22 PM

1,400 20
18
1,200
16
1,000 14
12
800 10 L /D
Bk b
8
600
6
400 4
200
0
24 28 32 36 40 44
Friction angle, deg
Fig. 17.2 Coefficients Bk (Source: Berezantzev, 1965)
Allowing a suitable factor of safety to Eq. 17.8 and comparing it with Eq. 17.7, the lower
value may be taken as the net safe base pressure, (Qp)ns.
The expression for skin resistance may be computed in the same manner as that for piles;
that is,
L
Qf = ∫ f πDsb dz (17.9)
0
where f = Kσ v′ tan δ and Dsb is the diameter of the shaft.
400
200
100
Nq 60
30
20
25 30 35 40 45 50 55
Friction angle, deg
Fig. 17.3 Bearing capacity factor N q (Source: Vesic, 1963)
M17_PURU1773_01_SE_C17.indd 644 2/12/2013 5:13:23 PM

For piers with casing, K < K 0, and for uncased piers, K = K 0. The value of δ is taken as
that of soil and casing in the first case and is equal to φ in the second case. For a uniform soil
condition,
Qf = (Kσ v′ tan δ )×(πDsb L) (17.10)
Applying the same factor of safety for skin friction, the net safe skin friction is given as
Qf
(Qf )s = (17.11)
F
where F is the factor of safety. Thus, the net allowable load on piers is
Qan = (Qp )ns + (Qf )s (17.12)
In granular soils, the value of φ depends on the confining pressure, and hence, a careful
assessment of φ has to be made.
In some conditions, the negative skin friction can occur as discussed for piles. As many
uncertain factors are involved, detailed estimates for skin friction are not very realistic.
In many circumstances, it can be ignored. However, a rough estimate may be made based
on the value of skin friction suggested by Terzaghi and Peck (1967), as given in Table 17.1.
When drilled caissons are sunk through sand, it is loosened by grabbing and surges into
the dredging wells. This effect is not considered, as the reconsolidation of sand is rapid and
the negative skin friction is again a relatively short-term effect.
17.2.4 Settlement of Drilled Piers

Drilled piers on sand or gravel will not be subjected to detrimental settlement as they are
normally constructed on dense deposits. Settlement on sands may be computed, as done in
shallow foundations, by obtaining or making an estimate of the stress–strain modulus E and
Poisson’s ratio ν and using Eq. 14.33. Settlement may be computed as done for shallow foun-
dations in clay. Since one of the purposes of using drilled piers is to place the large loads on
Table 17.1 Observed values of skin friction for piers

and caissons
Type of soil Skin friction (kN/m2)
Silt and soft clay 7–29

Very stiff clay 48–192
Loose sand 12–34
Dense sand 34–67
Dense gravel 48–96
Source: Terzaghi and Peck (1967).
M17_PURU1773_01_SE_C17.indd 645 2/12/2013 5:13:24 PM

firm stratum, long-term settlement may not occur at all. However, immediate settlement of
piers on saturated clays may be computed using the Burland et al. (1966) expression as
q
S = Db (I B ) (17.13)
qult
where
⎡ ⎛ γ ′L ⎞⎟⎤⎥
I B = ε1 ⎢⎢ 0.295 ⎜⎜⎜9 + ⎟⎟ (17.14)
⎢⎣ ⎜⎝ cu ⎟⎠⎥⎥⎦
where ε1 is the axial strain at a deviator stress of one-third to half of the peak value and
q = γ ′Db Bk .
17.2.5 Construction Procedures of Drilled Piers

Construction procedures of drilled piers shafts are of three major types and they may be
classified as (1) dry method, (2) casing method, and (3) wet method. Each procedure of con-
struction is explained as follows:
1. Dry method
Dry method of construction is the simplest type wherein it is to remove all weak soils up
to the foundation depth and construct the pier in the form of concrete column.
For construction of the shafts, excavations may be made by hand or machine. A hand-
excavated pier is called as Chicago caisson or Gow caisson depending on the procedure
adopted, and the machine-excavated piers so by the name of drilled caissons.
In the Chicago type, circular holes are excavated up to a certain depth by hand, and the
sides are then lined with vertical boards called laggings (Fig. 17.4a(i)). They are secured in
place by two circular steel rings. The excavation is continued up to the desired depths, and
Lagging Steel casing

Steel ring
(i) Chicago method (ii) Gow method
Fig. 17.4 (a) (i–ii) Methods of pier construction
M17_PURU1773_01_SE_C17.indd 646 2/12/2013 5:13:25 PM

then the bell of the pier is excavated. The hole is filled with concrete after completion of
the excavation. In the Gow method (Fig. 17.4a(ii)) the excavation is done by hand, and the
telescopic metal shells are used to maintain the shaft. The shells are removed, as concreting
progresses piers up to a depth of 30 m can be constructed by this method.
Caisson shafts may be excavated more efficiently by utilizing mechanical devices such
as an angle, a bucket, a chopping bit, or a chopping bucket. Open-helix auger is a common
excavation tool that is attached to a shaft referred to as the kelly and pushed in to the soil
and rotated. The augered soil is taken out, and an under-reaming tool is inserted into the
hole to shape the bell. When boulders or hard stratum is encountered, the caisson is usually
excavated by a chopping bucket or a chopping bit.
The sequence of operation of a vertical drilled shaft without a bell using mechanical
device (O’Neill and Reese, 1999) is shown in Fig. 17.4 (a(iii)–(vi)).
The dry method is adopted in soils and rocks that are above the water table and will not
cave in during the time of drilling the hole.
2. Casing method of construction

This method is adopted in situations wherein casing or excessive deformation is likely to
occur during the time of excavation. The sequence of construction (O’Neill and Reese, 1999)
is explained as follows (Fig. 17.4b):
(i) As in the case of dry method of construction, the excavation procedure is initiated
(Fig. 17.4b(i)).
(ii) In the case of casing, if caving is met with bentonite slurry is pumped into the
borehole (Fig. 17.4b(ii)). Drilling is continued until the excavation goes past the caving
soil and a firm impermeable soil or rock is reached.
(iii) After this stage, a casing is introduced into the hole (Fig. 17.4b(iii)).
(iv) Using a submersible pump the slurry is bailed out of the casing (Fig. 17.4b(iv)).
(v) Then a small drill that can pass through the casing is introduced into the hole and the
excavation is continued (Fig. 17.4b(v)).
(vi) If an enlarged base or bell is needed, an under-reamer is introduced (Fig. 17.4b(vi)).
(vii) If a reinforcing steel is needed, it is inserted up to the full length of the excavation.
Concrete is then poured into the excavation and the casing is gradually pulled out
(Fig. 17.4(vii)).
(viii) The completed view of the drilled pier is shown in Fig. 17.4 b(viii).
3. Slurry method of construction

In this method of construction, slurry is used to keep the borehole open during the full depth
of excavation. The sequences of construction (O’Neill and Reese, 1999) are explained as fol-
lows:
(i) Along with the slurry the excavation is continued to the entire depth (Fig. 17.4 c(i)).
(ii) The reinforcement cage is placed, if needed (Fig. 17.4c(ii)).
M17_PURU1773_01_SE_C17.indd 647 2/12/2013 5:13:25 PM

Surface
casing, if
required
Competent, Competent,
noncaving soil noncaving soil
Drop chute
(iii) (iv)
Competent,
noncaving soil
Competent, Surface
noncaving soil casing, if
required
(v) (vi)
Fig. 17.4 (a) (iii–vi) Dry methods of construction (Source: O’Neill and Reese, 1999)
(iii) Then the concrete is placed in the drill hole, displacing the volume of slurry
(Fig 17.4c(iii)).
(iv) The completed view of a shaft without a bell is shown in Fig. 17.4c(iv).
This method is also called as slurry displacement method.
A drilled caisson is usually inspected before the placement of concrete to ascertain
1. the correctness of alignment and dimensions,
2. the condition of the load-bearing stratum,
3. the risk of loss of ground and settlement of the adjacent area, and
4. the complete removal of the loose material.
Generally, the permissible misalignment is about 75 mm from the required location and
1% of the height or depth of the pier for verticality.
M17_PURU1773_01_SE_C17.indd 648 2/12/2013 5:13:26 PM

M17_PURU1773_01_SE_C17.indd 649
Drilling
Cohesive soil Cohesive soil slurry Cohesive soil Cohesive soil
Caving soil Caving soil Caving soil Caving soil
Cohesive soil Cohesive soil Cohesive soil Cohesive soil

Drilled Piers and Caisson Foundations
(i) (ii) (iii) (iv)
Level
of fluid
concrete
Drilling fluid Competent soil
forced from
space between
casing and soil
Caving soil
Competent soil Competent soil Competent soil
Competent soil
Caving soil Caving soil Caving soil
Competent soil Competent soil Competent soil
(v) (vi) (vii) (viii)
Figure 17.4(b) Casing method of construction (Source: O’Neill and Reese, 1999)
649
2/12/2013 5:13:30 PM
Drilling
Cohesive soil slurry Cohesive soil
Caving soil Caving soil
(i) (ii)
Cohesive soil
Caving soil
Cohesive soil Sump
Caving soil
(iii) (iv)
Figure 17.4c Slurry method of construction and Methods of construction of drilled shafts
(Source: O’Neill and Reese, 1999)
17.3 CAISSONS
17.3.1 Uses
Caissons are generally used for major foundation works because of the high construction
cost. In general, a caisson foundation is recommended and found to be advantageous when
(i) large-size boulders are encountered and (ii) a massive sub-structure is required to with-
stand large lateral stresses. Caisson foundation is used for the following works:
1. Structures for shore protection
2. Docks, wharfs, and quay walls
M17_PURU1773_01_SE_C17.indd 650 2/12/2013 5:13:31 PM

3. Water front structures, such as large pump houses subjected to heavy vertical and
horizontal loads
4. Bridge piers and abutments in rivers, lakes, etc.
Structures on land are generally founded on piles or drilled caissons, but for structures
in water, in particular if the bearing stratum is 15 m below the water level, caissons are
generally less expensive.
17.3.2 Types of Caissons

Caissons can be classified into three major types: (i) open caissons, (ii) box caissons, and (iii)
pneumatic caissons.
Open Caissons. These are concrete or masonry shafts which remain open both at the top
and at the bottom during construction (Fig. 17.5a). An open caisson of heavy mass concrete
or masonry construction, containing one or more wells for excavations, is called a monolith
(Fig. 17.5b). The conventional well foundation adopted in India is essentially an open cais-
son with minor changes in the construction procedure. The caisson is sunk into place, as
the soil is removed from the inside, till the well sinks to the required depth. Then, a bottom
concrete seal is made by depositing concrete. The well is pumped dry, after maturity of the
bottom concrete seal, and filled with concrete or sand.
Box Caissons. These are structures with a closed bottom (Fig. 17.5c). They are constructed
on land and transported and sunk in to the prepared foundations below the water level.
Pneumatic Caissons. These caissons have their top closed, and compressed air is used to
stop the entry of water into the working chamber (Fig. 17.5d). Thus, excavation and concret-
ing are done in a dry condition. The caisson is sunk as the excavation proceeds, and after
reaching the required depth, the working chamber is filled with concrete.
17.3.3 Bearing Capacity and Settlement of Caissons

Computation of bearing capacity and settlement of caissons is similar to that of drilled piers.
17.3.4 Construction Procedure of Caissons

Construction of caissons requires heavy engineering equipment. A variety of usual and
unusual problems have to be faced during the construction of caissons. Caissons may be
constructed in slipways, on barrages, or on sand islands. Sometimes false bottoms are made
to aid floating. For sinking the first few lifts of caissons, guide piles are commonly used. Dur-
ing the sinking process, if a hard or firm material is encountered under water, blasting may
be necessary.
Open-well-type caissons have been used in India for many centuries for the foundations
of river bridges. In the early days, the masonry of the wells was built on timber curbs and
the caissons sunk by hand excavation from within the wells. Nowadays, the design and con-
struction of well foundations in India have attained perfection. A detailed treatment of well
foundations, as designed and constructed in India, is given in the next section.
M17_PURU1773_01_SE_C17.indd 651 2/12/2013 5:13:31 PM

Dredging wells sealed at bottom Dredging well

after completion of sinking
(a) Open caisson (b) Monolith
Air locks
Air staft
Dredged bed
(c) Box caisson (d) Pneumatic caisson
Fig. 17.5 Types of caissons (Source: Thomlinson, 1986)
Open caissons and monoliths are provided with a cutting edge at the bottom, forming the
lowermost portion of the shoe. The shoe has vertical outer steel skin plates (Fig. 17.6) and
sloping inner steel haunch plates. Steel trusses are used in both directions for proper brac-
ing. Trusses prevent the distortion of the shoe during fabrication, during towing to site, and
during the early stages of sinking. After initial sinking, the space between the skin plates is
filled with concrete (called steining). Over this rigid shoe, the steining is extended by placing
reinforced concrete between form-work. Now the caisson sinks under its own weight while
the soil is being excavated from the dredging wells. As the sinking operation progresses,
additional lifts are successively installed. When the required foundation level is reached,
the bottom of the caisson is plugged by a concrete seal by depositing under water a layer
M17_PURU1773_01_SE_C17.indd 652 2/12/2013 5:13:32 PM

Pier
High water level
Concrete Dredging wells

cast in situ
Concrete
steining
Shoe in steel
construction Outer skin
plates
Cutting edge Elevation Haunch plates
Outer skin
plates Dredging
wells
Concrete
steining
Plan
Fig. 17.6 Details of an open caisson (Source: Tomlinson, 1986)
of concrete in the bottom of the wells. The wells are then pumped dry and more concrete is
placed. Caissons can be further filled with concrete or sand depending on the condition fol-
lowed, by a top concrete seal. During the process of grabbing under water in loose and soft
materials, there is a possibility of surging and inflow of materials beneath the cutting edge.
Further, as the sealing is done under water, open caissons have the disadvantage that the soil
or rock at foundation level cannot usually be inspected before placing the sealing concrete.
Open caissons are suitable in soft clays, silts, or gravels since excavation by grabbing will
be easiest.
Monoliths are unsuitable for sinking in deep soft deposits because of their weight. They
are usually used in quay walls where their heavy weight is favourable for resisting lateral
and impact forces.
Reinforced cement concrete is used for making box caissons. They are constructed on
land and floated to position after the concrete has strengthened. A box caisson during float-
ing should be sufficiently safe against the danger of tipping or capsizing. The stability of a
floating caisson can be analysed using the principles of hydrostatics. Box caissons can also
be floated in horizontal or inverted positions and placed at the required location in a normal
position. They can be founded on dredged gravel or rock formation, on crushed rock blanket
over rock surface, or on piled raft. Box caissons are not suitable in sites where erosion can
undermine the foundations. Figure 17.7 shows certain methods of founding box caissons.
Pneumatic caissons are used in situations where the soil flow into the excavated area
is faster than it can be removed. They are also used in varying soil conditions. Pneumatic
M17_PURU1773_01_SE_C17.indd 653 2/12/2013 5:13:32 PM

Original Crushed Removed

surface rock material
blanket Sand fill
Rock or dredged gravel Concrete cast under

water in soft material
(a) On dredged gravel or (b) On crushed (c) On pile raft
rock formation rock blanket
Fig. 17.7 Methods of founding box caissons (Source: Tomlinson, 1986)
caissons have several advantages over open or box-type caissons. For instance, excavations
can be carried out by hand in a dry working chamber, soil condition can be inspected at the
foundation level, and foundation concrete can be placed under ideal conditions. Compared
to open caissons, the process of sinking is slow and needs elaborate equipment.
A pneumatic caisson requires airlocking devices, a decompression chamber, a working
chamber, and a means for the workers to get to the chamber. There should be a provision
to remove the excavated material and sufficient continuous power to maintain constant air
pressure. The essential features of a pneumatic caisson are shown in Fig. 17.8. Since a pneu-
matic construction is expensive, the construction should start with the open method and con-
tinue as deep as practicable. When the condition requires a pneumatic operation, air shafts
are introduced and the roof of the working chamber is formed either by concrete or by steel
Hoisting rope
Muck lock
Man lock
Air supply
Skin plating
Muck
bucket
bracing
Ladder
Air shaft
Caisson shoe
Working chamber
Cutting edge
Fig. 17.8 General arrangement of a pneumatic caisson (Source: Tomlinson, 1986)
M17_PURU1773_01_SE_C17.indd 654 2/12/2013 5:13:33 PM

diaphragm. Then, water from the working chamber is removed by applying compressed
air, followed by pumping if necessary. The excavation proceeds and the excavated material
is removed from the air shaft by bucket devices or liquefied if feasible and pumped out,
or if the soil is loose and sandy, it can be blown out using air pressure. As the excavation
progresses, additional sections of shafts should be added, and the skin friction can be over-
come by installing jets on the shaft walls. If sinking stops due to build-up of skin friction,
it can be revoked by a process known as “blowing down”. In this process, the air pressure
is reduced to increase the effective weight of the caisson, thus increasing the sinking effect.
When sinking reaches the required depth, concrete is placed to fill any base defects and also
the remainder of the working chamber. The space between the roof of the working chamber
and the concrete is filled with high-pressure grout. The rest of the procedure is similar to
that for the open caisson. During the whole operation, the chamber pressure is maintained
around 100 kN/m2 above the atmospheric pressure. When chamber pressure of about 300
kN/m2 is required, the workers should not be allowed to stay for more than 1½ to 2 hours.
17.4 WELL FOUNDATIONS

Well foundations are constructed either on dry ground or over an artificially formed island.
The curbs are pitched in the correct position and then sunk into the ground to the desired
level by grabbing the soil through the dredge holes formed by the masonry or concrete of
steining. Caisson, in the Indian context, is a type of well foundation, and it is distinct owing
to the method of commencing construction. That is, caissons are foundations for which the
skin of the portion covering the curb and some length of steining of the well (to first cover
the depth of the standing water) is fabricated or cast outside, floated to the final location, and
lowered in position there. In India, this procedure of initial sinking is referred to as the cais-
son method.
Well foundations have all the advantages of open caissons. Additionally, in the Indian
environment there is a lot of skill available. Further, under the present Indian conditions,
well foundations are claimed to be the most economically viable method as cost of labour is
cheaper compared to mechanical equipment (Ponnuswamy, 1986).
17.4.1 Types of Wells

Wells have different shapes, and accordingly, they are named as circular wells, twin circular
wells, double-D wells, double octagonal wells, single and double rectangular wells, and
multiple dredged holed wells. Among these, most common types used in India are circular,
twin circular, and double D.
Circular Wells. These wells have uniform strength in all directions (Fig. 17.9a). As the
weight per square of peripheral surface is highest, the sinking process is easy. They are
simple in construction. Tilts and shifts can be controlled and corrected easily in this type.
Twin Circular Wells. In this type, two independent wells are placed close to each other
and combined with a common well cap (Fig. 17.9b). These wells have all the advantages of
circular wells. Further, the diameter and spacing of the two wells can be adjusted so as to
accommodate the width and length of the pier. The minimum clear space between the well
periphery is 0.6 to 1 m for a 6 to 7 m depth of well and 2 to 3 m for greater depth. This type
M17_PURU1773_01_SE_C17.indd 655 2/12/2013 5:13:33 PM

8,530
2,130
4,070
All dimensions
14,530
in millimetres
2,130 2,130
4,270
2,130
(a) Circular (b) Twin circular (c) Double D
Fig. 17.9 Types of wells (Source: Ponnuswamy, 1986)
of well is used with advantage when the sinking depth is less and hard foundation material
is available.
Double-D Wells. This type is commonly adopted for deep foundations and major bridges
with multiple land/line traffic (Fig. 17.9c). The length of the well is restricted to twice its
width. Because of the presence of two wide dredge holes, casting and sinking are done more
efficiently. In large-depth wells, the possibility of cracks is more due to the large bending
moment. Further, dredging of the comers of the wells also poses some problem.
17.4.2 Components of Well Foundations

A typical sectional elevation of a well foundation is shown in Fig. 17.10. The bottom of a well
structure is suitably tapered to end in a steel cutting edge. This tapered portion, called a well
curb, is sufficiently strengthened with heavily reinforced concrete so as to take heavy load-
ing. The main body of the well is called steining. The materials used for steining are brick or
stone masonry, mass or reinforced cement concrete. After completion of the sinking at the
stipulated level, the dredge hole is cleaned and filled with cement concrete. This bottom seal
is called bottom plug. The remaining portion of the well may be filled partially or fully with
saturated sand, water or left hollow. At the top finishing level, another plug is provided with
plain or reinforced cement concrete; this is referred to as the top plug. This is provided to
transmit the imposed load uniformly to the wall of the well. The topmost layer, called a well
cap, is provided to accommodate the pier based on its shape and transmit the load to the
steining. In certain cases, a lean top plug or no top plug is provided, but a well cap is invari-
ably provided.
Brief design procedure of different components of well foundation is presented below
based on the detailed treatment made by Ponnuswamy (1986).
17.4.3 Design of Wells

This basically involves finding the depth of the well, the size of the well, and the design of
the other components.
Depth of Scour. As well foundations are constructed in river beds, they should be taken
to a safe depth well below the anticipated scour level. Scour around piers depends on
M17_PURU1773_01_SE_C17.indd 656 2/12/2013 5:13:33 PM

Pier cap
Pier
Well cap
Top plug
Steining
Bond rod
Sand filling
Bottom plug
Well curb
Cutting edge
Fig. 17.10 A typical section of a well foundation
several factors like flood discharge, the angle of attack of the flow, flow obstruction, etc. It is
generally assumed that the flow at the high stage of the river is straight. The most common
method is Lacey’s formula, which is given as
q1/3
Ds = 0.473 (17.15)
sf
where Ds is the scour depth (m), q the design discharge (m3/s), sf the silt factor (= 1.76Dm),
and Dm the diameter of sand in the river bed (mm).
Grip of Foundation. The foundation should be taken well below the scour level to pro-
tect it from any movement due to the force of the stream flow and other external forces. The
Indian Road Congress (IRC) method (reported by Ponnuswamy, 1986) recommends a grip
of foundation (grip length) D equal to 1/3Ds below the maximum scour level. However, the
adequacy of the depth has to be checked and found stable against transverse/longitudinal
forces by developing sufficient skin friction and passive earth resistance.
The depth of the foundation can be reduced if rock is met with. In a majority of the cases,
wells are constructed on rocks and keyed to a minimum depth of 30 cm. If rock is available at
a shallow depth and the mass of the well is not adequate to withstand the effect of longitudi-
nal/transverse forces, the well is anchored down using mild steel (MS) rods or high-tensile
wires. In soft rocks, the well is taken in the strata up to a sufficient depth.
Size of the Well. The practice of the Indian Railways is to design the wells based on
point bearing resistance, whereas the IRC’s practice is to consider both the skin friction and
the point resistance. As the major portion of the load is transmitted by bearing, the size is
decided based on point resistance only. Further, the well has to feasibly accommodate the
pier with less cantilevering.
Safe bearing pressure for rocks can be found, as given in Chapters 15 and 23, based
on either the field tests or the laboratory compression tests. A factor of safety of 6 to 8 is
adopted. For disintegrated rocks and various soft varieties of rocks (where the recovery
M17_PURU1773_01_SE_C17.indd 657 2/12/2013 5:13:34 PM

ratio is less than 35%), the safe bearing capacity is computed considering such rocks as soils.
For bearing capacity and settlement of caissons, the procedure followed for piers can be
adopted. Instead, the generalized bearing capacity formula (Eqs. 15.13 and 15.14) for base
resistance may be used. The minimum factor of safety to be adopted for well foundations
on soils is 2.0.
The size of the dredge hole of a well varies. In small and shallow wells, the minimum
diameter of the dredge hole should be 1.8 m. In larger wells, where dredgers and chisels of
large sizes are involved or pneumatic sinking is resorted to, the minimum size of the dredge
hole should be 3 m. The final size is decided after satisfying the lateral stability condition of
the wells.
Steining. This is subjected to different types of stresses. At the sinking it is subjected to
water and earth pressure. At the dredging stage, inside surface is subjected to water pressure
while outside surface to the earth pressure. The net pressure diagram for design requirement
is shown in Fig. 17.11a.
Steining walls are subjected to vertical compressive forces. In situations where the stein-
ing is passing through a stiff clay followed by a soft stratum, there is a possibility of the bot-
tom portion of the well getting dropped at the soft layer interface due to inadequate tensile
strength of the well. Such dangers may be avoided by providing adequate bonding roads or,
alternatively, the well should be reinforced. Some of the thumb rules recommended by the
IRC for fixing the thickness of the steining are given below:
1. Cement concrete steining
(i) For circular and dumbbell-shaped wells
T = k(0.01DH + 0.1De ) (17.16)
where
k = 1.1 for sandy, silty, and soft clayey strata
= 1.25 for hard strata including hard clay, boulders, kankar, shale, etc.
DH = height of the well
De = external diameter of the well
(ii) For rectangular and double-D wells
T = k(0.01DH + 0.12) (17.17)
Bed level
De
Water level h
Di
W
h H
q O
R
g h¢ De – external diameter
w
k (g h + g h¢ ) D i – internal diameter
(a) Pressure distribution on steining (b) Force acting on curb
Fig. 17.11 Pressure distribution on well steining and curb (Source: Ponnuswamy, 1986)
M17_PURU1773_01_SE_C17.indd 658 2/12/2013 5:13:34 PM

where
k = 1.0 for sandy strata
= 1.1 for soft clay
= 1.15 for clay
= 1.20 for boulders, shale, kankar, etc.
2. Brick steining
⎛D D ⎞
T = k ⎜⎜ e + H ⎟⎟⎟ (17.18)
⎜⎝ 8 40 ⎠
where
k = 1.0 for sand
= 1.1 for soft clay
= 1.25 for hard clay
In order to keep on sinking continuously, the mass of the well should be greater than the
skin friction. Otherwise, the well will reach a floating condition, and continued scooping
may induce blowing of sand. This can be avoided by loading the wells with kentledges at
the top to assist in sinking.
For a light structure, brick steining is used. In heavier and deeper structures, reinforced
cement concrete or plain cement concrete with suitable bond rods is provided. It has been
reported (Ponnuswamy, 1986) that M10 or M15 concrete wherever used is normally ade-
quate.
Curb. The curb of a well transfers all the superimposed loads to the soil through the cut-
ting edge while sinking. The material used for curbs may be timber or reinforced concrete.
The present-day practice recommends heavy RCC well curbs with steel cutting edges for
any type of well. Hoop tension is caused in the well curb due to bearing pressure, and suit-
able reinforcement has to be provided to withstand the same. The forces acting on the well
are shown in Fig. 17.11b.
The total horizontal force on the well curb on both sides is
De + Di
W cot θ
2
where Di is the internal diameter of the well, W the weight of the well and curb per unit
length along the centre line of the steining, and θ the internal angle of the well. The force act-
ing on the well curb on one side is half of the above.
Well curbs should be reinforced to resist these forces. A typical arrangement line of rein-
forcement in a well curb is shown in Fig. 17.12.
Cutting Edge. The cutting edge is provided at the bottom of the well below the curb to cut
through the soil during sinking. It is generally made of steel and welded to an angle iron to
fit the outer dimensions of the well steining.
During routine sinking in sand or clay, only very little stress is induced. But while rest-
ing on rock or pushing through boulders or cutting through logs of wood, high stresses are
induced. The height he of cutting edge is given as
qu t
he = (17.19)
fc tan θ
M17_PURU1773_01_SE_C17.indd 659 2/12/2013 5:13:35 PM

Cover Cover
Bond rod
Circumferential rods
Single-edged stirrups
Triangle-shaped
MS plate Diaphragm
MS plate
MS angle
Fig. 17.12 A typical well curb (Source: Ponnuswamy, 1986)
where qu is the crushing strength of rock, t the thickness of the cutting edge, and fc the safe
compressive stress of concrete.
The value of θ is usually taken as 30°. The choice of this angle has been proved to be suit-
able for easy access to the cutting edge. The thickness of the outer plate should not be less
than 12 to 18 mm depending on the size of the well. Further, the unsupported plate height
is limited to 7.5 to 10 cm.
Bottom Plug. After final grounding of the well to the required foundation level, a con-
crete plug is provided (Fig. 17.13). The bottom plug transfers the entire load, viz., the weight
of the steining, the superimposed load, and the weight of saturated sand filled in the well, to
the ground. The bottom plug functions as an inverted dome supported along the periphery
Bottomless box of concrete

bag sunk up to rock
Rock Sand
Exposed at the level of

the cutting edge
Wall of bottomless boxes

of concrete bags
(a) Dressing and treating bottom of well on rock
Bottom Bottom
plug plug
Sand
Piles
Small-diameter wells
Filled with concrete
(b) Dressing bottom before plugging bottom
Fig. 17.13 Bottom concrete plug
M17_PURU1773_01_SE_C17.indd 660 2/12/2013 5:13:36 PM

Anchor rods
Pier
Well cap
Top plug
Sand filling
Fig. 17.14 Top plug and well cap
of the steining. As it is not feasible to provide reinforcement at the bottom, it is generally

made thick, and a rich concrete mix (M20) is used.
Sand Filling. The bottom plug concrete is cured, and after curing, the well is filled with
sand in a saturated condition. Sand filling, although not mandatory, provides stability to
the bottom of the well, helps in the elimination of tensile forces at the base, and also can-
cels the hoop stresses induced in the steining. In earthquake-prone areas, it is preferable to
make the well lighter by filling with saturated sand up to the top of well and then casting
the top plug. But on the road bridges in India, either the well is kept empty or the filling is
done only up to scour level.
Top Plug. A top plug is provided after the filling is completed. This enables transfer of the
load of the pier and the superimposed load to the well steining (Fig. 17.14). The thickness of
the top plug is generally kept greater than 50% of the smaller dimension of the dredge hole.
If a sand filling is provided, no reinforcement is needed, and a plain cement concrete (PCC)
mix of 1:2:4 is enough. In the case of larger dredger holes, it has been observed that the top
plug provided is not sufficient but needs a reinforced well cap to transfer the superimposed
load. In such cases, a lean concrete (1:3:6) top plug is provided, which forms an even base
for the well cap.
Well Cap. As the shapes of the pier and the well are different, the well cap (entablature)
forms an interim transition layer to accommodate the pier. The well cap is so designed that
the base of the pier is provided with a minimum all-round offset. Further, the centre of the
cap is made to coincide with that of the pier and not with that of the well. Such positioning
nullifies the effect of the minor shifts which might have occurred during well sinking.
17.4.4 Stability Analysis of Well Foundations (Approximate Solution)

A well foundation supporting a bridge pier is subjected to vertical and horizontal forces. The
forces causing instability (disturbing forces) of a well are self-weight of the well and its
superstructure, live loads, water currents and buoyancy, forces due to environmental factors
such as temperature, wind, and earthquake, breaking and tracking forces, forces on account
of resistance of the well walls, and base and skin friction.
The stability requirement is to find the depth of the well below the maximum scour level
(i.e., the grip length or grip of foundation) after allowing a suitable factor of safety after sat-
isfying the static conditions.
Terzaghi (1943) gave an approximate solution based on the analysis of the free rigid bulk
head. For the analysis, the critical combination of forces is considered.
M17_PURU1773_01_SE_C17.indd 661 2/12/2013 5:13:36 PM

PV Pier
PB PL
H
Direction of
flow of water
Scour line
D
Well
Fig. 17.15 Forces acting on a well
Resolving all the forces in the vertical direction, one resultant force is obtained, PV. Simi-
larly, resolving the forces in two horizontal directions across and along the pier, two resul-
tant horizontal forces, PB and PL, respectively, are obtained (Fig. 17.15).
The critical force system to be considered for stability analysis is the one in the direction
perpendicular to the water flow (i.e., in the direction of transverse axis of the pier). Thus, the
resultant vertical force PV and the resultant horizontal force PB are considered for analysis.
Considering a state of plastic equilibrium, the forces and the earth pressure distributions
acting on the well are shown in Fig. 17.16.
Pressure at any depth z below the scour level is
p = γ z (K p − Ka ) = γ zK ′ (17.20)
z = DPD = γ DK ′ (17.21)
The well is assumed to fail as soon as the soil reaction at the bottom is equal to PD. For
equilibrium at that instant,
(PB )max = resultant of total pressure per unit length
= area of ΔAEF − area of ΔBCF
= 12 γ D2 K ′ − 12 2γ DK ′D1
Therefore,
(PB )max = 12 γ DK ′(D − 2D1 ) (17.22)
Taking moment about E,

1 D 1 D
(PB )max × H1 = γ D2 K ′ − × 2Dγ K ′D1 × 1
2 3 2 3
M17_PURU1773_01_SE_C17.indd 662 2/12/2013 5:13:37 PM

(PB)max
r1
H1
B
D
D1
O
C E
r2 F
PD = g DK′ PD = gDK ′
Fig. 17.16 Forces on well considered for analysis
Solving for D1,
2D1 = 3 H1 + 9 H12 − 2D(3 H1 − D) (17.23)
Substituting Eq. 17.23 in Eq 17.22, an expression in depth D is obtained. This is the grip
length required to sustain the maximum horizontal force.
A safe depth can be obtained by reducing PD by a factor of safety, F, i.e., considering PD/F.
This theory is based on the further considerations that (i) the well is treated as a light bulk
head, (ii) Kp and Ka are Rankine’s earth pressure coefficients, and (iii) there is no friction at
base and wall. Omission of these frictional forces yields a conservative (PB)max .
If ρ1 and ρ2 are the horizontal displacements, then the angular deflection of the centre line
of the well, δ, is given as
1
tan δ = (ρ1 − ρ2 ) (17.24)
D
The above analysis may be applied to the force system of a well provided the movements
due to side friction and the resultant base reaction are ignored. This omission is on the safe
side. The possible vibrations due to traffic over the bridge and water currents may minimize
the side friction. Further, the deflection of a heavy well may be small, and thus the passive
resistance will also be less. Thus, Rankine’s coefficient of earth pressure is justifiable as it
gives a conservative result. Equation 17.22 represents the maximum equivalent resisting
force per unit length of well due to earth pressure.
17.4.5 Stability Analysis of a Heavy Well

In the Terzaghi approximate analysis, it is assumed that the bulkhead tends to rotate about
some point O above the lower edge and tends to transfer the soil from elastic to plastic equi-
librium.
M17_PURU1773_01_SE_C17.indd 663 2/12/2013 5:13:37 PM

In the case of a heavy well embedded in cohesionless soil, the well is assumed to invariably
rotate about its base, and the assumed pressure distribution is given in Fig. 17.17a. Taking
the moments about the base, the value of (PB)max is obtained as
1 D3
(PB )max = γ ′(K p − K a )
6 H +D
Normally, around the well, scouring takes place. Beyond the well surroundings, the
uncovered soil acts as a surcharge. The surcharge depth D2 is very difficult to assess and
may be assumed to be equal to half the normal depth of scour. The pressure distribution is
shown in Fig. 17.17b. The equivalent maximum resistance force is then given as
1 D2 (D + D2 )
(PB )max = γ ′K ′
6 ( H + D)
If d is the diameter or length of the well, the total resisting force after allowing a factor of
safety, F, is given as
(PB )max d
Pa =
F
(PB)Max.
(PB)Max.
H Bed
Max. scour level
level
y Scour level D2
H1
Assumed
pressure
variation
D W
D
1/2 g¢(Kp–Ka)g ¢D 2
D/3 D/3
Base
friction (Kp–Ka)g¢D R (Kp–Ka)g ¢(D + D2)
(a) Tiliting of a heavy well about base in (b) Effect of surcharge on a well tilting about
cohesionless soil its base
Fig. 17.17 Stability analysis of a heavy well
M17_PURU1773_01_SE_C17.indd 664 2/12/2013 5:13:38 PM

The factor of safety should not be less than 2.

The maximum pressure f at the base of the well for the no overturning moment condition is
W
f=
A
where W is the net direct load on the well base, after making allowance for buoyancy and
skin friction; A is the area of the well base; and z is the section modulus of the well base.
This maximum foundation pressure should be kept within the safe bearing capacity of the
soil, assuming no tension occurs at the base.
The maximum moment on the steining occurs where the resultant shear force is zero.
If the shear force is zero at a depth y below the maximum level, then
γ ′K ′y 2 d
Pa =
2F
That is,
2 FPa
y=
γ ′K ′d
17.4.6 Construction of Well Foundations

If the well has to be laid on a dry ground, the soil is excavated and levelled 15 to 30 cm above
the surface water level. The line of the cutting edge is marked with reference to the centre of
the pier and the axis. In case the well has to be in water, an artificial island is formed and
compacted. The marking is made over the compacted and levelled layer. Wooden blocks of
sand bags are placed tangentially or along the cutting line to prevent unnecessary settlement
of failure. The cutting edge is assembled on the wooden or sand bags which will fall inside
the dredge hole when grabbing starts.
The inner conical wooden shuttering is placed, followed by curb reinforcement. The cut-
ter steel shuttering is positioned, and the curb is concreted. The concrete mix should not be
leaner than M15. The inner conical shuttering is removed after 72 hours and the others after
48 hours.
The steining reinforcement or bond rods are placed, and additional shuttering is posi-
tioned and the next stage of steining concreted with a 1:3:6 mix; for brick or stone masonry
1:2 or 1:3 cement mortar is used. Two-metre height of steining or curb has to be built before
open sinking.
The sinking may be done by sending men down and excavating the soil with shovels.
The excavated soil is removed using baskets, and sinking is continued till the curb sinks to
a depth of 0.5 to 1.5 m. Additional steining can be raised in heights of 1.2 to 1.5 m. At least
24 hours’ timing is given before each stage of addition if ordinary cement is used, otherwise
a longer setting time has to be allowed. The steining heights can be increased to 3 m after
attaining a grip depth of about 6 m. The verticality of the well has to be ascertained at each
stage.
M17_PURU1773_01_SE_C17.indd 665 2/12/2013 5:13:39 PM

Manual grabbing and removal of earth can be resorted to only for small depths. For
greater depths and where subsoil water is encountered, dredgers have to be resorted to.
The most commonly used dredger is the bells dredger. The dredgers are lowered into the
well from overhead derricks or winches through pulley blocks are to be adopted. For the
purpose of hoisting the grab and removing the material, different types of derricks are used.
For minor jobs, timber Scotch Derrick is used. The dredging operation is stopped when the
required height less the minimum free board is attained. The steining height is raised up to
the free board level and then concreted.
The sinking process may be hastened by temporarily reducing the skin friction. The
outer surface of the steining is finished smooth. The surface may also be coated with coal
tar or bentonite solution. Alternatively, air or water jets may be used, which is an effective
method of reducing skin friction.
The rate of sinking has to be monitored, and an advance work chart should be available
for this purpose. The following rates of sinking (cm/day) have been recommended by
Ponnuswamy (1986):
Medium-sized well through sandy strata: 60 to 90 cm/day
Medium-sized well through clayey strata: 40 to 50 cm/day
Large-sized well through sandy strata: 50 to 60 cm/day
Large-sized well through clayey strata: 30 to 40 cm/day
Large-sized well through rocky/hard strata
by driving: 10 to 15 cm/day
by pneumatic sinking: 15 to 25 cm/day
After the required foundation level is reached, the bottom of the well has to be dressed
and treated before plugging. Figure 17.13a shows the condition of the well sunk in sandy
strata. When sinking is done by de-watering or by a pneumatic process, the bottom can be
dressed by sending men down to the bottom. Shallow wells and those just resting on rock
may be anchored with the help of mild steel dowels fixed to the rock before plugging. In case
there is difficulty in keying in the rock, the arrangement shown in Fig. 17.13b and c may be
adopted.
After curing of the bottom plug, the well is filled with sand or water; the top plug is pro-
vided followed by the well cap as discussed earlier.
17.4.7 Shifts and Tilts

In the process of sinking a well, it may get shifted from the true alignment or get tilted due
to non-uniform sinking. Shift and tilt of each well have to be measured regularly during the
entire operation. At each stage of casting of steining, observations have to be taken to this
effect.
As per the IRC (1986) norms, the tilt should not exceed 1 in 80 and the shift should not
be more than 5% of the maximum outside dimension of a well subject to a maximum of
150 mm.
Necessary corrective measures have to be taken while sinking proceeds to keep the tilts
and shifts within permissible limits. The following measures can be employed to contain the
tilts and shifts within the limits:
M17_PURU1773_01_SE_C17.indd 666 2/12/2013 5:13:39 PM

1. Regulation of dredging: The dredging is made at the higher side of the well. This may be
done manually after de-watering. If the well has sunk to a great depth, this method will
not be effective.
2. Eccentric loading: In this case, a bracket is provided at the top of the well on the higher side,
and sand bags or other heavy materials are placed on the bracket. This causes eccentric
loading, and minor tilts are rectified.
3. Blocking and hooking: Sometimes a wooden piece is placed temporarily under the cut-
ting edge of the lower side to avoid further tilt while other methods of rectification are
attempted at the higher side. Instead of a timber piece, the cutting edge on the lower edge
may be hooked, and the hook is pulled of with the help of a winch if necessary.
4. Pulling the well: This technique is effective only during the initial stages of sinking. This is
done by putting one or more steel ropes round the well and pulling it towards the higher
side. Alternatively, the well may be pushed from the lower side using hydraulic jacks.
5. Water jetting and excavation: This method may be resorted to at the high side to reduce
skin friction. This technique alone can be effective unless supported by other correction
methods. Skin friction can be reduced by releasing the earth pressure on the higher side
by making an open excavation.
POINTS TO REMEMBER
17.1 Drilled piers are structural members of relatively large-diameter massive struts con-
structed of concrete placed in a pre-excavated hole.
17.2 Caissons or well foundations are structural boxes or chambers that are sunk in place
through ground or water by simultaneously excavating below the bottom of the
unit, which thereby descends to the final depth.
17.3 Types of drilled piers are straight shafted pier, belled pier, and pier socked into rock.
17.4 Drilled piers and caissons derive supporting power from both skin friction and bot-
tom bearing as in a pile. The contribution by end bearing capacity is much more than
by skin friction.
17.5 Settlement of piers and caissons in sands and gravels is mostly immediate, but in
clays the settlement may be immediate or long term.
17.6 Types of caissons are open caisson, box caisson, and pneumatic caisson.
17.7 Caisson, in the Indian context, is a type of well foundation and it is distinct owing to
the method of commencing construction. Well foundations have all the advantages
of open caissons.
17.8 Types of wells are circular wells, twin circular wells, and double-D wells.
17.9 Components of a well foundation are cutting edge, well curb, bottom plug, staining,
top plug, and well cap.
17.10 Design of a well foundation involves the following factors: (i) depth of scour, (ii) grip
of foundation, (iii) size of the well, (iv) steining, (v) curb, (vi) cutting edge, (vii) bot-
tom plug, (viii) sand filling, (ix) top plug, and (x) well cap.
M17_PURU1773_01_SE_C17.indd 667 2/12/2013 5:13:39 PM

QUESTIONS
Objective Questions
1. The drilled pier derives its maximum supporting power from the end bearing.
2. Monolith is a type of open caisson with less mass concrete or masonry.
3. Sinking of a pneumatic caisson is slow and requires elaborate equipment.
4. Steining is subjected to water and earth pressure during the sinking stage.
5. Wells located in earthquake areas are completely filled with sand and compacted
so that the well is made heavy.
17.2 Machine-excavated piers are designated as
(a) Chicago caisson (b) Gow caisson
(c) Drilled caisson (d) Open caisson
17.3 The critical force system to be considered for stability of a well foundation is the one
in the
(a) Vertical direction
(b) Direction perpendicular to the water flow
(c) Direction parallel to the water flow
(d) Horizontal direction
17.4 The stability of a floating caisson can be analysed by the principle of
(a) Hydrodynamics
(b) Hydrolysis
(c) Hydrostatics
(d) Hydromechanics
17.5 Pneumatic caissons are preferred in situations where the soil flow into the excavated
area is ______ than it can be removed.
(a) Faster
(b) Slower
(c) Initially faster
(d) Initially slower
17.6 Identify the wrong choice:
A cast-in-place broad pile with diameter greater than 75 cm is called a
(a) Caisson (b) Drilled shaft (c) Drilled pier (d) Short column
17.7 Assertion A: Unlike a driven pile, a drilled caisson does not compact the surrounding
area.
Reason R: Skin friction along the shaft of a drilled caisson is relatively small.
Select the correct code:
M17_PURU1773_01_SE_C17.indd 668 2/12/2013 5:13:39 PM

17.8 Identify the incorrect statement:

(a) Settlement analysis is necessary for drilled caissons resting on medium and soft
clays.
(b) Drilled caissons carried to materials other than hard rock are often enlarged at the
bottom.
(c) The drilled caisson shaft is designed in accordance with the principles of the long
column.
(d) Drilled caissons carried to bedrock are generally safe.
17.9 Describe the method of analysis of a heavy well as right bulk head.
17.10 Briefly explain the procedure adopted in well sinking and bring out the problems that
are encountered in open sinking.
17.11 Discuss in detail the methods to correct tilts and shifts of wells during sinking.
17.12 Explain briefly the method of assessing the load-carrying capacity of a well founda-
tion in a saturated stiff clay.
EXERCISE PROBLEMS
17.1 A bridge pier is to be founded at 5 m below the bed level of a river. The 8.5 m diam-
eter pier has a gross load of 28,000 kN. The mean high and low water levels are
8.5 and 3.2 m above the bed level, respectively. The saturated unit weight of the clay is
21.2 kN/m3. Adhesion between the clay and the surface of the pier is estimated as
36 kN/m2, and this is effective for a depth of 3.5 m from the base of the pier. Compute
the factor of safety against general shear failure at low and high water levels of the
river. Use a value of Nc = 7.5.
17.2 A reinforced concrete floating caisson has outside dimensions of 5.5 m × 5.5 m × 5.5 m.
The thickness of the well and the bottom is 400 mm. Examine whether this can provide
stable flotation in fresh water.
17.3 An open caisson has an inside diameter of 3.5 m and a length of 15 m. The caisson is
made of concrete, and it is intended to sink it by its own weight. Determine the stein-
ing thickness necessary.
M17_PURU1773_01_SE_C17.indd 669 2/12/2013 5:13:39 PM

M17_PURU1773_01_SE_C17.indd 670 2/12/2013 5:13:39 PM
18
Ground Investigation
CHAPTER HIGHLIGHTS
Planning the ground investigation programme – Types of samples – Indirect
methods: geophysical methods, sounding methods – Semi-direct methods:
wash boring, rotary drilling, auger boring – Direct methods: sampling process,
sample disturbance, types of samples, accessible explorations, undisturbed
sampling – Routine field tests – Recording of field data – Location, spacing,
and depth of borings
18.1 INTRODUCTION
Before undertaking any civil engineering work, a thorough investigation of the ground
is essential. Ground investigation refers to the methodology of determining surface and
sub-surface features in the proposed construction area.
Information on surface conditions is necessary for planning the accessibility of the site,
for deciding the disposal of removed material (particularly in urban areas), for removal
of surface water in water-logged areas, for movement of construction equipment, and
other factors that could affect construction procedures.
Information on sub-surface conditions is a more critical requirement in planning and
designing the foundations of structures, de-watering systems, shoring or bracing of
excavations, the materials to be used in construction, and site improvement methods.
The above information will enable a civil engineer to plan, design, and execute a
construction project. In a nutshell, the purpose of a ground investigation is to
1. determine the geological conditions of rock and soil formation,
2. establish groundwater level,
3. select the type and depth of foundation,
4. determine the bearing capacity of the site,
M18_PURU1773_01_SE_C18.indd 671 2/12/2013 8:20:43 AM

5. evaluate the anticipated settlement of the structure,

6. locate and select the materials of construction, and
7. locate suitable transportation routes.
18.2 PLANNING THE GROUND INVESTIGATION PROGRAMME

Basically, in a ground investigation, one is interested in finding out details about sub-surface
strata and their engineering properties, such as strength, deformation, and hydraulic char-
acteristics. The programme should aim at obtaining the maximum information with mini-
mum investment. For successful operation of the programme, the civil engineer must have
an adequate knowledge of the current exploration techniques and laboratory tests.
The ground investigation, irrespective of the magnitude of the project, consists of four
phases, which are discussed below.
1. Available Information. This is the first phase in which collection of published geological
and topographical information of the area, hydrological data, details of existing or historic
development, local regulations for construction activity, etc. are made.
2. Reconnaissance. This is the phase during which a first examination of the area is made by
the engineer along with other specialists, such as the geologist, land surveyor, geotechnical
engineer, etc. At this stage, a thorough study of the existing structures for the type of con-
struction and defects, such as cracks and settlement, soil profiles in highway or railroad cuts
and quarries, erosion in existing cuts, high-water marks on bridge abutments, rock outcrops,
and history of flood and scour levels from the local people are collected.
3. Preliminary Investigation. This is an important phase of the entire programme. In this
stage, the engineer planning the investigation programme should undertake two obliga-
tory steps (Terzaghi and Peck, 1967). As the subsoils were formed under different geological
environments, the first step towards a ground investigation is a thorough understanding
of the geology of the site, which enables an efficient working out of the investigation pro-
gramme. The second step is to obtain more details about the subsoil strata (e.g., thickness of
individual strata) from one or two exploratory drill holes. All further steps depend on the
magnitude of the job and the character of the soil profile.
4 Detailed Investigation. Additional borings are planned from the data obtained from the
preliminary borings. If the subsoil is uniform in stratification, an orderly spacing may be
planned. Often, additional borings are made to locate weak soil or rock zones, outcrops,
etc., which may influence the design and construction of the project. Necessary in situ tests
should also be performed. Sufficient samples should be procured to obtain relevant parame-
ters for design and construction. Certain additional samples should be recovered to redefine
the design or construction procedure.
18.3 TYPES OF SOIL AND ROCK SAMPLES

Soil and rock samples may be classified into the following three groups, based on the
condition of the material.
Non-Representative Samples. These samples comprise mixtures of materials from
various soil or rock layers. Here, there are possibilities of removal or exchange of some
mineral constituents by washing and sedimentation. Such samples, also called wash or wet
M18_PURU1773_01_SE_C18.indd 672 2/12/2013 8:20:43 AM

Ground Investigation 673
samples, are not a true representation of the material found at the bottom of the borehole.
These samples are unsuitable for identification and laboratory tests.
Representative Samples. These samples consist of constituent minerals from each layer
and are not mixed with material from other layers. The structure and water content of the
sample are seriously altered. The soil samples are suitable for identification and classifica-
tion purposes but not suitable for major laboratory tests. These samples are also referred to
as dry samples but the samples do contain some moisture.
Undisturbed Samples. These are the types of samples in which the material has experi-
enced so little disturbance that it is suitable for all laboratory tests. Thus, these samples may
be used for permeability, consolidation, and strength tests. The term undisturbed is mislead-
ing to some extent, and hence, it is appropriate to replace it with undistorted (Hvorselv, 1984).
18.4 INDIRECT METHODS OF SUB-SURFACE EXPLORATION

Indirect methods consist of geophysical and sounding methods. In these methods, depths to
the principal strata are established, based on some physical properties of the material, and
the measurements are made on the ground surface. No samples are obtained in the geo-
physical method, but in some sounding methods, representative samples are obtained.
18.4.1 Geophysical Methods

This method involves the technique of determining underground materials by measuring
some physical property of the material. Geophysical methods can be used rapidly and eco-
nomically for large areas of great linear extent. These methods are more suitable for recon-
naissance exploration, for location of general deposits of construction materials, and in
identifying the depth of ground-water. The results obtained are generally satisfactory.
Two types of geophysical investigations have been found to be useful for civil engineer-
ing works. They are electrical resistivity and seismic refraction methods.
Electrical Resistivity Method. In this method the differences in the electrical resistivities
of the various strata are detected. In general, both soil and rock are poor conductors of elec-
tricity, but because of the dissolved salts in the pore water, the conductivity of the ground is
improved. An electric field is produced in the ground by means of two current electrodes,
and measurements are made of the current (I) flowing between these electrodes and of the
potential drop (voltage, V) between a pair of intermediate electrodes. A schematic diagram
of the equipment (Wenner configuration) is shown in Fig. 18.1.
Battery I Ampere Measurement affected

V mostly by properties of
Volts soil in shaded area
a a
Soil surface a
Current flow a
lines
Equipotential
surfaces
Fig. 18.1 Schematic diagram of the resistivity method – Wenner configuration
M18_PURU1773_01_SE_C18.indd 673 2/12/2013 8:20:43 AM

Resistivity
Station points
Fig. 18.2 Resistivity traverse plot
The resistivity (ρ) of the soil or rock between the two electrodes with spacing a is given as
ρ = 2π aV / I (18.1)
Two different field procedures, viz., electrical profiling and electrical sounding, are in use for
obtaining information about surface conditions. The first method is suitable for establishing
boundaries between different underground materials, and has practical applications in the
study of variation of sub-surface condition with depth and in detecting layered deposits and
water-bearing strata.
In the electrical profiling method, four electrodes are kept at a constant spacing and
moved across the area, and resistivity measurements are made. The information resulting
from a profile line can be plotted with station points as the horizontal axis and resistivity
along the vertical axis. A change in the plotted curve indicates a change in the underground
materials (Fig. 18.2). From the series of profile lines, boundaries of areas underlain by differ-
ent materials can be established on a map of the area.
In the electrical sounding method, the electrode spacing a is progressively increased to
pick up changes in resistivity with depth. For homogeneous materials, half the electric cur-
rent flows to a depth of a to 1.5a and half to a greater depth. On the basis of field measure-
ments, the resistivity depth curves are plotted with resistivity along the horizontal axis and a
along the vertical axis (Fig. 18.3). Since the depth is directly related to electrode spacing, the
series of resistivity data obtained will indicate changes of resistivity with depth, and hence
provide information about the layering of material.
Average resistivity values for various rocks and minerals are given in Table 18.1 (IS: 1892, 1979).
Seismic Refraction Method. This is based on the fact that the velocity of a longitudinal
or compression wave in a material is a function of the modulus of elasticity, Poisson’s ratio,
and density of the material. The method of seismic refraction comprises generating a sound
wave by exploding a small charge of high velocity dynamite (a sledge hammer or a falling
weight may also be used) and then recording its reception at a series of geophones located
at various distances from the shot point (Fig. 18.4). The geophones transform the vibrations
into an electric current and transmit them to a recording unit or oscillograph equipped with
a timing mechanism. Only the arrival time of the initial impulse at each geophone is utilized.
The direct waves travelling with velocity, VT, in the top layer arrive first at the geophones
nearer to the shot point. The waves with the higher velocity, VB, reach the farther detectors
M18_PURU1773_01_SE_C18.indd 674 2/12/2013 8:20:44 AM

Resistivity
Top
soil
Sand
Electrode spacing a
Hard
sand- Hard
stone
Soft
sand-
stone
Soft
clay
Fig. 18.3 Resistivity–depth curve
Table 18.1 Resistivity values for rocks and minerals
Material Mean resistivity (ohm m)

Limestone (marble) 1012
Quartz 1010
Rock salt 106–107
Granite 5,000–106
Sandstone 35–4,000
Moraines 8–4,000
Limestones 120–400
Clays 1–120
Source: IS: 1892 (1979).
Shooting distance
Direct waves
arrive first Refracted waves arrive first
Shot Seismometers
point x y (Geophones)
Dry
loose
top soil
Stiff clay
Fig. 18.4 Seismic refraction method
M18_PURU1773_01_SE_C18.indd 675 2/12/2013 8:20:45 AM

ve
wa 1/V T
ec
t wave
ir = f r acted
x
D ope R e
S l
First arrival time curve

Time
1/V B
e=
Slop
y
Distance
Fig. 18.5 Distance–time plot
first by travelling downwards in the lower velocity material (say top layer), horizontally in
the higher velocity material, and then return to the surface.
The first arrival times are plotted against distance from the shot point (Fig. 18.5). This
gives two time/distance straight lines with slopes 1/VT and 1/VB. Let the distance of the
intersection point from the shot point be x; then, the thickness of the top layer is given by
1 ⎡ V − VT ⎤ 1/ 2
hT = ( x ) ⎢ B ⎥ (18.2)
2 ⎢V + V ⎥
⎣ B T⎦
This method can only be used when the wave velocity is successively higher in each layer.
Complications may be encountered when there is no sharp contrast in the velocities of the lay-
ers. In such cases, the travel–time diagram will be a curve. When inclined strata are encountered,
only the average depth can be determined. However, if the positions of shot point and detectors
are reserved, the actual depth and dip of the strata can be found. Further, in a multilayered series
of strata a blind layer may occur when its wave velocity is less than that in the overlying layer.
The range of velocities for different materials is given in Table 18.2 (IS: 1892, 1979).
Table 18.2 Range of shock wave velocities

for different materials
Materials Velocity (m/s)
Sand and top soil 180–365

Sandy clay 365–580
Gravel 490–790
Glacial till 550–2,135
Rock talus 400–760
Water in loose materials 1,400–1,830
Shale 790–3,350
Sandstone 915–2,740
Granite 3,050–6,100
Limestone 1,830–6,100
Source: IS: 1892 (1979).
M18_PURU1773_01_SE_C18.indd 676 2/12/2013 8:20:46 AM

18.4.2 Sounding Methods

Soil sounding or probing consists in forcing a rod, a rod enclosed in a sleeve pipe or a
cone or a sample into the soil and observing the penetration or withdrawal resistance.
Variation in this resistance indicates the existence of different soil strata, and the numeri-
cal values of the resistance permit an estimate of some of the physical properties of the
strata.
The oldest and simplest form of soil sounding consists in driving a rod into the ground
by repeated blows of a hammer. The penetration resistance of the rod may be directly
correlated with some physical properties of the soil based on local experience. Variations
in the diameter of the rod, in the driving force, and in the method of driving may affect the
penetration resistance, and hence, the assessed physical property. Since the friction acting
on the rod is cumulative with depth, the penetration resistance does not directly represent
the strength or density of strata. Methods are available to separate the resistance offered by
skin friction and point bearing. Comparisons of penetration resistance by rotation with the
properties of strata are also being used.
Standardized methods of sounding are the standard penetration and cone penetration
tests and are explained at the end of this chapter.
18.5 SEMI-DIRECT METHODS OF SUB-SURFACE EXPLORATION

Semi-direct methods are common boring and drilling methods combined with intermit-
tent sampling. The depths of different layers are ascertained by the rate of advancement
of boring tools or by means of non-representative samples obtained in the course of bor-
ing operations. Borings provide access to a particular layer for sampling. In general,
only major changes in the character of the sub-surface materials can be detected by
this method. The common boring methods are wash boring, rotary drilling, and auger
boring.
18.5.1 Wash Boring

This method requires equipment to lift, rotate, and drop a bit, and a water supply unit. Due to
the up and down movement of the chisel, the soil is loosened and broken up by the water jets.
Water is pumped through a hollow drilling rod and released under pressure. It is continuously
pumped till the soil particles are washed to the surface through the annular space between the
rod and side of the borehole. The water–soil suspension is collected in a sump. The settled mate-
rials may be used for visual inspection. Wash boring can be used in most types of soils.
An advantage of the method is that it produces a clear and fairly undisturbed environment at the
bottom of the borehole for obtaining undisturbed samples using samplers. The details of wash
boring are illustrated in Fig. 18.6.
18.5.2 Rotary Drilling

Although this method is primarily meant for rocks, it is also used in soils. In rotary drilling,
the borehole is advanced by rapid rotation of the drilling bit, which cuts, chips, and grinds
the material at the bottom of the holes into particles. Water or drilling mud is allowed
M18_PURU1773_01_SE_C18.indd 677 2/12/2013 8:20:46 AM

Derrick,
4 legs of pipe
Rope Swivel Tee, replaced by

Suction driving head when
hose driving casing
Pump Tub
Weight for driving drill rods

when wash point is replaced
Casing by sampling spoon.
Larger weight
used for driving casing
Wash pipe (drill rods)
Chopping bit, replaced by

sampling spoon during
sampling operations
Fig. 18.6 Typical arrangement for wash boring (Source: IS: 1892, 1979)
through the hollow rod, which then passes through narrow holes in the bit (Fig. 18.7). The
drilling fluid cools and lubricates the drilling tool, and, as in wash boring, the loose debris
are carried to the surface. The drilling mud also stabilizes the sides of the uncased borehole.
Cables
Tower mast
Water swivel Swivel hose
Stand pipe
Kelly
Yoke and Kelly drive

Drive rod
Rotary Stand pipe hose
drive
Hoisting drum Mud pump
Suction hose
Base plate
Slips
Return flow ditch
Hydraulic cylinder Spider

Casing
Drill pipe
Drill collar
Drill bit
Fig. 18.7 Typical arrangement for rotary drilling (Source: IS: 1892, 1979)
M18_PURU1773_01_SE_C18.indd 678 2/12/2013 8:20:47 AM

Open-hole drilling and core drilling are the two types of rotary drilling. Open-hole
drilling is used in soils and weak rocks, and in other circumstances, as explained above.
In core drilling, which is used in rocks and stiff clays, the bit cuts an annular hole in the
material and an intact core is removed as a sample.
Rotary drilling is best suited for boring of diameter more than 100 mm. The rate of
progress is greater in soils and rocks, and a uniform, clean hole with less disturbance of the
soil is generally produced. This method is not suitable for coarse gravel, boulders, and badly
fissured rocks.
18.5.3 Auger Boring

Hand-operated auger methods are suitable for identifying various types of soils with depth
and also for getting information about the depth to the groundwater table. For deeper
boring, solid or hollow-stem, continuous-flight augers (rotary augers) are frequently used.
Such auger drilling requires rotary drill equipment to provide both a push against the drill
head and a rotation. As the drill advances, additional auger flights are added and soil is
brought to the surface in a disturbed form. Figure 18.8 shows various augers.
In auger boring it is possible to identify even disturbed soils. Since the borehole is kept
dry, auger boring is particularly suitable for advancing borings above the water table to
obtain undisturbed, partially saturated samples. It further facilitates the determination of
free groundwater level.
18.6 DIRECT METHODS OF SUB-SURFACE EXPLORATION

Direct methods are boring and sampling methods which continuously provide representative
or undisturbed samples. All accessible exploration, such as test pits, trenches, large diameter
(a) Hand auger (b) Short-flight auger (c) Continuous-flight auger
Fig. 18.8 Types of augers
M18_PURU1773_01_SE_C18.indd 679 2/12/2013 8:20:47 AM

boring, shafts, drifts, etc., are grouped under direct methods. These accessible explorations
allow direct examination of strata in situ.
18.6.1 Sampling Process

Soil samples are recovered from the ground by utilizing special techniques and sampling
equipment. Typically, in a sampling process, a borehole is first advanced to the required
depth by using a suitable boring method and the bottom of the borehole is then cleaned.
Further, a soil sampler is advanced by driving it with a drop hammer or by pushing it with
a hydraulic piston or jack, and then the sampler is brought to the surface.
Thereafter, some soil is removed from each end of the sampler, a thin layer of molten wax
is applied to it and the ends of the tube are then covered by protective caps. All samples
(both disturbed and undisturbed) should be clearly labelled with details about project name,
date of sampling, borehole number, depth, and method of sampling. Extra care has to be
exercised in handling, transportation, and storage of samples prior to testing. The use of a
proper sampler and driving technique suited to the particular type of soil will reduce the
sample disturbance.
18.6.2 Sample Disturbance

It is impracticable to obtain a sample that is totally undisturbed, especially from deep holes.
During the boring and sampling processes, disturbances are bound to take place despite the use
of elaborate and careful sampling techniques. In the case of clayey soils, there is a possibility of
swelling due to stress relief. Soft clays are more susceptible to sample disturbances and the effect
is more in low plastic clays. The disturbance caused in driving the sampling tool depends on the
thickness of the tube and the manner in which it is driven. The rotation of the sample to free it
from the ground causes disturbance. Thus, extension collars are provided at the top and bottom
of the sampler to avoid disturbance.
The speed and continuity of motion with which the sampler is forced into the soil have
a great influence on the degree of disturbance. The various methods used to force a sam-
pler into the soil are hammering, jacking, fast pushing, single blow, and shooting. Ham-
mering eliminates the entrance of excess soil and is suitable only in hard or dense coarse
soils. It should not be used in soft or loose soils to obtain undisturbed samples. Slow jacking
allows plastic deformation and volume changes to take place. Fast pushing, single heavy
blow, or shooting produces longer and less disturbed samples when special vents are pro-
vided in the sample to accommodate the escape of water and air at high velocities.
The volume of soil displaced by the sample or the volume of the sample is represented by
a factor called the area ratio (Ar). It is given as
De2 − Di2
Ar = ×100% (18.3)
Di2
where De is the external diameter of the sampler and Di the internal diameter.
To reduce sample disturbance, this ratio should be as low as possible (generally, <10%)
but, at the same time, maintaining sufficient strength in the wall of the tube. Further friction
between the soil and inside tube wall would compress the sample. On the other hand,
M18_PURU1773_01_SE_C18.indd 680 2/12/2013 8:20:47 AM

the presence of large stones, roots, and other materials may tear the sample, causing it to
lengthen. A measure of such a disturbance is the recovery ratio (Lr), which is given as
Length recovered
Lr = (18.4)
Length pushed
Lr < 1.0 indicates compression and Lr > 1 shows expansion. Samples with a recovery ratio
much different from 1.0 are substantially disturbed. Thus, the use of an appropriate type of
sampler may reduce the sample disturbance to a great extent.
18.6.3 Types of Samplers

Open-Drive Sampler. It consists of a steel tube with a screw thread at each end. The lower
end is generally fitted with a cutting shoe but sometimes with an extension piece. The upper
end is fitted with a sampler head which incorporates a non-return valve. The non-return
valve allows air and water to escape while the sample enters the sampler and closes as the
sampler is raised to the surface, thus retaining the sample within the tube (Fig. 18.9). This is
the simplest and most common type of sampler.
Thin-Walled Sampler. To obtain undisturbed samples in soft to firm clays and plastic
silts, thin-walled samplers may be used (IS: 11594, 1985). No separate cutting shoe is attached
to the lower end, but the sampler’s lower end is itself machined to serve as a cutting edge
(Fig. 18.10). Good-quality undisturbed samples are possible if Ar < 10% and the soil is not
disturbed during the boring operation. This sampler can be used more conveniently in trial
pits and shallow boreholes.
Drill rod
Air release
valve
Sample
head
Sampling
tube
Casing
Dc
Shoe
Dw
Fig. 18.9 Open-drive sampler
M18_PURU1773_01_SE_C18.indd 681 2/12/2013 8:20:48 AM

Drill rod
Ball check valve
Vents
Sampler head
Set screens
Casing
Sampling tube
Cutting edge with

inside clearance
Fig. 18.10 Thin-walled sampler
Split-Spoon Sampler. It consists of a longitudinally split tube or barrel fitted with a shoe
and a sampler head with provision for air release (IS: 9640, 1980). The splitting aspect of the
sampler permits it to be opened for a sample examination and for onward transmission in
sample containers to laboratories (Fig. 18.11). Samples obtained using this sampler are rated
Coupling Head Split Liner Cutting
body shoe
(a) Assembly of split-spoon sampler
Square threads to suit Ball 25

'A' of coupling 4. Holes 10 20 32
30 Pin 3 20
50.8±0.2
38=0.2
45
41
51
20
22
12 56 23 20
75
55 100 25
(b) Head (c) Cutting shoe

All dimensions in mm
Fig. 18.11 Details of split-spoon sampler (Source: IS: 9640, 1980)
M18_PURU1773_01_SE_C18.indd 682 2/12/2013 8:20:49 AM

as representative. This sampler is suited for sands and is used only in the standard penetration
test (SPT). Split-soon samplers may be provided with a liner, which is a thin metal or plastic
tube fitted within the split spoon, in which case it is called a composite sampler. The purpose
of the liner is to protect the sampler during handling, shipping, and storage.
Piston Sampler. For very soft alluvial silts and clays, piston samplers are quite useful.
These consist of a thin-walled tube which includes a piston device that serves to push the
thin-walled tube into the undisturbed soil from the bottom of the boring (IS: 10108, 1982).
The piston is locked in the lower position and the sampler is lowered to the bottom of the
borehole. The piston is provided with a seal which prevents the entry of water and debris.
When it is unlocked, the tube is driven down into the soil to the full length of travel of the
piston. The whole assembly is withdrawn to the surface after locking the piston at the top
of the tube (Fig. 18.12). The sampler is separated from the sample head and the piston. It is
then sealed at both ends.
18.6.4 Accessible Explorations

Accessible exploration permits a direct visual examination of the subsoil and affords the
most complete information about the ground. For major works, the locations of these should
be decided in consultation with an engineering geologist. These are useful techniques, pro-
vided the overburden and groundwater conditions permit, and are universally adopted to
obtain the required information for prospecting of materials for dam and embankment con-
struction or concrete aggregates.
Trial Pits. Another method of sub-surface exploration is excavation of trial pits or test
pits. It is one of the most dependable and informative methods of investigation. It is limited
to a depth of 4 to 5 m. Relatively undisturbed samples can be obtained from walls or the bot-
tom of pits by pushing a thin-walled steel tube. In cohesive soils, block samples can be cut by
To the tower Static force to push

Piston extension rod the sampling rod
(through drill rod)
Drill rod
Casing pipe
Connection between
rod and head of
sample by piston
rod lock
Piston rod
Sampling tube
Piston
Piston in
stationary
position
(a) During lowering of (b) During penetrating the

sampler tube in soil
Fig. 18.12 Schematic diagram explaining principle of operation of piston sampler (Source: IS:
10108, 1982)
M18_PURU1773_01_SE_C18.indd 683 2/12/2013 8:20:49 AM

Original ground
surface
150 mm × 150 mm stud frames
to be spaced as nature of
ground requires
3,000 mm Backfill and tamp

Top 900 mm of crib
50 mm × 100 mm each
Ground Corner length optional
surface
Excelsior/Hay/
non-cohesive soil
(a) Sheeting and bracing (b) Typical test pit cribbing

for test pits
Fig. 18.13 Arrangement for protecting test pits (Source: IS: 4453, 1980)
hand from the sides or bottom. Trial pits are suitable for all types of soils, and permit a most
detailed examination of the soil formation for the entire depth.
Deeper pits have to be sheeted and braced (Fig. 18.13) or cribbed to prevent collapses
(IS: 4453, 1980). Ventilation of deep test pits is necessary to prevent accumulation of dead air.
This is done by providing pipes starting slightly above the floor and extending about 1 m
above the top of the pit. Special precautions have to be exercised if the presence of obnoxious
gases is anticipated (IS: 3764, 1966). A de-watering system has to be used if pits are to extend
below the water table.
Trenches. These are similar to test pits. They provide a long continuous exposure of the
surface of the ground along a desired line or section. They are best suited for exploration on
slopes. Necessary safety precautions have to be taken, as in deep test pits.
Drifts or Tunnels. These are employed to find the nature of strata and the structure of
particular geological formations. They are used to estimate the minimum excavation limits
to reach fresh and sound rock. Further, tunnels are helpful in loading buried channels, faults,
and other zones of weakness. Drifts are also used for conducting some in situ tests, such as
plate bearing test, jack and shear tests, etc. Drifts are usually provided with a low outward
slope for easy draining.
Generally, a rectangular section with minimum dimensions of 1.5 m width by 2 m height
is followed in hard rock (IS: 4453, 1980). An arched roof may be provided in soft rock. Neces-
sary supports should be provided wherever the ground is unstable (Fig. 18.14). Rock bolts
may be used to hold together the joined blocks of rocks. As the excavation in rocks is slow
and costly, this is resorted to only in major works. Ventilation by air from a compressor or
a blower may be provided for removing foul air or blast gases. Adequate lighting arrange-
ments are also provided for examination of the stratum.
M18_PURU1773_01_SE_C18.indd 684 2/12/2013 8:20:50 AM

Section-YY Wooden wedges

Wooden sleepers X
250 × 150
250
2100
125
Y 1500 Y
X Section-XX
Over break Drift 1500 × 2100
Fig. 18.14 Typical method of supporting weak zones in drift (Source: IS: 4453, 1980)
Shafts and Headings. Shafts or deep pits are advanced by hand excavation with suitable
sheeting. Headings or adits are horizontally excavated from the bottom of shafts, from the
surface of steeply sloped grounds, or from quarry faces. Shafts and heading are not exca-
vated from below the water table. These are very costly and used only in special investiga-
tions, such as pilot tunnels, mineral exploration surveys, etc.
Shafts may be circular (or rectangular) in section with minimum dimensions of 2.4 m ×
2.4 m diameter, so as to provide ample room for movement of men and machinery (IS: 4453,
1980). As in deep open pits, dead air or blast gases may be removed by stove pipes starting
above the floor and extending 1 m into open air above the mouth of the shaft. Air from a
compressor or blower may also be used. A pumping system should be used when water is
encountered.
18.6.5 Undisturbed Sampling of Soils

Undisturbed samples of soils are required for a number of tests. As discussed earlier, it is
impossible to get a truly undisturbed sample. But a minimally disturbed sample is possible
if certain procedures and precautions are observed. Although undisturbed sampling is
somewhat expensive and time consuming, these samples are more valuable. Sampling tech-
niques differ for fine and coarse-grained soils and are discussed below.
Undisturbed Sampling of Fine-Grained Soil. A thin-walled sampler with stationary pis-
ton is recommended for clay and silt formations (IS: 10108, 1982). A thin-walled sampler with
stationary piston, operated hydraulically is illustrated in Fig. 18.15. In this sampler, the piston
is positioned at the required depth and the sampler is pushed down – hence the name station-
ary piston. The sampler head is connected tightly with a drill rod. The thin-walled sampler
is connected to another piston head which moves in the pressure cylinder. The piston rod is
hollow and a hole is provided in it which helps to release the pressure and thereby avoids
over-driving. Operation of the system hydraulically minimizes the disturbance to the soil.
A borehole is made to the desired depth and cleaned. The borehole walls may be pro-
tected against cave-in by casing or bentonite mud. The fixed piston is kept at its lowest
point, thus closing the lower end and preventing the entry of any foreign matter into the
sampler. The different stages of sampling are illustrated in Fig. 18.15. The sample is teared at
M18_PURU1773_01_SE_C18.indd 685 2/12/2013 8:20:50 AM

Drill rod
Sampler head Ball check
Piston
Air vent
Pressure Water under
cylinder pressure
Hollow Water return
piston rod circulation
Hole in
piston rod
Fixed piston
Thin-walled
sampling
tube
Soil sample
(a) Sampler is set in drilled (b) Penetration of (c) Pressure is released

hole sampler tube through holding piston
into soil
Fig. 18.15 Diagrammatic sketch of hydraulically operated piston (Source: IS: 10108, 1982)
its bottom without causing any shock to the sample. In very loose sandy silt and silty sands,
particularly below the water table, a core catcher (Fig. 18.16) may be used to avoid loss of
samples while lifting.
Undisturbed Sampling of Sands. Cohesionless soils are always problematic as far as
undisturbed sampling is concerned. Thus, in situ tests are preferred in cohesionless soils.
Under favourable conditions, freezing is supposed to be the best method. By freezing,
the lower part of the sample is solidified, which makes it easier to retain in the sampler.
This is a very expensive technique. Two other methods of undisturbed sampling in unce-
mented sands are stationary piston sampling with drilling fluid circulation and compressed
Sampling tube
Threads
Core catcher
Cutting edge
Fig. 18.16 Core catcher fixed inside the cutting edge of the sampler (Source: IS: 10108, 1982)
M18_PURU1773_01_SE_C18.indd 686 2/12/2013 8:20:51 AM

air technique (IS: 8763, 1978). However, the sample obtained may be considered only as
relatively undisturbed. These samples are normally used to determine the in situ density.
The first method works on the same principle as explained in the previous section but
with an additional provision for circulation of the drilling fluid. In this method, a partial
vacuum is created above the sample while withdrawing. The reader may refer to IS: 8763
(1978) for details.
The second method is important as it is suitable for sampling of sand below the water
table. Compressed air is used in this technique and keeps the groundwater separated from
the sample. This is necessary to avoid dispersion of sampled sand (Fig. 18.17). In a bore-
hole, the sampler is pushed into the soil at the required depth with the help of a drill rod,
a spacer block, and a shackle arrangement. The spacer block located above the bell pre-
vents over-driving and allows the correct sample length. Now compressed air is forced into
the bell, which in turn closes the diaphragm check valve and hence an excess pressure of
140 kN/m2 is maintained inside the bell. This enables the water to be expelled from the bell.
After the complete expulsion of water, the sampler along with the soil is withdrawn into the
bell. The complete assembly is raised to the surface by means of a cable. During the process
of withdrawal, water is continuously poured to keep the drill hole full and the air pump is
also kept working. The spacer block is removed, the sampler is pushed out of the bell, and
Lifting cable
Compressed
air line Lowering peg
Adapter rod, 38 mm
Spring
Air Shackle
nipple
Socket block
159
Guide rod
775
Removable spacer
Sealing ring (anus type)
152
Bronze bushing
Steel weight
Relief valve
Water exit port
Rubber diaphragm
Set screw
787
Rubber sealing ring

521
Sampling tube
63 mm OD, 1.7 mm thick
Compressed air bell
Sample
Casing pipe 152 mm
Fig. 18.17 General layout of sand sampler with auxiliary bell for compressed air (Source:
IS: 8763, 1978)
M18_PURU1773_01_SE_C18.indd 687 2/12/2013 8:20:52 AM

the sampling tube is disconnected. A filter plug is placed at the lower end, the suction is
released, and the undisturbed sample obtained.
Undisturbed Sampling from Accessible Explorations. Undisturbed samples may be
obtained from accessible explorations, particularly from open pits and trenches. For this
purpose, a pillar of dimensions 40 cm × 40 cm may be left if the sample is strong and undis-
turbed at the centre of the pit to extract the undisturbed sample of required size. If the sam-
ple is weak or to be transported to a far-off place, additional protection is required. A box
with open ends is placed around the sample with a gap of about 25 mm all around, and the
gap is filled with paraffin wax. The sample and the box are removed and additional wax is
poured on the top and bottom, which prevents evaporation of moisture from the sample.
18.7 ROUTINE FIELD TESTS

Reliable results can be obtained from a carefully obtained sample and appropriate testing
techniques in the laboratory. Even then, all the environmental conditions (e.g., in-place stress
condition, moisture, etc.) cannot be simulated in the laboratory, and evidently, the soil sam-
ples undergo some disturbance during sampling and handling, which may greatly affect the
test results. In addition, some valuable data can be obtained only by field testing. Three rou-
tine field tests are identified, viz., penetration test (standard penetration or cone penetra-
tion), vane shear test, and tests for groundwater observations. One or more of these are
performed in every sub-surface exploration in addition to sampling. They are briefly
explained below.
18.7.1 Standard Penetration Test (SPT)

This test is a standardized method of sounding (IS: 2131, 1981). It consists of driving a stan-
dard sampler (a split-spoon sampler), adopting a standard method of driving (a 63.5 kg
weight with a free fall of 75 cm), and expressing the number of blows (as-value) required to
push the sampler to a standard depth (300 mm).
Where a casing is used, it is driven just above the level at which the test is made. In cohe-
sionless soils, the sample casing should be advanced by slowly turning or pushing so as to
avoid change in the density of the soil. The borehole is cleared using augers, bailers with flap
valves, or by wash boring.
The split-spoon sampler is lowered to the bottom of the borehole and driven with the
hammer to seat the sampler up to a depth of 150 mm. The sampler is further driven to a
depth of 300 mm or 50 blows. The number of blows required to effect each 150 mm of pen-
etration is recorded. The total number of blows (N values) required for the second and third
150 mm of penetration is termed the penetration resistance.
The sampler is raised to the surface. A typical sample, or samples, from the opened split is
placed in jars, which are sealed with an identifying label. The test is repeated at every change
in the stratum or at intervals of not more than 1.5 m, whichever is less. The intervals may be
increased to 3 m if the vane shear test is performed in between.
The relative density of granular soils can be reliably assessed from SPT values. The cor-
relation of N values with the shear strength of cohesive soils is not very reliable. Several
factors affect the blow count, and necessary judgement has to be exercised while adopting
M18_PURU1773_01_SE_C18.indd 688 2/12/2013 8:20:52 AM

Effective vertical overburden pressure, kN/m2

98
196
294
392
490
0.4 0.8 1.2 1.6 2.0
Correction factor
Fig. 18.18 Correction chart for N-value in cohesionless soil for overburden (Source: IS: 2131, 1981)
the results. In cohesionless soils, an increase in depth increases the overburden pressure;
consequently, the number of blows needed to drive the sampler increases for a cohesionless
soil of the same relative density. Hence, the N value has to be corrected for overburden pres-
sure in cohesionless soils. The corrected N value (N′) is given as
N ′ = ( Nobserved )(correction factor) (18.5)
The correction factor is given in Fig. 18.18, which is as per IS: 2131 (1981). The value obtained
has to be further corrected for dilatancy. If the stratum consists of fine sand and silt below
the water table, the revised N′ value (N″) is given as
N ′′ = 15 + 12 ( N ′ − 15) (18.6)
18.7.2 Cone Penetration Test (CPT)

The CPT is another important sounding method. Two penetration tests have been standard-
ized, viz., dynamic CPT (IS: 4968 – Parts 1 and 2, 1976) and static CPT (IS: 4968 – Part 3, 1976).
The dynamic cone penetration value (qcat) is the resistance offered by a 60° cone for a 300
mm penetration when driven by a 65 kg weight with a 750 mm free fall.
The 50 mm diameter or 60° cone is shown in Fig. 18.19. The cone is threaded to the driving
rod. The hammer head is joined to the other end of the rod and a guide rod, 150 cm long,
is connected to the hammer head. The assembly is kept vertical as illustrated in Fig. 18.20.
The cone is then driven into the soil using the 65 kg hammer falling through the free fall
of 750 mm. The number of blows required for every 100 mm penetration is recorded. The
process is repeated till the cone is driven to a depth of 300 mm.
The dynamic cone is a simple device by which the continuous soil resistance of a location
can be obtained. Further, another advantage is that no borehole is needed. In the attempt to
M18_PURU1773_01_SE_C18.indd 689 2/12/2013 8:20:53 AM

Square threads of
'A' rod coupling 32
40 mm
45° to 60°
60° ± 15′
50 ± 005 mm
Threaded cone
Fig. 18.19 Dynamic cone (Source: IS: 4968 – Part 1, 1976)
65 kg hammer
1750 mm
Guide rod
Driving head
Driving rod A
Arrangement
for keeping
rod vertical
Ground
level
Cone
Cone adopter
Fig. 18.20 Typical assembly of equipment for cone penetration test (Source: IS: 4968 – Part 1, 1976)
arrive at a correlation between dynamic cone penetration and standard penetration, a wider
cone of 62.5 mm with use of bentonite slurry has been suggested (IS: 4968 – Part 2, 1976).
Among the sounding tests, the static cone test is the best and can give more reliable values
in locations below the water table and where SPT fails. Basically, the test consists of pushing
the cone first and then the cone and friction jacket, thus finding the frictional resistance. Tolia
(1978) reviewed the factors affecting the static cone and the influence of overburden pressure
on static CPT values.
18.7.3 Vane Shear Test

The principle of the vane shear test and the laboratory method of determining the same have
been discussed in Chapter 7. The vane shear test is suitable for saturated clays of soft medium
M18_PURU1773_01_SE_C18.indd 690 2/12/2013 8:20:53 AM

consistency. This is more suitable for soils which are fissured or highly susceptible to
sampling disturbance.
The height of the vane is twice its diameter. The overall diameter of the vane should be
37.5, 50, 65, 75, or 100 mm. The area ratio (Ar) of the vane is given as
8t(D − d) + πd 2
Ar = ×100% (18.7)
πD2
where t is the thickness of the vane blade (mm), D the overall diameter of the vane (mm),
and d the diameter of central vane rod (mm).
As per Indian Standards (IS: 4434, 1978) the area ratio shall not exceed 18% for the 37.5
mm vane and 12% for the 50, 65, 75, and 100 mm diameter vanes.
The instrument should be capable of applying a torque to the vane and measuring the
same. The torque applicator should be capable of controlling the speed at the rate of 0.1° per
second.
There are two methods of testing, viz., testing from the bottom of a borehole and direct
penetration from the ground surface. The first test method is explained below.
An arrangement (diagrammatic) for testing from the bottom of the borehole is shown
in Fig. 18.21. About 5 minutes after insertion of vane, the torque is applied. The maximum
torque applied is noted. Just after this, the vane is rotated through a minimum of 10 revo-
lutions. After about a minute the remoulded strength is determined. The shear strength is
computed from Eq. 9.18.
Torque measuring
instrument
Ground level
Intermediate guides
at 5 m intervals
Borehole
casing
Bottom guide
Vane rod
Penetration
as required Vane rod sleeve
(5 × DIA of
borehole min)
Vane
Fig. 18.21 Diagrammatic vane test arrangement (for test from bottom of borehole) (Source: IS:
4434, 1978)
M18_PURU1773_01_SE_C18.indd 691 2/12/2013 8:20:54 AM

18.7.4 Groundwater Observations

There are two aspects to groundwater observations: viz., determination of levels and pore
water pressure and permeability tests. Field permeability tests have been discussed in
Chapter 5. Observations of water table and piezometric surface levels are discussed below.
Borehole observation is the simplest technique. Boreholes drilled for a sub-surface inves-
tigation can be kept open for 24 hours. The level of water is normally determined by lower-
ing a chalked tape or a tape with a float or by an electrical switching device which is actuated
on contact with water.
In a cohesive soil stratum, the stabilization of the water table may take time. In such situ-
ations, the location may be ascertained by adopting the extrapolation method. In this case, a
plot of water level versus time is made and the groundwater level is estimated by extrapolat-
ing the curve until it becomes parallel to the time axis (Fig. 18.22). If several levels are noted
at equal time intervals the following computational method is used.
Let the rise in water level from time t0 to t1 be h1
And let t1 − t0 = t2 − t1 = t3 − t2 (Fig. 18.22). The depths of the observed water level below the
stabilized groundwater level are
h12 h22 h32
D0 = , D1 = , D2 =
h1 − h2 h1 − h2 h2 − h3
Observations are repeated till a satisfactory estimate of the location of the stabilized
groundwater level has been made.
Pore water pressures may be determined directly by sinking open piezometers or a stand
pipe. In an unconfined aquifer, knowing the height (h) of water in the stand pipe, the pore water
pressure (uw) can be determined from uw = γwh. In a confined aquifer, the water level auto-
matically rises to a level corresponding to the piezometric surface, representing the pore water
pressure in the confined aquifer. For soils of low permeability or where a more rapid response
is required, closed piezometers like hydraulic or pressure transducer piezometers are used.
Stabilized ground- Dw
water level
Water level above
a given datum
D2 D
Estimated ground- 1
water level above t3
t2 D0
the datum Water level
in borehole t1
at time t0
Elapsed time t1 – t0 = t2 – t1 = t3 – t2
(a) (b)
Fig. 18.22 Location of groundwater level (Source: Whitlow, 1983)
M18_PURU1773_01_SE_C18.indd 692 2/12/2013 8:20:55 AM

18.8 RECORDING OF FIELD DATA

Data obtained from exploratory work must be recorded accurately to derive maximum
benefit from the effort. The following information should be routinely recorded in the field
notes:
1. name and address of the project,
2. date of work and names of personnel involved,
3. location of boring or test pits with respect to an established coordinate system,
4. elevation of ground surface,
5. water table elevation,
6. elevation of upper level of each soil layer,
7. elevation at which soil samples were taken,
8. record of SPT or CPT results,
9. a field classification of each layer, and
10. description of drilling equipment and changes encountered in drilling.
After the completion of laboratory tests, the information obtained from the field is assem-
bled and summarized to obtain the details of a soil profile, which is termed a boring log. A
typical boring log of a deposit in Calcutta (Som, 1975) is shown in Fig. 18.23.
Water Cohesion,
N Values content, w, % c, kN/m2
15 35 55 75
0 10 20 30 40 25 45 65 0 24 48 72 96
Fill
0
Soft to firm
2 brownish grey
clayey silt
4
Soft to firm bluish
6 80
grey clayey silt
8
Peat 8.3 – 9.0 m
10 Soft to firm bluish
Depth, metres
2100
grey silty clay
12
Consistency
14 changes to stiff
beyond 10.60
16
Med. dense clayey 81 20 m
18 sandy silt with 65 21.5 m
traces of mica
20
22 Yellowish brown 65 25 m
24
medium to fine
sand
(a) Soil profile (b) Details of N, w, and c with depth
Fig. 18.23 Boring log for a deposit in Calcutta (Source: Som, 1975)
M18_PURU1773_01_SE_C18.indd 693 2/12/2013 8:20:55 AM

18.9 LOCATION, SPACING, AND DEPTH OF BORINGS

There is no hard and fast rule to decide the boring location or spacing. The number of
borings, or the spacing between borings for a project is related to the type, size, and weight
of the structure planned. In general, borings should be located to obtain maximum informa-
tion from the minimum number of boreholes. Primarily, as a few as two or three boreholes
may be made to get a confident assessment of sub-surface conditions.
The type of structure often decides the number and the depth of borings. Usually, for a
building project, boreholes are made at the building corners, at the centre of the site and
locations where heavily loaded columns or machinery pads are proposed. Generally, at least
one boring should be taken to a deeper stratum, preferably up to the bedrock if practicable.
Other borings may be taken at least to the significant stress level. This significant depth
corresponding to the associated stress is taken to be approximately two times the least
width. It is not advisable to terminate the borings on soft or organic soils, if they are encoun-
tered. For highways and runways, sewer and water lines, borings are taken along the centre
line. For airports, additional borings are needed for parking stands, taxiways, and terminal
buildings. Boreholes for highways and runways are taken up to the bedrock to determine
the soil stratification, so that some of the soils can be used for filling and construction. Water
and drainage works also consider the water table location and rock line. Generally, for a
preliminary survey of long-distance projects, such as highways and dams, the spacing of
boring required is from 150 to 300 m. The final programme may eventually require borings
of 30 to 70 m spacing. Boring in river beds for bridge piers and abutments, should ascertain
the scour and competent soil depth.
Specific recommendations are made by Indian Standards regarding the type, extent, and
details of surface explorations and the number, depth, and spacing of boreholes for the fol-
lowing civil engineering works:
1. Foundations of Multi-Storeyed Buildings (IS: 1892, 1979)
2. Earth and Rockfill Dams (IS: 6955, 1973)
3. Power House Sites (IS: 10060, 1981)
4. Canals and Cross Drainage Works (IS: 11385, 1985)
5. Ports and Harbours (IS: 4651 – Part 1, 1974)
POINTS TO REMEMBER
18.1 Ground investigation consists of four phases, viz., collection of available information,
reconnaissance, preliminary investigation, and detailed investigation.
18.2 Non-representative samples comprise mixtures of materials from various soil or rock
layers. Representative samples consist of constituent minerals from each layer and
are not mixed with the materials from other layers. Undisturbed samples are types
of samples in which the material has experienced such little disturbance that they are
suitable for all laboratory tests.
M18_PURU1773_01_SE_C18.indd 694 2/12/2013 8:20:55 AM

18.3 In indirect methods (geophysical and sounding methods) depths to principal strata
are established, based on some physical properties of the material, and the measure-
ments are made on the ground surface.
18.4 In semi-direct methods (borings and rotary drilling) the depths of different layers are
ascertained by the rate of advancement of boring tools or by means of non-representative
samples obtained in the course of boring operations.
18.5 Direct methods include boring and sampling methods, which continuously provide
representative or undisturbed samples.
18.6 Sample disturbance is caused due to the sample thickness, method of driving the
samples, and rotation of samples during removal.
18.7 Accessible explorations permit a direct visual examination of the subsoil and afford
the most complete information of the ground.
18.8 Routine field tests include the standard penetration test, cone penetration test, vane
shear test, and groundwater observations.
QUESTIONS
Objective Questions
1. In the electrical resistivity method, differences in the electrical potential of the
strata are detected.
2. Non-representative samples comprise mixtures of materials from various soil or
rock layers.
3. A value of recovery ratio greater than one indicates compression and less than one,
expansion.
4. The cone penetration test is another important sampling method.
5. The field vane shear test is most suitable for saturated soft clays.
18.2 The actual planning of a sub-surface exploration programme includes
(1) Collection of all available information
(2) Reconnaissance of the area
(3) Preliminary site investigation
(4) Detailed site investigation
(a) 1, 2, 3, and 4, are correct
(b) 1 and 2 are correct
(d) 2, 3, and 4 are correct
18.3 The basic requirement in a seismic refraction method is that the wave velocity in the
upper layer must be ______ that in the lower layer.
(a) Less than
(b) Greater than
M18_PURU1773_01_SE_C18.indd 695 2/12/2013 8:20:55 AM

(c) Equal to
(d) Twice
18.4 The standard penetration test is most frequently used to measure the
(a) Shear strength of soft clays
(b) Undrained strength of fissured clays
(c) Relative density of granular soils
(d) Consistency of clays
18.5 Which of the following pairs is correctly matched?
1. Wash boring Can be conveniently used even below the water table
2. Percussion boring Only method suitable for drilling base holes in bouldery
and gravelly strata
3. Auger boring Samples recovered have high value
4. Rotary boring Useful only for sands and clays
Codes:
(a) 1 and 2 (b) 2 and 3 (c) 3 and 4 (d) 4 and 1
18.6 The degree of disturbance for a soil sample is usually expressed by
(a) Void ratio (b) Area ratio
(c) Recovery ratio (d) Consolidation ratio
18.7 Identify the incorrect statement. Undisturbed samples are obtained from
(a) Thin-walled tube samplers (b) Piston samplers
(c) Split-spoon samplers (d) Hand-trimmed samplers
18.8 Samples of highly fissured soils can be obtained from
(a) Accessible explorations (b) Open-drive sampling
(c) Split-spoon sampling (d) Thin-walled sampling
18.9 What sort of expertise would be needed to choose the borehole depth and number in
a ground investigation?
18.10 How do you analyse the samples obtained after sampling for strength if they are fine
sand, over-consolidated clay, and soft sandstone?
18.11 Explain the basic differences in explorations for foundations (including abutments)
and construction materials.
18.12 Discuss the various stages of sample disturbance.
18.13 Discuss the factors which are relevant to the planning of a well-balanced exploration
programme.
18.14 Compare soil boring and sampling methods with test pits.
18.15 Explain the following terms which are used in subsoil exploration.
1. Area ratio
2. Recovery ratio
3. Non-representative sample
M18_PURU1773_01_SE_C18.indd 696 2/12/2013 8:20:55 AM

4. Representative sample
5. Undisturbed sample
18.16 Briefly describe the types of sub-surface information that seismic refraction studies
can provide.
18.17 If you are in charge of subsoil exploration of important structures, how do you decide
the depth of exploration. List the factors you will consider and their importance.
18.18 Discuss briefly the methods of taking undisturbed samples in (i) non-cohesive soils
and (ii) cohesive soils.
18.19 What steps will you take to economically investigate a site which is suspected to be
erratic in nature?
18.20 If a thin-walled sampler is pushed fast into the following soils, what will be the effect
of the disturbance: (i) sensitive clay, (ii) sandy silt, and (iii) sandy clay.
EXERCISE PROBLEMS
18.1 An open drive sampler has an outside diameter of 76 mm and an inside diameter of
72 mm. What is the area ratio of the sampler?
18.2 During a sampling operation, the open-drive sampler is advanced 600 mm and the
length of the recovered sample is 525 mm. What is the recovery ratio of the sample?
18.3 In a standard penetration test, the following observations were taken at a depth of 4 m
below the ground level:
First 15 cm 31 blows
Second 15 cm 32 blows
Third 15 cm 23 blows
Fourth 15 cm 36 blows
Estimate the corrected SPT value for overburden if the average unit weight of the soil
at 4 m depth is 20 kN/m2.
M18_PURU1773_01_SE_C18.indd 697 2/12/2013 8:20:56 AM

M18_PURU1773_01_SE_C18.indd 698 2/12/2013 8:20:56 AM
19
Soil Improvement
CHAPTER HIGHLIGHTS
Soil improvement techniques – Surface compaction – Drainage methods: well-
point systems, deep-well drainage, vacuum de-watering system, de-watering
by electro-osmosis – Vibration methods: vibro-compaction, vibro-displacement
compaction – Pre-compression and consolidation – Grouting and injection –
Chemical stabilization – Soil reinforcement – Geotextiles and geomembranes –
Other methods
19.1 INTRODUCTION
In situ soil characteristics of a construction site are different from those desired and,
almost always, far from ideal for a designated need. With increased urban develop-
ment, sites with favourable foundation conditions became rare. At times, the civil engi-
neer has been forced to construct structures at sites selected for reasons other than soil
conditions. Thus, it is increasingly important for the engineer to know the degree to
which soil properties may be improved or if there is an alternative that can be thought
of for the construction of an intended structure at the stipulated site.
If unsuitable soil conditions are encountered at the site of a proposed structure, one
of the following four procedures may be adopted to ensure satisfactory performance of
the structure (Mitchell, 1976):
1. Bypass the unsuitable soil by means of deep foundations extending to a suitable
bearing material.
2. Redesign the structure and its foundations for support by the poor soil, a procedure
that may not be either feasible or economical.
3. Remove the poor material and either treat it to improve and replace it, or substitute
it by a suitable material.
4. Treat the soil in place to improve its properties.
M19_PURU1773_01_SE_C19.indd 699 2/12/2013 8:20:20 AM

Further, in the case of existing structures exhibiting foundation distress, in-place

foundation treatment may be used as corrective measures. Nowadays, various processes
are available by which the characteristics of the construction site can be improved to facili-
tate construction operation, to allow increased bearing pressures, or to reduce settlements.
The techniques involved in the attainment of the required improvement facilities are referred
to as geotechnical processes.
19.2 IMPROVEMENT TECHNIQUES

Soil improvement in its broadest sense is the alteration of any property of a soil to improve
its engineering performance. This may be either a temporary process to improve the con-
struction of a facility or may be a permanent measure to improve the performance of the
completed facility. The result of an application of a technique may be increased strength,
reduced compressibility, reduced permeability, or improved groundwater condition. Soil
improvement techniques may be classified based on the nature of the process involved,
material used, the desired result, etc. The various techniques discussed are surface compac-
tion, drainage methods, grouting and injection, chemical stabilization, thermal stabilization,
soil reinforcement, and application of geotextiles and geomembranes.
The factors that must be considered in the selection of the best technique in any case
include the following (Mitchell, 1976):
1. Soil type – sand, clay, organic, etc.
2. Area and depth of treatment required depend on the geometric characteristics of the soil
deposit and load distribution
3. Type of structure and load distribution
4. Soil properties – strength, compressibility, permeability, etc.
5. Permissible total and differential settlements
6. Material availability – stone, sand, water, admixture, stabilizers, etc.
7. Availability of skills and equipment
8. Environmental considerations – waste disposal, erosion, water pollution, etc.
9. Local experience and preferences
10. Economics
The discussion on different improvement techniques in the following sections is directed
primarily at their use as a permanent measure of soil improvement. However, some of them
are well suited to expedite or facilitate construction as well, for example, pumping to control
water flow into excavations, the use of grounting to reduce liquefaction or settlement, the
use of electro-osmosis for seepage control, and temporary slope stabilization by freezing.
19.3 SURFACE COMPACTION

One of the most widely used and the oldest techniques of soil densification is compaction.
Construction of a new road, a runway, an embankment, or any soft or loose foundation site
needs a compacted base for laying the structure. If the depth to be densified is less, then
compaction on the surface of the soil alone may solve the problem. Surface compaction
needs less skilled labour and is usually the most economical of the techniques available.
M19_PURU1773_01_SE_C19.indd 700 2/12/2013 8:20:20 AM

Soil Improvement 701
The field engineer can vary the content, amount of compaction, and type of compaction
to attain the required density. The usual surface compaction devices are rollers, tampers, and
rammers. All conventional rollers like smooth wheel, rubber-tyred, sheeps-foot, vibratory,
and grid rollers can be used (as discussed in Chapter 4). A particular type of roller has to
be chosen to suit a particular job. In order to achieve the required density, only the required
number of passes should be allowed. Mere increase in the number of passes will not increase
the density beyond a certain depth. For all practical purposes, granular soils can be surface
compacted using vibratory rollers up to a depth of about 2 m.
In cohesive soils, the required percentage compaction can be obtained using any of the
rollers and tampers, but vibrations are not effective. In the cases of sub-grades and base
courses for heavy duty roads and airfields, the heavy rubber-tyred rollers may be used with
advantage. In order to efficiently bond each layer for water-retaining structures a sheeps-foot
roller may be used.
19.4 DRAINAGE METHODS

Groundwater causes some of the most difficult problems in excavation work. The presence
of water increases the pore water pressure and decreases the shear strength. Further, heavy
inflow of water to the excavations is liable to cause erosion or collapse of the sides of open
excavations. Sometimes, there can be instability of the base due to upward seepage. How-
ever, from a knowledge of the soil and groundwater conditions, it is feasible to adopt certain
methods to control the groundwater and ensure a safe and economical construction scheme.
19.4.1 Well-Point Systems

Well points are small well screens of sizes 50 to 80 mm in diameter and 0.3 to 1 m in length.
The well-point system of ground water lowering comprises the installation of well points
around the excavation. The well points are attached to a header pipe and in turn connected
to a well-point pump. The water flowing by gravity to the fiter well is drawn by the vacuum
up to the header main and discharged through the pump. A typical installation of a well-
point system is shown in Fig. 19.1.
Valve
Header Connection Pump suction level
main
Original water level

5–5.5 m
max
Coarse sand filter
Riser pipe
Lowered water level
Fig. 19.1 Single-stage well-point installation
M19_PURU1773_01_SE_C19.indd 701 2/12/2013 8:20:20 AM

If the site is accessible and the water-bearing strata to be drained are not too deep, the
well-point system will be the most suitable method. The equipment is reasonably simple
and cheap and can be installed rapidly. Well points are generally used where the water
table does not have to be lowered too much. In the case of large excavations or where the
depth of excavation below the water table is more than 10 m or there is artesian pressure,
deep wells and turbine pumps have to be adopted. The added advantage with the well-point
system is that the water is filtered as it is removed from the ground and carries few little or
no particles with it.
A single well point of 50 mm has a capacity of about 10 l/min. Permeability of the soil
and the time available to effect the drawdown are the factors that govern the spacing of well
points. In a highly permeable gravel medium, the spacing is about 0.3 m, whereas in fine
to coarse sands, a spacing of about 0.75 to 1 m is satisfactory. In low permeability mediums
like sandy silts, the spacing may be of the order of 1.5 m. A well-point equipment of nor-
mal size comprises 50 to 60 points to a single 150 or 200 mm pump with a separate 100 mm
jetting pump.
19.4.2 Deep-Well Drainage

Where the soil formation becomes more pervious with depth, large diameter deep wells are
suitable for lowering the groundwater table. In contrast to a well-point system, deep-well
de-watering can be installed outside the zone of construction operation, and drainage
effected to the depth of excavation required. Deep wells may be combined with the well-
point system on certain occasions for lowering the groundwater table (Fig. 19.2). Deep wells
are also suitable for relief of the artesian pressure in some field conditions. The cost of instal-
lation of a deep-well system is relatively high. Therefore, the process is generally restricted
to jobs which have a long construction period, such as dry docks.
The procedure in installing a deep well is to sink a cased borehole having a diameter
that is about 300 mm larger than the well casing. The diameter of the latter depends on the
size of the submersible pumps. The inner well casing is inserted after the completion of the
borehole. A perforated screen is provided over the length where de-watering of the soil is
Discharge pipe
Header main
Lowered water
level
Deep well Deep-well

pump Well point
Fig. 19.2 Deep-well de-watering
M19_PURU1773_01_SE_C19.indd 702 2/12/2013 8:20:20 AM

Rising main
Unperforated
Inner casing
casing
Soil backfill
Outer
well casing
(withdrawn)
Perforated
Graded filter
casing
material
Lowered water level
Mesh filter
screen
Unperforated
Submersible pump
casing
Soil backfill
Soil collected
in sump
Fig. 19.3 Details of deep-well installation (Source: Tomlinson, 1986)
required. Graded gravel filter material is placed between the well casing and the outer bore-
hole casing over the length to be de-watered. The outer casing is withdrawn in stages as the
filter material is placed. The space above the screen is backfilled with any available material.
The details of the completed installation are shown in Fig. 19.3. Deep wells are spaced at 10
to 100 m intervals depending on the situation.
19.4.3 Vacuum De-watering System

Gravity methods are not very effective in fine-grained soils with permeability in the range of
0.1 to 10 × 10−3 mm/s. Such soils can be stabilized by means of a vacuum well or well-point
system. A vacuum de-watering system consists of wells or well points with the screen and
riser pipes surrounded with a free drainage sand filter extending to within a few metres of
the surface. A bentonite or impervious soil seal is provided at the remaining portion of the
hole. By maintaining a vacuum in the well screen and the sand filter, the hydraulic gradient
producing flow towards the well or well point is increased. For proper de-watering, closer
spacing of wells is essential (Fig. 19.4).
19.4.4 De-watering by Electro-osmosis

Fine-grained soils are more troublesome to drain because of capillary forces acting on the
pore water. If the vacuum well-point system is ineffective, application of an electrical gradi-
ent may be made. When a direct electric current is passed through a saturated soil, water
moves towards the cathode. If the water is removed at the cathode, the soil decreases in
volume resulting in increased shear strength. This process is called de-watering by electro-
osmosis. Casagrande (1952) has shown that the electro-osmotic flow is dependent on the
M19_PURU1773_01_SE_C19.indd 703 2/12/2013 8:20:21 AM

Atmospheric
Header pressure
Original Seal
water level
Silt
Clayey silt
Sand
Sandy silt filter
Vacuum
Silt
Silty sand
Water level in filter

Well point
Fig. 19.4 Vacuum de-watering system (Source: Tomlinson, 1986)
porosity of the soil and the electric potential. A comparison of electro-osmotic flow with
hydraulic flow through a single capillary is shown in Fig. 19.5.
The general layout of the electrodes depends upon the purpose for which they are
intended. Figure 19.6 shows electrode arrangements for two field situations. Sheet piles
of any shape and old pipes of 25 to 50 mm diameter can be used as anodes. Since the
Resisting force Resisting force

Double
Double
layer
layer
+ + +
Moving force
Free Moving force
water Free water
Velocity Velocity
Double
Double
+ + +
layer
layer
Resisting force Resisting

force
(a) Electro-osomotic (b) Hydraulic
Fig. 19.5 Comparison of electro-osmotic and hydraulic flows
M19_PURU1773_01_SE_C19.indd 704 2/12/2013 8:20:22 AM

D D
+ Well cathode Well cathode

+
+
D 2D 3D
+
2D
Iron pipe 2D 3D
as anode
Sheet piling as anode
(a) Sheated excavation (b) Cuttings
Fig. 19.6 Electrode arrangements
anodes corrode considerably in the course of a few weeks of electro-osmotic treatment, they
should be replaced as soon as the current drops to less than 30% of the initial consumption.
Perforated tubes form cathodes and the cathode wells are connected to a pumping system.
Electro-osmosis is resorted to only to remedy a difficult situation where other methods
have failed. This method has a high cost of installation and initial running cost. But the
power consumption, and hence running cost, decreases considerably after the ground is
stabilized.
19.5 VIBRATION METHODS

19.5.1 Vibro-Compaction
Vibro-compaction methods can be effectively used for rapid densification of saturated non-
cohesive soils. Vibrations and shock waves in loose deposits of such materials cause lique-
faction followed by densification and settlement accompanying the dissipation of excess
pore water pressures. The effectiveness of these methods decrease with increase in the per-
centage of fines in the soil, since the fines reduce the permeability of the material, which is a
prime factor for liquefaction. Similarly, the effect is less in partially saturated soils because of
surface-tension forces.
Blasting. This technique consists of detonating a certain amount of explosive charge at
a certain depth of the soil required to be compacted. Localized spontaneous liquefaction
develops in saturated deposits due to sudden shock waves and soil grains get displaced. In
dry soils too the shocks place the particles in a more irreversible way and the new density
attained is permanent.
Pipes of about 100 mm in diameter are driven into the soil strata to the required depth.
After removing the soil from inside the pipe, sticks of dynamite and an electric detonator are
wrapped in water-proof bundles and lowered through the casing (Fig. 19.7). A wad of paper
M19_PURU1773_01_SE_C19.indd 705 2/12/2013 8:20:22 AM

Casing Connecting wire
Sand
Plug
Detonator
Dynamite
sticks
Fig. 19.7 Installation of explosives (Source: Mitchell, 1970)
or wood is placed against the charge of the explosives to protect it from misfire. The casing is
withdrawn, and in order to obtain the full force of the blast, the hole is backfilled with sand.
The electrical circuit is closed and the charge is fired. The surface settlements are measured
by taking levels or from screw plates embedded at certain depths below the ground surface.
The correct amount of charge has to be used so that it is just enough to shatter the soil
particles uniformly and at the same time prevent the formation of craters. Layman (1942) has
suggested an empirical formula which can be taken as a rough guide.
W ≈ 164 CR3
where W is the weight of explosive (N), C the coefficient (=0.0025 for 60% dynamite), and R
the radius of the sphere of influence (m).
To avoid cratering, the minimum depth of charge should be greater than R. Charge
spacings less than 3 m should be avoided; values of 3 to 8 m are typical. The centre of charges
should be located at a depth of two-thirds the thickness of the layer to be densified. It is found
that repeated blasts of small charges are more efficient than a single large blast (Hall, 1962).
If the depth of the deposits to be densified is 10 m or less, compaction is carried out in single
tier only; for deeper deposits, more tiers need to be used. Charges should be exploded from
the bottom-most tier in an upward direction in a uniform manner. The uppermost portion of
the stratum is always loosened and can be compacted by any surface compaction device. The
typical grid spacing and firing pattern as suggested by Mitchell (1970) is given in Fig. 19.8.
The blasting technique is less expensive and involves less time, labour, and equipment.
Although blasting is one of the most economical stabilization methods, it suffers from the
disadvantages of non-uniformity, potential adverse effects on adjacent structures, and the
dangers associated with the use of explosives in populated areas.
Vibrating Probe. Vibrating probe, also known as Terra-probe, is a patented process used
to densify loose sands. A 760 mm open-ended tubular probe is vibrated into the ground from
a vibratory pile driver operating at 15 Hz. Vibrations of 10 to 25 mm amplitude are made in a
M19_PURU1773_01_SE_C19.indd 706 2/12/2013 8:20:22 AM

7
6 3
5m
1
12 9
2
10 11
NOTE
First blast
8 Second blast
5 4 Third blast
5m Settlement stakes
(a) Grid spacing (b) Firing pattern
Fig. 19.8 Typical charge spacing and firing (Source: Mitchell, 1970)
vertical mode. After reaching the planned penetration depth, the probe is withdrawn slowly
while vibrations continue. Effective treatment can be obtained between depths of 4 m below
the ground surface and about 20 m. The operation is most efficient where groundwater is
within 2 to 3 m of the surface. Water jets can be attached to the probe, or ponding of the sur-
face can be done to assist the penetration and densification.
Test sections of the order of 10 to 20 m on a side are desirable to evaluate the effectiveness
and to determine the required spacing in any given case. A square pattern is often used, with
a fifth probe at the centre of each square giving more effectively increased densification than
a reduced spacing. The density achieved by this process is generally lower. Since the proce-
dure does not require a sand fill, it can be applied effectively for offshore sites.
It is essential that lift thickness, soil type, and roller type be matched. If lift thickness is
too great, then low-density layers will form alternately between high-density layers. If lift
thickness is too small, then much of the effort is lost through repeated over-compaction of
near-surface layers.
19.5.2 Vibro-Displacement Compaction

The methods described in this section are similar to those dealt with in the preceding section
except that the vibrations are supplemented by active displacement of the soil and, in the
case of compaction piles and vibro-flotation, by backfilling the zones from which the soil has
been displaced.
Compaction Piles. Driving displacement piles at close spacings can densify highly per-
meable and partially saturated soils. One effective procedure is to drive a pipe pile with a
false bottom to the desired depth. The pipe is then filled with sand or other backfill materi-
als in lifts. Each lift of materials is compacted concurrently with withdrawal of the pipe pile.
The backfill not only gets compacted but also expands laterally at the bottom, forming a
broadened base which is also referred to as a sand pile. Compaction piles are economical for
smaller sites of moderate depths up to 15 m.
In another rapidly developing technique, soft cohesive strata or loose non-cohesive depos-
its are replaced by granular material which is compacted by ramming or vibration. Such
compacted piles are referred to as granular piles or stone columns. Different installation
M19_PURU1773_01_SE_C19.indd 707 2/12/2013 8:20:23 AM

Power supply
Water pump
Follow- A
up pipe Cylinder of
compacted material,
Vibrating added from the surface to
unit compensate for the loss of
volume caused by the inc-
rease of density of the
A compacted soil
B
Cylinder of compacted
B material, produced by a
single vibrofloat compaction
Fig. 19.9 Vibro-flotation equipment (Source: Brown, 1976)
techniques are being adopted (Baumann and Bauer, 1974; Datye and Nagaraju, 1977; Ranjan
and Rao, 1983). But most of the granular piles are installed adopting the vibration technique
through a vibroflot (discussed in the next section). A comprehensive review and adoptability
of stone columns in ground improvement are presented by Ranjan (1989).
Vibroflotation. Vibroflotation is a technique for densifying in situ non-cohesive soils with
simultaneous vibration and saturation. This principle of densification was first published
in 1936 by Steuerman in a Russian journal and later applied in Germany during 1939 for
improvement of foundation soils for buildings.
The equipment required for vibroflotation involves a vibroflot probe, accompanying
power supply, water pump, crane, and front-end loader (Fig. 19.9). The vibroflot probe is an
essential piece of equipment consisting of a cylindrical penetrator, about 0.38 m in diameter
and about 2 m in length, with an eccentric weight inside the cylinder developing a horizon-
tal centrifugal force of about 100 kN at 1,800 rpm. A typical vibroflot consists of two parts.
The lower part is the horizontal vibrating unit which connects to the upper part of the follow
up pipe, the length of which can be varied depending on the compaction depth (Fig. 19.10).
The water pump provides water to jet the vibroflot into the ground as the vibroflot is low-
ered with the crane. The front-end loader is used to supply the backfill material even as the
in situ soils are densified.
The probe is freely suspended from a crane. Each compaction sequence has four basic
steps (as suggested by Brown, 1976, and Vibroflotation Foundation Co., USA) (Fig. 19.11).
They are as follows:
M19_PURU1773_01_SE_C19.indd 708 2/12/2013 8:20:23 AM

Upper water jet
Outside jets
Follow-up pipe
Universal joint
connecting rod
1.5 m
Flexible hose
Follow-up
Motor (electric or
pipe
hydraulic)
extensions
3.5 m Outside jets
Lower
follow-up
Eccentric shaft pipe with
2m universal
joint
Vibrator
Vibrator
Fig. 19.10 100 HP vibroflot (Source: Brown, 1976)
Loose
sand
Densified
sand
Step 1 Step 2 Step 3 Step 4
Fig. 19.11 Vibroflotation compaction process (Source: Brown, 1976)
M19_PURU1773_01_SE_C19.indd 709 2/12/2013 8:20:25 AM

1. The vibroflot is positioned over the spot to be compacted and its lower jet is then fully
opened.
2. Water is pumped in faster than it can drain away into the subsoil. This creates a momen-
tary “quick” condition beneath the jet, which permits the vibroflot to settle due to its own
weight and vibration.
3. Water is switched from the lower to the top jets and the pressure is reduced enough to
allow water to be returned to surface, eliminating any arching of backfill material and
facilitating the continuous feed of backfill.
4. Compaction takes place during the 0.3 m per minute lifts, which return the vibroflot to
the surface. First, the vibrator is allowed to operate at the bottom of the crater. As the par-
ticles densify, they assume their most compact form. By raising the vibrator step by step
and simultaneously backfilling with sand, the entire depth of the soil is compacted into a
hard core.
Most vibroflotation applications have been to depths less than 20 m, although depths of
30 m have been attained successfully. The maximum depth appears limited mainly by the
ability of the crane to pull the vibroflot out of the ground. The factors contributing to suc-
cessful densification are
1. Equipment capacity
2. Probe spacing and pattern
3. In situ soil
4. Vibroflot withdrawal procedure
5. Backfill material
6. Workmanship
The two most important factors are the grain-size distribution of the soil and the nature
of backfill material used. The range of the grain-size distribution of in situ soils suitable for
vibroflotation (Brown, 1976) is shown in Fig. 19.12. The technique is best suited for den-
sifying very loose sands below the water table that have grain-size distributions falling
Gravel Sand Fines

Coarse Fine coarse Medium Fine Silt Clay
100
80
Percentage finer
60
A B C
40
20
0
10.0 1.0 0.1 0.01 0.001
Grain size, mm
Fig. 19.12 Soils suitable for vibroflotation (Source: Brown, 1976)
M19_PURU1773_01_SE_C19.indd 710 2/12/2013 8:20:25 AM

entirely within Zone B. Layered clays, fine particles, cementation, and organics in the in
situ soil pose a variety of difficulties for compaction by this technique. Soils with the grain-
size distribution entirely within Zone C are very difficult to compact by vibroflotation. In
general, the densities achieved and the zone of compaction decrease with increasing silt
and clay contents. The fines and organics apparently damp out vibrations, stick the sand
particles together between particles, and thus restrict the relative movement of particles
necessary for densification.
Clay layers present in in situ soil also reduce the zone of compaction. Gravel, dense sand,
and cemented sands are represented by Zone A. These soils have a reduced rate of probe
penetration and the effect becomes still less when the water table is located at a greater
depth. Hence, under these conditions, vibroflotation might prove to be uneconomical in the
long run.
The suitability of backfill material depends on the gradation. Brown (1976) has developed
a rating system to judge the suitability of the backfill material. The rating system is based on
a suitability number defined as
3 1 1
Suitability number = 1.7 2
+ 2
+
(D50 ) (D20 ) (D10 )2
where D10, D20, and D50 are the particle sizes corresponding to 10%, 20%, and 50% finer.
Table 19.1 gives the rating description. The withdrawal of the probe is also affected by the
quality of the backfill material.
With conventional vibroflotation equipment, minimum relative densities in excess of 70%
can be obtained for spacings up to about 2 m. Continuous square or triangular patterns are
often used. Typical patterns for spread footings for allowable soil pressures up to 300 kN/m2
are shown in Fig. 19.13.
Heavy Tamping. The most basic and simplest way of compacting loose soil is by repeated
dropping of a weight on to the ground. Although this technique was used long ago, it has
undergone rapid development after 1975 (Menard and Broise, 1975). The method, also
known as deep dynamic compaction or deep dynamic consolidation, consists of allowing
a very heavy weight (up to 400 kN) to fall freely on the ground surface from a height of 15
to 40 m. This leaves an impression on the ground. The tamping is then repeated either at
the same location or over other parts of the area to be stabilized. In the case of non-cohesive
Table 19.1 Backfill evaluation criteria
Suitability number Description of rating
0–10 Excellent
10–20 Good
20–30 Fair
30–50 Poor
>50 Unsuitable
Source: Brown (1976).
M19_PURU1773_01_SE_C19.indd 711 2/12/2013 8:20:25 AM

1.83 m 2.3 m
1.37 m + + 1.83 m to
+ + 1.68 m 1.88 m 2.13 m
+
1.83 m
2.13 m
+ + + +
2.13 m 3.05 m to
1.83 m 2.13 m
2.9 m 3.5 m
+ + + +
+ Centre of vibroflot penetration

Typical footing sizes
Fig. 19.13 Typical vibroflotation patterns for footings (Source: Brown, 1976)
soils, the impact energy causes liquefaction, followed by settlement as water drains. Fissures
formed around the impact points sometimes facilitate drainage in some soils. This method
has been successfully used to treat various types of soils and fill deposits up to 20 m thick.
This method can be adopted for densifying soils both above and below the water table. This
technique produces equal settlements faster than if a static load is applied.
Since the variables involved are many, no satisfactory ideal rigorous model is available
yet to understand the behaviour of dynamic consolidation. Only empirical information is
available with regard to the depth of penetration of the compaction. Dobson and Slocombe
(1982) have given an expression for the range of the effective depth in terms of the energy
of impact as
1.26 wh < D < 3.16 wh
where D is the effective depth (m), w the weight being dropped (kN), and h the height of
drop (m).
As the analytical understanding of this process is in the formative stage, it is recom-
mended to have a small test section at the site under consideration for necessary prelimi-
nary field evaluation (Koerner, 1985). Because of high-amplitude, low-frequency vibrations,
a sufficient minimum distance should be maintained depending on the type of structure or
facility.
19.6 PRE-COMPRESSION AND CONSOLIDATION

19.6.1 Pre-loading and Surcharge Fills
In this process, an earthfill or some other material is placed over the selected site. The amount
of fill is sufficient enough to produce a stress in the soil equal to the one anticipated from the
final structure. The soft soil is allowed to consolidate prior to construction. Since the con-
solidation takes a very long time, this method is suitable only for stabilization of thin layers.
If the thickness of fill placed for pre-loading is greater than that required for the final
ground surface elevation, the excess fill is termed surcharge fill. Thus, both permanent fill
M19_PURU1773_01_SE_C19.indd 712 2/12/2013 8:20:26 AM

and surcharge fill together cause a given amount of settlement in a shorter time than can
the permanent fill alone, and hence the time required for stabilization is drastically reduced.
Conventional consolidation theories can be followed to estimate the time required for consol-
idation. Both the primary consolidation and most of the secondary compression settlements
can be taken out in advance by surcharge fills. Secondary compression settlements may be
the major part of the total settlement of highly organic deposits or old sanitary landfill sites.
The rate of pre-load and surcharge fill placement has to be controlled depending on the
bearing capacity of the soil. If the bearing capacity of the soil is inadequate, layers of fill can
be placed only after a sufficient gain in shear strength is obtained. Geotechnical analyses can
be carried out for predictions of the rates of consolidation, strength, and strength gain. The
predicted values have to be checked from field measurements like piezometer readings and
in situ strength tests.
The two main requirements for pre-loading are enough space and availability of fill mate-
rial. Heaping of fill is the most common method of pre-loading although pre-loading can be
successfully effected by the weight of water or by lowering the water table. Among the fill
materials, granular soil is the most desirable because it does not turn into mud during rains.
Ores and industrial products are generally satisfactory, but clayey soils are less desirable.
Pre-loading has been used successfully on virtually every type of naturally laid or man-
made soil. Natural soils include loose sands and silts, soft silty clays, organic silts, and
erratic alluvial deposits, whereas man-made fills may be miscellaneous depositions such
as uncompacted dredged materials, industrial wastes, and rubbish sites previously used as
urban dumps. Deposits to be handled with care are thick homogeneous layers of plastic clay
and sanitary land fills.
The main advantage in the pre-loading is that the construction equipment needed is the
same as that for simple earth-moving jobs. The additional equipment required for follow-
up of pre-loading are relatively simple and inexpensive. By measuring ground movements,
the effect of pre-loading can be immediately assessed. Pre-loading ensures uniformity of
improvement because it eliminates local inhomogenities and it reduces considerably the
danger of liquefaction by earthquake in sands. Compared with other methods of improv-
ing ground support, pre-loading costs much less (about 10% to 20% and about 20% to 40%
with vertical drains). A detailed treatment of this subject can be found in Stamatopoulos and
Kotzias (1985) where several field examples are given.
19.6.2 Vertical Drains

For deep clay deposits, pre-loading alone will take more time because of the long drainage
path available for consolidation. An efficient way to do this is by providing vertical drains.
Vertical drains are continuous vertical columns of pervious material installed in clayey soil
for the purpose of collecting and discharging the water expelled during consolidation. Verti-
cal drains with pre-loading will rapidly accelerate consolidation. A further aspect of the
drains is that they reinforce the ground in which they are installed.
Vertical drains are mainly of two types: (i) sand drains, made by filling a cylindrical hole
with sand and (ii) pre-fabricated drains, also known as “wickdrains” or simply “wicks.”
These drains can also be of flexible corrugated plastic pipe, wrapped inside a filter. For want
of modern installation machines, these techniques are not yet available in India. However,
one particular type of wick called “sandwick” (sand packed in filter stocking) has been used
M19_PURU1773_01_SE_C19.indd 713 2/12/2013 8:20:26 AM

successfully in several installations in India (Dastidar et al., 1969; Som, 1975). Another type
of drain called rope drain has been developed and used in several projects by the Central
Buildings Research Institute (Mohan et al., 1977; Sengupta et al., 1980). The rope material
consists of natural fibres such as coir. In such drains, the drainage capacity is the major con-
straint.
Until about 1950, most installed vertical drains were sand drains. Since 1980, fabricated
drains have become popular because they cost less and can be installed quickly. In the
technically advanced countries, it has been reported that the cost of pre-fabricated drains
is about one-third the cost of sand drains. However, in India, only sand drains have been
widely used (Datye and Nagaraju, 1975, 1976). Figure 19.14 shows a typical arrangement of
vertical drains with drainage blanket and surcharge fill.
The holes required for installing sand drains are guarded against collapse by pipes or
mandrels which are inserted by jetting, driving, rotating, or vibrating. The soil that origi-
nally occupied the space where the hole is being made is excavated by washing or augering
(referred to as non-displacement drains) or, alternatively displaced downwards or sideways
by driving closed bottom pipes (referred to as displacement drains). Under Indian conditions,
displacement drains up to 400 mm in diameter can be installed by the equipment commonly
used for driven cast in place of concrete piles (Datye, 1982). Adopting a modified technique,
Datye (1982) reported that over 10,000 drains 200 mm in diameter have been successfully
installed. The usual installation consists of 200 to 450 mm diameter sand drains installed
at spacings of 2 to 5 m. Displacement drains are less expensive than augered or bored non-
displacement drains.
In the installation of sand drains, the soil adjacent to the well is disturbed, causing a
reduction in permeability. This effect is referred to as smear. Methods are available to account
for smear effects based on the permeability and thickness of the smear zone.
Vertical drains are ineffective in fibrous organic deposits and in clay deposits with abun-
dant pervious inclusions. Also, in sensitive clays, especially when drain installation is by
the displacement methods, soil disturbance may result in high initial pore water pressures
Pre-loading
Drainage
blanket
Vertical
Flow drain Lower boundary
lines compressible sand
Section AA
r4 = Equivalent radius
2r
A 60
60° A
2r
Plan
Fig. 19.14 Arrangement of sand drains
M19_PURU1773_01_SE_C19.indd 714 2/12/2013 8:20:26 AM

and a zone of low permeability around the drain. However, it is certainly true that in many
instances vertical drains have shortened the time required for soil stabilization.
19.6.3 Dynamic Consolidation

Heavy tamping, also referred to as dynamic consolidation, has also been used for cohesive
soils. Pre-construction settlements are usually of the order of two or three times the settle-
ment predicted for the construction itself. This process also reduces the secondary compres-
sion. Treatment time is much less compared to surcharge loading with sand drains and the
bearing capacity of clay increases by about 100% to 150%.
Because of the low permeability of cohesive soils, this technique takes more time than
for non-cohesive soils. Each location is subjected to several blows and a rest period of 1 to
4 weeks is given before repeating the process. Drainage is facilitated by the radial fissures
that form around impact points and by the use of horizontal and peripheral drains. Gener-
ally, the settlement is immediate. Because of time lapse, to be provided between successive
cycles, only a limited area can be treated economically.
19.6.4 Electro-osmotic Consolidation

The application of electro-osmosis for de-watering purposes has been discussed earlier. The
removal of water is an effective means of consolidation in fine-grained soils. Similar to a one-
dimensional hydraulic field, a one-dimensional direct current field is formed due to the
application of a current. The water flow rate is high in the initial stages, but decreases with
time. It ceases when a hydraulic gradient, induced by water content variation, tending to
cause a flow from the cathode towards the anode exactly balances the electricity-induced
hydraulic gradient causing a flow from the anode towards the cathode. When this condition
is satisfied, there is an increase in the effective stress. The amount of consolidation associated
with this effective stress increase is obtained from a void ratio versus pressure relationship
for the soil, determined in the usual manner. The rate of consolidation is governed by the
same relationships that apply to consolidation under directly applied loading.
The efficiency of the method is decreased by gas generation and drying and fissuring at
the electrode. The treated soils will have non-uniform changes in properties between elec-
trodes because the induced consolidation depends on the voltage, and the voltage varies
between the anode and cathode. In order to achieve a more uniform stress condition, rever-
sal of the polarity of the electrodes may be desirable. Pre-loading or surcharge fill techniques
may be combined with electro-osmosis to accelerate the consolidation.
19.7 GROUTING AND INJECTION

Grouting is a process whereby stabilizers, either in the form of a suspension or solution, are
injected into sub-surface soil or rock for one or more of the following applications:
1. control of groundwater during construction;
2. void filling to prevent excessive settlement;
3. strengthening adjacent foundation soils to protect them against damage during excava-
tion, pile driving, etc.;
M19_PURU1773_01_SE_C19.indd 715 2/12/2013 8:20:26 AM

4. soil strengthening to reduce lateral support requirements;

5. stabilization of loose sands against liquefaction;
6. foundation underpinning; and
7. reduction of machine foundation vibrations.
That the basic function of grouting is to serve should always be kept in mind when
considering different aspects of the subject. The three basic functions are the following
(Koerner, 1985):
1. Permeation or penetration: The grout flows freely with minimal effect into the soil voids or
rock seams.
2. Compaction or controlled displacement: In this condition the grout remains more or less
intact as a mass and exerts pressure on the soil or rock.
3. Hydraulic fracturing: Hydraulic fracturing or uncontrolled displacement occurs when the
grouting pressures are greater than the tensile strength of the soil or rock being grouted;
then, the latter material fails, and the grout rapidly penetrates into the fracture zone.
19.7.1 Suspension Grouts

Suspension grouts consist of solid particles like soil, cement, lime, asphalt emulsion, etc. car-
ried in water. The solution grouts are numerous, viz., aqueous, non-aqueous, colloidal, etc.
Particles in a suspension grout are of silt size, and hence these materials cannot be injected
into the pores of soils finer than medium to coarse sand sizes. For successful grouting of
soils, it has been established (Kravetz, 1958) that
(D15 ) formation
GR = > 20
(D85 ) grout
where GR is the groutability ratio, D15 the particle size corresponding to 15% finer of the
formation, and D85 the particle size corresponding to 85% finer of the grout.
The criterion basically limits the use of suspension grouting to permeation of sands and
gravels. Other considerations that must be taken into account in grouting design are the
grout’s setting time and its stability.
Grouting with Soil. Soil itself can be used to fill up some of the volumes in coarse-grained
deposits. Even fine sands and silts used for this purpose settle down quite quickly after
injection. The soil to be used as a grout should be a very fine-grained soil. Bentonite clay is
the commonly used material. Viscosity, strength, and flow properties of a bentonite clay can
be adjusted to suit the situation.
No flow of soil-grout occurs when the water-to-soil ratio is kept very low. Pressure is then
exerted by the grout against the soil mass from the outlet of the grout pipe. This causes the
densification and movement of the grout to adjacent areas. This technique was originally
known as mudjacking and quite often used to raise pavement slabs or to underpin shallow
building foundations.
To prevent a blow-out during grouting, the grouting pressure is generally limited to about
20 kN/m2 per metre of depth. Higher pressures are used when grouting under heavy struc-
tures or in other situations where greater confinement exists. Sometimes, higher grouting
pressure can be used by working from the outside of the area to be treated. In certain cases,
M19_PURU1773_01_SE_C19.indd 716 2/12/2013 8:20:26 AM

high pressures are deliberately applied to widen or increase the fracture, thus providing
added channels through which the grout can flow.
Grouting with Soil–Cement Mixes. Soil in combination with a stabilizing material, e.g.,
cement, would do better than soil alone. Grouts may have different properties depending
on the amount and type of soil, cement, and water they contain. The viscosity of a grout
depends on solid-to-water ratios and different cement-to-soil ratios.
In soil–cement systems, volumes of soil between four and six times the loose volume of
cement are common. The volume of mixing water varies from about three-fourth to twice
the volume of clay per bag of cement in cement–clay grouts, and from about one-third the
loose volume of sand per bag of cement to an equal volume of the same in cement–sand
grouts. Water–cement ratios in the range of 0.5:1 to 5:1 have been in use. The lower this ratio,
the less likely cement segregation and filtering will be, but injection will be more difficult
and the friction losses in the pumping system will be greater.
High viscosity systems with very low water content can be used as displacement grouts.
The advantage of these soil-cement mixes over soil alone is the permanence of the grout but
their dis-advantage is an increase in cost. The other factors, such as equipment, pressure,
and pumping rates, however, are roughly the same.
Grouting with Cement. Cement grouting has been widely used, more often in seepage
cut off beneath dams, but also in groundwater control in certain cases. Cement grouts are
usually made from Portland cement and water. Sometimes the cement particles come out
of the suspension before complete curing of the grout has occurred. This phenomenon is
referred to as bleeding. This behaviour is more pronounced when water content and fluidity
of the grout are greater.
When selecting a particular cement grout for use, one would obviously like to know its
final strength, flow rate, set time, shrinkage, permeability, and durability. Other ingredients
that are sometimes used in cement mixes are fine sand, clay (as discussed in the previous
section), fly ash, fluidizers, accelerators, and retarders or expansion additives.
Grouting with Lime. Grouting with lime is a special form of grouting – pressure injected
lime has been increasingly applied, especially for the stabilization of expansive soils for foun-
dations of light structures. In this process, a lime slurry, containing 3 to 4 N weight of lime per
litre of water plus a surfactant, is injected under high pressure (350 to 1,400 kN/m2). Treat-
ment locations may be spaced at 1 to 2 m laterally and 0.3 to 0.5 m vertically up to the depth
of seasonal moisture variation. Pumping of the grout is continued at each depth until refusal,
or until the slurry runs out at the ground surface. About 120 litres of grout per metre depth
is quite usual.
The method is best suited for expansive soils with cracks, fissures, slickensides, fractures,
and root holes. These passages, in conjunction with channels formed by hydraulic fracturing
under higher injection pressure, provide channels for the slurry throughout the soil. Lime
reacts with the soil adjacent to the cracks forming moisture barriers to protect the unre-
acted soil blocks against volume change. Free penetration of grout into soil pores is difficult
because of the large lime particle size and small bore size.
Displacement Grouting. Displacement or compaction grouting is a specialized technique
used for controlled densification of in situ soils at depth. This technique is not the same as
conventional grout filling within a soil mass by penetrating it with a cementing material.
The basic concept of compaction grouting is that of injecting a growing “bulb” of grout that
M19_PURU1773_01_SE_C19.indd 717 2/12/2013 8:20:27 AM

Grout injection pipe
Growing bulb of
grout
Radial densification
of soil particles
Fig. 19.15 Compaction grout concept
acts as a radial hydraulic jack, displacing the surrounding soil particles and thus radially
compacting the soil from the point of injection (Fig. 19.15).
Most displacement grouts are composed of a cement–sandy-loam mixture containing
three to five sacks of cement per cubic metre of soil. After thorough mixing, the materi-
als are pumped using a mud jack. Equipments are available to pump a zero slump grout
up to 30 m depth under a high pressure of 2,700 kN/m2. Most applications of a displace-
ment grout are for correction of differential settlements. This technique is more suitable in
partially saturated cohesive or organic soil masses, silts, sands, and soils containing void
pockets (Mitchell, 1970).
19.7.2 Solution Grouts

The advantages of solution grouting that serve to offset its high cost are (i) absence of par-
ticulate material, (ii) low viscosity, and (iii) control over setting time. Solution grouting is
done using “one-shot” or “two-shot” systems. In the one-shot system, all required chemicals
are injected together after re-mixing. Setting times are controlled by varying the catalyst
concentration according to the grout concentration, water composition, and temperature.
The two-shot system, wherein one chemical is injected followed by injection of a second
chemical which reacts with the first to produce a precipitate in the soil pores, may also be
used. Two-shot systems are slower and require higher injection pressures and more closely
spaced grout holes.
The electro-osmotic driving force was used by Karpoff in 1953 as a means for injecting
a stabilizer into a soil. In this method, the stabilizer is introduced at the anode and car-
ried towards the cathode by electro-osmosis. For this direct current, electrical gradients of
the order of 50 to 100 V/m were required. Such techniques may be attractive in cases in
which the permeability of the soil is so low as preclude injection under high pressures or in
situations where large pressures cannot be used. The method is likely to be expensive and
effective only in special cases (Mitchell, 1970).
M19_PURU1773_01_SE_C19.indd 718 2/12/2013 8:20:27 AM

19.8 CHEMICAL STABILIZATION

Chemical stabilization in the form of lime, cement, fly ash, and a combination of the
above is widely used in soil stabilization to (i) reduce the permeability of the soil, (ii) improve
shear strength, (iii) increase bearing capacity, (iv) decrease settlement, and (v) expedite
construction. Many cases of successful chemical stabilization attempts have been described
in the literature. Chemical stabilization may be used for surface soils more successfully.
Surface treatments are common in connection with the sub-grades or bases for pavement
construction. Mixture of soil and chemicals are mixed either mechanically in place or
by batch process. The optimum benefit of using these agents in stabilization must be
determined by laboratory testing. The general principles of these admixtures as stabilizers
are discussed below.
19.8.1 Lime Stabilization

Lime is an effective agent to be mixed with fine-grained soils with high plasticity. Hydrated
high calcium lime [Ca(OH)2], calcitic quick lime [CaO], monohydrated dolomitic lime
[Ca(OH)2, MgO], and dolomitic quick lime are the commonly used limes for stabilization.
Addition of lime to soil causes chemical reactions, such as cation exchange and flocculation–
agglomeration. In these reactions, the monovalent cations in clays are replaced by divalent
calcium ions. Further, these reactions make highly active clays inactive, changing the texture
of the clay soil. In general the reactions reduce the plasticity of the soil and thus improve the
strength and deformation properties of the soil.
The percentage of lime for effective action is in the range of 5% to 10%. The first 2% to
3% lime by weight has a substantial influence in improving the workability and property of
the soil. Lime stabilization can be applied in the field by any one of the following processes:
(i) the in situ material (and borrowed material, if necessary) is mixed with the proper amount
of lime at the site and then compacted after addition of moisture, (ii) the soil and the amount
of lime with water are mixed at a plant and then hauled back to the site for compaction,
and (iii) lime slurry can be poured through small drill holes or pressure injected to a depth
of 3 m at a spacing of 2 m depending on the situation. Both hydrated lime and quick lime
can be used. Quick lime has to be used cautiously as it may affect the tools and working
personnel. The curing period has been reported to affect the strength of lime-stabilized soils
considerably (Ramana Sastry, 1989).
19.8.2 Cement Stabilization

Cement has been widely used to stabilize sandy and clayey soils. Similar to lime, cement has
the effect of decreasing the plasticity and increasing the strength of the soil. Stabilization is
most effective in soil with less than about 40% of it has particle size of 2 μm. The percentage
of cement (by volume) required for the stabilization of sandy soil is in the range of 6 to 10,
for clays and silts of low plasticity it is 8 to 12, and for highly plastic clays it is about 10 to 14.
Clays with calcium ions are easily stabilized by cement, whereas clays with sodium are
better stabilized by lime stabilization.
M19_PURU1773_01_SE_C19.indd 719 2/12/2013 8:20:27 AM

Cement stabilization as a surface treatment is used in highway and runway sub-grades.

For field compaction, the mixing may be done at the site or in a mixing plant and then
transferred to the site. The mixed soil is then compacted to the required density with a
predetermined quantity of water.
19.8.3 Fly Ash Stabilization

Fly ash is a by-product of the pulverized coal combustion process and has silica, alumina,
and various oxides and alkalis as its constituents. It is fine-grained and pozzolanic in nature.
Fly ash reacts actively with hydrated lime and hence is used in combination with lime as a
stabilizer. A mixture of about 10% to 35% of fly ash and 2% to 10% of lime forms an effective
stabilizer for the stabilization of highway bases and sub-bases. Soil–lime–fly ash mixes are
compacted under controlled conditions with an adequate quantity of water.
19.9 SOIL REINFORCEMENT

Soil reinforcement is not a new concept but the principles already exist in nature and are
demonstrated by animals, birds, and plants. The scientific basis for the modern concept of
soil reinforcement lies in the idealization of the problem of soil reinforcement, wherein a
weak soil is reinforced by high-strength thin horizontal membranes (Westergaard, 1938).
The modern form of soil reinforcement was first applied by Vidal (1969). According to Vidal’s
concept, the interaction between the soil and the reinforcing horizontal membranes is solely
due to friction generated by gravity. The first major retaining walls using the Vidal concept
wre built in France in 1968. The new technology has been widely used in Europe and the
USA. However, this technique has not yet become popular in India, except in a few cases
(e.g., Nagaraj et al., 1982), the constraining factor being the non-availability and cost of
reinforcing materials (Datye, 1982).
Reinforced soil is somewhat analogous to reinforced concrete. But a direct comparison
between the functions of reinforcement in the two cases is not valid. The mode of action of
reinforcement in soils is not through carrying developed tensile stresses, as in reinforced
concrete, but rather of anisotropic reduction of the normal strain rate (Jones, 1985).
Jones (1985) identifies several soil reinforcement field applications, viz., in bridge works,
dams, embankments, foundations, highways, housing, industries, military applications,
railways, root pile systems, pipe works, and waterway structures. Some of the applications
are shown in Fig. 19.16.
A wide variety of materials are in use as reinforcing materials. Early structures were formed
using organic materials such as timber, straw, or reed for reinforcement. As these materials
are less durable, new materials such as steel, concrete, glass fibre, rubber, aluminium, and
thermo plastics have been used successfully.
Durability of reinforced soil is the basic requirement; hence sufficient judgement has to
be exercised while selecting the reinforced material. Resistance to corrosion is the most sig-
nificant factor affecting the durability. High alloy steel, aluminium, glass-fibre reinforced
plastics (GRP), and geosynthetics (discussed in the next section) are non-corrosive and in
general have long lives.
M19_PURU1773_01_SE_C19.indd 720 2/12/2013 8:20:27 AM

Bridge structure
Reinforcement
Reinforcement
(b) Embankment reinforced to produce stability
Tank
(a) Bridge abutment and support
to bankseat Reinforcement
Highway
Reinforcement
(c) Reinforced tank foundation
Highway
Reinforcement
Reinforcement
(d) Stepped highway structure (e) Cutting formed using soil nailing
Fig. 19.16 Some applications of soil reinforcement (Source: Jones, 1985)
19.10 GEOTEXTILES AND GEOMEMBRANES

19.10.1 Geotextiles
Geotextiles are porous fabrics manufactured from synthetic materials that are primarily
petroleum products and others, such as polyester, polyethylene, polypropylene and polyvi-
nyl chloride, nylon, fibre glass, and various mixtures of these. They are manufactured in
thicknesses ranging from 10 to 300 mils (1 mil = 0.0254 mm), widths up to 10 m, and roll
lengths up to about 600 m. Geotextiles are termed filter fabrics and have permeabilities com-
parable in range from coarse gravel to fine sand.
Geotextiles are manufactured in a variety of patterns, the most common methods being
(Koerner, 1985)
s woven-made from continuous mono-filament or slit-film fibres;
s non-woven-made from continuous or staple fibres joined at the fibre cross-over by
mechanical, thermal, or chemical bonding;
s grid-made from a sheet of polymer, punched and then elongated in at least one
direction; and
s hybrid combinations of any of the above materials and techniques.
Geotextiles have been used in a variety of civil engineering works. Thus, in the selec-
tion of a proper geotextile, due importance has to be given to the major function that the
geotextile is intended to perform. The four major functions are soil separation, filtration,
drainage, and reinforcement. They are explained briefly below.
M19_PURU1773_01_SE_C19.indd 721 2/12/2013 8:20:27 AM

Geotextile
Top granular soil
Subsoil Drain pipe
Geotextile
envelope
(a) Separation of two materials (b) Separation of soil from

discharge element
Fig. 19.17 Geotextiles as separators (Source: Zanten, 1986)
Geotextiles as Separators. There are situations in which two groups of particles of

different size ranges cannot be placed together, as there is a danger of small-size particles
migrating to the voids of large-size particles. For example, in the construction of highways,
a clayey sub-grade can be kept separate from a granular base course (Fig. 19.17). In such
cases, the usual procedure is to provide a soil containing grain sizes intermediate between
the soils. If there is a wide variation of grain sizes between the two soils, more than one inter-
mediate grain-size soil has be provided. The obvious alternative in such a situation is the
use of a geotextile. The advantages of using geotextiles as separators over single or multiple
soil layers are the following: (i) simplicity in the construction, (ii) less excavation required,
(iii) less weight placed, and (iv) short time required for construction.
Geotextiles as Filters. Geotextiles can be used more effectively as filters. Geotextile filtra-
tion occurs in fabrics where the flow of water transports some of the fine-grained particles
of the protected soil onto the surface of the geotextiles. This modification in the soil and void
of the fabric attains equilibrium after sometime. Only then does clear water start passing
through the fabric. For relatively thin geotextiles, most of the filtering takes place within
the soil, upstream from the fabric. This application of geotextiles has been widely used.
Figure 19.18 signifies some of the applications of geotextiles as filters.
Geotextiles as Drains. Geotextiles themselves function as drains because they have a higher
water transporting capacity than that of the surrounding material. Geotextile drainage occurs
Geotextiles
Geotextiles
Clear Turbid
water water
Shell Core Shell
(a) Transition filter (b) Silt curtain filter
Fig. 19.18 Geotextiles as filters (Source: Zanten, 1986)
M19_PURU1773_01_SE_C19.indd 722 2/12/2013 8:20:28 AM

Geotextiles
Surcharge Retaining
wall
Geotextiles
Drain pipe
(a) Vertical and horizontal drainage (b) Drainage behind walls
Fig. 19.19 Geotextiles as drains (Source: Zanten, 1986)
either cross-plane, when functioning primarily as a filter, or in-plane, when water is transmitted
within the geotextile structure itself. In the latter case, a bulky geotextile or a composite system
is needed. Figure 19.19 represents typical cases of application of geotextiles as drain.
Both the filtration and drainage systems have the following advantages: (i) faster
installation, (ii) less soil to excavate and dispose of, (iii) less load, and (iv) greater system
stability.
Geotextiles as Reinforcement. Since the tensile strength of soil is less, geotextiles, which
have high tensile strengths, can contribute to the load-bearing capacity of the soil. Thus,
geotextiles perform the function of reinforcement in soils. This application has solved
many construction problems on soft and compressible soils. The most important use has
been in road construction on soft sub-grades. Geotextiles are also used to reinforce walls
and embankments. Figure 19.20 represents some of the applications of geotextiles as
reinforcement.
General Applications of Geotextiles. Geotextiles can also be used in the following situ-
ations (Zanten, 1986):
1. Bank and bed protection: In this application, the geotextile protects the underlying mate-
rial against excess pore water pressure. The filtering material is water permeable during
its life time. This construction functions in a subsidiary way as a separating sheet and
reinforcement.
Backfill
Geotextiles
Wall
Embankment
Geotextiles
(a) Reinforcement in embankment (b) Fabric reinforced wall
Fig. 19.20 Geotextiles as reinforcement (Source: Zanten, 1986)
M19_PURU1773_01_SE_C19.indd 723 2/12/2013 8:20:29 AM

2. Embankments, slopes, and foundations: Here, geotextiles function as reinforcement to

increase the resistance of soils to shear stresses, and as separation and filter media.
3. Vertical drainage: In this case, the geotextile is used to accelerate the consolidation by
reducing the excess pore water in the surrounding soil. This is achieved with drains with
a low entrance resistance and a large discharge capacity.
4. Horizontal drainage: In this application, the geotextile serves as a selective filter that pre-
vents soil particles of a certain size from migrating the drainage medium, and it also acts
as a means of reducing the entrance resistance in connection with the discharge of water.
5. Road and railway construction: Here the geotextile is used in the foundation to separate
the subsoil from the road foundation material, to increase the load-carrying capacity
of the subsoil, and to drain surface water into the subsoil.
19.10.2 Geomembranes
Geomembranes are thin materials with very low permeability. They are flexible and are
manufactured from synthetic or bituminous products. They may be strengthened, if neces-
sary, with a fabric or film. Geomembranes differ from geotextiles with reference to the rate
of permeability. The permeability is high (as in sand) in geotextiles and very low in geomem-
branes (as in bentonite or colloidal clay). For all practical purposes, geomembranes may be
considered to be impermeable to both gases and fluids. This makes them ideal for construct-
ing waterproof or gasproof barriers between adjacent bodies of soil, or soil and fluid.
Geomembranes are used in the following situations (Zanten, 1986):
1. Sealing against fluid percolation – e.g., on sea coasts, river banks, shipping canals and
locks, reservoirs, terrain bunding, etc. The function of geomembranes in these cases is to
form a barrier between the water and the surroundings and to ensure that water transport
is reduced to a minimum.
2. Buffers against pollutants – e.g., permanent or temporary storage of waste products,
waste-water treatment plants, basins for use in emergencies, roads in areas used for
extraction of groundwater, etc. In these cases, the function of the geomembrane is to cre-
ate a barrier between two media and prevent any mixing of these media.
As stated above, geomembranes are manufactured from synthetic (thermoplastic) or
bituminous products. Some of the synthetic materials used are high-density polyethylene,
low-density polyethylene, polyvinyl chloride, etc.
19.11 OTHER METHODS

19.11.1 Thermal Methods
Of the two possible temperature extremes, hot or cold for stabilizing fine-grained soils, heat is
rarely used. But it is technically feasible to stabilize saturated clays by heat. A temperature of
100°C causes the fusion of clay particles into a solid material like brick. The burning of liquid or
gas fuels in boreholes, or injection of hot air into 0.15 to 0.20 m diameter boreholes, can produce
1.3 to 2.5 m diameter strengthened zones after continuous treatment for more than 10 days. Dry
or partly saturated weak clayey soils and loess are well-suited for this type of treatment. The
economics of using heat, however, precludes its use in most construction projects.
M19_PURU1773_01_SE_C19.indd 724 2/12/2013 8:20:29 AM

Ground freezing appears to be gaining popularity in recent years. Frozen soil is far
stronger and less pervious than unfrozen ground. Stabilizing soils by freezing has the fol-
lowing advantages: (i) high strength of stabilized soil, (ii) low permeability in the stabilized
soil, (iii) barrier to seepage flow, (iv) protection from soil deformation, (v) noiseless opera-
tion, and (vi) applicability to a wide range of soils. Ground freezing is accomplished by
bringing a refrigerant into the proximity of soil pore water. The pore water may be stationary
or moving with a velocity of 2 m per day. The pore water around the refrigerant pipes freeze,
and continued freezing yields a continuous wall of ice.
Ground freezing has applications in temporary underpinning and excavation stabiliza-
tion. Other applications are in backfreezing of soil around pile foundations in permafrost
and in maintenance of frozen soil under heated buildings on permafrost. Design on frozen
ground involves properties of the frozen ground, heat flow, transfer of water to ice, and
design of a refrigeration system.
19.11.2 Moisture Barriers

Expansive clays and partly saturated, lightly cemented soils, such as loess, can provide
excellent foundation support for light structures, provided they are maintained relatively
dry. Moisture barriers are sometimes used to keep water from critical zones under
foundations.
Geomembranes discussed earlier can act as moisture barriers more effectively than
sand or gravel blankets, and trenches may also act as effective moisture barriers, provided
they can be maintained at a low degree of saturation. The hydraulic conductivity of such
unsaturated coarse layers is so low as to effectively prevent significant water transfer across
the layer.
19.11.3 Pre-wetting
One technique for the stabilization of expansive soils that can be effective under light struc-
tures, such as dwellings, is to flood the area prior to construction. Successful ponding is facil-
itated initially by the natural soil. After successful pre-wetting, the soil has a water content
closer to that to be attained after construction; hence, subsequent volume changes are small.
Lime treatment of the surface layer to a depth of 0.3 to 0.5 m after ponding may be ben-
eficial. This treatment provides a working platform for construction and an impermeable
moisture barrier to retard subsequent desiccation of the pre-wetted soil.
19.11.4 Addition or Removal

The engineering property of a soil can be significantly changed by adding some selected soil
or by removing some selected fraction of the soil. Sometimes poor soil may be removed and
replaced with the same soil treated by compaction, with or without admixture. In general,
the addition–removal technique of soil stabilization may work out economical in the absence
of groundwater. Invariably, all the inorganic soils can be processed and treated to form an
acceptable construction medium. This technique is not suitable for highly organic soils,
peats, and sanitary fills. No generalized procedure can be given for this technique as it
depends entirely on the problem and the site condition.
M19_PURU1773_01_SE_C19.indd 725 2/12/2013 8:20:29 AM

POINTS TO REMEMBER
19.1 Soil improvement in its broadest sense is the alteration of any property of a soil to
improve its engineering performance. This may be either a temporary process or a
permanent measure to improve the performance of the completed facility.
19.2 Surface compaction is one of the most widely used techniques and is also one of the
oldest techniques of soil densification.
19.3 Drainage methods like well-point systems, deep-well drainage, etc. are adopted to
control the groundwater entry into the construction site, as a temporary or permanent
measure, thereby ensuring a safe and economical construction scheme.
19.4 Vibration techniques produce shock waves which cause liquefaction followed by den-
sification and settlement accompanying the dissipation of excess pore water pressure.
These methods will be effective only in coarse-grained soils with less fines.
19.5 Compression of fine-grained soils is effectively done by pre-loading and surcharge
fills, by installing vertical drains, and by dynamic consolidation.
19.6 Grouting is a process whereby stabilizers, either in the form of suspension or solution,
are injected into sub-surface soil or rock fissures to control groundwater intrusion,
prevent settlement, increase the strength of soil, etc.
19.7 Chemical stabilization uses lime, cement, fly ash, and a combination of the above for
soil stabilization.
19.8 Geotextiles are porous fabrics manufactured from synthetic materials which are
nowadays used for four major functions, viz., soil separation, filtration, drainage, and
reinforcement.
19.9 Geomembranes are flexible materials with very low permeability, which are manufac-
tured from synthetic or bituminous products. They are used for sealing against fluid
percolation and as buffers against pollutants.
QUESTIONS
Objective Questions
1. Well points are generally recommended where the water table needs to be lowered
only by a small depth.
2. When an electrical gradient is applied in a saturated soil, water moves towards the
anode.
3. Fine particles reduce the permeability of a material which is a prime factor for
liquefaction.
4. Vibratory rollers are the best and most economical for attaining high density in
cohesionless soils.
5. Suitability of a backfill material in vibroflotation is independent of the gradation of
the material.
M19_PURU1773_01_SE_C19.indd 726 2/12/2013 8:20:29 AM

19.2 Fine-grained soils are more troublesome to drain because of

1. Small-sized grains which form small voids
2. High surface area which attracts more water
3. Presence of clay minerals
4. Capillary forces acting on the pore water
19.3 Indicate the incorrect statement. Injection method is used to
(a) Increase the strength
(b) Reduce the permeability
(c) Increase the grain size
(d) Reduce the compressibility of the soil
19.4 Generally, grouting can be used if the permeability, k, of the deposit is
(a) <10−3 m/s
(b) >10−3 m/s
(c) <10−5 m/s
(d) >10−5 m/s
19.5 Stone columns are not applicable in deposits of
(a) Highly organic silts or clays
(b) Loose sands
(c) Inorganic clays
(d) Inorganic silts
19.6 To avoid cratering in a blasting technique the minimum depth of charge should be
______ the radius of the sphere of influence.
(a) Equal to
(b) Greater than
(c) Less than
(d) Twice
19.7 The vibro-flotation technique is best suited for densifying
(a) Very loose sands below the water table
(b) Layered clays and fine cemented particles
(c) Organic clays with fine silt
(d) Coarse sands and gravels
19.8 The installation of sand drains causes the soil adjacent to the well to
(a) Reduce in shear strength
(b) Increase in porosity
(c) Reduce in permeability
(d) Increase in compressibility
19.9 Compare the use of sheep’s foot and vibratory rollers in the surface compaction of
granular soils.
19.10 In electro-osmotic stabilization, what are the different types of anodes used? What
types of electrodes are used in the marine environment?
M19_PURU1773_01_SE_C19.indd 727 2/12/2013 8:20:29 AM

19.11 Why is cratering undesirable in the process of densifying granular soils by blasting
with explosives?
19.12 How does deep dynamic compaction densify granular soils? Describe the influence of
water content in the process.
19.13 Evaluate the technique of vibroflotation, compaction piles, and terraprobe with
reference to equipment, time for compaction, and the maximum density achievable.
19.14 Considering the groutability of various types of soils, what method do you recom-
mend to grout fine-grained soils?
19.15 Discuss the advantages of using fly ash in cement grouting over naturally available
soils.
19.16 Describe with illustrations the differences between geotextiles and geomembranes.
19.17 Discuss the various methods used to control groundwater in excavations of soft clays.
Indicate their relative suitability.
19.18 Describe a method suitable to stabilize a highway foundation in hilly terrain with
high rainfall.
19.19 Two earth dams, each 100 m high, are to be constructed, on foundations whose soil
properties are shown below. Discuss and suggest a foundation stabilizing technique
in each case.
Dam A B
Coefficient of permeability 120 m/year 0.6 m/year

Nature of soil Medium sandy soil Dr = 58% Soft clay
wL = 60%
Shear parameters c = 0, φ = 32° c = 0.04 N/mm2, φ = 2°
19.20 How can the horizontal spacing of reinforcing strips be designed for the material in a
retaining wall?
19.21 A highway alignment passes through a region where the subsoil is a highly compressible
clay. Describe any one technique by which the consolidation of this clay can be has-
tened so that the construction of the road can be completed early.
EXERCISE PROBLEMS
19.1 A grain-size analysis of a backfill material for a vibroflotation work yielded the
following characteristic grain sizes: D10 = 0.10 mm, D20 = 0.18 mm, and D50 = 0.55 mm.
Find the suitability number and give the rating of the material.
19.2 A surcharge fill has a volume of 6,000 m3 and is placed at a dry unit weight of 20.0 kN/m3.
The borrow source for the fill has a dry unit weight of 15.6 kN/m3 and G = 2.68. Estimate
the volume of material required from the borrow to make the surcharge fill.
M19_PURU1773_01_SE_C19.indd 728 2/12/2013 8:20:29 AM

20
Embankment Dams
CHAPTER HIGHLIGHTS
Types of embankment dams – Components of embankment dams – Other
embankment details – Flow nets for earth dams – Design criteria for earth
dams: safety against over-topping – Control of seepage and pressure in earth
dams – Stability of upstream and downstream slopes – Selection of dam sec-
tion: Earth dams – Rockfill dams
20.1 INTRODUCTION
Embankment dams are water impounding structures. These are flexible structures
which can deform slightly to conform to the deflection of the foundation. Embankment
dams are primarily made out of earth and rock fragments. Thus, the term embankment
encompasses both earth and rockfill dams. These dams are also referred to as dykes or
simply embankments or banks. Earth dams that are used to confine flood waters are
called levees or guide banks. Dams that are both safe and economical can be constructed
at a given site with the available materials.
Since the dawn of history, man has built dams of earth and rocks. Numerous tanks of
reservoirs found in South India are over 2,000 years old. With modern technology earth
dams of varying heights have been constructed in different parts of India, such as Ram
Ganga Dam, Kishan Dam, Kothar Dam, Nagarjuna Sagar Dam. It is often said that the
largest structure ever built by man is an earth dam.
20.2 TYPES OF EMBANKMENT DAMS

Based on the type of construction, embankment dams may be classified as rolled fill or
placed earthfill dams and hydraulic fill dams. In the rolled fill type of construction, the
major portion of the embankment is constructed in successive layers which are mechanically
M20_PURU1773_01_SE_C20.indd 729 2/12/2013 8:19:44 AM

compacted. Materials needed for construction are invariably procured from the construction
sites, known as borrow areas. Materials from borrow areas are transferred to the embankment
location and spread to the required thickness; the proper amount of water is added and then the
material is compacted by power-operated rollers. The materials used are unprocessed natural
materials.
Some small quantities of specific materials required for drains, filters, etc. are transported
to the site or processed by screening or otherwise at the site.
In hydraulic fill dams the embankment materials are suspended in water. The soil–water
suspension (generally with about 85% water) is pumped to the required site and allowed
to settle. With proper control of the suspension and the settling process, a fairly uniform
construction can be achieved. However, because of segregation problems for coarse materials
this method may not be suitable. The placed earthfill type is widely used.
Embankment dams are of three types, viz., diaphragm, homogeneous, and zoned.
20.2.1 Diaphragm Type

In this type of dam, a major portion of the embankment is constructed from pervious
materials. A water barrier is formed using a thin diaphragm of impermeable material. The
diaphragm may be located in the centre as a vertical core or placed as a blanket on the
upstream face. The material used for the diaphragm may be earth, cement concrete, bitumen,
etc. If an earth core is provided to serve as a diaphragm, the thickness should be less than
3 m or less than the height of the embankment.
20.2.2 Homogeneous Type

A purely homogeneous dam is constructed using a single kind of material excluding the
material used for slope protection. Many of the low- to moderate-height dams are essentially
homogeneous. The material selected for such dams should be sufficiently impervious and
for stability requirements the slope should be relatively flat. Homogeneous dams of 6 to 8 m
height have to be provided with some type of downstream drain, which helps to reduce the
pore water pressure in the downstream portion of the dam and control any seepage. The
types of drains provided in homogeneous dams are toe drains, horizontal blanket drains,
and chimney drains.
20.2.3 Zoned Type

This is a more common type of dam constructed using, basically, pervious and impervious
materials. The impervious material, called the core, is placed at the centre and is flanked by
zones of pervious materials called shells or casings (discussed in the next section).
The central core is supported and protected by the shells. The upstream shell affords sta-
bility against sudden drawdown and the downstream one acts as a drain to control the line
of seepage. The materials for the pervious zones may be sands, gravels, cobbles, or rocks or
mixtures of these materials. The width of the core is controlled by the availability of material
and design requirements, such as stability and seepage.
If a variety of soils is available in a location, there will be an ample range of material avail-
able to choose from for different sections of embankment. In such situations, there will be no
M20_PURU1773_01_SE_C20.indd 730 2/12/2013 8:19:44 AM

Embankment Dams 731
constraint on the height of the dam and the zoned type will invariably yield a more economical
section. If the major part of the dam is composed of rock, it is classified as a rockfill dam.
20.3 COMPONENTS OF EMBANKMENT DAMS

Each embankment dam consists of three basic components, viz., foundation, shell, and core
(Fig. 20.1). Depending on the type of dam, additional appurtenances are added to enable the
basic components to function efficiently.
20.3.1 Foundation
The foundation of a dam is the sole supporting medium resisting the vertical and horizontal
forces. Depending on the foundation material, i.e., whether it is soil or rock, the foundation
may allow or resist the flow of water. Rocks form the best foundation material provided they
are free from faults, joints, or seams of soft shale or clay, etc. Sands and gravels also provide
good support for dams, but adequate steps should be taken to control the seepage. Fine
sands with relative densities less than about 65% should be compacted by vibration so as to
avoid liquefaction. Clay foundations pose serious stability and settlement problems unless
measures are taken to accelerate the consolidation. Because of their low shear strength they
require a flat slope.
20.3.2 Casing
As pointed out earlier, the shell imparts stability and protects the core. All relatively pervi-
ous materials which are not prone to cracking under normal atmospheric conditions are
suitable for casing. Table 20.1 shows (IS: 8826, 1978; 1498, 1970) the suitability of different
soils for use as shell and core.
Core
Transition filter
Transition filter
Top
Wave protection rip-rap Sod or rip-rap for
width
Upstream erosion protection
impervious
Filter
blanket Free board Relief
Internal drain
Top Down- Filter wells
Upstream stream Toe drain
shell Toe
shell
Cut-off
Foundation
Impervious stratum
Note : Not all of the above ordinarily would be incorporated in any one dam
Fig. 20.1 Parts of an earth dam
M20_PURU1773_01_SE_C20.indd 731 2/12/2013 8:19:46 AM

Table 20.1 Suitability of soils for construction of earth dams
Relative suitability Homogeneous dams Zoned earth dams Impervious blanket
Core Casing
Very suitable GC GC SW, GW GC
Suitable CL, CI CL, CI GM CL, CI
Fairly suitable SP, SM, CH GM, SM, SC, CH SP, GP CH, SM, SC
Poor – ML, MI, MH – –
Not suitable – OL, OI, OH, Pt – –
Source: IS: 8826 (1978); IS: 1498 (1970).
The upstream and downstream slopes of a casing have to be decided based on availability
of material, foundation condition, height, and type of dam. The upstream ranges from 2:1 to
4:1 for stability requirements, and a flat one is chosen for material with low permeability. The
usual downstream slope varies from 2:1 to 2.5:1. Goel et al. (1980), based on statistical shear
strength data, have recommended design slopes for dams (up to 15 m high) which confirm
the slope ranges given above.
20.3.3 Core
The core acts as an impermeable barrier and prevents the free seepage of water through the
body of the dam. Soils with high degrees of compressibility, swelling, shrinking, and organic
content are not suitable for cores. The Indian Standards’ recommendation for cores are listed
in Table 20.1.
Based on the availability of material, topography of the site, and diversion considerations,
the core may be positioned either centrally or inclined upstream. The top level of the core
should be fixed at 1 m above the maximum water level. This condition is imposed to prevent
seepage by capillary syphoning. The minimum top width of the core should be 3 m and
the final thickness has to be decided based on the practical considerations given below (as
suggested by IS: 8826, 1978): (i) availability of suitable impervious material; (ii) resistance
to piping; (iii) permissible seepage through the dam; (iv) availability of other materials for
casing, filter, etc.
20.3.4 Other Embankment Details

In Fig. 20.1, the three basic components and other appurtenances are schematically repre-
sented. All the appurtenances are not ordinarily provided in one dam. In IS: 8826 (1978)
typical earth rockfill dam sections are given with more details.
Top Width. The top width of the dam is fixed depending on the work-space require-
ments. A minimum crest width of 6 m is recommended in IS: 8826 (1978). The following
empirical expression may be used to find the crest width b in metres for a dam of height H:
H
b= + 3 for very low dams (20.1)
5
M20_PURU1773_01_SE_C20.indd 732 2/12/2013 8:19:47 AM

Embankment Dams 733
b = 0.55 H + 0.2 H for dams with H < 30 m (20.2)

1/ 3
b = 1.65( H + 1.5) for dams with H > 30 m (20.3)
Free-Board. This is the vertical distance between the reservoir level and the crest level.
The free-board is said to be minimum for the maximum reservoir level and normal for the
normal reservoir level. The primary function of a free-board is to save the embankment from
over-topping by severe wave action and large inflow flood. For a rational determination of
free-board, details of height and action of waves are required. Arthur (1973) recommends
normal and minimum free-boards based on fetch∗ (Table 20.2).
Cut-off. On a permeable foundation, if the under seepage is very high it is necessary to
extend the core down into the foundation up to the impervious layer. Such a provision not
only reduces the loss of stored water but also prevents sub-surface erosion by piping. The cut-
off should be positioned such that the centre line is in the upstream of the impervious core.
While deciding the width, the following Indian Standards recommendations may be
considered: (i) provide sufficient working space for compaction equipment and to carry out
certain grouting and (ii) provide safety against piping. To satisfy the second condition, a
bottom width of 10% to 30% of hydraulic head has to be provided. A minimum width of 4 m
is recommended. Side slopes of at least 1:1 or flatter may be provided.
Internal Drain. This is essential primarily to carry away the seepage passing through the
core or the cut-off. Secondly, this is necessary to prevent undue saturation of the upper part
of the downstream shell by rain or spray falling on the dam. An internal drainage system
consists of inclined and horizontal filters, rock toe, toe drain, etc. For design details of
internal drainage systems refer to IS: 9429 (1980).
Transition Filter. This is needed between a coarse and fine material to act as a transition
medium to prevent the migration of fine material into coarse material. In zoned dams, this is
provided between the core and the shell. The transition filter may be omitted when the seepage
gradient is low and no wide variation exists in the grain sizes of core and shell.
For the reasons mentioned above, all the internal and toe drains require protective filters.
Slope Protection. Upstream slope protection is ensured by providing rip rap. This prevents
erosion or wash by waves of upstream materials. Usually, it starts from above the maximum
water level to just below the minimum.
Table 20.2 Normal and minimum free-boards
Fetch (km) Normal free-board (m) Minimum free-board (m)
<1.60 1.22 0.91
1.60 1.52 1.22

4.03 1.83 1.52
8.07 2.44 1.83
16.13 3.05 2.13
Source: Arthur (1973).
∗
Fetch is the distance over which the wind can act on a body of water.
M20_PURU1773_01_SE_C20.indd 733 2/12/2013 8:19:47 AM

The downstream slope is protected by providing rip rap or sod (turfing). The conventional
practice is to provide the entire downstream slope with turfing, which prevents the slopes
from rain cuts. In certain cases, even paving is resorted to, but it is more expensive. Sometimes
berms are employed on the downstream face to accommodate access for maintenance. This
also forms a break in the slope to reduce the rain wash. For design details of slope protection
refer to IS: 8237 (1976).
Upstream Blanket. The horizontal upstream impervious blanket is provided to increase
the path of seepage when full cut-off is not practicable on pervious foundations. The blanket
is connected to the core of the dam and may or may not be provided with a partial cut-off. The
permeability of the blanket material should be far less than that of the foundation material.
Table 20.1 gives Indian Standards recommendations. For design details refer to IS: 8414 (1977).
Relief Wells. These are vertical or nearly vertical holes arranged in one or more lines or
in a pattern so as to reduce the internal pressure and collect the seepage. Relief wells are also
used as a corrective device in completed dams that later develop unforeseen seepage pres-
sures. These wells may be installed either downstream or inside the structure.
20.4 DESIGN CRITERIA FOR EARTH DAMS

More than 200 earth dams have been studied by Middle Brook and the causes of failure have
been attributed to the following:
Types of failure Percentage
Over-topping 30
Seepage through foundations 25
Upstream and downstream slope slides 15
Conduit leakage 13
Slope paving 5
Miscellaneous 7
Unknown 5
In a broader sense, the above failures may be grouped under three main classes, viz.,
hydraulic failures (40%), seepage failures (30%), and structural failures (30%).
Based on the knowledge gained from the studies of failures of old and new dams, the fol-
lowing design criteria are laid out:
1. There should be no danger of over-topping of the dam.
2. The seepage flow through the dam and foundations should be controlled.
3. There should be no free passage of water from the upstream to the downstream face.
4. The upstream and the downstream slopes should be safe against sliding.
5. The crest and upstream and downstream faces should be protected from wave action,
rain, and frost action.
20.4.1 Safety Against Over-Topping

The causes of over-topping may be either inadequate provision of spillway or the free-board.
The provision of spillway capacity is based on the design flood, which is estimated from the
M20_PURU1773_01_SE_C20.indd 734 2/12/2013 8:19:47 AM

Embankment Dams 735
principles of hydrology. Generally, for a 30 m dam a 100-year flood and for dams higher than
30 m a 250-year flood may be considered. The free-board requirements have already been
discussed.
20.4.2 Control of Seepage and Pressure in Earth Dams

Uncontrolled seepage flow affects the earth dams broadly in two ways, viz., excess leakage
and excess pore pressure or gradients. Seepage control is the correction of the conditions
which lead to these troubles. While applying corrective measures care should be taken that
a correction at one point does not bring in or increase the trouble in some other point. Data
needed for analysis for these aspects are obtained from flow nets.
Flow Nets for Earth Dams. In Chapter 6, we discussed seepage under impermeable struc-
tures, such as flow beneath concrete dams, sheet pile walls, etc. These are termed confined
flow conditions whereas the flow through a permeable structure, such as an earth dam, is an
example of unconfined seepage.
In an earth dam, the phreatic surface constitutes the top flow line, and its position must
be located first. The different boundary conditions for an earth dam are shown in Fig. 20.2.
Seepage through the dam takes place below the phreatic line. As the pressure on the phreatic
line is equal to the atmospheric pressure, the total head is equal to the elevation head. The
soil is homogeneous and isotropic except at the rock filter; hence, the conventional prop-
erties of flow net are satisfied. The phreatic line takes different shapes depending on the
materials used in the dams. Figure 20.3 shows different phreatic lines for various earth dam
cross sections.
It has been mathematically shown that the basic shape of the phreatic surface is that of a
parabola and it deviates only at the upstream and downstream faces, depending on the cross
section of the dam. Thus, the entry and exit conditions are discussed first.
As the upstream entry surface is an equipotential line with maximum pressure head, the
flow lines must intersect it at right angles as long as the equipotential line with maximum
head (i.e., the upstream face of the dam) is less than 90º (Fig. 20.4a). When an upstream
coarse filter is provided, the equipotential line with maximum head may be greater than or
equal to 90º. In such cases, the phreatic surface’s entry is horizontal since water at zero pres-
sure cannot flow upwards (Fig. 20.4b).
Phreatic line
Upper flow boundary
Q
Equipotential lines Flow lines

R
Rock-filter toe
P
S
Flow line Impervious
Fig. 20.2 Boundary conditions and flow net for a homogeneous dam
M20_PURU1773_01_SE_C20.indd 735 2/12/2013 8:19:47 AM

Phreatic line
Rock toe filter
Impervious
Phreatic line
Horizontal drain
Impervious
Phreatic line
Impervious
Coarse soil
Phreatic line
Impervious
Phreatic line
Coarse soil
Impervious
Fig. 20.3 Phreatic lines for various dam cross sections
M20_PURU1773_01_SE_C20.indd 736 2/12/2013 8:19:48 AM

Embankment Dams 737
Phreatic line
°
90
Soil
N
or
m
al
b < 90°
(a) b < 90°
Phreatic line
Normal
Coarse Soil
filter
b = 90°
(b) b = 90°
Phreatic line
al
m
or
N
Horizontal
Coarse Soil
filter
b > 90°
(c) b > 90°
Fig. 20.4 Phreatic line details at entry to seepage zone (Source: Whitlow, 1983)
M20_PURU1773_01_SE_C20.indd 737 2/12/2013 8:19:48 AM

The theoretical parabola may have to be modified at the downstream or exit surface
also, depending on the conditions at the toe. No correction to the basic parabola is needed
when the downstream exit is a horizontal filter. When the toe filter is angular with α < 180º
(Fig. 20.5), a correction is made to relocate the phreatic line (as proposed by Casagrande,
1937). If the basic parabola cuts the downstream slope at S and the exit surface intersects the
Δa
base at F, the correct point is located at R, adopting the ratio (Fig. 20.6). In case no
a + Δa
toe filter is provided and the base is impermeable, the phreatic line makes a tangent with the
Δa
downstream face while exiting. This point can be corrected and relocated adopting
ratio (Fig. 20.7). a + Δa
Now, let us consider the properties of a parabola. The parabola is the locus of all points
equi-distant from a fixed point, called the focus, and a line, called the directrix (Fig. 20.8).
By definition, FA = AB, FD = DC, and FO = OE, taking the vertex O as the origin and x
and z as coordinates:
p
DC = + x
or 2
2
⎛ p⎞
(FD)2 = z 2 + ⎜⎜ x − ⎟⎟⎟
⎜⎝ 2⎠
or
⎛p ⎞2 ⎛ ⎞2
⎜⎜ + x ⎟⎟ = z 2 + ⎜⎜x − p ⎟⎟ (Therefore, DC = FD)
⎜⎝ 2 ⎠⎟ ⎜⎝ 2 ⎟⎠
Simplifying, the equation of the parabola is given as
z 2 = 2 px (20.4)
where 2p is the parameter of the parabola.
Horizontal Under-drainage. Figure 20.9 represents the condition of discharging into a
horizontal toe filter. Draw the dam and its appurtenances to a suitable scale. Assume the
focus point F as the inner end of the filter. Locate point D taking DC = 0.3BC. Locate the
directrix by extending BC and cutting by an arc with D as centre and DF as radius; i.e., make
DF = DE. Draw a vertical tangent EH to get the directrix. As all the points on the parabola
are equidistant from the focus and directrix, from any point N if a vertical is drawn at X,
then FX = XX′. Similarly establish other points. After constructing the parabola, draw a
normal at C and join C to C′ by a smooth curve such that at the entry point the phreatic line
is perpendicular.
After establishing the phreatic line, the rest of the flow lines and equipotential lines are
drawn. Then,
N
q = kH f
Nd
Sloping Discharge Faces. In Fig. 20.5 several conditions of discharging at the down-
stream portions of an earth dam are shown. In all cases, the intersection point of the bottom
flow line with the discharge face is considered as the focus F.
After deciding the focus point F, the parabola as described earlier is developed. Consider-
ing Fig. 20.4a, point D and the point on the directrix E are found. The parabola is drawn as
before. Let S be the point of intersection of the parabola with the downstream face. Now FS,
M20_PURU1773_01_SE_C20.indd 738 2/12/2013 8:19:49 AM

Embankment Dams 739
Phreatic line
Parabola
S Δa Directrix
R a
a = 60°
F O E
p/2 p/2
Phreatic line
(a) a = 60°
Parabola
S Directrix
Δa
R
a a = 90°
F O E
Phreatic line p/2 p/2
(b) a = 90°
Parabola
S Directrix
R Δa
a = 135°
a
F O E
p/2 p/2
(c) a = 135°
Fig. 20.5 Phreatic line details at exit from seepage zone (Source: Cernica, 1982)
M20_PURU1773_01_SE_C20.indd 739 2/12/2013 8:19:50 AM

04
03
a + Δa
02
Δa
01
0
30° 60° 90° 120° 150° 180°
Slope of discharge face a
Fig. 20.6 Slope of discharge face versus Δa/(a + Δa)
Phreatic line
0.3 BC
B D C
C Directrix
Parabola
S T
R J
Δa a K
A a FO E
Impervious
p/2 p/2
Fig. 20.7 Phreatic line for filterless toe
D
C
Z A B
Directrix
F O E
Focus
X
p/2 p/2
Fig. 20.8 The basic parabola
M20_PURU1773_01_SE_C20.indd 740 2/12/2013 8:19:50 AM

Embankment Dams 741
0.3 BC Basic parabola

p
B D E Directrix
C
C
X X′
h
J K Filter
m
a
A FOH
N
a = 180° p/2 p/2
Fig. 20.9 Construction of parabola for dam with horizontal drain
which is equal to a + Δa is measured. α is found and from Fig. 20.6 Δa /( a + Δa) is read.
Knowing a + Δa a and hence the break out point R are found. The transition section between
R and the parabola can be sketched by eye.
After drawing the phreatic line, the flow net can be completed and the discharge
calculated.
For the condition α > 30°, if the constructed flow net looks like confocal parabolae, the
discharge may be calculated theoretically. We know that
dz
q = kiA = k
A
dx
Considering unit thickness of the dam, A = z × 1 = z.
From Eq. 20.4, z2 = 2px, or z = 2 px , or A = 2 px .
Therefore,
⎛ p ⎞⎟
q = k ⎜⎜⎜ ⎟⎟( 2 px )= kp
⎜⎝ 2 px ⎟⎟⎠
From the properties of a parabola,

FD = DC = m + p
Also, (FD)2 = h2 + m2, where m = x − p/2 and h = z. That is, m + p = h 2 + m2 .

Solving for p we get p = h 2 + m2 − m . Therefore,
q = k h 2 + m2 − m (20.5)
Control of Quantity of Seepage. Excess seepage may be caused due to highly permeable
dam material, short seepage paths, and defects, such as fissures and cracks brought in by
uneven settlements. The seepage of reservoir water through the body of the dam, apart from
causing excessive water loss (IS: 9429, 1980), creates the following problems:
1. Finer particles migrate and clog soil structure, thus preventing seepage and causing
sloughing and weakening of soil strength.
M20_PURU1773_01_SE_C20.indd 741 2/12/2013 8:19:52 AM

2. Finer particles move through coarser particles or contacts and start internal erosion or
piping.
3. Seepage flow causes pore pressures and reduces the effective strength of the soil.
Similar effects are also experienced due to flow beneath a dam. Although the control of
seepage through and underneath embankments may be treated separately, it should be real-
ized that an effective treatment of seepage requires that we consider the embankment, its
foundation, and the abutting or adjoining structures as a unit (IS: 8414, 1977).
The discharge is reduced by using a low permeable material for the dam, providing a core
in the earth structure and cut-offs in the foundations, and by increasing the seepage path by
the inclusion of an impervious upstream blanket.
Keeping in mind the guidelines given in Table 20.1, a low permeability material has to
be chosen depending on the availability of material in the field. If a suitable material is not
available, then by proper selection and by mixing different soils available at the site, it is
often possible to obtain a composite soil with sufficiently low permeability.
An effective way of reducing leakage through an embankment is by providing a rela-
tively impervious core (Fig. 20.10a). The impervious core may be made of wood, steel,
concrete, masonry, or soil and should remain intact and impervious throughout the life of
the structure. Generally, clay cores are preferred. A core may be placed near the upstream
(called the sloping core) or at the centre. An upstream core reduces the pore pressure in the
downstream part of the embankment and increases its safety. As far as stability is concerned,
the upstream core is less stable, particularly during a sudden drawdown, and more volume
of soil is required. The central core requires minimum core material and is more stable dur-
ing a sudden drawdown.
As discussed earlier, the provision of a cut-off in the foundation not only reduces the
loss of water but also controls piping. A cut-off may be partial or complete, depend-
ing on whether its depth of penetration is partial or full (Fig. 20.10b and c). A properly
constructed complete cut-off can reduce the seepage to a negligible amount, whereas
a partial cut-off is less effective. All the materials used for cores may also be used for
cut-offs.
Another efficient way of controlling seepage and gradient is by providing an imper-
vious upstream blanket (Fig. 20.10d). By this provision, the length of the seepage path
is increased, thus reducing the gradient and seepage. The length of blanket required is
estimated using a flow net. Blankets are particularly useful when fissures and cracks
exist in the foundation. However, the blanket itself can crack due to settlement of the
foundation (Fig. 20.10e). Ordinarily, the blanket is effective in reducing the seepage only
by 50%. Blankets are of compacted impervious soil, as suggested by Indian Standards
(Table 20.1). Sometimes, various chemical additives are also employed to reduce the per-
meabilites.
Grouting is also an effective way of controlling excessive seepage flow. Foundation grout-
ing is resorted to in specially drilled holes for the purpose of sealing off or filling joints,
seams, fissures, or other openings encountered.
Control of Pressures and Gradients. Excess hydrostatic pressures cause boiling and piping,
especially at points where there is less weight of structures to resist them. The methods of
reducing excess pore pressures and gradients are (i) changing the direction of the seepage,
M20_PURU1773_01_SE_C20.indd 742 2/12/2013 8:19:52 AM

Embankment Dams 743
Core
Cut-off
Cut-off
(a) Core type dam (b) Complete cut-off (c) Partial cut-off
Blanket Blanket
(d) Upstream blanket (e) Cracking of blanket

due to settlement
Fig. 20.10 Measures of reducing seepage losses
(ii) installation of internal drains, (iii) incorporating relief wells, (iv) increasing the external
load, and (v) providing cut-off walls and an upstream blanket.
Among the above-mentioned methods, internal drainage is the most effective. In principle,
such a provision short circuits the seepage, reduces the excess pressures, changes the direction
of movement, and hence shifts the point of high gradients to a safer place inside the structure.
In Fig. 20.11, different measures of correcting excessive uplift pressures and gradients are given.
The trench drain (Fig. 20.11a) lowers the line of seepage in the homogeneous dam and
prevents pore pressures and loss of strength in the casing of a zoned dam. In a foundation
severely damaged by cracks and fissures, a continuous drain, as shown in Fig. 20.11b, is
provided. In a highly pervious downstream shell and in small dams, toe drains (Fig. 20.11c)
are provided. Apart from reducing the pressures at the downstream slope, it prevents
saturation of the soil at the toe of the dam due to rainfall.
Relief wells are an important adjunct to most of the preceding basic schemes for seepage
control and pressure relief. They are provided in earth dams where there are seams or pockets
of pervious water-bearing strata at great depth which cannot be intercepted by other means
(Fig. 20.11d). Usually, they are used in nearly all cases with upstream impervious blankets.
Besides, they are also used along with other schemes to provide additional assurance that
excess hydrostatic pressures do not develop. Continuous observation and maintenance of
relief wells is essential so as to ensure the satisfactory performance of the overall system as
regards seepage and pressure control.
M20_PURU1773_01_SE_C20.indd 743 2/12/2013 8:19:53 AM

Line of seepage
Trench Continuous
drain drain
Conduit
(a) Trench drain (b) Continuous drain
Line of seepage
Toe drain Relief

Stiff well
clay
Stiff clay
Dense sand
(c) Toe drain (d) Relief well
Fig. 20.11 Measures for reducing excess uplift pressures and gradients
Another method for correcting pressures is to increase the downward load at the
downstream side. Blocks of concrete have been used on the downstream top of a pervious
stratum. In earth dams, a large over-sized toe drain can serve the same purpose.
All the drainage systems require protective filters to prevent the movement or erosion of
the soil.
20.4.3 Protection Against Free Passage of Water Through Dams

The free passage of water from the upstream to the downstream will cause serious danger to
the stability of a dam. Such free passage may be caused due to the following:
1. water moving along the exterior surfaces of pipes or conduits which are embedded in the
body of the dam,
2. failure to form sufficiently strong bonds between successive layers of the dam,
3. failure to bind the lower layers of the dam properly to the foundation,
4. wrong placing of a pervious material in an otherwise impervious section in such a manner
as to make a blind drain from the upstream to the downstream,
5. water following the smooth surfaces of concrete abutments or other concrete structures, and
6. burrowing animals.
Of the above-mentioned causes, the first four can be effectively controlled by strictly
following the specifications during construction.
M20_PURU1773_01_SE_C20.indd 744 2/12/2013 8:19:54 AM

Embankment Dams 745
Water has a tendency to follow the smooth surface of abutments, pipes, conduits, etc. The
best practice is to place pipes and conduits in trenches excavated in the original foundation
material. Rocky abutments should be suitably shaped and prepared such that a perfect bond
is obtained at the contact between the impervious core and the rock. Overhangs should
be removed and the vertical surface excavated to form moderate slopes. Indian Standards
(IS: 8826, 1978) recommends a wider impervious zone and thicker transitions at the abutment
contacts to increase the length of the path of seepage.
Burrowing animals, such as musk rats and land squirrels, are responsible for piping fail-
ures in small dams. However, in modern dams, this danger is quite remote because (i) the
core material is so densely compacted that it may be difficult to burrow into it and (ii) the
crest width and the free-board are generally ample and animal holes do not penetrate to a
great depth.
Apart from the important design criteria discussed so far, there are some special design
requirements, such as control of cracking and stability in earthquake regions and at junc-
tions. A detailed discussion of these factors is beyond the scope of this book. The reader may
refer to some advanced book on earth dams and to IS: 8826 (1978).
20.4.4 Stability of Earth Dam Slopes

In the design of earth dams, both safety and economy call for thorough soil studies of the
foundation and of the materials of construction, combined with stability computations.
The present-day stability analyses are based on the results of studies of actual slides in
old dams. Stability analysis has to be performed for ascertaining the factor of safety for a
new structure or for the redesign of an old structure. An effective stress method of analysis is
recommended by Indian Standards (IS: 7894, 1975). Depending on the type of failure surface,
either the circular arc method or sliding wedge method is adopted.
The stability of an earth dam with respect to slope or foundation failure depends on the
magnitude and distribution of pore water pressure at all times. While designing an earth
dam, both the upstream and downstream slopes have to be checked for adequate stability.
For design purposes, the distribution of pore water pressure during construction, after the
reservoir has been filled, during steady state of seepage, and during sudden drawdown are
critical and have to be considered in the slope stability analysis.
The following are the conditions identified by Indian Standards (IS: 7894, 1975) as critical
for the stability of an earth dam:
Case I – Construction condition with or without partial pool (for upstream and down-
stream slopes)
Case II – Reservoir partial pool (for upstream slope)
Case III – Steady seepage (for downstream slope)
Case IV – Sudden drawdown (for downstream slope)
Case V – Steady seepage with sustained rainfall (for downstream slope)
Case VI – Earthquake condition (for upstream and downstream slopes).
The methods of analysis and the above-mentioned conditions are discussed below.
Methods of Analysis. The circular arc method adopted for the analysis is the method
of slices. This method was discussed in Chapter 13. The second method, the sliding wedge
method, is most suitable for earth dams and is discussed below.
M20_PURU1773_01_SE_C20.indd 745 2/12/2013 8:19:54 AM

The sliding wedge method divides the sliding mass into two or three wedges. In the
three-wedge system, the upper and lower wedges are called the driving or active wedge and
resisting or passive wedge, respectively, and the middle segment is referred to as the sliding
block. In a two-wedge system there is no middle segment. For earth dam this method is most
frequently applied under two circumstances, as shown in Fig. 20.12.
The factor of safety may be computed by adopting the methods of slices with side forces
(discussed in Chapter 13), the only difference being that the number of slices, instead of
being large, is only two or three.
Now, an alternative method, more suitable in practice, as mentioned in Fig. 20.12, is con-
sidered (Sherard et al., 1972). Let us assume that sufficient movement has taken place to
keep the active and passive wedges at failure. The active (Pa) and passive (Pp) forces acting
on the planes bc and de are computed assuming no shear forces act on the planes. Now, the
factor of safety is defined with reference to the stability of the central block. The unbalanced
force acting on the central wedge is
(20.6)
P1 = Pa − Pp
Active Central Passive

wedge block wedge
a c
W1
Weak layer Assumed
(soft clay or silt failure surface
or fine sand with Foundation
high pore pressure W2 W3
Pa e f
b Pp d
(a) Dam on weak foundation material
Driving Resisting
wedge wedge
Possible alternate shape of

failure surface for driving
Strong base wedge
(sand, gravel
Assumed
or rock)
failure surface
(b) Dam on strong foundation material
Fig. 20.12 Sliding wedge method
M20_PURU1773_01_SE_C20.indd 746 2/12/2013 8:19:55 AM

Embankment Dams 747
The force available to resist the movement of the central block is

′
P2 = C bd + (W2 − U bd ) tan φbd (20.7)
Therefore, the factor of safety

P2
F= (20.8)
P1
Construction Condition or End of Construction. This is the condition in which there is
development of pore pressure due to overlying fill. Due to non-availability of sufficient time,
the dissipation of pore pressure is partial and its estimation involves many uncertainties.
Both upstream and downstream slopes have to be analysed.
In principle, total and effective stress methods of analysis can be applied, provided a correct
estimation of undrained strength (with its associated pore pressure for the total stress analy-
sis) and pore pressure (only in the effective stress analysis) can be made. In the effective stress
analysis, a pore pressure value (in terms of pore pressure ratio, ru) may be determined in the
laboratory or assumed based on experience (Bishop and Bjerrum, 1960). Such an assumed
pore pressure in the design can be measured during construction and modified accordingly.
This is the advantage of the effective stress method over the total stress method (Bishop, 1957).
The pore water pressure at any point is
uw = (uw )0 + Δuw (20.9)
where (uw)0 is the initial pore water pressure and Δuw the pore water pressure in the undrained
condition. This may be represented in terms of the overall pore pressure coefficient B as
uw = (uw )0 + BΔσ1 (20.10)
Dividing by γh and letting ru = uw /γ h , we have
(uw )0 BΔσ1
ru = + (20.11)
γh γh
For the no surcharge condition, Δσ1 = γ h . Therefore,
(uw )0
ru = +B (20.12)
γh
For higher initial water content, (uw)0 = 0 and B is high; then,
ru ≈ B (20.13)
The value of B is obtained from a triaxial test corresponding to the stress condition in the
field. For the end of construction condition, a factor of safety of 1.3 may be allowed.
Reservoir Partial Pool Condition. This condition corresponds to the initial partial pool filling.
It has been presumed that a condition of steady seepage has developed at some intermediate
stages. For this condition, the upstream slope is critical. Minimum safety factors are computed
for three reservoir level conditions, viz., one-thirds, two-thirds, and full reservoir levels.
It has been suggested (IS: 7894, 1975) that all zones above the phreatic line are to be taken
as moist and those below are to be taken as submerged, while calculating the resisting and
driving forces. The partial condition is critical for high dams where the range of drawdown
is small compared to the height of the dam.
M20_PURU1773_01_SE_C20.indd 747 2/12/2013 8:19:56 AM

Steady-state Seepage Condition. Conditions of steady seepage are established after

a sufficient lapse of time from the time of filling of the reservoir. For steady seepage, the
stability of the downstream slope is always analysed by effective stress analysis and the pore
pressure is estimated from a flow net (Fig. 20.13a). Values of ru up to 0.45 are used in homo-
geneous dams. A least factor of safety of 1.5 is adopted for this condition.
Sudden Drawdown Condition. Upstream slopes of earth dams and natural slopes adja-
cent to the reservoir can undergo rapid drawdown if there is a sudden fall in the reservoir
level (Fig. 20.13b). Generally, high pore pressures result during drawdown, unless there is
an equal adjustment of pore pressure with the drawdown level.
An effective stress analysis can be adopted by calculating pore water pressure from a flow
net or from a reasonable estimation depending on the time availability for consolidation.
If the consolidation time is much less than the drawdown time, a flow net may be used to
compute the pore water pressure. As the flow net pattern changes with dissipation, it is suffi-
cient to construct the flow net immediately after the drawdown (Lambe and Whitman, 1979).
If the consolidation time is much longer than the drawdown time, the change in pore water
pressure due to the change in water load may be assumed to take place under undrained
conditions. The flow net for the steady seepage condition is considered (Fig. 20.13b). The pore
water pressure at a point P on the trial slip circle is given as (Bishop and Bjerrum, 1960).
(uw )0 = γ w ( h + hw − h ′) (20.14)
hw h
Equipo-
tentials
Trial slip
surface P
uw = gwhw
Equipotential passing
through P
(a) At steady seepage
Water level before drawdown
hw h′
Phreatic line
New water
h
level
Equipotential
P
(b) After rapid drawdown
Fig. 20.13 Pore pressures in steady seepage and drawdown conditions
M20_PURU1773_01_SE_C20.indd 748 2/12/2013 8:19:57 AM

Embankment Dams 749
Let σ1 be γh. Drawdown causes the piezometric level to fall to hw, and thus, the change in
total major principal stress is given by
Δσ1 = −γ w hw
and
Δuw = BΔσ1 = −Bγ w hw
uw = (uw )0 + Δuw
where uw is the pore water pressure at P immediately after drawdown; that is,
uw = γ w [h + hw (1 − B) − h ′]
Dividing by the overburden pressure, γsath, we get
uw γ ⎡ h h′ ⎤
= w ⎢1 + w (1 − B) − ⎥ (20.15)
γ sat γ sat ⎢⎣ h h ⎥⎦
When B = 1 and h ′ ≈ 0 , a conservative value for ru is obtained. A pore pressure ratio
of 0.3 to 0.4 is the typical value for a rapid drawdown condition. A factor of safety of 1.2 is
acceptable for this condition. Morgenstern (1963) presented stability coefficients for a rapid
drawdown condition.
Depending on the value of the coefficient of permeability of the shell material, Indian
Standards (IS: 7894, 1975) has recommended that the pore water pressure in the casing may
be allowed in the analysis in the following manner:
1. Full pore water pressures shall be considered if the coefficient of permeability is less than
10–4 cm/s.
2. No pore water pressures shall be considered if the coefficient of permeability is more than
10–2 cm/s.
3. A linear variation from full to zero pore water pressure shall be considered for the coef-
ficients of permeability lying between 10–4 cm/s to 10–2 cm/s.
For the core material, the recommendation is to allow full pore water pressures for the
core zone lying in the drawdown range.
For a zoned dam, the pore water pressure can be determined (IS: 7894, 1975) from the
formula (Fig. 20.14)
uw = γ w [hc + hr (1 − m)h] (20.16)
where uw is the drawdown pore water pressure at any point, hc the height of core material at
the point, hr the height of shell material at the point, m the volume of water draining out
from the shell per unit volume, and h the drop in the head under steady seepage condition
at the point.
As in the case of a homogeneous dam, the drawdown pore water pressures can be deter-
mined from the flow net.
Steady Seepage with Sustained Rainfall Condition. This condition is critical again for
the downstream slope. For this, a partial saturation of shell material due to rainfall is arbi-
trarily assumed. Accordingly, during analysis, the shell and other material lying above the
phreatic line shall be considered as moist for computing driving forces and buoyant for
resisting forces.
M20_PURU1773_01_SE_C20.indd 749 2/12/2013 8:19:58 AM

Centre of assumed
failure surface
Phreatic line
Water level before drawdown
New h
hr
water
level hc
Shell Core
Assumed failure surface Equipotential line
Fig. 20.14 Criterion for drawdown pore pressure in compressible core
The saturation for the downstream shell material shall be assumed, based on Indian
Standards recommendations (IS: 7894, 1975), as
1. 50%, if the coefficient of permeability is 10–4 cm/s or less,
2. 0%, if the coefficient of permeability is 10–2 cm/s or more, and
3. the percentage shall vary linearly from 50% to 0% for the coefficients of permeability lying
between 10–4 cm/s and 10–2 cm/s.
Earthquake Conditions. For this case, both the upstream and downstream slopes are
critical. The reader may refer to IS: 1893 (1975) for other details.
20.4.5 Protection of Crest, Upstream, and Downstream Faces

Some type of surfacing should be provided on the crest to protect it against damage by wave
splash and spray, rainfall, and wind and traffic wear and tear. The conventional method is to
place a 30 cm thick layer of selected fine rock or gravelly material with a suitable crown for
surface drainage. A parapet wall is provided on the upstream side and the downstream edge
is protected by a curb.
The upstream slope should be protected from destructive wave action. The different
types of surface protection of the upstream slope include stone rip-rap (either dry-dumped
or hand placed), concrete pavement, and steel facing. Sometimes sacked concrete or willow
mattresses are used for relatively small and unimportant dams. The rock for rip-rap should
consist of hard, dense, durable boulders or rock fragments from the quarries.
The downstream slope is protected by providing a heavy layer of coarse gravelly material,
or it may be surfaced with top soil and planted with native vines, shrubs, or grass. If sufficient
rock or cobble is available, it is preferable to provide a downstream rock or cobble fill. This,
in addition to its primary function of providing a stabilizing weight, also furnishes a protec-
tive covering for the underlying earth slope. On high dams, the effects of surface run-off
may be minimized by the use of berms or shoulders at intervals on the slope to collect and
dispose off the run-off water.
M20_PURU1773_01_SE_C20.indd 750 2/12/2013 8:19:58 AM

Embankment Dams 751
20.5 SELECTION OF DAM SECTION

For the selection of an earth or rockfill dam section, Indian Standards (IS: 8826, 1978) has
provided some recommendations which are discussed below.
20.5.1 Earth Dam

For all site conditions, no single type of cross section of an earth dam section is suitable. The
following factors have to be considered while deciding the earth section:
1. availability of construction materials, their quantity, and nearness to site;

2. condition of foundation and cut-off requirements;
3. types of construction machinery;
4. construction schedule and diversion considerations;
5. climatic conditions and their interference in placement water content;
6. safety requirements as regards to stability and seepage.
A homogeneous section is generally preferred in a location where adequate construction

material is available. If different types of soils are available in a location, a zoned dam is
preferred. In zoned earth dams, the weaker materials are often utilized for random zones.
Random zones are those that are generally provided below the minimum drawdown level
on the upstream side and on the downstream of the inclined filter. The planned section is
often altered or modified due to one reason or another.
In order to accommodate for post-construction vertical deformation resulting from
compression of the embankment and foundation settlement, an extra height of the dam is
provided. This extra height is provided as a longitudinal camber with zero at the abutments
and maximum at the centre of the gorge. Generally, a provision of increase in height of 1% to
2% of the embankment height is made to account for this.
20.5.2 Rockfill Dams

Rockfill dams primarily consist of an impervious membrane for water tightness and support-
ing rockfill. The merit of each rockfill dam section has to be analysed for the intended site.
The separation of the rockfill zone into several different zones and specification of a different
gradation for each zone is considerably expensive. It is generally satisfactory to use quarry
run rock in the entire rockfill zone. This ensures a reasonably uniform rock embankment with
free drainage characteristics.
WORKED EXAMPLES
Example 20.1 Compute the factor of safety with respect to effective stress for the slip sur-
face shown in Fig. 20.15. The shear strength parameters of the soil are c′ = 15 kPa, φ′ = 30°,
and γ =20 kN/m3. The groundwater level is also shown in the figure.
M20_PURU1773_01_SE_C20.indd 751 2/12/2013 8:19:59 AM

Equipotential G.W.L
line
8m h
hw′
a
l
1
2
17 m
3
4 11 m
5
6
7
Fig. 20.15
Solution
For this problem, the Fellenius method of slices may be adopted. The sliding mass is divided
into seven slices, as shown in Fig. 20.15. The width, height, length, and angle of inclination
of the base are determined and tabulated as below:
Slice b h l α W = bhγ N = hw U= N′ = N′ tan φ′ T = W c′l
no. (m) (m) (m) (°) (kN) W cos α (m) γwhwl N − U (kN) sin α (kN)
(kN) (kN) (kN) (kN)
1. 3.6 4.0 6.8 58 288.0 152.6 1.2 80.0 72.6 41.9 244.2 102.0
2. 4.4 7.2 5.8 41 633.6 478.2 4.0 227.5 250.7 144.7 415.7 87.0
3. 4.0 8.0 4.4 24 640.0 584.7 5.8 250.3 334.4 190.1 260.3 66.0
4. 4.0 7.4 4.0 13 592.0 576.8 5.8 227.5 349.3 201.7 183.2 60.00
5. 4.0 5.8 4.0 0 464.0 464.0 5.0 196.1 267.9 154.7 0 60.00
6. 4.0 3.4 4.2 −13 272.0 265.0 2.4 98.9 166.1 95.9 −61.2 63.00
7. 2.0 1.0 1.0 −22 60.0 55.6 0.5 34.4 20.9 12.1 −22.5 15.0
Σ 841.1 969.7 453.0
F=
∑ c ′l + ∑ N ′ tan φ ′
∑ W sin α
453 + 841.1
F= = 1.34
969.7
M20_PURU1773_01_SE_C20.indd 752 2/12/2013 8:19:59 AM

Embankment Dams 753
Example 20.2 The natural moisture content of a borrow pit soil is 12.5% with a unit weight
of 18.6 kN/m3 with a specific gravity of 2.68. Compute the compacted volume of an embank-
ment per 1.0 m3 of a borrow pit soil. The void ratio of the proposed embankment is 0.42.
Solution
Dry unit weight of ⎫⎪⎪ γ 18.6
⎬= =
borrow pit soil, γd ⎪⎪⎭ 1 + ω 1 + 0.1125
= 16.53 kN/m3.
Cγ w
Void ratio of borrow pit eb = −1
γd
2.68 ×10
= − 1 = 0.62.
16.53
Volume of soil solids is same both in borrow pit and embankment, i.e.,
vb v
vs = = e .
1 + e b 1 + ee
vb
∴ Volume of embankment = (1 + ee )
1 + eb
1
= (1 + 0.42)×
(1 + 0.62)
= 0.877 m3.
Hence for 1 cu m of borrow pit soil, 0.877 cu m of embankment can be constructed with a
void ratio of 0.42.
POINTS TO REMEMBER
20.1 Embankment dams are flexible structures primarily made out of earth and rock
fragments and used to impound water.
20.2 Embankment dams are classified as rolled fill or placed earthfill dams and hydrau-
lic fill dams. Rolled fill dams are constructed by compacting soils in successive
layers with the available materials in and around the construction site. Hydraulic
fill dams are constructed by pumping the soil water suspension to the required site
and allowing it to settle.
20.3 Embankment dams are of three types, viz., diaphragm type, homogeneous type, and
zoned type. In the diaphragm type, the embankment is constructed of pervious mate-
rials with a water barrier on the upstream side. Homogeneous dams are constructed
from a simple kind of material. Zoned dams are more common types of dams with
pervious materials as shells and impervious materials as cores.
20.4 The components of an earth dam are the foundation, casing, core, top width, free-
board, cut-off, internal drain, transition filter, slope protection, and relief wells. All the
components are not provided in a single dam.
M20_PURU1773_01_SE_C20.indd 753 2/12/2013 8:20:01 AM

20.5 The salient features in the design of earth dams are (i) there should be no danger
to the dam by over-topping, (ii) seepage through the body of the dam and founda-
tion should be controlled, (iii) upstream and downstream slopes should be protected
against sliding, and (iv) crest, upstream, and downstream faces should be protected.
20.6 In earth dams, the phreatic surface constitutes the top flow line which has to be located
based on the boundary condition.
20.7 It has been mathematically shown that the basic shape of the phreatic surface is that
of a parabola, and it deviates only at the upstream and downstream faces depending
on the cross section of the dam.
20.8 Excess seepage may be caused due to highly permeable dam material, short seepage
paths, and defects such as fissures and cracks caused by uneven settlement.
20.9 Excess seepage through the body or foundation of a dam is controlled by providing (i) an
impervious core, (ii) partial or free cut-off in the foundation, or (iii) upstream blanket.
20.10 Excess uplift pressures and gradients are reduced by providing (i) trench drains,
(ii) continuous downstream drains, (iii) toe drains, or (iv) relief wells.
20.11 Stability of the slopes of an earth dam are critical under the following conditions:
(i) construction condition with or without partial pool (for upstream and down-
stream slopes), (ii) reservoir partial pool (for upstream slopes), (iii) steady seepage
(for downstream slopes), (iv) sudden drawdown (for upstream slopes), (v) steady
seepage with sustained rainfall (for downstream slopes), and (vi) earthquake condi-
tions (for upstream and downstream slopes).
20.12 The circular arc method or sliding wedge method is adopted for stability analysis of
slopes.
20.13 Crest, upstream, and downstream slopes should be protected from destructive wave
action. The crest is protected by surfacing. The upstream slope is protected by rip-rap,
concrete pavement, or steel facing. The downstream slope is protected by a heavy
layer of coarse gravelly material or surfaced with top soil and planting shrubs or grass.
20.14 The factors to be considered while selecting an earth or rockfill dam are (i) availability
of construction materials, their quantity, and proximity to the site; (ii) condition of
the foundation and cut-off requirements; (iii) types of construction machinery;
(iv) construction schedule and diversion considerations; (v) climatic conditions and
their interference in placement water content; and (vi) safety requirements as regards
stability and seepage.
QUESTIONS
Objective Questions
1. In an earth dam, the phreatic surface constitutes the bottom flow line.
2. The upstream slope of an earth dam during steady seepage conditions is an equi-
potential line.
3. The shell imparts stability and protects the core.
M20_PURU1773_01_SE_C20.indd 754 2/12/2013 8:20:01 AM

Embankment Dams 755
4. Cut-off is a barrier to reduce seepage of water through the core and casing.
5. The material used for an impervious blanket should have far less permeability than
the foundation soil.
20.2 Select the type of soil most suitable for the core of a zoned dam from the following:
(a) SC (b) GC (c) CH (d) CI
20.3 A transition filter is provided between a coarse and fine material to
(a) Give sufficient support for the core and shell.
(b) Prevent the seepage of water completely into the core.
(c) Prevent the migration of fine material to coarse material.
(d) Reduce the excess hydrostatic pressure.
20.4 In a homogeneous dam sudden drawdown of a reservoir level causes instability to the
______
(a) Downstream slope
(b) Upstream slope
(c) Both upstream and downstream slopes
(d) None of the slopes
20.5 The extra height provided in the crest of the dam is to allow for
(a) Compression of fill material
(b) Settlement of foundation
(c) Compression of fill material and settlement of foundation
(d) Extra safety against wave action
20.6 In an earth dam the critical condition(s) for which the stability has to be checked
during construction with or without partial pool is/are
(a) Downstream slope
(b) Upstream slope
(c) Upstream and downstream slopes
(d) None
20.7 In the stability analysis of an upstream slope for the sudden drawdown condition,
generally, no pore pressure is considered in the shell when the coefficient of perme-
ability is greater than
(a) 10–3 cm/s (b) 10–4 cm/s (c) 10–1 cm/s (d) 10–2 cm/s
20.8 The sliding wedge method of analysis is generally applicable in the circumstances
where
(a) One or more horizontal layers of weak soil exist in the upper part of the foundation.
(b) More than one type of material is used in the body of the earth dam.
(c) The dam material is a homogeneous one and resting on a pervious foundation.
(d) The dam material is a homogeneous one and resting on a very stiff soil.
20.9 Explain the design considerations for upstream and downstream slopes of an earth dam.
20.10 What is a rockfill dam? How is it different from an earth dam?
M20_PURU1773_01_SE_C20.indd 755 2/12/2013 8:20:01 AM

20.11 Explain why a thin core dam may be economical even when ample supplies of both
pervious and impervious soils are available.
20.12 Under what conditions would it be advisable to place a blanket layer of coarse-grained
material between an embankment and its foundation?
20.13 Define seepage force, piping, and roofing as understood in the study of stability of an
earth dam resting on a relatively porous medium.
20.14 What is meant by construction pore pressure in earth dams? What are the methods
adopted to reduce the same?
20.15 An earth dam foundation consists of clay having very low shear strength. It is pro-
posed to construct a dam of 10 m height. Suggest a method of foundation treatment
and design considerations so as to avoid a base failure.
20.16 What is the significance of a filter in an earth dam? Discuss the design criteria of filters.
20.17 State some general principles of soil selection in the design and construction of
embankments.
20.18 Bring out the role of the nature of embankment soil on the cracking phenomena of
embankments.
20.19 Write a detailed note on the construction of rockfill dams.
20.20 List the field tests you would conduct on a composite rolled fill construction. Explain
the importance of each test.
EXERCISE PROBLEMS
20.1 Construct a flow net and estimate the seepage of the homogeneous earth dam sketched
in Fig. 20.16. The average coefficient of permeability is 4 × 10–7 m/s.
15 m
5m
2:1 2:1
40 m
Impervious
45 m
Fig. 20.16
20.2 Make a flow net for the earth dam given in Fig. 20.17 and estimate the seepage loss.
20.3 Figure 20.18 shows a section through an earth–rock dam. The clay is isotropic with a
permeability of 10–8 m/s. Assuming the rockfill permeability to be infinite, determine
the quantity of seepage through the core per metre width of the dam.
M20_PURU1773_01_SE_C20.indd 756 2/12/2013 8:20:01 AM

Embankment Dams 757
8m
4m
2.5:1
2.5:1 Toe drain
20 m
90°
20 m
Fig. 20.17
Clay core
Rockfill 6m
2.5 2.5
1 Rockfill
1 20 m
17 m 1 1
1.5 1.5
Impervious base
Fig. 20.18
20.4 Calculate the minimum length of under drain needed for the cross section of an earth
dam shown in Fig. 20.19, such that the top seepage line shall not be nearer than 7.2 m
from the downstream surface of the dam.
Top width
7.2 m
48.8 m 52.4 m
45° 30° 20.0 m

Under drain
L=?
Fig. 20.19
20.5 A reservoir embankment 30 m in height consists of compacted earth with c = 35 kN/m2,

φ = 15°, and γ = 19.6 kN/m3. The embankment has a slope of 20°. Compute the factor
of safety if a sudden drawdown of the reservoir occurs. Use Taylor’s stability charts.
20.6 A 50 m high homogeneous earth dam with an upstream slope of 2.5 to 1 is founded
on a rock base. The estimated pore pressure coefficient ru = 0.30 at the end of con-
struction. The soil parameters of the earth dam material are c′ = 22 kN/m2, φ′ = 27°,
M20_PURU1773_01_SE_C20.indd 757 2/12/2013 8:20:02 AM

and γ = 21.8 kN/m3. Choose a trial circular slip surface passing through the toe and
determine the factor of safety using Fellenius method of slices.
20.7 The downstream slope of an earth dam is shown in Fig. 20.20. Find the factor of safety
against sliding along the slip surface shown. The approximate pore pressure distribu-
tion is also shown.
19 m
m
10 m 37
r= 12.5 m
2 C 20 m
1
g = 19 kN /m3
c¢ = 15 kPa
f¢ = 28° b
uw = 15 kPa Slip surface
Assumed pore water

pressure distribution
(Centre C)
Fig. 20.20
20.8 The upstream slope of an earth dam under steady seepage conditions is shown in
Fig. 20.21. The relevant parameters of the dam are e = 0.60, G = 2.68, c′ = 17 kN/m2,
and φ′ = 26°. Find the factor of safety against sliding along the slip surface using the
ordinary method of slices.
Centre line
m
78
r=
12 m
6m
2:1
34 m
2 :1
Nf = 3
Nd = 12
Fig. 20.21
M20_PURU1773_01_SE_C20.indd 758 2/12/2013 8:20:03 AM

21
Dynamic Loading of Soil
CHAPTER HIGHLIGHTS
Earthquake: seismic waves, magnitude – Other dynamic loads – Theory
of vibration: free and forced vibrations with and without damping – Types
of machines and machine foundations – Dynamic bearing capacity of
shallow foundations – Design requirements – Methods of analysis for block
foundation – Liquefaction of soils
21.1 INTRODUCTION
Elastic waves moving through a soil mass produce ground motion, which is transmit-
ted through foundations to a structure as vibrations. The vibrations are caused by earth-
quakes or from construction activity such as rock blasting, pile driving, etc. On the
other hand, vibrations also result from operating machinery and are transmitted
through the foundations to the soil. Vibrations developed by operating machinery pro-
duce several effects which must be considered in the design of foundations. As large
machines are usually supported on the soil and the impulses are directly transmitted to
the soil, the design of machine foundations involves the problem of soil dynamics. A
detailed review of the effects of dynamic loading on soil properties is given by Prakash
(1981). Liquefaction of soils is an important phenomenon that occurs due to large vibrat-
ing forces.
21.2 EARTHQUAKES
When friction between rocks on either side of a fault is adequate to prevent the rocks
from slipping easily or when the stressed rock is not already fractured, some elastic
deformation occurs before failure. When the stress exceeds the rupture strength of the
rock (or the friction between rocks along an existing fault), sudden movement occurs
M21_PURU1773_01_SE_C21.indd 759 2/12/2013 8:19:22 AM

along the fault, resulting in an earthquake. After the rupture, the rocks snap back to the previ-
ous dimensions due to elasticity, which is referred to as elastic rebound. The stress release and
the movement are reflected as the relative displacement of the rocks on either side of the
fault following the earthquake. This released energy is propagated in the form of seismic
waves, which pass on the energy through the earth media. During this process, all the struc-
tures erected on the earth’s surface are subjected to vibrations.
The point inside the earth’s surface at which the first movement or break occurs during
an earthquake is called the focus or biocentre of the earthquake. The point on the earth’s sur-
face directly above the focus is the epicentre. The position of the focus is determined from a
seismograph record.
21.2.1 Seismic Waves

The major seismic waves are body waves and surface waves. Body waves consist of P- and
S-waves. P-waves are compressional waves travelling through the earth, similar to sound
waves, which travel through the air. S-waves are shear waves involving a side-to-side slid-
ing motion of the material. Surface waves travel along the surface, which is somewhat anal-
ogous to the movement of surface waves on water. Surface waves cause rocks and soils to be
displaced in such a way that the ground surface undulates. Vertical ground motions are
caused by Rayleigh waves, while Love waves cause horizontal motions. Most of the dam-
ages due to earthquake are caused by surface waves.
21.2.2 Magnitude
Magnitude of an earthquake is a measure of the amount of ground shaking (i.e., the amount
of vertical motion) based on the amplitude of elastic wave it generates. Richter’s magnitude
scale, named after Prof. Charles Richter, a geologist, is most often used. The Richter scale
starts from 2, and there is no upper limit. Table 21.1 gives the description of an earthquake
in relation to its magnitude on the Richter scale.
The Richter scale is a logarithmic one; that is, an earthquake of magnitude 4 causes 10
times as much ground movement as one of magnitude 3, one hundred times as much as one
of magnitude 2, and so on.
Table 21.1 Magnitude of an earthquake
Magnitude Description
>8.0 Great earthquake

7.0–7.9 Major earthquake
6.0–6.9 Destructive earthquake
5.0–5.9 Damaging earthquake
4.0–4.9 Minor earthquake
3.0–3.9 Small earthquake, usually felt
2.0–2.9 Detected but not felt
Source: Montgomery (1990).
M21_PURU1773_01_SE_C21.indd 760 2/12/2013 8:19:22 AM

Dynamic Loading of Soil 761
21.3 OTHER DYNAMIC LOADS

Apart from earthquakes, there are other sources of dynamic loads. Soils may also be sub-
jected to vibrations due to bomb blasts, driving of piles in the vicinity, wind and water action
on structures, landing of aircraft, etc. Some of the above sources are natural, and some are
man-made.
The other major man-made sources of dynamic loads are machines, machine tools,
engines, generators, turbines, forging and foundry machinery, vehicle and conveyers, min-
ing, and transmission equipment, which cause different vibration problems. These dynamic
forces and moments should be contained by proper means, otherwise it may lead to undesir-
able noise, discomfort, instability of structures, etc.
21.4 THEORY OF VIBRATIONS

The behaviour of a structure subjected to dynamic load is better understood by studying the
mechanisms of vibrations caused by the dynamic load. The pattern of variation of a dynamic
load with respect to time may be either transient or periodic. Transient vibrations may have
non-periodic time history, e.g., vibrations caused by earthquakes, bomb blasts, quarry blasts,
etc. The periodical motions can be resolved into sinusoidally varying components, e.g.,
vibrations due to reciprocating machine foundation.
A structure subjected to a dynamic load vibrates in one of the following four ways of
deformation or a combination of one or more:
1. Extensional (Fig. 21.1a)
2. Shearing (Fig. 21.1b)
(a) Extensional (b) Shearing (c) Bending (d) Torsional
Fig. 21.1 Different types of vibrations (Source: Swami Saran, 1999)
M21_PURU1773_01_SE_C21.indd 761 2/12/2013 8:19:22 AM

3. Bending (Fig. 21.1c)

4. Torsional (Fig. 21.1d)
21.4.1 Harmonic Vibrations

In order to understand the reasons for the additional precautions needed for oscillating
machinery, it is essential to have a clear concept of the basic theory of harmonic vibrations.
Many of the foundation problems arising from oscillating machinery can be analysed by
considering a simple harmonic application of force. Some of the terms relevant to machine
foundation vibrations are discussed below.
The motion of a point in a straight line is said to be in simple harmonic motion such that the
acceleration of the point is proportional to the distance of the point from some fixed origin
and is always directed towards the origin.
A foundation system is said to be in free vibration when it is disturbed and then left free
to vibrate about some mean position. Thus, if an elastic system vibrates under the action
of inherent forces in the system and in the absence of any externally applied force, the fre-
quency of vibration is termed natural frequency. Vibrations that result from exciting agencies
are referred to as forced vibrations.
The foundation system is said to be at resonance when the frequency of the exciting agency
(operating frequency) is close to the natural frequency. Under the condition of resonance,
large forces and amplitudes of motion (the distance that a body moves from its position of
rest when subjected to vibration) can be generated.
Damping is associated with energy dissipation and is the internal resistance offered by a
foundation system to the vibration of a machine.
Degrees of freedom of a system is defined as the number of coordinates required to describe
the displaced position of the system.
21.4.2 Free Vibration of a Spring–mass System

In a single-degree-of-freedom system, there is a single-point mass and only one position
coordinate is needed to define the state of motion. A mass suspended by a spring and set into
a vertical motion is said to be in simple harmonic motion with one degree of freedom.
Consider a spring–mass system with a spring of stiffness k (Fig. 21.2a). After the addition
of a mass m of weight W, the mass attains a position called equilibrium position, as shown in
Fig. 21.2b. Then, the deflection
W
δstat = (21.1)
k
where k is the spring constant.
If the mass in this position is pulled down by z (Fig. 21.2c) and further to a maximum posi-
tion zmax = A (Fig. 21.2d) and released, the peak-to-peak displacement is referred to as double
amplitude (Fig. 21.2e). Forces acting on undamped vibrating mass are shown in Fig. 21.2f.
On release of the external force, the mass starts to oscillate between two extreme points
and continues to oscillate as there is no resistance to these oscillations.
In the case of a weightless spring, the equation of motion can be written as
mz + k z = 0 (21.2)
M21_PURU1773_01_SE_C21.indd 762 2/12/2013 8:19:22 AM

k
+
Double
zmax amplitude
0=z m
z
m dstat = A k1 + k d stat
Equilibrium m
z m m
position
m mg (= W )
(a) (b) (c) (d) (e) (f)
Fig. 21.2 Spring–mass system: (a) unstretched spring; (b) equilibrium position; (c) mass in oscillat-
ing position; (d) mass in maximum downward position; (e) mass in upward position;
(f) free-body diagram of mass corresponding to (c)
A solution to this equation is obtained (for the solution, refer Converse, 1962; Prakash,
1981), and after satisfying Eq. 21.1, the expression for ωn, the circular natural frequency of
the system, is given as
k
ωn = (21.3)
m
One cycle of motion is completed when ωnT = 2π, where T is the period. Then,
2π m
T= = 2π (21.4)
ωn k
The natural frequency, fn, the number of cycles executed in unit time, is given as
1 1 m
fn = = (21.5)
T 2π k
The period Tn is the free period of the spring–mass system and depends on both the
spring constant and the mass. With reference to foundations on soil, the foundation and
attached masses are taken as m and the earth as the spring. In the above treatment, the mass
of the spring (earth) is neglected, and hence this simple theory represents approximately the
behaviour of foundation on soil.
21.4.3 Free Vibration with Viscous Damping

The governing equations for the free-vibration condition are based on the assumption that
there was no loss of energy in the oscillating system due to internal friction or external forces
opposing the motion. In practical problems, there is loss of energy and decrease in ampli-
tude due to damping. In one type of damping, the force Fd is taken directly proportional to
the velocity of the oscillating mass, which is defined as viscous damping. That is,
Fd = − cz (21.6)
where c is the coefficient of viscous damping.
M21_PURU1773_01_SE_C21.indd 763 2/12/2013 8:19:23 AM

Sign convention
z, z, z
c
+
k
kz + cz
Equilibrium
position z
m
m
Fig. 21.3 Spring–mass–dashpot system
Consider a spring–mass–dashpot system (Fig. 21.3). The free-body diagram of a displaced

mass under damped vibration condition is also shown in Fig. 21.3. The equation of motion
may be written as
mz + cz + k z = 0 (21.7)
A solution to this equation (for the solution, refer Converse, 1962; Prakash, 1981) is based
on two conditions. For one condition, the value
⎛ c ⎞⎟2
⎜⎜ ⎟ = k = ωn2 (21.8a)
⎜⎝ 2m ⎟⎠ m
This condition is referred to as critical damping condition and Eq. 21.8a is rewritten with
c = cc as
cc = 2 m ωn (21.8b)
The ratio of actual damping (c) to critical damping (cc) is referred to as the damping factor,
ζ. That is,
c
ζ= (21.8c)
cc
For the second condition, k / m < (c / 2m)2 , the solution yields an expression for ωnd, the
circular natural frequency for the damped free vibration, as
ωnd = ωn 1 − ζ 2 (21.8d)
Then, the period is given by

2π
T= (21.9)
ωn 1 − ζ 2
Figure 21.4 represents a typical damped oscillation for ζ < 1.0.
21.4.4 Forced Vibrations with Viscous Damping

Consider a spring–mass–dashpot system subjected to a force F (Fig. 21.5a), where
F = F0 sin ωt (21.10)
M21_PURU1773_01_SE_C21.indd 764 2/12/2013 8:19:25 AM

2
1
7 8 9 10
0 5 6
1 2 3 4
Fig. 21.4 Typical damped oscillation
Sign convention
z, z, z
c
+
k
kz + cz
Equilibrium
position z
m m
F0 sin wt F0 sin wt
(a) Spring–mass–dash (b) Free-body
pot diagram
Fig. 21.5 Forced vibrations with viscous damping
where ω is the frequency of the force of excitation. The free-body diagram is shown in
Fig. 21.5b. The equation of motion is
mz + cz + k z = F0 sin ωt (21.11)
A solution to this equation (for the solution, refer Converse, 1962; Prakash, 1981) gives the
following expressions:
F0 / k
z0 = (21.12)
(1 − r 2 ) + (2 ζ r )2
⎡ 2ζ r ⎤
θ = tan−1 ⎢ ⎥ (21.13)
⎢⎣ 1 − r 2 ⎥⎦
where z0 is the maximum displacement of the forced vibration, θ the phase angle between
the applied force and the displacement, and r = ω / ωn, the frequency ratio.
The factor F0/k represents the static deflection of the system (δstat) under a static load
equal to the dynamic force F0. Thus,
z0 1
= (21.14)
δstat (1 − r ) + (2 ζ r )2
2 2
M21_PURU1773_01_SE_C21.indd 765 2/12/2013 8:19:26 AM

10 x
0.0
0.06
8
Magnification, N
0.08
6
0.10
4
0.15
0.20
2 Actual damping
x=
Critical damping
1.00
0
0 1.0 2.0 3.0
w
Frequency ratio, w
n
Fig. 21.6 Frequency ratio versus magnification ratio (Source: Leonards, 1962)
Equation 21.14 is called the magnification factor (N). This factor signifies the increase in
the maximum deformation of a forced vibration from that caused by a static load equal in
magnitude to that of the dynamic force.
For the undamped vibration condition, c = 0, and N becomes
1
(21.15)
(1 − r 2 )2
For low values of ω, the increase in amplitude is less, as N is close to unity. For ω = ωn , N is
infinite and the amplitude is infinite if sufficient damping is not available in the system. For
ω > ωn, the amplification gradually decreases. By providing adequate damping, the ampli-
tude decreases depending on the magnitude of the damping ratio.
Figure 21.6 shows the magnification factor N versus frequency ratio r. To avoid large
amplitudes, r should be kept less than 0.5 or greater than 1.5.
21.5 TYPES OF MACHINES AND MACHINE FOUNDATIONS

Machines generate different periodic forces. They can be broadly categorized under three
heads (Prakash, 1981):
1. Reciprocating machines: These machines produce unbalanced force and work with low
operation frequency∗ (speed) in the order of 600 revolutions (rev)/min. While analysing
∗
The rotating speed of the main drive in rpm or the frequency of the periodic force acting on the system.
M21_PURU1773_01_SE_C21.indd 766 2/12/2013 8:19:27 AM

Machine Machine
Motor Motor
Pedestal
(a) Block-type (b) Box- or caisson-type

foundation foundation
Beam and slab
Wall and columns
Footing
(c) Complex-type foundation
Fig. 21.7 Typical machine foundations (Source: Prakash, 1981)
the foundation, the unbalanced force is considered to be sinusoidal. Compressors and

reciprocating engines fall under this category.
2. Impact machines: These machines produce impact loads and operate with a speed range
of 60 to 150 blows per minute. In these machines, dynamic load reaches a maximum in a
very short time and attenuates immediately. Machines like forging hammers, stamping
press, etc., are included in this category.
3. Rotary machines: These machines operate at frequencies ranging from very low to high.
They are classified (IS: 2974 – Part 3, 1975) as low (frequency up to 1,500 rpm), medium (fre-
quency 1,500 to 3,000 rpm), and high (frequency greater than 3,000 rpm). Machines which are
included in this category are crushing mills, pumps, motor generators , turbo generators, etc.
Machine foundations can be broadly classified into three types: (i) block foundations,
(ii) box or caisson foundations, and (iii) complex foundations (Fig. 21.7). Block foundation is the
simplest type, consisting of a pedestal resting on a footing. Block foundations have a large mass
and, hence, less natural frequency (defined in the next section). This type of foundation is suit-
able for compressors and reciprocating foundations. If more machines of a similar type are to be
arranged in a machine shop, it can be done economically by providing a single, continuous mat.
In the box or caisson type of foundation, the mass is reduced, and by virtue of this, the
natural frequency is increased. Such foundations can be used for medium-speed machines
like forging hammers.
High-speed machines like steam turbines require a system of wall columns and beam
slabs. Here, the elements of the systems are relatively flexible compared to block or box-type
foundations.
21.6 DYNAMIC BEARING CAPACITY OF SHALLOW FOUNDATIONS

Dynamic loads may act in different directions. The dynamic loading in the vertical direction
only is considered here. Such loadings may induce large, permanent deformations in foun-
dations. The most important factors that need consideration while analysing a foundation
M21_PURU1773_01_SE_C21.indd 767 2/12/2013 8:19:27 AM

under dynamic loading are magnitude and nature of the loading pulse, duration of the
pulse, and the strain rate response of the soil during deformation (Das, 1983). Only very
limited data are available on dynamic bearing capacity.
Vesic et al. (1965) conducted laboratory model tests on dry and saturated dense sands
under varied strain rates of 25 × 10−5 to 250 mm/s. This corresponds to a varying loading
velocity of 14.4 to 19.75 mm/s. Based on the results of Vesic et al. (1965) and other similar
works, Das (1983) has shown a gradual decrease of (qf / 12 γ B) with loading velocity up to
a minimum value, followed by an increase. This has been attributed to a reduction of the
friction angle by about 2°. Thus, it has been recommended by Vesic (1973) that for dynamic
conditions the φ value can be reduced by 2° and the reduced value can be used for bearing
capacity under dynamic conditions in sand. Heller (1964) has suggested that foundations on
sand are subjected to an acceleration level of amax ≤ 13g under general shear and amax > 13g
under punching shear. Such a prediction of bearing capacity in sands may be unreliable in
loose, submerged sands as they are susceptible to liquefaction under dynamic loads.
For footings on saturated clays, the conventional bearing capacity equation can be used
for the φ = 0 condition, but an appropriate cohesion value has to be introduced. Generally,
undrained cohesion increases with strain rate. Carroll (1963) suggested that undrained cohe-
sion under dynamic condition can be taken as 1.5 times the undrained cohesion under static
condition. This modified cohesion can be used in the bearing capacity equations to obtain
the dynamic bearing capacity in clays.
21.7 DESIGN REQUIREMENTS

The conventional considerations of safe bearing capacity against shear failure and allowable
settlement are insufficient to ensure a satisfactory machine foundation design. In the earlier
designs, the static load was increased by multiplying by a dynamic factor, and conventional static
foundation analysis was done for the increased static load without any knowledge of the safety
factor. Because of this uncertainty about the magnitude of the dynamic factor, erroneous behav-
iour of foundations was observed in practice. The other approach is to increase the mass of the
foundation so as to absorb the vibrations. Cozens (1938) suggested desirable ratios of foundation
mass to mass of the engine (Table 21.2). Evidently, this approach does not consider the type of soil
and its environment, the frequency level or the amplitude of the machine. However, the data
available in Table 21.2 can be used to obtain an initial rough estimate of the size of the foundation.
Thus, the design of a machine foundation has to satisfy the following additional criteria
(Prakash, 1981):
1. Resonance should be avoided; that is, the natural frequency of the foundation–soil sys-
tem should not be equal to the operating frequency. This can be achieved if the natural
frequency of the system lies outside the zone of resonance.
2. The amplitude of vibration at the operating frequency should be less than the limiting
amplitude. The limiting amplitudes are usually specified by the manufacturers or in
Codes of Standards. Frequency and amplitude requirements for machine foundations are
given in Table 21.3, based on data from the Indian Standards (IS: 2974, 1979).
3. The vibration of the machine should neither cause damage to machines and structures nor
harm the health and comfort of the people. The degree of severity of vibration depends on
the operating frequency and amplitude of motion.
M21_PURU1773_01_SE_C21.indd 768 2/12/2013 8:19:27 AM

Table 21.2 Foundation mass required to absorb vibrations
Type of engine Ratio of foundation mass to

mass engine
Steam engine (single cylinder) Not less than 4.0:1

Steam engine (compound) 3.75:1
Steam engine (triple expansion) 2.50:1
Gas engine (single cylinder) 3.00:1
Gas engine (two cylinders) 3.00:1
Gas engine (four cylinders) 2.75:1
Gas engine (six cylinders) 2.25:1
Gas engine (eight cylinders) 2.00:1
Diesel engine (two cylinders) 2.75:1
Diesel engine (four cylinders) 2.40:1
Diesel engine (six cylinders) 2.10:1
Diesel engine (eight cylinders) 1.90:1
Vertical steam engine (compound) coupled to generator 3.80:1
Vertical steam engine (triple expansion) coupled to generator 3.50:1
Horizontal steam engine (cross-compound) coupled to generator 3.25:1
Horizontal steam turbine coupled to generator 3:1 to 4:1 (if of small output)
Vertical gas engine coupled to generator 3.50:1
Vertical diesel engine coupled to generator 2.60:1
Source: Couzens (1938).
Table 21.3 Foundation frequency and amplitude requirements for various machines
Reciprocating machines
fop should be greater than 2fn or less than 0.5fn in important machines; otherwise, fop should be greater
than 1.5fn or less than 0.6fn
Drop and forge hammers
fn should be greater than 2.5fi or less than 0.7fi
Peak vertical amplitude of the foundation should not exceed 1.2 mm. Peak vertical amplitude of the
foundation block should not exceed 0.8 mm if the foundation is on sand below the water table
With important structures nearby, maximum peak velocity of the foundation is 3 mm/s
Rotary-type machines
For high-speed machines, fn should be no closer than 20% to fop. Permissible amplitudes at beating level:
1. For fop < 3,000 rpm
Vertical, 0.04–0.06 mm
Horizontal, 0.07–0.09 mm
2. For fop > 3,000 rpm
Vertical, 0.02–0.03 mm
Horizontal, 0.04–0.05 mm
Foundation vibrations should not exceed 50% of the above figures. For low-frequency machines ( fop <
1,500 rpm), fn must be considerably greater than fop, and foundation peak amplitude must not exceed 0.3 mm
Note: fop is the operating frequency of the machine, fi is the frequency of impact, and fn is the natural
frequency of the machine–foundation system.
Source: IS: 2974 (1979); compiled by Moore (1985).
M21_PURU1773_01_SE_C21.indd 769 2/12/2013 8:19:27 AM

250
A B C D
200
150
100
75
50
25
Amplitude of vibrations, μm (±)
20
15
10
7.5
D′
5.0
C′
2.50
2.00 B′
1.50
1.00
0.50
0.25
3 5 10 20 30 50 100 160
Disturbing frequency, Hz
Line ADD′ limit to avoid damage to buildings
Line ACC′ limit to avoid service discomfort to persons
Line ABB′ limit to ensure reasonable comfort to persons
Note: These limits do not include any factor of safety
Fig. 21.8 Amplitude limits of a foundation block (Source: IS: 2974 – Part 1, 1982)
A chart has been provided by the Indian Standards (IS: 2974 – Part 1, 1982), which gives
various limits of frequency and amplitude for different conditions (Fig. 21.8).
The main steps to be followed in the design of a satisfactory machine foundation are as
follows:
1. The design criteria to be adopted are finalized (as discussed above).
2. The dynamic loads which are to be transmitted to the foundation are computed.
3. The appropriate soil parameters under static and dynamic conditions are estimated from
field and laboratory tests.
4. An appropriate method of analysis is chosen to compute the natural frequency and the
vibration amplitude at operating frequency for each of the relevant modes of vibration.
This is accomplished for an assumed foundation size, which is modified if necessary to
satisfy the design criteria.
Moore (1985) suggests the following factors to be considered while proportioning the
foundation and finalizing the details:
M21_PURU1773_01_SE_C21.indd 770 2/12/2013 8:19:27 AM

1. The centre of gravity of the machine and foundation should be as close as possible to the
centroid of the foundation base area.
2. For low-speed machines ( fop < 5 Hz), the foundation has to be designed such that the
natural frequency is much higher than the operating frequency. In case of high-speed
machines (fop > 8 Hz), the natural frequency should be maintained at a lower level.
3. In order to reduce or isolate transmission and vibrations from the machine foundation to
other foundations and parts of the building, joints or vibration isolators are provided.
4. The foundation has to be designed such that it can be modified at a later date if some
unforeseen vibration problems develop.
21.8 METHODS OF ANALYSIS FOR BLOCK FOUNDATION

In general, a machine foundation may undergo six independent displacement rotations,
which are referred to as six degrees of freedom. They are translations in the x (lateral), y
(longitudinal), and z (vertical) directions (Fig. 21.9) and rotations about the x-axis (pitching),
the y-axis (rocking), and the z-axis (yawing) (Fig. 21.9).
Of these six types of motion, some are independent and some are coupled. Translation
along the z-axis and rotation about the z-axis can occur independently. But translation about
the x-axis (or y-axis) and rotation about the y-axis (or x-axis) are coupled motions. Hence, in
the analysis of a block foundation, one should consider four types of motions.
The theory of vibrations, based on one degree of freedom, is sufficient for the solution
of many problems (Converse, 1962). Hence, analysis pertaining to vertical vibration only is
considered.
A variety of theoretical approaches are available in the literature for determining the
dynamic response of machine foundations. Two principal methods are common. The first
considers the soil as an elastic half-space with the foundation on the surface. The second
Vertical
Z
Yawing
X Y
l
ina
gitud
Lon
Pitching
Y La
ter
Rocking al X
Fig. 21.9 Modes of vibration (Source: IS: 2974 – Part 1, 1982)
M21_PURU1773_01_SE_C21.indd 771 2/12/2013 8:19:28 AM

considers the foundation–soil system as a damped or undamped vibrating spring–mass

system. The elastic half-space concept is more rational but complicated and beyond the
scope of this book. In the spring–mass system, the soil behaviour is considered to be linear
and elastic. Damping does not affect the resonant frequency, but it has a considerable effect
on resonant amplitudes. As a machine foundation is designed avoiding the zone of reso-
nance, the effect of damping on amplitudes computed at operating frequency is also small
compared to that at resonance. Hence, for no damping condition, the relevant expressions
for natural frequency and amplitude are given below.
For defining the soil constant, for use in the analysis, the soil is idealized as linear and
elastic. Further, the soil constant varies with the stress produced and the elastic deformation
formed below the block. This depends on the mode of vibration.
The soil below the base of the foundation block undergoes uniform compression when
the block is subjected to vertical oscillations. Accordingly, the coefficient of elastic uniform
compression (Cu) is defined as the ratio of uniform compression (p) and elastic settlement (Se);
that is,
p
Cu = (21.16)
Se
But by definition the spring constant in the vertical direction, kz, is given as
Load pA
kz = = (21.17)
Elastic deformation Se
The equivalent spring constant of soil, kz, is given as
k z = Cu A (21.18)
where A is the area of the test plate.

The coefficient of elastic uniform compression can be obtained from field tests (Prakash,
1981) or Barkan’s (1962) values may be taken (Table 21.4) for preliminary designs.
Table 21.4 Recommended design values of Cu
Soil Soil group Permissible static Cu (kg/cm3)

load (kg/cm2)
I Weak soils (clays and silty clays with sand, in a plastic state, Up to 1.5 Up to 3.0
clayey and silty sands, also of categories II and III with
laminae of organic silt and peat)
II Soils of medium strength (clay and silty clays with sand, 1.5–3.5 3–5
close to the plastic limit, sand)
III Strong soils (clays and silty clays with sand, of hard consistency,
gravels and gravelly sands, loess and loessial soils)
IV Rocks >5.0 >10.0
Source: Barkan (1962).
M21_PURU1773_01_SE_C21.indd 772 2/12/2013 8:19:28 AM

Let us consider a machine foundation of area A subjected to an unbalanced vertical force

of F0 sin wt and placed at a depth of Df. For the sake of analysis, Df = 0 (this is a conservative
assumption) is taken, and the equation of motion of this system is given as
mz + k z z = F0 sin ωt (21.19)
The circular natural frequency, ωnz , of the system is
Cu A
ωnz = (21.20)
m
The amplitude of motion Az is given as

F0 sin ωop t
Az = (21.21)
m(ωn2z − ωop
2
)
where ωop is the circular operating frequency.
Equations 21.20 and 21.21 assume that the foundation rests on the surface. But under
field conditions, foundations are always embedded in the soil partially or fully. The effect of
embedment influences both the resonant amplitude and the frequency. In general, embed-
ment causes a reduction in the amplitude at the resonant peak followed by a marginal increase
in the resonant frequency. The two factors which contribute to this behaviour are the devel-
opment of frictional force on the vertical sides of the foundation and the surcharge. Several
analytical methods have been reported in the literature to study the effect of embedment with
particular reference to vertical vibrations (e.g., Anandakrishnan and Krishnaswamy, 1969,
1973; Baranov, 1967; Ramiah et al., 1977; Krishnaswamy (1975), based on field vibratory tests,
has shown that the amplitude of motion at resonance can be reduced by suitably adjusting
the intensity of surcharge around the footing. The other notable factor which influences the
vertical vibrations is soil moisture. It has been reported that soil moisture has a tendency to
reduce the spring constant (Krishnaswamy and Anandakrishnan, 1975).
21.9 LIQUEFACTION OF SOILS

Liquefaction denotes a condition where a soil will undergo continued deformation at a con-
stant low residual stress or with no residual resistance. Liquefaction occurs mostly due to
earthquake forces which induce high pore water pressure resulting in low confining pres-
sure. Liquefaction often appears in the form of sand foundations. When soil fails due to liq-
uefaction, the structures founded on such soil sink. Liquefaction is more pronounced in
sandy soils.
The shear strength of sand depends solely on internal friction. The effective shear strength
for sands is
τ = σn′ tan φ ′ (21.22)
where τ is the shear strength of sand, σn′ is the effective normal stress on any plane, and φ′ is
the angle of internal friction.
M21_PURU1773_01_SE_C21.indd 773 2/12/2013 8:19:29 AM

Ground vibrations in sand tend to reduce the volume; if there is no room for reduction in
volume, the pore water pressure increases dramatically and is expressed as
τ dyn = (σn′ − udyn ) tan φdyn

′ (21.23)
where τdyn is the shear strength of soil under dynamic condition, udyn is the excess pore
water pressure due to dynamic loading, and φdyn is the angle of internal friction under
dynamic conditions.
It can be observed that because of increase of pore water pressure, the effective normal
stress decreases, resulting in shear strength reduction. However, in sands under dynamic
′ is almost equal to φ′.
conditions, φdyn
For a zero shear strength condition, that is,
τ dyn = 0 ; σ n′ = udyn
or
udyn
=1 (21.24)
σn′
Increase in pore water pressure results in reduction in shear strength. Complete transfer
of inter-granular stress from soil grains to water is known as complete liquefaction. If the
transfer of stress is partial, then it is called partial liquefaction.
In case of complete liquefaction, the effective stress reduces to almost zero, and the sand–
water mixture behaves as a viscous material followed by starting of consolidation resulting
in surface settlement.
Liquefaction of sand can develop at any zone of a deposit. Liquefaction of the upper
layers may occur not as a direct result of the ground motion but because of development of
the liquefaction condition in an underlying zone of the deposit. When liquefaction develops
at a deeper level, the excess pore water pressure in the liquefied zone will dissipate due to
flow of water in an upward direction. If the hydraulic gradient is greater than the critical
gradient, then the upward flow of water induces a quick or liquefied condition in the surface
layers of the deposit. The onset of liquefaction at one zone of a deposit may lead to liquefac-
tion of other zones.
POINTS TO REMEMBER
21.1 When stress along a plane (or fault) in the ground exceeds the rupture strength of
rock, a sudden movement occurs along that plane of fault resulting in an earthquake.
21.2 After rupture, due to the elastic property of rocks, the ruptured portion pushes back
to the original position, which is referred to as elastic rebound.
21.3 The point inside the earth’s surface at which the first movement or break occurs
during an earthquake is called the focus or biocentre. The point on the earth surface
directly above the focus is the epicentre.
M21_PURU1773_01_SE_C21.indd 774 2/12/2013 8:19:30 AM

21.4 The major seismic waves are body waves and surface waves.
21.5 Magnitude of an earthquake is a measure of the amount of ground shaking based
on the amplitude of elastic waves. Richter’s scale represents the magnitude of an
earthquake.
21.6 Other dynamic forces are due to bomb blasts, machineries, pile driving, landing of
aircraft, etc.
21.7 Many of the foundation problems arising from oscillating machinery can be analysed
by considering the simple harmonic application of force.
21.8 A foundation is said to be in free vibration when it is disturbed and then kept free to
vibrate about some mean position. Vibrations that result from exciting agencies are
referred to as forced vibration.
21.9 In an elastic system, vibrations which are taking place due to inherent forces in the
system and are free from any external forces are termed natural frequency.
21.10 The foundation system is said to be at resonance when the frequency of the exciting
agency (operating frequency) is close to the natural frequency.
21.11 Damping is associated with energy dissipation and is the internal resistance offered
by a foundation system to the vibration of a machine.
21.12 Machines are broadly classified as (i) reciprocating machines, (ii) impact machines,
and (iii) rotary machines.
21.13 The design of a machine foundation has to satisfy the following conditions: (i) reso-
nance should be avoided, (ii) the amplitude of vibration at the operating frequency
should be less than the limiting amplitude, and (iii) vibrations of the machine should
neither cause damage to the machine and structures nor harm the health and comfort
of the people.
21.14 Liquefaction denotes a condition where a soil will undergo continued deformation at
a constant low residual stress or with no residual resistance.
QUESTIONS
Objective Questions
21.1 The major seismic waves are
(a) Body waves and surface waves
(b) Tension waves and translation waves
(c) Compound waves
(d) Rayleigh waves
21.2 An earthquake is said to be damaging when the Richter number is
(a) >8 (b) 6 to 6.9 (c) 5 to 5.9 (d) <4
21.3 When the frequency of the exciting agency (operating frequency) is close to the natu-
ral frequency, then the foundation system is said to be at
(a) The free vibrations stage (b) The forced vibrations stage
(c) Resonance (d) Low amplitude
M21_PURU1773_01_SE_C21.indd 775 2/12/2013 8:19:30 AM

21.4 Indicate the wrong statement.

1. In viscous damping, the damping force is taken directly proportional to the veloc-
ity of the oscillating mass.
2. The ratio of the actual damping to the critical damping is referred to as damping
factor.
3. Magnification factor is the ratio of maximum displacement and static deflection of
the system.
4. In order to avoid large amplitudes, the frequency ratio should be unity.
21.5 A rigid-body displacement of a block foundation produces two independent motions
and ______ coupled motions.
(a) Four (b) Three (c) Two (d) Six
21.6 Ground vibration in saturated sand causes the pore water pressure to
(a) Remain constant (b) Decrease
(c) Increase steadily (d) Increase suddenly
21.7 In case of complete liquefaction, the effective stress
(a) Reduces to almost zero (b) Increases suddenly
(c) Remains constant (d) None of the above
21.8 How are earthquakes caused?
21.9 What is an elastic rebound?
21.10 Name the two seismic waves. Explain briefly the importance of Rayleigh waves.
21.11 How is the magnitude of an earthquake measured?
21.12 Discuss the different types of vibrations.
21.13 Bring out the difference between free and forced vibrations.
21.14 What is a spring–mass system? How is damping represented?
21.15 Briefly explain the method of analysis of a block foundation.
21.16 What are the limits of amplitude of vibration with reference to human beings?
21.17 Explain liquefaction of sands during an earthquake.
M21_PURU1773_01_SE_C21.indd 776 2/12/2013 8:19:30 AM

22
Environmental Geotechnology
CHAPTER HIGHLIGHTS
Environmental cycles – Natural cycles: oxygen cycle, nitrogen cycle, carbon
cycle – Environmental imbalance: solid and liquid wastes – Petroleum contami-
nation – Acid rain and acid drainage – Scrap wastes – Arid lands and deserts
– Wetlands – Coastal margins – Soil erosion – Land subsidence – Birth of envi-
ronmental geotechnology – Contaminated soils: identification, classification,
permeability, electrical properties, shear strength, consolidation – Applications
– Load–environment design criteria
22.1 INTRODUCTION
Environmental geotechnology is an interdisciplinary science which covers soil and rock
and their interactions with various environmental cycles. This field has grown rapidly
since 1986.
Air, water, and soil are in eternal interaction with each other. Therefore, the phases
of the pollution process should not be viewed in isolation but should be taken in inte-
gration with the other phases (Fang, 1997). Hence, designs that are based only on the
physico-mechanical behaviour of the soil are not adequate, and the role of the environ-
mental aspects has to be included in the design.
22.2 ENVIRONMENTAL CYCLES

Environmental cycles consist of atmosphere, biosphere, hydrosphere, lithosphere, and
geo-microbiosphere.
The atmosphere is the layer of air surrounding the earth. The biosphere represents all
plant and animal life. The hydrosphere deals with water in oceans. The lithosphere refers to
soil and rock which contain combined silicon, oxygen, aluminium, and other elements.
M22_PURU1773_01_SE_C22.indd 777 2/12/2013 8:19:11 AM

Atmosphere
Air Pollution cycle
Environmental
Biosphere geotechnology
Geo-microbiosphere
Water
Land
Lithosphere Hydrosphere
Fig. 22.1 Environmental cycles and their interactions (Source: Fang, 1997)
Geo-microbiosphere is a new term which does not come under the ecosphere. It is the study
of the characteristics of tree and vegetation roots and bacterial activities in the groundwater
and soil and subsequent response to the engineering behaviour of the soil–water system.
Figure 22.1 shows the environmental cycles and their interactions.
22.3 NATURAL CYCLES

Many elements in the ecosphere are exchanged. These exchanges are caused from physical
or physico-chemical interactions among the atoms and molecules that exist in each sub-
sphere. Such cyclic exchanges are called natural cycles, viz., oxygen cycle, nitrogen cycle,
and carbon cycle.
22.3.1 Oxygen Cycle

In the earth, oxygen is available along with silicon (Si) and aluminium (Al) as aluminium
silicates (rocks). It may also be found in combination with metals, viz., metallic oxides,
carbonates, sulphates, nitrates, and phosphates. The three substances, viz., water, carbon
dioxide, and molecular oxygen, in combination with molecules of the living systems form
the oxygen cycle.
M22_PURU1773_01_SE_C22.indd 778 2/12/2013 8:19:11 AM

Environmental Geotechnology 779
22.3.2 Nitrogen Cycle

Nitrogen is the fourth abundant element next to carbon, hydrogen, and oxygen. Nitrogen
cycle involves the transfer of nitrogen between atmosphere, biosphere, lithosphere, and
hydrosphere in various chemical forms.
22.3.3 Carbon Cycle

In the biosphere, the carbon cycle consists of essential elements. The main function of the cycle is
the oxidation of carbon-containing materials to atmospheric carbon dioxide, with photosynthesis
returning it to the biosphere. Three important features of the cycle are as follows: (i) Fixed carbon
content in plants is oxidized back to carbon dioxide, (ii) There is a large availability of carbon as
carbonate in the sedimentary rocks and (iii) Carbon gets modified by fossil fuel combustion.
22.4 ENVIRONMENTAL IMBALANCE

Man-made environments and natural causes bring in an environmental imbalance. Man-made
environments include solid and liquid wastes, petroleum contamination, acid rain, rubber
tyres, etc. Natural causes include erosion, subsidence, landslides, land usage, etc. (Fang, 1997).
22.4.1 Solid and Liquid Wastes

Wastes may be of two forms: solid wastes and liquid wastes. Urban refuse is the major solid
waste. With time, the solid wastes generate gases and liquid wastes, which are referred to as
leachates. In solid and liquid wastes, hazardous and toxic wastes constitute 16%, sludge 14%,
and nuclear wastes 12%.
22.4.2 Petroleum Contamination

Oil spills occur accidentally during transportation, both on land and on sea from storage tanks
or during oil drilling. Underground storage tanks of petroleum products are not leak-proof, but
25% of them leak petroleum liquids into the ground. When a leak or oil spill occurs, soil around
the point of leakage gets contaminated and requires special remediation and reclamation.
22.4.3 Acid Rain and Acid Drainage

Acid rain is formed by the emission of various pollutants into the air. The major sources are fos-
sil fuel power plants, automotive exhausts, and industrial facilities. These emissions contain
primarily sulphur dioxide, nitrogen, and other gases, which combine with oxygen and water
vapour in air to form sulphuric acid, nitric acid, hydrochloric acid, and other acids. As acids are
suspended high above the earth, they are carried far off by wind and are dropped or washed out
from the air thus causing acid rain depositions. Also, acid rains are not uniformly distributed.
Acid drainage is due to water draining from old abandoned coal mines. The iron sul-
phides or pyrites in the mines get oxidized to sulphuric acid, which dissolves in water and
drains through the mines. This acid water then enters streams, lakes, or ground soil–water
systems, thereby resulting in pollution.
M22_PURU1773_01_SE_C22.indd 779 2/12/2013 8:19:11 AM

22.4.4 Scrap Wastes

The major scrap waste is from rubber tyres. Only a small amount is used for recycling. In
most occasions, tyres pile up, particularly in developed countries, and cause fire hazard and
health hazard.
22.4.5 Arid Lands and Deserts

It is estimated that deterioration of productive land to desert-like conditions is increasing and
is at a rate of 14.8 million acres per year. The deserts thus formed constitute about 35% of the
earth’s surface, i.e., about 45 million square miles. Because of the loss of productive land due
to desertification, more people are likely to be underfed and die of starvation. This should be
avoided.
22.4.6 Wetlands
Urban development invariably uses large areas of wetlands for residential, industrial, and
commercial purposes. Although additional useful lands are created for urban development,
there are many detrimental effects. For example, various ecological effects may be caused by
the construction of highway earth fills on wetlands, viz., inhibition of storm water and tidal
distribution, increased water turbidity, alteration of water circulation patterns, removal of
natural filtration systems, etc. Thus, construction of any structure on a specific wetland or on
a problematic soil deposit is a complex problem.
22.4.7 Coastal Margins

Construction of structures on coastal lands and other activities on the area may lead to many
changes in the coastal environments. Some of the factors are given below:
1. Construction of waterworks for irrigation, storm protection, and power may reduce the
coastal land area, which may cause interception of water and sediments and shifting of
channel locations.
2. Construction of industries on coastal land may result in the production of efflu-
ents which are let into coastal land or sea causing increased pollution of near-shore
environments.
3. Construction and activities on coastal land leads to loss of wildlife habitat.
4. Further constructions lead to increase of rate of sea level.
22.4.8 Soil Erosion

Soil erosion is caused by the drag action of wind, rainfall, or wave action on the surface of an
unprotected soil surface. Soil erosion causes soil particle detachment and transport. The
sediments formed due to soil erosion (i) restrict the efficiency of drainage and irrigation
ditches, fills, and reservoirs, (ii) pollute surface water, (iii) damage agricultural lands, and
(iv) upset the natural ecology and environment.
M22_PURU1773_01_SE_C22.indd 780 2/12/2013 8:19:11 AM

22.4.9 Land Subsidence

Application of a load on a soil mass causes deformation due to (i) immediate deforma-
tion of soil structure, (ii) escape of pore water due to consolidation, (iii) time-dependent
deformation, and (iv) a combination of all the above. Thus, subsidence depends on the
applied load, soil properties, drainage condition, stress history, and environmental
factors. Ground subsidence may also be caused due to natural causes such as tectonic
movements, which include tectonic creep and tectonic earthquake. Ground surface
movement may be caused by tree roots, floods, de-watering, etc.
22.5 BIRTH OF ENVIRONMENTAL GEOTECHNOLOGY

Environmental geotechnology has come into being because of two major reasons, viz.,
population growth and rising living standards.
An increase in population demands a need for more land. Many soil deposits which were
claimed to be unfit for residential housing or for other construction projects have been com-
pelled to be used now because of increase in population. Such soil deposits consist of land-
fills, wetlands, collapsible soil regions, mining subsidence areas, etc. Construction in such
problematic soils requires additional techniques and use of new materials along with the
conventional construction techniques and materials.
Rising living standards in a rapidly growing urban area causes hazardous pollution of
air, water, and land and urban refuse, which collectively endanger the global environment.
Both problematic soil deposits and pollution problems have challenged the conventional
soil mechanics concepts and methods of analysing soil behaviour under varied environ-
mental conditions. Thus, environmental geotechnology has come into being to tackle the
problems both on sound theoretical grounds and on practical considerations.
The birth of environmental geotechnology is based on practical and logical reasons
and stems to provide (i) better productivity, (ii) healthier economy, and (iii) safe and
secure environment. Environmental geotechnology is an emerging discipline comprising
geotechnical engineering, environmental science and engineering, and soil agricultural
science.
22.6 CONTAMINATED SOILS

Soils polluted due to various environmental reasons have to be identified before taking up
construction activity. Further, apart from conventional tests, special tests have to be con-
ducted so as to have a thorough knowledge of the behaviour of the soil.
22.6.1 Identification
During reconnaissance and field investigation, colour, odour, and presence of organic matter
and other debris help in the proper identification of contaminated soils. Sometimes, water
bubbles are formed by certain polluted waters. Cracking patterns and soil colour are affected
by contaminated soil water.
M22_PURU1773_01_SE_C22.indd 781 2/12/2013 8:19:11 AM

22.6.2 Classification
The basic concept of classification of contaminated soil is based on the particle size. Even a
small percentage of clay and colloids may contribute to the entire solid surface. Thus, the
interaction of soil surface and the pore fluid depends on the pore fluid characteristics and the
physico-chemical properties of the particle surface. The controlling parameters for proper
classification, considering the environmental factors, are 2 μm fraction, specific surface, pH
in pore fluid, ion-exchange capacity, absorption and adsorption, and dielectric constant.
22.6.3 Permeability
Permeability of a contaminated soil is significantly affected by pore fluid characteristics.
Fang (1997) has reported that permeability increases when (i) the percentages of multivalent
cations in the pore fluid are increased, (ii) the organic contents are increased, and (iii) the salt
concentration of a pore fluid is increased.
22.6.4 Electrical Properties

Engineering properties of a contaminated soil–water system are closely related to the electri-
cal properties of the system. In general, they depend on the electric surface charge, dielectric
constant, and redox reaction. The pH in contaminated pore fluids will change the ion-exchange
reactions, which in turn influences the electrical properties of the soil–water system.
22.6.5 Shear Strength

Temperature and electrolyte concentration have significant effect on the shear strength of
contaminated soils. Soils formed from clay–water slurries, with a high salt concentration
and subsequently leached, may show high sensitivity. Contaminated pore fluids due to
industrial and hazardous/toxic waste may have adverse effects on shear strength.
22.6.6 Consolidation
As for shear strength, the effect of pore fluid and temperature may show different settlement
characteristics in contaminated soil. More recently, Terzaghi’s concept of consolidation has
been extended to contaminated soils taking into account the electrokinetic, thermal, and
electromagnetic effects.
22.7 APPLICATIONS
All civil engineering structures are founded directly or indirectly on ground soil or rock. The
stability of the structure, both sub-structure and superstructure, depends on the stability
characteristics of the soil, which is influenced to a great extent by local environmental
factors. Foundation is the most important component of a structure whose performance in
turn depends on the nature of ground or its improvement.
The current practice of foundation is based mostly on the load and settlement criterion.
Now, the practice has to be modified to take into account the environmental factors into
M22_PURU1773_01_SE_C22.indd 782 2/12/2013 8:19:11 AM

design. Thus, the modified design criteria should have a safe load–environment factor
design.
If the existing ground does not satisfy the load–settlement–environmental factors, measures
must be taken to improve the factor that will make the ground suitable for use. As discussed
in Chapter 19, ground improvement comprises improving mechanical, hydrological, physico-
chemical, biological, or a combination of these properties or any other properties of nature or
man-made soil deposits. The purposes of ground improvement are as follows:
1. To strengthen ground soil before failure occurs,
2. To strengthen ground soil for the full life period,
3. To strengthen ground soil after premature or unexpected failure, and
4. To provide temporary ground improvement.
For deciding an effective ground improvement system, the following factors need to
be considered: (i) sensitivity of soil to environment, (ii) ground soil–water pollution, (iii)
adverse ground conditions, and (iv) selection of construction materials.
22.8 LOAD–ENVIRONMENT DESIGN CRITERIA

Conventional design approach is based on a safe load which could avoid a collapse, and the
structure will be allowed to settle within the permissible limit. But the ground soil is
extremely sensitive to the local environments such as temperature, moisture content, the
characteristics of the pore fluid, weather conditions, groundwater fluctuations, and degree
of groundwater and air pollution. Thus, civil engineering design of structures is not simple
but interdisciplinary in nature and demands the help of all related fields so as to provide a
safe and economical structure to have a long life.
Loads to be considered for a load–environment design criteria are, static loads, dynamic
loads, and environmental loads.
Static load, also called as dead load, is the weight of the structure and all materials
permanently attached to it. Dynamic loads are live load and impact load. Major live loads
are human occupancy, partition walls, furniture, warehouse goods, and mechanical equip-
ment. Local Building Code provides the live load requirements depending on the type of
structure.
Environmental loads include earthquake, wind, snow, water pressure, and others. Forces
due to water flow, ice flow, and wave forces have to be considered in the design of water-
front structures.
POINTS TO REMEMBER
22.1 Environmental geotechnology is an interdisciplinary science which covers soil and

rock and their interactions with environmental cycles.
22.2 Environmental cycles consist of atmosphere, biosphere, hydrosphere, lithosphere,
and geo-microbiosphere.
22.3 Natural cycles are oxygen cycle, nitrogen cycle, and carbon cycle.
M22_PURU1773_01_SE_C22.indd 783 2/12/2013 8:19:11 AM

22.4 Man-made environments and natural causes bring in an environmental imbalance.

22.5 Man-made environments include solid and liquid wastes, petroleum contamination,
acid rain, acid drainage, scrap wastes, etc.
22.6 Natural causes include acid lands and deserts, wetlands, coastal margins, soil
erosion, land subsidence, etc.
22.7 Two major reasons, viz., population growth and rising living standards, have shown
the way for the development of the field of environmental geotechnology.
22.8 Soils polluted due to various environmental reasons have to be identified before
taking up any construction activity.
22.9 In addition to the conventional design requirements such as a safe load to avoid
collapse and not to exceed permissible settlement, environmental effects on the soil
have to be considered.
22.10 If the existing ground is not in a position to satisfy the load–settlement–environmental
factors, the ground should be improved by any ground improvement techniques.
QUESTIONS
Objective Questions
22.1 Many elements in the ______ are exchanged.
(a) Hydrosphere (b) Ecosphere
(c) Microsphere (d) Lithosphere
22.2 Indicate the wrong statement. Man-made environments include
(a) Solid wastes (b) Liquid wastes
(c) Petroleum contamination (d) Land subsidence
22.3 Assertion A: Nitrogen is the fourth abundant element next to carbon, hydrogen, and
oxygen.
Choose the correct statement:
Reason R: In the biosphere, carbon cycle, like nitrogen cycle, is of essential elements.
(a) A and R are true, and R is the explanation of A.
(b) A and R are true, and R is not the explanation of A.
22.4 Identify the wrong statement. Sediments formed due to erosion
(a) Resist the efficiency of drainage systems
(b) Damage agricultural lands
(c) Upset the natural ecology and environment
(d) Improve surface water quality
22.5 Choose the wrong term. For design of water-front structures, the following forces
should be considered:
(a) Water flow (b) Ice flow
(c) Atmospheric flow (d) Wave force
M22_PURU1773_01_SE_C22.indd 784 2/12/2013 8:19:11 AM

22.6 Select the wrong statement.

(a) Dynamic loads are live load and impact load.
(b) Environmental design criteria consider safe load, dynamic load, and environmental
load.
(c) Ground improvement may be for permanent or temporary purposes.
(d) Contaminated soils generally have high shear strength.
22.7 Explain clearly the natural cycles.
22.8 What are the factors involved in environmental imbalance?
22.9 Briefly explain the importance of land subsidence.
22.10 How are contaminated soils identified?
22.11 Briefly explain the need for consideration of environmental factors.
22.12 What are the additional factors to be considered in the load–environment design
criteria?
M22_PURU1773_01_SE_C22.indd 785 2/12/2013 8:19:11 AM

M22_PURU1773_01_SE_C22.indd 786 2/12/2013 8:19:11 AM
23
Introductory Rock Mechanics
CHAPTER HIGHLIGHTS
Index properties of rocks – Classification of rocks – In situ state of stress –
Mechanical properties of rocks
23.1 INTRODUCTION
Rock mechanics is an inter-disciplinary subject covering various fields, such as engi-
neering geology, mining, petroleum, and civil engineering. The common man believes
that rocks are strong materials with reliability and permanence and need no treatment
if they are used as foundations. But rocks present in nature have several geological dis-
continuities, such as faults, joints, fissures, etc. The behaviour of a rock mass subjected
to a change in stress is governed by the mechanical properties of the rock material and
the geological discontinuities.
Several major failures of dams and rock slopes have occurred in the past due to
inadequate analysis and understanding of rock behaviour. Nowadays, tall structures,
high dams, lengthy tunnels, and deep mines are constructed, which induce a large
stress change. Thus, for an exact analysis and proper prediction of the behaviour of
civil engineering structures on rock and the use of rock as a construction material, a
complete understanding of the engineering properties and behaviour of rock masses
is essential. The basic knowledge of soil behaviour, gained from the preceding chap-
ters, forms a sound foundation for understanding the rock behaviour dealt in this
chapter.
23.2 INDEX PROPERTIES OF ROCKS

Rocks occur with a variety of structures, fabrics, and components. In order to describe a
rock quantitatively, we need to know about certain properties of the rocks which may
be designated as index properties of rocks. Such properties are porosity, density or unit
M23_PURU1773_01_SE_C23.indd 787 2/12/2013 8:18:27 AM

weight, permeability, strength, durability, and sonic velocity (Goodman, 1980). These
properties are explained in the following paragraphs.
23.2.1 Porosity
Porosity of a rock, as in soils, is defined as the fraction expressing the proportion of void
space to total space in the rock and is represented as a percentage. It identifies the relative
proportion of solids and voids. The porosity of a rock depends on the particles-size distribu-
tion, sorting, grain shape, fabric, degree of compaction and cementation, solution effects,
and mineralogical composition. The highest porosity is commonly attained when all the
grains are of the same size. Large ranges of porosity can be obtained with irregularly shaped
grains.
In sedimentary rocks, the average porosity is about 15%. In these rocks, there is a decrease
in porosity with age and depth. Among the rocks, chalk is the most porous with a porosity of
50%. In unweathered igneous rocks, the porosity is usually less than 1% or 2% which, with
the progress of weathering, increases to about 20%. Thus, the measurement of porosity in
such rocks serves as an accurate index to rock quality.
Porosities of rocks are determined by adopting a variety of techniques, such as measured
density, measured water content after saturation in water, mercury content after saturation
with mercury using a pressure injector, and measured solid volume and pore air volume
using Boyle’s law. The standard saturation method (Franklin, 1970) and the air porosimeter
(Ramana and Venkatanarayana, 1971) are the conventional methods used for determining
porosity.
23.2.2 Density
The dry density or unit weight of a rock is its specific weight (kg/m3 or kN/m3 as in soil).
From a knowledge of specific gravity of rock grains, the rock density or unit weight can be
approximately found. For example, a rock with a specific gravity of 2.8 has a unit weight of
28 kN/m3. The grain specific gravity can be determined by grinding the rock and adopting
methods used for soils. Rocks exhibit a large range of density values compared to soils. The
density of a rock is easily measured from the volume and mass of a carefully drilled core
specimen. Table 23.1 lists the densities of some common rocks.
A knowledge of rock density has wide application in civil engineering and mining prac-
tices, e.g., in the selection of aggregates for roads, slabs and beams, and gravity dams and in
the analysis of underground openings; besides, in oil shale deposits, it directly correlates to
the oil yield and in coal deposits, it directly reflects the ash content.
23.2.3 Permeability
Water is always present and stored in the rock masses. It percolates through pores when
subjected to hydraulic gradient. As in soils, permeability in rocks is defined as the ability of
a rock to allow the passage of fluids into or through it, without impairing its structure. Per-
meability is selected as an index property of rock since it gives information about the inter-
connection between the pores or fissures.
M23_PURU1773_01_SE_C23.indd 788 2/12/2013 8:18:27 AM

Introductory Rock Mechanics 789
Table 23.1 Dry densities of some typical rocks
Rock name Dry density (g/cm3)
Coal 0.7–2.0
Oil shale 1.6–2.7
Rock salt 2.1
Shale 2.25–2.62
Gypsum 2.30
Rhyolite 2.37
Syenite 2.60–2.70
Granite 2.65
Basalt 2.77
Quartz, mica schist 2.82
Diorite 2.85
Amphibolite 2.99
Gabbro 3.00
Source: Goodman (1980).
Darcy’s law is valid in most of the rocks. If the permeant is other than water and the tem-
perature varies considerably, a more useful form of Darcy’s law is
K dp
q= A (23.1)
ηw dx
where q is the discharge, K the permeability coefficient independent of permeant (length2), p
the fluid pressure (equal to γh), ηw the viscosity of the permeant, and A the area of cross section.
As in soils, permeability of rocks can be assessed in the laboratory from a constant or
falling head test. An alternative approach is to allow a radial flow in a hollow specimen
(Bernaix, 1969) and measure the discharge either from inner surface to outer surface or from
outer surface to inner surface. For radial flow in an unfractured rock with inter-connected
pores, the permeability is given by
qγ w ⎛R ⎞
k= log ⎜⎜⎜ 2 ⎟⎟⎟ (23.2)
2πLΔp ⎝ R1 ⎟⎠
where R1 and R2 are the radii of inner and outer surfaces, L the length of cylinder, and Δp the
difference between external and internal water pressure.
Berniax (1969) showed that porous rocks are unaffected by pressure changes. Further, it
was shown that fissured rocks exhibit higher permeability in divergent flow than in conver-
gent flow. It is presumed that some amount of hydraulic fracturing∗ would have occurred
in divergent flow. Goodman and Sundaram (1980) reported similar behaviour based on the
tests on tuff, schist, sandstone, and limestone. Some typical values of permeability of rocks
are presented (Serafim, 1968) in Table 23.2.
∗
It is the process of initiation of a crack caused by water pressure.
M23_PURU1773_01_SE_C23.indd 789 2/12/2013 8:18:27 AM

Table 23.2 Coefficients of permeability (to

water) of various rock materials
Rock materials k (cm/s)
Sandstone 1.6 × 10−7 ⯝10−10
Siltstone 10−8 ⯝10−9
Granite 2 × 10−10 ⯝ 5 × 10−11
Slate 1.6 × 10−10 ⯝ 7.0 × 10−11
Calcite 9.3 × 10−8 ⯝ 7.0 × 10−10
Limestone 1.2 × 10−7 ⯝ 7.0 × 10−10
Dolomite 1.2×10−8 ⯝ 4.6 ×10−9
Hard mudstone 2 × 10−6 ⯝ 6 × 10−7
Source: Stagg and Zienkinwicz (1968).

The measurement of permeability in rocks has a significant bearing on many engineering
and mining problems, viz., pumping of water, oil, or gas into or out of a porous formation,
storing fluids in mined caverns for energy conversion, disposing brine waste in porous for-
mation, assessing the water tightness of a reservoir, and predicting water inflows in a tunnel
(Goodman, 1980).
23.2.4 Strength
A strength index is useful only if the results from tests on rocks are reproducible and inex-
pensive. The point load test described by Franklin and Broch (1972) is identified as a strength
index test. In this test, a specimen is subjected to compression between two cone-shaped
platens (Fig. 23.1), which causes tensile stresses normal to the axis of loading and the point
load strength (tensile strength) σpl is given as
P
σpl = (23.3)
D2
where P is the load at rupture and D the distance between the point loads.
The point load strength index is reported as the point load strength of a 50 mm diameter core.
The point load strength test is simple and quick and can be carried out in the field itself.
Rocks showing strong anisotropy, such as slates, schists, laminated sandstones, etc., have
Fig. 23.1 Point load test on rock prism (Source: IS: 10082, 1981)
M23_PURU1773_01_SE_C23.indd 790 2/12/2013 8:18:29 AM

Table 23.3 Point load index
Description Point load strength index (MPa) Equivalent uniaxial compressive

strength (MPa)
Extremely high strength >10 >160
Very high strength 3–10 50–160
High strength 1–3 15–60
Medium strength 0.3–1 5–16
Low strength 0.1–0.3 1.6–5.0
Very low strength 0.03–0.1 0.5–1.6
Extremely low strength <0.03 <0.5
Source: Franklin and Broch (1972).
to be tested both along and parallel to bedding planes. Franklin and Broch (1972) have pro-
vided a point load strength index for easy identification of rocks (Table 23.3). The uncon-
fined compressive strength of rocks (qu) is correlated to the point load strength test as
qu = 24σpl(50) (23.4)
where qpl(50) is the point load strength corresponding to a 50 mm diameter core.
23.2.5 Slaking and Durability

Hydration, slaking, solution, oxidation, and other processes produce changes in the
properties of rocks. Such changes affect the durability of rocks. As there are many paths of
deterioration, some index to represent the degree of durability of rocks is required.
One such index test is the slake durability test suggested by Franklin and Chandra (1972).
The test essentially consists of rotating about 500 g of rock broken into 10 lumps and loaded
in a drum which is turned at 20 rpm in a water bath for 10 minutes. The drum is 140 mm in
diameter and 100 mm long with a 2 mm opening sieve mesh forming the cylindrical walls.
After 10 minutes, the percentage of rock retained inside the drum, on a dry weight basis, is
reported as the slake durability index (Id). Gamble (1971) recommended a second 10 minute
cycle and, based on the test results, proposed a Slake Durability Scale (Table 23.4).
Table 23.4 Gamble’s Slake Durability Scale
Group name Percentage retained after one Percentage retained after two
10-minute cycle (dry weight basis) 10-minute cycles (dry weight basis)
Very high durability >99 >98
High durability 98–99 95–98
Medium high durability 95–98 85–95
Medium durability 85–95 60–85
Low durability 60–85 30–60
Very low durability <60 <30
M23_PURU1773_01_SE_C23.indd 791 2/12/2013 8:18:29 AM

Table 23.5 Description of rate and amount of slaking
Amount of slaking Liquid limit (%)

Very low <20
Low 20–50
Medium 50–90
High 90–140
Very high >140
Rate of slaking Change in liquidity index
Slow <0.75
Fast 0.75–1.25
Very fast >1.25
Morgenstern and Eigenbrod (1974) used a water absorption test to assess the amount and
rate of slaking of argillaceous materials. They found that non-cemented claystone or shale
absorbed water faster and that all materials eventually reached a final water content equal to
their liquid limits. Materials with high liquid limits exhibited a high degree of slaking than
those with low liquid limits. They have also shown that the rate of slaking is reflected by the
rate of water content change following soaking. The rate of slaking is given by the change in
liquidity index (ΔIL) following 2 hours of immersion in water; that is,
Δw
T ΔI L = (23.5)
wL − w p
where Δw is the change in water content of the rock after soaking for 2 hours.
Table 23.5 provides the rate and amount of slaking as suggested by Morgenstern and
Eigenbrod (1974), reported by Goodman (1980).
23.2.6 Sonic Velocity

The effect of discontinuities, such as fissures, in a rock mass can be estimated by comparing
the in situ wave velocity with the laboratory sonic velocity of an intact core obtained from
the same rock mass. The movement of stress waves in a rock mass depends on the elastic
property and density of rocks.
In practice, a network of fissures in the specimen superimposes an overriding effect.
Under these conditions, the sonic velocity can be utilized to index the degree of fissuring
with rock specimens (Goodman, 1980).
Fourmaintraux (1976), reported by Goodman (1980), proposed a classification scheme
based on longitudinal wave velocity for fissuring in rock specimens. Assuming the speci-
men has no pores or fissures, the longitudinal wave velocity (Vl*) is calculated from the
known mineral composition. Thus,
1 C
=∑ i (23.6)
Vl i Vl , i
M23_PURU1773_01_SE_C23.indd 792 2/12/2013 8:18:30 AM

100
90
75
Sl N
igh on-f
M tly iss
od
50 er fis ure
at su d
IQ%
ely re
St f iss d
ro ur
ng ed
Ve ly
25 ry fis
fis stro s ur
su n ed
re gly
d
0 10 20 30 40 50 60
n%
Fig. 23.2 Classification scheme for fissuring in rock specimens (Source: Goodman, 1980)
where Vl,i is the longitudinal wave velocity in the ith mineral constituent and Ci the volume
proportion.
Now the actual velocity of longitudinal waves (Vl) is measured. A quality index (IQ) is
defined as
⎛V ⎞
IQ = ⎜⎜⎜ l* ⎟⎟⎟×100% (23.7)
⎜⎝ Vl ⎟⎠
Further, Fourmaintraux (1976) showed the dependence of IQ on the porosity of the rock.
Thus,
IQ = (100 − 1.6np )% (23.8)
where np is the porosity of the non-fissured rock expressed in percentage.

Fourmaintraux (1976) presented a plot of IQ versus porosity (Fig. 23.2) as a basis for
describing the degree of fissuring of a rock specimen.
23.3 CLASSIFICATION OF ROCKS

In general, rock classification is required to help in making an initial assessment of a problem
and to examine the areas where additional information has to be supplemented in order to
obtain the required answer. Many civil engineering works require rock classification, which
includes the assessment of slope stability, open and sub-surface excavations, foundation sta-
bility, and the selection of rocks as construction materials.
Geological classifications of rocks are based on a genetic point of view, viz., the conven-
tional geological classification: igneous, metamorphic, and sedimentary. Deere (1979) and
Goodman (1980) emphasized that the geological classification should also consider the
lithology of rocks, which refers to mineralogy, texture, fabric, and structural anisotropy.
M23_PURU1773_01_SE_C23.indd 793 2/12/2013 8:18:30 AM

Goodman (1980) presented a geological classification based on lithology and divided the
rocks into the following classes and sub-classes:
1. Crystalline texture
(i) Soluble carbonates and salts, e.g., limestone, dolomite, marble, rock salt, trona,
gypsum
(ii) Mica or other planar minerals in continuous bands, e.g., mica schist, chlorite schist,
graphite schist
(iii) Banded silicate minerals without continuous mica sheets, e.g., gneiss
(iv) Randomly oriented and distributed silicate minerals of uniform grain size, e.g., gran-
ite, diorite, gabbro, syenite
(v) Randomly oriented and distributed silicate minerals in a background of very fine
grain and with vugs, e.g., basalt, rhyolite, other volcanic rocks
(vi) Highly sheared rocks, e.g., serpentinite, mylonile
2. Clastic texture
(i) Stably cemented, e.g., silica-cemented sandstone and limonite sandstones
(ii) With slightly soluble cement, e.g., calcite-cemented sandstone and conglomerates
(iii) With highly soluble cement, e.g., gypsum-cemented sandstone and conglomerates
(iv) Incompletely or weakly cemented, e.g., friable sandstone, tuff
(v) Uncemented, e.g., clay-bond sandstones.
3. Very fine grained rocks
(i) Isotropic, hard rocks, e.g., hornfels, some basalts
(ii) Anisotropic on a macro-scale but microscopically isotropic hard rocks, e.g., cemented
shales, flagstones
(iii) Microscopically anisotropic hard rocks, e.g., slate, phyllite
(iv) Soft, soil-like rocks, e.g., compaction shale, chalk, marl
4. Organic rocks
(i) Soft coal, e.g., lignite and bituminous coal
(ii) Hard coal
(iii) Oil shale
(iv) Bituminous shale
(v) Tar sand
For engineering purposes, mere geological names are insufficient and should be accompa-
nied by an engineering classification. Such a classification may be concerned with the physical
properties of rocks and the nature of discontinuities within the rock masses such that system-
atic predictions can be made about their engineering behaviour. One such classification has
been suggested by Deere and Miller (1966) for intact rocks. This classification is based on
two important engineering properties of intact rocks, viz., the uniaxial compressive strength
and the modulus of elasticity. Here the tangent modulus is used which is being computed
at a stress level equal to one-half the ultimate strength of the rock. The uniaxial compressive
strength is that determined on specimens with a length diameter ratio of at least two.
Based on uniaxial compressive strength, rocks are divided into five categories, and
designated as A, B, C, D, and E, as shown in Table 23.6. The second element modulus of
elasticity has been used as a ratio of the modulus to the uniaxial compressive strength,
M23_PURU1773_01_SE_C23.indd 794 2/12/2013 8:18:30 AM

Table 23.6 Strength classification of intact rocks
Class Description Uniaxial compressive strength

(MPa)
A Very high strength >224
B High strength 112–224
C Medium strength 56–112
D Low strength 28–56
E Very low strength <28
Source: Deere and Miller (1966).
Table 23.7 Modulus ratio classification of intact rocks
Class Description Modulus ratio

H High modulus ratio >500
M Medium modulus ratio 200–500
L Low modulus ratio <200
Source: Deere and Miller (1966).
called the modulus ratios designated as H, M, and L (Table 23.7). The strength categories
follow a geometric progression. Combining these two letters a rock may be classified as
AM, CH, BH, DL, etc.
A classification chart with the value of compressive strength and modulus in logarithmic
scale has been provided by Deere and Miller (Fig. 23.3). The strength classification is shown
across the top of the chart and the modulus ratio is obtained from the plotted position with
respect to the diagonal lines. The zone above the upper line with a modulus ratio of 500:1
represents the zone of high modulus ratio (H), the zone below the lower line with a modulus
ratio of 200:1 shows a zone of low modulus ratio (L), and the stippled zone is referred to as M –
the zone of medium modulus ratio. It is observed that different rock types occupy different
unique positions in the chart. Deere and Miller (1966) attributed this characteristic variation
in properties to the sensitiveness of rocks to mineralogy, fabric, and direction of anisotropy.
This engineering classification for intact rock is considered as a workable and useful system.
Engineers are concerned with a classification system which considers various aspects of
both the rock material and the rock masses. Such a classification system is referred to as the
geomechanics classification system. Terzaghi (1946) was one of the earlier researchers who
recognized the significance of discontinuities, their spacing, and the filling materials. The
basic idea of classification of rock masses was given by John (1962), who considered the
compressive strength, degree of jointing, and amount of alteration. The geomechanics clas-
sification system of Wickham et al. (1968), Bieniawski (1973, 1974a, b, 1975, 1976), and Barton
et al. (1975) are very well received. Bieniawski’s system may be rated high.
Bieniawski’s system suggests a general rock rating (RMR) increasing with rock quality
from 0 to 100. This system is based on five universal parameters: strength of rock (15),
drill core quality (20), groundwater condition (10), joint and fracture spacing (30), and
M23_PURU1773_01_SE_C23.indd 795 2/12/2013 8:18:30 AM

E. Tangent Modulus at 50% ultimate strength
E D C B A
Very low Low Medium High Very high
16 strength strength strength strength strength
2 1
8 5
Young's modulus (Et), MPa × 104
3
4a 6
tio
4 8
ra
7 9
us
ul
od
2 tio
m
ra 4b
gh
us
Hi
l
du
H.
o tio
1 m ra
um us
ed
i ul 10
1 od
0: .M m
0.5 50 M w
Lo
L.
1
0:
0.25 20
75 125 25 50 100 200 400

Uniaxial compressive strength, MPa
Metamorphic: 1. Quartzite, 2. Gneiss, 3. Marble, 4a. Schist, Steep foliation, 4b. Schist, flat
foliation Igneous: 5. Diabase, 6. Granite, 7. Basalt and other flow rocks
Sedimentary: 8. Limestone and dolomite, 9. Sandstone, 10. Shale
Fig. 23.3 Engineering classification of intact rock based on uniaxial compressive strength and mod-
ulus ratio (Source: Deere and Miller, 1966)
joint characteristics (25). The figures in parentheses refer to the maximum rating value of
each parameter. In order to determine the RMR, the individual rating of the parameters
are summed. The RMR values recommended by Bieniawski (1974a, b) are presented in
Table 23.8.
Table 23.8 Geomechanics classification of rock masses
Class Description of rock mass Rock mass rating (RMR)

I Very good rock 81–100
II Good rock 61–80
III Fair rock 41–60
IV Poor rock 21–40
V Very poor rock 0–20
M23_PURU1773_01_SE_C23.indd 796 2/12/2013 8:18:31 AM

23.4 IN SITU STATE OF STRESS

The natural state of a stress at a point within a rock mass depends on the geologic processes
that have acted on the mass. It is impossible to delineate with sufficient accuracy the various
events which have brought about the in situ state of stress. However, it is often possible to
estimate the stresses and their directions. But the margin of error cannot be ensured unless
substantiated by field measurements.
A knowledge of rock stresses has been helpful in several civil engineering works, e.g., in
choosing the orientation for a cavern, in deciding the layout of complex underground works,
in fixing the location of pressure tunnels and penstocks, and in the economical planning for
a large surface excavation.
23.4.1 Vertical Stress

The vertical normal stress can be generally taken as equal to the weight of the overlying rock
and on an average can be taken as 0.027 MPa/m depth (Goodman, 1980). In the near-
horizontal profiles, the principal stress directions are vertical and horizontal and are gener-
ally assumed to be true with depth also. This has been widely accepted in practice. However,
this is not true beneath a valley side wherein one principal stress is normal to the slope and
equals zero while the other two principal stresses lie in the plane of the slope. In situ stresses
approach the strength of rock beneath V-shaped valleys.
The conventional vertical total stress equation
σv = γ z
has been supported by Brown and Hoek (1978) based on stress measurements. Effects of
geological structure could alter the vertical stress and the direction of principal stresses. In
major works, wherever geological heterogeneities can be expected, a thorough investigation
has to be made using numerical methods like finite element method.
23.4.2 Horizontal Stress

The magnitude of horizontal stress (σh) is estimated (as done in soils) using the lateral stress
coefficient (K)
σh = Kσ v
The expression is based on one-dimensional loading of an elastic half-space with zero

horizontal strain. Such a relationship has no bearing in a rock mass since it has experienced
cycles of loading and unloading. However, the horizontal stress may lie between Kaσv and
Kpσv (Fig. 23.4) where Ka and Kp correspond to conditions for normal and reverse faulting.
For no fault condition, the assessment is difficult.
Based on published data on in situ stresses, Brown and Hoek (1978) suggested a hyper-
bolic relation for the limits of K (z) as follows:
100 1500
0.3 + < K < 0.5 + (23.9)
z z
M23_PURU1773_01_SE_C23.indd 797 2/12/2013 8:18:31 AM

sv
Ka s v
(a) Normal fault
sv
Kp s v
(b) Reverse fault
Fig. 23.4 Stresses required to initiate normal and reverse faults (Source: Goodman, 1980)
where K is the ratio of average horizontal stress to vertical stress. It has been observed that
the actual measured data are consistently found to be inverse with depth (Goodman, 1980).
In a broader sense, one can estimate the variation of horizontal stress with depth without
any field measurements.
23.4.3 Effective Stress in Rock Masses

Terzaghi’s effective stress equation for saturated soils is an excellent approximation for soils
although it is unlikely to be an exact statement. Skempton (1961) remarked that it is not nec-
essarily true for rocks which are commonly subjected to very high stresses. Skempton sug-
gested generalized effective stress expressions for saturated rocks in the form
1. For shear strength:
⎛ tan ψ ′ ⎞⎟
σ ′ = σ − ⎜⎜⎜1 − ac ⎟⎟ uw (23.10)
⎜⎝ tan φ ′ ⎟⎠
2. For volume change:

⎛ C ⎞
σ ′ = σ − ⎜⎜⎜1 − s ⎟⎟⎟ uw
⎝ C⎠ (23.11)
where ψ′ is the angle of intrinsic friction of solids, φ′ the angle of shearing resistance, Cs
the compressibility of solid particles, C the volume compressibility, and ac the mineral
contact area ratio.
For saturated rocks, tan ψ′/tan φ′ is in the order of 0.1 to 0.3. This means that even if ac
approaches 0.5 (rocks with very low porosity), Eq. 23.10 reduces to
σ ’ = σ −(0.85 to 0.95)uw (23.12)
M23_PURU1773_01_SE_C23.indd 798 2/12/2013 8:18:32 AM

This shows that Terzaghi’s effective stress equation used in soils is not an unaccept-
ably poor approximation for rocks. Further, if the compressibility of grains is low, Eq. 23.11
reduces to Terzaghi’s equation. Thus, Skempton (1961) concluded that Terzaghi’s effective
stress equation is valid in rocks too if the grains are incompressible and the yield stress of
grains is independent of pressure.
If this is not true, the effective stress can be obtained from Eq. 23.13 (a generalized form
of Eqs. 23.10 and 23.11):
σ ′ = σ − k uw (23.13)
where k is a constant depending on the area of contact between grains and their mechanical
properties (Jaeger and Cook, 1969). Terzaghi (1945) suggested a similar expression for rocks
where k is assigned a value lower than unity for rock material and unity for rock masses
(when the surfaces of joints are considered).
However, Terzaghi’s effective stress law used for soils has been confirmed to be valid for
rocks also [see, e.g., Brace and Byerlee, 1966 (reported by Serafim, 1968); Goodman, 1980;
Hubert and Rubey, 1959].
Based on Bishop’s (1955) concept for effective stress in partially saturated soils, Skempton
(1961) expanded Eqs. 23.10 and 23.11 for partially saturated materials as
1. For shear strength:
⎛ tan ψ ′ ⎞⎟
σ ′ = σ − ⎜⎜⎜1 − ac ⎟⎟ Sχ uw (23.14)
⎜⎝ tan φ ′ ⎟⎠
2. For volume change:
⎛ C ⎞
σ ′ = σ − ⎜⎜⎜1 − s ⎟⎟⎟ Sχ uw (23.15)
⎝ C⎠
where
ua − uw
Sχ = 1 + (1 − χ) (23.16)
uw
As the parameter χ has serious limitations, the application of Eqs. 23.14 and 23.15 for rocks
has not yet been explored.
23.4.4 Measurement of In Situ Stresses

Among several techniques available for measurement of in situ stresses, three techniques,
viz., hydraulic fracturing, flat jack method, and overcoring method, are the best known. All
the techniques involve perturbing the rock in order to create a response that can then be mea-
sured and analysed. In the hydraulic fracturing method, water pressure is used to create a
crack in a borehole. From the knowledge of the tensile strength of the rock and the inferred
concentration of stress, the initial stress in the plane normal to the bore hole is assessed. In the
flat jack method, a slot is cut in the rock and then reloaded with the help of a jack. The in situ
stress is related to the pressure needed to balance the deformation which occurs as a result of
slot cutting. In the overcoring method, a large core sample is drilled, whereby a condition of
stress release is created. The radial displacement or surface strains are monitored in a central
M23_PURU1773_01_SE_C23.indd 799 2/12/2013 8:18:33 AM

parallel borehole. An analytical treatment based on an unloaded thick-walled cylinder model

gives the required in situ stress. The reader may refer to Goodman (1980) for the details of the
techniques.
23.5 MECHANICAL PROPERTIES OF ROCKS

Engineering properties of rocks and rock masses are needed in the analysis of every rock
mechanics problem. This necessitates a knowledge of in situ stresses, modes of failure of
rocks, the stress–strain behaviour of rocks, and the time-dependent behaviour of rocks. With
this knowledge (based on a particular criterion of failure), a solution to the problem can be
arrived at. Methods of assessing in situ stresses have been discussed in the previous chapter.
The other aspects shall be discussed in this chapter.
23.5.1 Modes of Failure of Rock

As the in situ load configuration is complex, no single mode of rock failure can be identified.
Depending on the field situation, the failures may be due to flexure, shear, compression, and
tension. A flexure failure refers to failure by bending followed by tensile cracks. Such a fail-
ure may occur in the layers of a mine roof (Fig. 23.5a). Shear failure refers to rupture along a
surface when the shear stresses have become critical. Slopes cut in weak material, such as
weathered clay, shales, and crushed rock of fault zones exhibit shear failure (Fig. 23.5b).
Crushing or compression failure occurs when high stresses are transferred through short-
ened rocks with less bearing area. A typical example is the failure of a mine pillar due to over
extraction of ore (Fig. 23.5c). Failure due to direct tension occurs in sedimentary rocks on the
flank of an anticline and sometimes in rock layers resting on convex upward slope surfaces
(Fig. 23.5d and e).
23.5.2 Behaviour of Rocks in Uniaxial Compression

Uniaxial or unconfined compression test is the oldest, simplest, and most frequently used
strength test for rocks. In this test, right circular cylinders or prisms of rocks are compressed
parallel to their longitudinal axis. The specimen is compressed between two hard discs or
end platens. The rigidity of the platens prevents the lateral expansion at the ends. The length
to diameter ratio of 2 to 3 (IS: 9221, 1979) and preferably 2.5 is recommended (Dhir and
Sangha, 1973). The diameter of the specimen shall not be less than 10 times the largest min-
eral grain in rock. As per Indian Standards recommendation, a 45 mm diameter specimen is
preferred and in no case the diameter should be less than 35 mm. The load on the specimen
is applied continuously with a stress rate of 0.50 to 1.0 MPa/s (IS: 9143, 1979). The uniaxial
compression (qu) is given as the ratio of peak load (P) to the initial cross-sectional area (A),
that is, qu = P/A. Table 23.9 gives some representative values of qu of some rocks (Jaegar and
Cook, 1969). The stress–strain behaviour of rocks under uniaxial compression is discussed
below.
Wawersik and Fairhurst (1970) explained the stress–strain behaviour of rocks tested
under an unconfined condition. They divided the stress–strain paths between zero stress and
the maximum compressive stress into three regions (Fig. 23.6). Region A shows a steadily
M23_PURU1773_01_SE_C23.indd 800 2/12/2013 8:18:33 AM

(a) Flexure
(d) Direct tension
(b) Shear
(c) Crushing and tensile

cracking followed by shear
(e) Direct tension
Fig. 23.5 Examples of failure models involving breakage of rock (Source: Goodman, 1980)
Table 23.9 Typical values of uniaxial compression, modulus of elasticity,

and Poisson’s ratio for different rocks
Rock type qu (MPa) E (MPa) ν
a a
Granite 229.1 55,890 0.11a
Quartzite 460.2a–200.1d 78,660a–77,970d –
Diabase 487.1a 99,360a –
Marble 152.9b–77.3d 47,610a–64,860d –
Granite 173.2b–587.9c 44,160b–82,800c 0.20c
Shale 171.8c 67,620c 0.23c
Dolerite 331.2d 84,180d –
Sandstone 37.0d 9,660d –
Limestone 224.3d 53,130d –
a
Brace (1964) and Walsh (1965).
b
Cook (1965).
c
Cook et al. (1966).
d
Wiebols et al. (1968).
Source: Jaeger and Cook (1969).
M23_PURU1773_01_SE_C23.indd 801 2/12/2013 8:18:34 AM

Class I
Stress B
Class II
Strain
Fig. 23.6 Three regions of the stress–strain curve (Source: Wawersik and Fairhurst, 1970)
increasing slope which is a typical one for highly porous rocks and those containing numer-
ous small fractures.
The stress–strain path over region B is almost linear, which is a characteristic one for
brittle rocks. Over region C, the slope of the curve decreases, depicting dissipation of energy.
Finding notable differences in the post-peak region, they identified two classes of failure.
In Class I, the failure propagation is stable, i.e., such rocks can retain some strength even
after the peak strength. In Class II, the failure is unstable, that is, the rock has lost virtually
all its strength. The fracture of such rocks cannot be controlled. Stages in the development of
fracturing with increasing uniaxial compressive stress is shown in Fig. 23.7.
Based on a comprehensive study on 28 rock types, Deere and Miller (1966) classified the
uniaxial stress–strain curves into six types (Fig. 23.8). The Type I curve represents a nearly
straight-line behaviour of a brittle material until a sudden explosive failure occurs. Such
failures are typical of basalts, dolerites, quartzites, and strong limestones. Softer limestones,
siltstones, and tuff exhibit a continuous increase in elastic yield, which is characterized by a
Type II curve. The stress–strain curve represented as Type III is typical of sandstone, granite,
some dolostones and dolerites, and schist scored parallel to foliation. The S-shaped curve
with an extremely stiff central portion is identified as Type IV which is typical of meta-
morphic rocks such as marbles gneiss. Schist scored perpendicular to foliation has the long
Fig. 23.7 Stages in the development of fracturing with increasing unconfined compressive stress
(Source: Wawersik and Fairhurst, 1970)
M23_PURU1773_01_SE_C23.indd 802 2/12/2013 8:18:34 AM

Type I elastic Type II elastic–plastic
s
e Basalt e Siltstone
Type III plastic–elastic Type IV
plastic–elastic–plastic
s
e Sandstone s e Marble
Type V Type VI
plastic–elastic–plastic elastic–plastic–creep
s
e Schist e Rock salt
Fig. 23.8 Typical stress–strain curves for rock in uniaxial compression loaded to failure (Source:
Stagg and Zeinkiewicz, 1968)
sweeping S-shaped curve of Type V. The Type VI curve, which is characteristic of rock salt,
has an initial straight-line portion, followed by an inelastic deformation and continuous
creep. Types III, IV, and V curves are all characterized by initial concave upward portion
followed by a steeper linear section as cracks close up. The upper part of these curves shows
varying degrees of inelastic yielding as failure is approached.
23.5.3 Tensile Strength of Rocks

Direct measurement of tensile strength has to be obtained by subjecting a cylindrical speci-
men to direct extension. Suitable devices to hold the specimen without causing stress con-
centrations or bending stresses are yet to be developed. Alternatively, the ends of the
specimen may be cemented with epoxy resin direct to steel end-platens (Fairhurst, 1961).
Specially shaped specimens have also been used (Brace, 1964).
The tensile strength of rocks is lower than their compressive strength. Further, an accurate
determination of tensile strength is not generally warranted. Thus, the tensile strength of a
rock is usually measured by rapid indirect means (IS: 10082, 1981).
M23_PURU1773_01_SE_C23.indd 803 2/12/2013 8:18:34 AM

The Brazilian test, suggested for cylindrical concrete specimens, is a convenient one. In
this test, a cylindrical rock specimen of thickness (t) and diameter (D) is loaded in a diametri-
cal plane with a load P (Fig. 23.9a). The specimen fails by splitting along the diametrical
loading and the tensile strength (sBt) is given as
P P
P P
Desirable mode of failure Undesirable mode of failure
P (a) Brazilian test P
1
D
10
P P
D
Undesirable mode of
Desirable mode of
failure
failure
(b) Ring test
Total P Steel packing over

load double layer of
adhesive paper strip
D′
Effective
P area
(c) Line load test
(d) Flexure test
Fig. 23.9 Tests for tensile strength of rocks [Source (for a, b, and c): IS: 10082, 1981]
M23_PURU1773_01_SE_C23.indd 804 2/12/2013 8:18:38 AM

2P
σBt = (23.17)
πDt
As per Indian Standards (IS: 10082, 1981), the specimen diameter shall not be less than
45 mm and the thickness shall be approximately equal to half the diameter. A loading rate
of 200 N/s is recommended, and the load has to be measured with 1% accuracy.
It has been reported that the Brazilian test yields higher tensile strength than the direct
tension tests. The extra strength has been attributed to the effect of fissures (Tourenq and
Denis, 1970, reported by Goodman, 1980).
In the ring test, which has come into vogue recently (IS: 10082, 1981), a hollow specimen is
tested (Fig. 23.9b). The testing apparatus and procedure are same as in the Brazilian test. As
the central hole acts as a stress raiser, the vertical splitting increases considerably. Further, the
cracks simultaneously initiate from the top and bottom of the hole and propagate towards
loading points. The tensile strength is obtained from the expression
2P[(6 + 38 (D0 / Di )2 ]
σri = (23.18)
πDi t
The ring test is considered to be an appropriate test as it is based on fracture mechanics
consideration.
Another indirect test recommended by Indian Standards is the line load test. In this test, a
rock prism is loaded along a line with point loads (Fig. 23.9c). The rate of loading is 200 N/s
and the tensile strength is given as
P
σlt = (23.19)
πD ′t
where D′ is the square base in millimetres. Equation 23.19 is applicable when the specimen
rests directly on the lower platen. This value is increased by 1.96 times if the specimen is
supported on a packing of steel bar.
The flexural test is another indirect tensile strength test. In this test, a cylindrical speci-
men of rock is loaded at three points with the bottom of the core supported on points near
the ends (Fig. 23.9d). If P is the failure load, L the length between the supports, and D the
diameter of the specimen, then the flexural tensile strength σft is given as
8 PL
σft = (23.20)
πD3
The flexural test also gives higher tensile strength than does the direct tension test.
The tensile strength test from point load has been discussed earlier. The point load test is
the only test suitable for field testing of cores (IS: 8764, 1978). As discussed earlier, the test
result can be used for evaluating the uniaxial compressive strength of unweathered rock
from the expression
qu = 22 σ(pl)50 (23.21)
Based on the brittle failure theory, the ratio of compressive strength to tensile strength is
taken as 8:1, but in practice the ratio is found to be 15:1 to 25:1 (Bell, 1983). Hendron (1968)
recommends a tensile strength of 5% to 10% of uniaxial compression strength for most rock
mechanics problems.
M23_PURU1773_01_SE_C23.indd 805 2/12/2013 8:18:38 AM

23.5.4 Behaviour of Rocks in Triaxial Compression

The mechanical properties of rocks can be better understood from the results of a triaxial
compression. In principle, the triaxial apparatus for a rock specimen is similar to that
used for soil specimen. The major difference is the magnitude of lateral pressure and the
application of deviatoric stress which demand special type of equipment. The rock
specimen is enclosed in an impervious jacket before applying the confining pressure.
Hydraulic oil is used as the confining fluid and the jacket is an oil-resistant rubber
(e.g., polyurethane). For a short duration test, a bicycle inner tube may be used. Many
varieties of triaxial cells are available in different rock mechanics laboratories. The reader
may refer to Hoek and Franklin (1968) and Vogler and Kovari (1978) to know the details
of equipments and testing.
At the maximum load stage, the stress conditions are σ1 = P/A and σ3 = p, where P is
the maximum load along the axis of the specimen and p is the pressure in the confining
medium.
Most rocks are appreciably strengthened due to confinement. This is especially pro-
nounced in highly fissured rocks. The effect of confining pressure on Berea sandstone
and gabbro (after Serdengecti and Boozer, 1961; reported by Hendron, 1968) is shown
in Fig. 23.10. An increase in the confining pressure causes an increase in the peak and
residual strength and strain to failure. Further, the inelastic strain before failure increases
with confining pressure in sandstones (Fig. 23.10a).
With increase in the confining pressure, the initial tangent modulus also increases in
weaker rocks but the effect has been reported to be less in strong hard rocks (Hendron,
1968). But in gabbro, the strength increases and the initial elastic properties are not changed
by the increase in the confining pressure (Fig. 23.10b). Further, in gabbro the increase in
the confining pressure did not improve the inelastic action and ductile behaviour before
ultimate failure. As for soils, the triaxial compression results are represented by Mohr’s
circles.
42.0
s 3 = 140,000 kN /m2
(s 1 – s 3), kN / m2 × 104
(s 1 – s 3), kN / m2 × 105
Stress difference
31.5 10.5
Stress difference
s 3 = 140,000
kN /m2
21.0 7.0
s 3 = 35,000 kN /m2 s 3 = 70,000
kN /m2
10.5 3.5
s3 = 0 s 3 = 35,000 kN /m2
0 0
0 1.0 2.0 3.0 4.0 0 10 20 30 40

(a) Berea sandstone (b) Gabbro
Fig. 23.10 Variation of stress–strain and strength properties (Source: Hendron, 1968)
M23_PURU1773_01_SE_C23.indd 806 2/12/2013 8:18:39 AM

23.5.5 Failure Theory Applicable to Rocks

The Mohr–Coulomb failure criterion is the simplest well-known criterion of failure for rocks
(Goodman, 1980). This criterion is applicable for both peak and residual strength of rocks. For
peak shear stresses, this is shown in Fig. 23.11 and stated in terms of normal and shear stresses as
τ p = si + σn tan φ (23.22)
where τp is the peak shear stress or shear strength of rock, φ the angle of internal friction, σn
the normal stress on the shear plane, and si the shear strength intercept.
The physical interpretation for Eq. 23.22 is that when the difference between applied
shear stress (τp) and the frictional resistance associated with the normal stress on the failure
plane (σn tan φ) becomes equal to a constant si, failure is said to occur. But for a tensile nor-
mal stress, it is not justifiable to accept frictional resistance. However, the minimum σ3 may
be tensile as long as σn is compressive. Hence, it is justifiable to extend the failure envelope
up to the tensile region where σ3 becomes equal to the uniaxial tensile strength −σt and the
minor principal stress can never become less than −σt.
Although the Mohr–Coulomb criterion is easy to work, a more accurate failure criterion
can be obtained for rocks by fitting in an envelope to Mohr circles corresponding to stresses
at failure (Fig. 23.12). Such an envelope is generally curved downwards. In most of the rocks,
the failure envelope lies between a straight line and a parabola (Hoek, 1968; Jaegar and Cook,
1969). In practice, depending on the rock type, an empirical curve fitting may be resorted to
without substantially violating the Mohr–Coulomb criterion (Eq. 23.25). For a straight-line
failure condition, σ1 and σ3 are related as
⎛ φ⎞ ⎛ φ⎞
σ1 = 2si tan ⎜⎜⎜ 45° + ⎟⎟⎟ + σ3 tan 2 ⎜⎜⎜ 45° + ⎟⎟⎟ (23.23)
⎝ 2 ⎠ ⎝ 2⎠
For an uniaxial compression, σ3 = 0 and σ1 = qu, then
⎛ φ⎞
qu = 2si tan ⎜⎜⎜ 45° + ⎟⎟⎟ (23.24)
⎝ 2⎠
Mohr–Coulomb failure envelope
t
f
Tension
cut-off
S1
- T0 s3 s3 s3 s1 s1 s1
Fig. 23.11 Mohr–Coulomb failure criterion with a tension cut-off
M23_PURU1773_01_SE_C23.indd 807 2/12/2013 8:18:40 AM

t Empirical failure envelope
C D
Fig. 23.12 An empirical criterion of failure: A, direct tension; B, Brazilian; C, unconfined; D, triaxial
compression
Hence, Eq. 23.23 reduces to

⎛ φ⎞
σ1 = qu + σ3 tan 2 ⎜⎜⎜ 45° + ⎟⎟⎟ (23.24a)
⎝ 2⎠
Dividing both sides of Eq. (23.24a) by qu, we have
σ1 ⎛ φ ⎞⎛ σ ⎞
= 1 + tan 2 ⎜⎜⎜ 45° + ⎟⎟⎟⎜⎜⎜ 3 ⎟⎟⎟ (23.24b)
qu ⎝ 2 ⎠⎜⎝ qu ⎟⎠
Equation 23.24b is a non-dimensional one and has been generalized by Bieniawski (1974a,
b), by a power law to include the tension cut-off σ3 = σt, as
σ1 ⎛ σ3 ⎞⎟M
= 1 + N ⎜⎜⎜ ⎟⎟ (23.25)
qu ⎜⎝ qu ⎟⎠
The constants N and M have to be determined by fitting a curve to the family of points
⎡⎛ σ ⎞ ⎛ σ ⎞⎤
⎢⎜⎜ 3 ⎟⎟ , ⎜⎜ 1 − 1⎟⎟⎥
⎢⎜⎜⎝ q ⎟⎟⎠ ⎜⎜⎝ q ⎟⎟⎠⎥⎥
⎢⎣ u u ⎦
23.5.6 Shear Strength of Rocks

The shear strength of rocks is reasonably found from triaxial compression and adopting
the failure criterion as discussed above. The above discussion has been made keeping in
view the peak strength parameters. Accordingly, si and φ are the peak shear strength
parameters.
It has been reported that the Mohr–Coulomb envelope is also applicable for residual con-
dition. As discussed in Chapter 7 for soils, the shear strength intercept for rocks at residual
M23_PURU1773_01_SE_C23.indd 808 2/12/2013 8:18:41 AM

strength may also approach zero (i.e., si = 0) and the residual friction angle (φr) may lie
between zero and peak friction angle.
The presence of water in rocks reduces their strength. Some rocks on saturation weaken
due to chemical deterioration of cement or clay binder. But the major reduction in strength
is due to the water pressure in pores and fissures. In such situations, effective stresses have
to be considered while defining the shear strength parameters.
Several unconfined shear test procedures, such as single, double, punch, and torsional
shear tests, are available for rocks. In all these shear tests, the results depend on the testing
technique and the rock material. A schematic testing arrangement and the relevant formula
are presented by Bell (1983). It is recommended by Goodman (1980) that the shear strength
of rocks should be assessed only from the triaxial compression test.
23.5.7 Elastic Properties of Rocks

The elastic behaviour of a material depends on its homogeneity, isotropy, and continuity.
Homogeneity in rocks means the uniform distribution of constituent particles and isotropy, in
terms of a rock, can be defined as a measure of its directional properties. Continuity refers to
the amount of fracture and pore space in a particular mass of rock. Thus elastic properties of
rocks depend on the size of the rock mass, its constituent minerals, and the degree of fracture.
Modulus of elasticity is the most important elastic property of a rock. Crystalline rocks
have an S-shaped stress–strain curve (Fig. 23.13), and at low stress level the modulus of elas-
ticity increases with the increase in stress. The initial tangent modulus, the tangent modulus
and the secant modulus for a rock are shown in Fig. 23.13. Dhir and Sangha (1973) recom-
mended the 50% tangent modulus (the tangent modulus at half the failure load) provides
the best value.
Farmer (1968) distinguished three types of elastic behaviour in rocks, viz., quasi-elastic,
semi-elastic, and non-elastic. The initial tangent modulus ranges from 6 × 104 to 11 × 104 MPa
in quasi-elastic rocks, between 4 × 104 and 7 × 104 MPa in semi-elastic rocks, and it is less than
5 × 104 MPa in non-elastic rocks. The quasi-elastic rocks are fine-grained, compact, and mas-
sive and show a near linear stress–strain curve up to failure (Fig. 23.13a). Semi-elastic rocks
are coarse-grained igneous rocks and fine-grained compacted sediments, and the stress–strain
Stress
Stress
Stress
Ei = (6 to 11) × Ei = (4 to 7) × E < 5 × 104

104 MPa 104 MPa MPa
Strain Strain Strain
(a) Quasi-elastic (b) Semi-elastic (c) Non-elastic
Fig. 23.13 Typical stress–strain relationships for rocks (Source: Farmer, 1968)
M23_PURU1773_01_SE_C23.indd 809 2/12/2013 8:18:41 AM

curve shows a slope decrease with increasing stress (Fig. 23.13b). Less cohesive rocks with
large pore space fall under the third category, whose stress–strain curve is shown in Fig. 23.13c.
Another important elastic property of rocks is the Poisson’s ratio (ν). Deere and Miller
(1966), based on the studies on rocks, showed that ν varied between 0.125 and 0.341. In rocks
subjected to uniaxial compression, ν remains more or less constant with increase in stress,
and ν reaches the theoretical maximum of 0.50 at the failure range. The trend is opposite in
rocks subjected to uniaxial tension. The ν values of some rocks are given in Table 23.9.
The laboratory uniaxial compression test is itself used to obtain circumferential and axial
deformations or strains. The data needed to evaluate them are obtained by electrical resis-
tance strain gauges, compressometers, optical devices, or other suitable means. While choos-
ing the measuring devices, it should be noted that at least two circumferential and two axial
strain measurements can be found for each increment of load (IS: 9221, 1979). The gauge
length for strain measurements should be at least five times the grain-size diameter. The
axial strain (εa) and the diametric strain (εd) are then calculated. Thus, Poisson’s ratio is
εd
ν=
εa
The modulus of elasticity E may be calculated from a stress–strain curve using any one of
the methods explained earlier (Fig. 23.14). It is determined in the field by adopting uniaxial
jacking test (IS: 7317, 1974).
23.5.8 Hardness
Hardness of a mineral is defined based on its capability to resist scratching and relative hard-
ness has been used as the index for identification. For this, Mohr’s hardness scale has been
in use. But in rocks, the condition is much more complex and no clear definition is yet avail-
able. Deere and Miller (1966) defined rock hardness as a combined resistance to displace-
ment and penetration of particles. Further, they noted that both hardness and toughness
depend on the same factors.
Tangent modulus, Et
P
B
Stress, s
S
Secant
modulus
Es
Initial tangent modulus, Et
O R Strain
Fig. 23.14 Representative stress–strain curve for rock in uniaxial compression
M23_PURU1773_01_SE_C23.indd 810 2/12/2013 8:18:42 AM

Average dispersion
strength for most rocks, MPa 32
30
Unit weight (g), kN/m3

350 29
±200
300 28
Uniaxial compressive strength, MPa
250 26
±100
200 24
±80
150 ±60 22
21
100 ±40 20
90
80
70
60
50
40
30
±20
20
Hammer Vertical Downwards
15
10
0 10 20 30 40 50 60
Schmidt hardness (R) L-hammer
Fig. 23.15 Correlation chart for Schmidt (L) hammer relating unit weight of rock, compressive
strength, and rebound number (Source: Deere and Miller, 1966)
Thus attempts were made to determine the hardness of rock either from the measured
indentation (Brinell, Rockwell, or Vickers) or from the rebound tests (Shores or Schmidt).
Deere and Miller (1966) used both Shores’s scleroscope and Schmidt rebound hammer and
presented correlation charts with uniaxial compressive strength. Figure 23.15 shows a cor-
relation chart for Schmidt (L) hammer. It is possible to use this hardness number and the dry
unit weight for predicting the uniaxial compressive strength within 75% confidence limits
(Hendron, 1968).
POINTS TO REMEMBER
23.1 Rock mechanics is an interdisciplinary subject which covers various fields like geol-
ogy, mining, petroleum, and civil engineering.
23.2 In order to identify the quality of a rock, certain properties like porosity, density, per-
meability, strength, durability, and sonic velocity have been grouped and designated
as index properties.
M23_PURU1773_01_SE_C23.indd 811 2/12/2013 8:18:42 AM

23.3 Porosity, density, and permeability are as defined in soils and have the same charac-
teristics and significance as in soils.
23.4 The maximum point load obtained from a test on rock prism is designated as the
strength index of a rock.
23.5 The results from a slake durability test, called the slake durability index, represent
the degree of durability of rocks.
23.6 Comparison of in situ sonic wave velocity with intact core helps in identifying dis-
continuities in rock.
23.7 Classification of rocks should consider lithology in addition to conventional geo-
logical classification. Such a classification divides the rocks as (i) crystalline texture
rocks, (ii) clastic texture rocks, (iii) very-fine-grained rocks, and (iv) organic rocks.
23.8 State of stress at a point in a rock mass depends on the geologic processes that have
acted on the mass. Knowledge of rock stresses is helpful in several civil engineering
works.
23.9 Skempton suggested generalized effective stress expressions for saturated rocks for
shear strength and volume change separately.
23.10 Hydraulic fracturing method, flat-jack method, and overcoring method are the best
known methods for measurement of in situ stresses.
23.11 As the in situ load configuration is complex, no single mode of rock failure can be
identified. Based on the ground condition, the failures may be due to flexure, shear,
compression, and tension.
23.12 Uniaxial or unconfined compression test is the oldest, simplest, and most frequently
used strength test for rocks.
23.13 The Brazilian test is the convenient test to determine the tensile strength of rocks.
23.14 The Mohr–Coulomb failure criterion is the simplest well-known criterion of failure
for rocks and applicable both for peak and residual strengths.
23.15 Shear strength of rocks is reasonably found from triaxial compression and adopting
the Mohr–Coulomb failure criterion.
23.16 Modulus of elasticity is the most important elastic property of rock. The tangent
modulus determined at half of failure load provides the best value.
23.17 Rock hardness is a combination of resistance to displacement and penetration of
particles.
QUESTIONS
Objectives Questions
1. In sedimentary rocks, the porosity varies from close to zero to as much as 90%.
2. Uniaxial compressive strength is identified as a strength index test.
3. Most rocks are significantly strengthened by confinement.
4. Engineering classification of intact rocks is based on the modulus of elasticity and
Poisson’s ratio of the rocks.
5. Terzaghi’s effective stress law used for soils is not valid for rocks.
M23_PURU1773_01_SE_C23.indd 812 2/12/2013 8:18:42 AM

23.2 Pororsity in a rock is found to be maximum when the grains are

(a) Of the same size (b) Of varying sizes
(c) In a specified size range (d) Arranged in a parallel array
23.3 The rate of slaking of a rock is governed by the change in
(a) Plasticity index (b) Flow index
(c) Liquidity index (d) Toughness index
23.4 The average uniaxial compressive strength of an intact rock is found to be 85 MPa.
Based on this value, the rock may be classified as
(a) Low strength (b) High strength
(c) Very high strength (d) Medium strength
23.5 In a ring test on a hollow specimen for tensile strength determination, cracks initiate
(a) From the top and bottom and propagate towards the centre
(b) From the top only immediately below the loading
(c) From the top and bottom of the hole and propagate towards loading points
(d) From the bottom only and propagate towards the hole
23.6 The pores of a rock are filled with mercury by injecting it at high pressure. Derive an
expression for porosity in terms of the volume of mercury filling the pores, specific
gravity of component grains, and density of water.
23.7 Explain the factors which affect the shear strength of a rock mass? Why cannot the
strength of a core sample be used for the mass?
23.8 How do modes of failure of rock specimens vary with effective confining pressure?
Illustrate your answer by stress–strain curves.
23.9 For a three-point bending test of a core sample with circular cross section, derive an
expression for the modulus of rupture.
23.10 In a homogeneous rock mass, how does erosion of top layers affect the ratio of hori-
zontal stress to vertical stress?
EXERCISE PROBLEMS
23.1 A sandstone is composed of a mixture of 73% quartz and 27% pyrite grains with a poros-
ity of 16%. Determine the dry density of the sandstone if the specific gravity of quartz and
pyrite grains are 2.65 and 5.0, respectively (Hint: Specific gravity, G, of the solid part of a
rock is given as
m
G = ∑ G jVj
j=1
where Gj is the specific gravity and Vj is its volume percentage of component j).
23.2 A sandstone core, 85 mm in diameter and 172 mm long, consists of quartz and feld-
spar grains. The sample was saturated with water and found to weigh 22.3 N when
M23_PURU1773_01_SE_C23.indd 813 2/12/2013 8:18:42 AM

wet and 21.1 N after oven drying. Determine the wet and dry unit weights and its
porosity.
23.3 Triaxial shear tests conducted on samples of rock revealed the shear strength param-
eters as si = 12,600 kN/m2 and φ = 38°. An unlined tunnel is made in the rock where
the confining pressure was zero and the water pressure was 850 kN/m2. Under such
a condition, what will be the compressive strength of the rock?
23.4 The triaxial strength of the rock specimen may be approximated by the following
expression:
qf ⎛ σ ′ ⎞n
= 0.1 + k ⎜⎜⎜ m ⎟⎟⎟
qc ⎜⎝ qc ⎟⎠
′
where qf is the maximum shear stress, σm the effective mean normal stress, qc the uni-
axial compressive strength, k = 0.78, and n = 0.90.
Plot Mohr’s failure envelope and estimate shear strength parameters for a range of
effective normal stresses of 250 to 500 kN/m2. The uniaxial compressive strength may
be taken as 2,000 kN/m2.
23.5 A porous rock under triaxial test conditions yielded the shear strength intercept as
1.2 MPa and the angle of internal friction as 33°. Compute the uniaxial compressive
strength and estimate the tensile strength of this rock.
M23_PURU1773_01_SE_C23.indd 814 2/12/2013 8:18:42 AM

24
Pavements
CHAPTER HIGHLIGHTS
Components of pavements; Types and requirements; subgrade; pavement
design – Wheel load – Flexible and rigid pavement; Methods – Group
Index – CBR – Mc Leod – Burmister – Westergaard’s – IRC
24.1 INTRODUCTION
A pavement is the main structure of a road or a runway, which is relatively a stable layer
or a crust constructed over the natural soil. The natural soil is called a subgrade. The
main function of a pavement is to support and distribute the wheel loads of vehicles
through the underlying layer to the subgrade.
24.2 COMPONENTS OF PAVEMENT

Subgrade is the bottom-most level of a pavement, which is termed as the formation level
over which other layer are laid, namely, sub-base, base and surface courses (Figure 24.1).
Surface course
Pavement
Base course
Formation
Sub-base course
level
Soil subgrade
Fig. 24.1 Components of a pavement
M24_PURU1773_01_SE_C24.indd 815 2/12/2013 8:18:02 AM

As heavy load of vehicles has to be carried, surface layer is the strongest layer provided
at the top. The courses of decreasing strength are provided in order, such that the load is dis-
tributed smoothly on to the subgrade.
Sub-base, base and surfacing courses are also called soling, metalling and wearing
courses, respectively. Sometimes a strong subgrade is available the sub-base may be
omitted.
24.3 TYPES OF PAVEMENT

Pavements may be classified into two major categories, namely, flexible pavement and rigid
pavement. The basic difference between these two types is the way by which they distribute
the load over their subgrade.
The design of flexible pavement is based on the principle that the wheel loads of vehi-
cles are distributed to the natural soil through the successive layers of granular mate-
rials. The intensity of load decreases with depth as the area of dissipation is increased.
The strength of the subgrade decides the thickness of the flexible pavement. Water-bound
Macadam (WBM) roads, gravel roads, earth roads, soil-stabilized roads, bitumen roads
etc., consisting of layers of road-making materials compacted to form an elastic bed, are
grouped under flexible pavements.
The design of rigid pavement is based on the strength of the structural slab, which tends
to distribute the load over a wide area of the soil. The pavement slab is of Portland cement
concrete, which has high rigidity that resists the deformation of the surface. For this reason,
minor variation in subgrade strength does not affect the structural capacity and the perfor-
mance of the rigid pavement.
A third category of pavement in between these two may be termed as semi-rigid. Cement
grouted, lean cement concrete, soil–cement pavement etc. may fall under this group.
24.4 REQUIREMENTS OF PAVEMENT COMPONENTS

Selection of pavement components for the selected pavement such as type of pavement, base
course, sub-base and surface course is governed by the following factors:
1. Type of subgrade soil.
2. Drainage conditions.
3. Type and volume of traffic.
4. Availability of construction material within a reasonable distance from the site.
5. Availability of plant and equipment.
6. Climatic condition.
7. Altitude at which the pavement is proposed.
8. Fund availability for construction and maintenance.
However, of the above factors, the important factor is the traffic volume or intensity of
traffic. Each pavement has its own life space depending on the thickness of pavement sec-
tion, type of soil subgrade and the traffic volume.
M24_PURU1773_01_SE_C24.indd 816 2/12/2013 8:18:02 AM

Pavements 817
24.5 SUBGRADE
24.5.1 Preparation
Preparation of subgrade includes all operations before the construction of a pavement struc-
ture. Depending on the topography, the subgrade may be situated on the existing ground
level or as embankment or as cutting.
In all cases, the side clearance is the first aspect. The site should be cleared off of the top
soil, grass, roots, rubbish and other organic matters.
Next the grading operation is reworked so as to bring the vertical profile of the subgrade
to the designed grade and camber. Necessary equipment may be used for removing the
boulder. Levelling and finishing it is essential that the subgrade in compact to the standard
density at an appropriate moisture content before the construction of a pavement structure.
24.5.2 Embankment and Cutting

Embankment is constructed when the grade line of a highway is required to be raised above
the existing ground level. Necessity of raising the ground level for the construction of high-
way may aid due to the following reasons:
1. In order to make the subgrade level high above the ground-water level.
2. In order to prevent the capillary water to enter the subgrade by capillary action or surface
water through permeation.
3. In order to satisfy the vertical alignment of the designed highway.
The design and construction of embankment comprises of:
1. deciding the height of the embankment,
2. finalizing side slopes and their short- and long-term stability,
3. embankment material,
4. compaction requirement and compacting with suitable equipment and
5. check for settlement and overall stability
Figure 24.2 shows the cross-section of a typical embankment
Top width
Shoulder Highway
Side slope Side slope

embankment
1:n 1:n
Height of
Embankment
foundation Levelled
natural
ground
Fig. 24.2 Details of highway embankment
M24_PURU1773_01_SE_C24.indd 817 2/12/2013 8:18:02 AM

Cut slope Cut slope
Side drain Highway

Shoulder
Fig. 24.3 Details of highway in cutting
Sometimes the grade of natural ground may not be as per design requirement, then it has
to be properly graded by cutting. In this case, apart from the preparation of subgrade, the
stability of side slopes and drainage arrangements are to be made.
Figure 24.3 shows a highway in cutting. Stability of cuts may be made by adopting the
following measures:
1. By maintaining adequate flat slopes.
2. By consolidating the cut slopes.
3. By providing steps at suitable intervals as the sloping cut edge.
4. By constructing breast walls against cut slopes.
5. By developing grass turf.
24.6 PAVEMENT DESIGN

Pavement design comprises of two parts, namely, (i) mix design of materials to be used in
each component layer and (ii) thickness design of the pavement and the component layer.
Depending on the type of materials to be used, the load and other conditions, the mix design
is made.
Various factors to be considered for the design of pavements are:
1. Design wheel load.
2. Properties of subgrade and pavement materials.
3. Other factors.
24.6.1 Design Wheel Load

Thickness design of pavement primarily depends upon the design wheel load. The design
wheel load depends on the (i) maximum wheel load, (ii) contact pressure, (iii) equivalent
single-wheel load and (iv) repetition of loads.
In order to know the way by which load is distributed, the wheel-load configuration
should be known. Indian Roads Congress (IRC) has specified the maximum legal load as
8,170 kg and maximum single-wheel load as 4,085 kg. The configuration or shape of wheel
load gives an idea how the load of a vehicle acts on the surface of the pavement. Figure 24.4
shows the typical wheel-load configuration of a tractor-trailer unit.
M24_PURU1773_01_SE_C24.indd 818 2/12/2013 8:18:03 AM

Pavements 819
Tandem axle
Single axle
Tractor Trailer
Fig. 24.4 Wheel configuration of tractor trailer unit
While total load influencing the thickness of the pavement, the tyre pressure influences
the quality of surface course. The wheel load is assumed to be distributed over a circular
area. This distributed load is the tyre pressure, which is differently referred to as inflation
pressure or contact pressure. Contract pressure, p, is given by the relationship
Contact pressure, p = hA
where h = load on wheel and
A = contact area or area of imprint.
The contact pressure is focused to be more than the tyre pressure when the tyre pressure
is less than 7 kg/cm2.
In order to have a maximum wheel load, dual wheel assembly is provided to the rear
axles of the load vehicles. Because of this, the load die to both wheels are not to be transfer
led to the pavement. However, there will be overlap pressure after a certain depth. The
actual effect is in between a single-wheel load and a double-wheel load, which is called the
equivalent single-wheel load (ESWL).
The deformation of a pavement or subgrade may be small due to the application of a
single-wheel load. As the pavement experiences repeated loads, the accumulated defor-
mation at any time may be freely elastic or partially plastic. This plastic deformation may
increase with more repetition and ultimately may result in pavement failure. For design
purposes, it is necessary to convert various wheel loads to a single-wheel load. Equivalent
wheel loads accounting for repetition of loads are those which require save thickness and
strength of pavement.
24.6.2 Properties of Subgrade and Pavement Components

The two strength characteristics of subgrade or pavement components are California bear-
ing ratio (CBR) and elastic modulus.
CBR has been used for the design of flexible and rigid pavements. It is an ad hoc property
of a subgrade and other materials, which shows relative significance and do not provide
absolute measure.
Elastic modulus of subgrade soil are determined using the results from a plate-bearing
test. Burmister’s elastic two-layer systems can be used to compute elastic modulus of pave-
ment materials for maximum deformation.
M24_PURU1773_01_SE_C24.indd 819 2/12/2013 8:18:03 AM

24.6.3 Other Factors

Other factors, which may influence the design of pavements, include variation in moisture
condition, frost action and variation in temperature.
24.7 DESIGN OF FLEXIBLE PAVEMENTS

Flexible pavements consist of different thickness of layers of different materials. Due to the
wheel loads, the maximum intensity of stress occurs in the top layer of the pavement. The
magnitude of the stress decreases with depth. Hence, superior quality of materials is used in
the top layers followed by less quality materials. However, in the design process, it is the
basic requirement that under the application of the load none of the layer is overstressed.
That is, no section of the pavement undergoes excessive deformation, which may result in
localized depression or settlement.
There is no rational design method of flexible pavement. In recent years, there has been
a considerable shift in the design concepts of flexible pavements. The flexible pavement
design methods may be classified under three distinct groups.
Empirical Methods: These are based on soil classification, physical or strength para-
meters of soils and other factors such as climate and moisture.
Sub-empirical Methods: These are based on stress–strain functions, performance and
serviceability concepts.
Theoretical Methods: These are based on theoretical analysis and mathematical compu-
tations, considering stress–strain behaviour of soils.
Some of the methods, which are in use, are discussed below.
24.7.1 Group Index Method

This method of flexible pavement design is based on the soil properties such as (i) particle-
size distribution, (ii) liquid limit and (iii) plasticity index.
Group index is claimed to be an inverse measure of the thickness of pavement. The higher
is the group index, the lower is the strength and greater is the thickness requirement for
a given traffic volume. According to this method, subgrades have been divided into five
categories based on group index (Table 24.1).
Table 24.1 Subgrade quality and group index
Subgrade quality Value of group index

Excellent 0
Good 0–1
Fair 2–4
Poor 5–9
Very poor 10–20
M24_PURU1773_01_SE_C24.indd 820 2/12/2013 8:18:03 AM

Pavements 821
Table 24.2 Classification of traffic
Total volume of traffic Number of vehicles

(commercial vehicles) per day
Height Less than 50
Medium 50–300
Heavy Over 300
Traffic is divided into three groups as given in Table 24.2.

In order to design the pavement thickness by this method, the following steps are fol-
lowed:
Step I: The value of the Group Index is found using the details of the soil.
Step II: The anticipated traffic is estimated and classified as light, medium or heavy
as indicated in Fig. 24.5 (a).
Step III: The appropriate design curve is chosen from Fig. 24.5 (b) and the total thick-
ness of pavement (surface, base and sub-base courses) is found for the
corresponding Group Index.
Step IV: The thickness of sub-base or base only is found from appropriate curve to
find the thickness.
As discussed earlier, the method is only empirical, which considers only soil properties
and not the strength and compressibility characteristics of the soil. These curves were pre-
sented as only approximate and may be modified by the engineer concerned to suit the local
conditions based on his/her experience.
24.7.2 California Bearing Ratio (CBR) Method

California Bearing Ratio method was made during the Second World War on the basis of
survey of the existing methods of pavement design performed by Corps of Engineers, USA.
The method was adopted for the design of military airport pavement. Later, California State
Highway Department carried out CBR tests on the there existing pavement layers including
subgrade, sub-base and base courses. Based on the exhaustive study on the pavements,
which have performed well and those which have failed, an empirical design chart was
developed correlating the thickness of the pavement and the CBR value.
This design chart simply specifies that a material with a particular CBR requires a certain
thickness of pavement layer as a cover for a given traffic load. It is evident that a higher load
needs a thicker pavement layer.
Primarily, the design charts (Figure 24.6) were prepared by California State Highway
Department for wheel loads of 3,175 kg and 5,443 kg, representing light and heavy traffic.
Subsequently, the design curve for medium traffic of 4,082 kg wheel load was obtained by
interpolation.
M24_PURU1773_01_SE_C24.indd 821 2/12/2013 8:18:03 AM

General Group Daily volume of commercial traffic

evaluation index range Light Medium Heavy (More
of subgrade of subgrade (Less than 50) (50 to 300) than 300)
30 cm
Surface and base
thickness vary with
30 cm volume of truck traffic
Excellent 20.5 cm
(A-1-a) 10 cm
15 cm
Good 0–1 0
Select sub-base
Fair 2–4 thickness, vary
10 cm 10 cm 10 cm
with subgrade
Poor 5–9 characteristics
20 cm 20 cm 20 cm
Very poor 10 – 20 30 cm 30 cm 30 cm
(a)
20
E
15
F G H
Group index
of subgrade
10
5 C D
B
A
0
Excellent
(A-1-a) 0 5 10 15 20 25 30 35 40 45 50 55 60
Thicknesses cm
(b)
Combined thickness of surface, base and sub-base
Thickness of surface and base
Note:
A – Thickness of selected material sub-base only.
B Light traffic
Combines thickness of surface, road base and
C – Medium traffic
selected material sub-base
D Heavy traffic
E – Thickness of additional road base which may be substituted
F Light traffic
G – Combined thickness of surface and road base (no sub-base) Medium traffic
H Heavy traffic
Fig. 24.5 Design chart by Group Index value
M24_PURU1773_01_SE_C24.indd 822 2/12/2013 8:18:03 AM

Pavements 823
California bearing ratio - percent

2 4 6 8 10 20 40 60 80
0
10
3175 kg wheel load
(light traffic)
20
Depth of construction - cm
4082 kg
wheel load
30
5443 kg wheel load
(heavy traffic)
40
50
60
70
Fig. 24.6 CBR design chart (California State Highway Department)
On similar grounds, Indian Roads Congress (IRC: 37–1970) provided design charts
similar to the one followed in UK, for different traffic volumes from 15 to over 4,500 com-
mercial vehicles per day exceeding 8 tons loader weight (Figure 24.7).
In order to use the chart, the following steps are followed:
1. CBR value should be found for a soaked specimen of the subgrade soil.
2. Considering the design wheel load or the anticipated traffic, the appropriate design curve
is chosen.
3. Corresponding to the CBR value and the chosen curve, the total thickness of flexible pave-
ment needed to cover the subgrade is found.
4. If a superior material is to be used as a sub-base, whose CBR value is found correspond-
ing to this CBR value of sub-base material, the required thickness of construction (over
this material) on the already chosen traffic curve is used.
5. Then the thickness of the sub-base course is the total thickness minus the thickness over
the sub-base.
The CBR method of flexible pavement design can be used to use materials of different
CBR values as layers with appropriate thickness.
24.7.3 Mc Leod Method

The Canadian Department of Transport under the direction of Mc Leod conducted exten-
sive plate-bearing tests to investigate the stability of air fields and highway pavements.
M24_PURU1773_01_SE_C24.indd 823 2/12/2013 8:18:04 AM

California bearing ratio percent

2 3 4 5 6 7 8 910 20 30 40 60 80 100
0
10
20
Depth of construction - cm
30
A
B Traffic classification
40
No. of commercial vehicles per day
Curve
C exceeding 3 tonnes laden weight
50
D A 0 – 15
B 15 – 45
60 E
C 45 – 150
F D 150 – 450
70
E 450 – 1500
G
F 1500 – 4500
80
G Above 4500
Fig. 24.7 CBR design chart (source: IRC 37–1970)
The plate-bearing tests were repetitive and were conducted on surface, base course,
sub-base and subgrade. Based on these results Mc Leod developed a design procedure,
which goes by his nature as Mc Leod method.
Mc Leod’s empirical design formula is given as:
P
T = K lo g 10 (24.1)
S
where T = required thickness of gravel base
P = gross wheel load, kg.
S = total subgrade support, kg (for the same contact area, deflection and number
of repetitions of load, P)
K = base course constant.
It has been found that the base course constant, K, is a function of the loaded area.
Figure 24.8 shows the relationship between the plate diameter and base course constant.
The thickness may be calculated from the above equation for a given wheel load P and the
subgrade support determined from a plate-bearing test.
The subgrade support, S, is computed from the support corresponding to 0.5 cm deflec-
tion after ten repetitions using a 30 cm diameter plate.
Figure 24.9 represents the perimeter of the area of the plate to the ratio of subgrade sup-
port (kg/cm2) at a particular deflection for any size of the plate to the subgrade support (in
kg/cm2) at 0.50 cm deflection on a 30 cm diameter plate.
M24_PURU1773_01_SE_C24.indd 824 2/12/2013 8:18:04 AM

Pavements 825
225
200
175
Base course constant K
150
125
100
75
50
25
0
25 50 75 100 125
Bearing plate diameter in cm
Fig. 24.8 Relation between plate diameter and base course constant
0.04
0.08
0.12
0.16
0.20
0.24
4
37.5 cm dia
45 cm dia
30 cm dia
20 cm dia
at deflection
Subgrade support (kg/cm2) at 0.5 cm

‘N’ for bearing plate of any diameter
deflection on 30 cm dia plate
ion
3 lect
Def cm
(kg/cm2)
For highway wheel loading 1750

cohesive subgrade soils at 15 5
12
ten repetitions of load 100
Subgrade support
2 075
050
1
025
012.5
Ratio of
0
Perimeter area ratio, P/A
Fig. 24.9 Relationship of subgrade support with P/A ratio
M24_PURU1773_01_SE_C24.indd 825 2/12/2013 8:18:05 AM

The design unit subgrade support is obtained by multiplying the contact pressure of
the design load by the above ratio. The value of the subgrade support, S, in the product
of the unit support and the contact area.
24.7.4 Burmister’s Method

Boussinesq’s stress considers a semi-infinite homogenous elastic medium. As flexible pave-
ment consists of layers of different elastic modulii this theory cannot be used. However,
Burmister’s layered system can be applied to flexible pavement to determine the thickness.
The layers of flexible pavement are subgrade, sub-base and base course, which may be
assigned elastic modulii as Es, Esb and Eb. In Burmister’s layered system Eb > Esb > Es is
considered. The effective reinforcing property of the pavement layers is logically utilized in
Burmister’s approach.
The assumptions made in applying Burmister’s layer theory to flexible pavements are as
follows:
1. All the materials used in the pavement structure including the subgrade are isotropic,
homogenous and elastic.
2. The surface layer is infinite in horizontal direction and finite in vertical direction; the
underlying layer is infinite in both the directions.
3. The layers have continuous perfect contact.
4. The top layer is free of shearing and normal stresses outside the loaded area.
5. Poisson’s ratio, v, is 0.5 in both the layers.
Figure 24.10 compares the vertical stress distribution of Boussinesq’s and Burmister’s
analyses. If Es is the modulus of elasticity of subgrade and Ep is that of the pavement; then,
as per Boussinesq’s theory, Es = Ep, whereas as per Burmister’s theory Ep > Es.
From Fig. 24.11, it could be seen that the stress is reduced from 70 to 30% in the second
layer, subgrade due to the introduction of first layer, pavement, of h = a having elastic modu-
lus 10 times as that of second layer, say Ep/Es = 10. Thus Burmister’s approach of reinforcing
action may be utilized in the pavement system.
A deflection factor, F2, has been introduced in two-layer system, which is dependent on
Es/Ep and h/a.
The relationship between two-layer deflection factor, F2, and the ratio Es/Ep and thickness
of reinforcing layer is shown in Fig. 24.11.
The displacement equation given by Burmister is rewritten as:
pa
For flexible plate, Δ = 1.5 F2 (24.2)
Es
pa
For rigid plate, Δ = 1.18 F2 (24.3)
Es
The above approach has been successfully used in the design of airfield pavements. The
layered system approach can also be utilized in the design of highway pavement. The diam-
eter of the loaded plate may be taken as 30 cm and the design deflection may be considered
as 0.50 or 0.25 cm.
M24_PURU1773_01_SE_C24.indd 826 2/12/2013 8:18:06 AM

Pavements 827
Radius a
Uniform pressure, p
modulus E1
90
Layer 1
80 h
90 70
60
80 50
70 40
30 a
60
50
40
20 Perfectly rough
interface
30 2a
modulus E2
Layer 2
20 Vertical stress σz,
as percentage of p
3a
10
10
4a
Uniform medium Two layer system
E1 E1
= 1 (Boussinesq case) = 10, υ = 1 , a = 1
E2 E2 2 h
Fig. 24.10 Comparison of vertical stress distribution by Boussinesq’s and Burmister’s approaches
1.0
0.8 1/2
0.6
0.5
1/5
0.4
Two layer deflection factor F2
0.3 1/10
1/2
0
0.2
1/5
0
1/1
00
0.1 1/2
00
0.08 1/5
0
0.06 1/1 0
000
0.05 1/2
0.04 00
0
1/5
0.03 000
1/1
000
0.02 0
0 0.5a 1.0a 1.5a 2.0a 3a 4a 5a 6a
Thickness of reinforcing layer
Fig. 24.11 Relationship of F2 and h in a two-layer system (Burmister’s method)
M24_PURU1773_01_SE_C24.indd 827 2/12/2013 8:18:07 AM

24.8 DESIGN OF RIGID PAVEMENTS

Cement concrete pavement is the classic example of a rigid pavement which provides an
ideal traffic surface. The cement concrete pavement may rest directly on a properly prepared
subgrade and distribute wheel loads over to the soil or on bridge decks.
Minor irregularities in the subgrade support are covered up by the flexural strength and
load-transfer capacity of the pavement in shear. In case the subgrade is not proper and sat-
isfactory, a sub-base course is provided to take the poor and non-uniform condition of the
subgrade (Fig. 24.12).
24.8.1 Stresses in Concrete Pavements

Stresses in concrete pavement slab are caused due to wheel loads, cyclic changes in tem-
perature, changes in moisture and volumetric changes in subgrade or base course.
Stresses Due to Wheel Loads Rigid pavements are designed primarily as the basis of
their resistance to bending. Stresses causing failure in rigid pavements are in general tensile.
As the concrete is laid as a large slab with longitudinal and cross joints, the induced stress
due to wheel load is not the same at all points unlike in flexible pavements. The critical
points where stresses are to be assessed are interior, edge and corner of a slab.
Stresses Due to Cyclic Changes in Temperature Cyclic changes in air temperature

causes warping stresses due to temperature gradient and change in mean temperature
causes expansion and contraction. The surface of the slab is subjected to a wide range of
temperatures, whereas the subgrade or base remains relatively at constant temperature. Re-
peated change of temperature in night and day cause cracks on the slab and ultimately the
slab breaks and get separated.
Stresses Due to Changes in Moisture Content The concrete shrinks when it

dries and expands when it is kept wet. Each drying and wetting cycle causes a
shrinkage and expansion, respectively. Stresses produced due to moisture gra-
dient between top and bottom of slab are similar to change in temperature.
Base course Cement concrete

(optional)
Subgrade
Fig. 24.12 Components of a rigid pavement
M24_PURU1773_01_SE_C24.indd 828 2/12/2013 8:18:07 AM

Pavements 829
As the magnitude of stress produced due to moisture change is less, compared to other
stresses, it is ignored.
Stresses Due to Volumetric Changes of Subgrade Due to changes in moisture content,

because of capillary waters or otherwise, subgrade soils are subjected to volume changes.
Clayey soils, particularly swelling soils, are much more affected than other soils.
Such variation in moisture in the subgrade causes non-uniform swelling and shrinking
resulting in non-uniform support under the slab. Further, their non-uniform subgrade sup-
port may give rise to high stress concentration at certain parts of the pavements.
Sometimes, non-uniformity of subgrade may lead to subsidence causing lateral displace-
ment of pavement slab. The movement may be accelerated due to further shrinking and
swelling.
Resultant Critical Stress and Location Stresses, discussed above, in a combined form
may cause distress on the pavement at a particular location as a resultant stress. The worst
condition depends on the time of variation of temperature and moisture as loading is a con-
tinuous process.
The three critical locations of stresses in rigid pavement are loading at corner, loading at
edge and loading at interior. The resultant stress varies on the atmospheric condition, day
and night times, moisture variation in subgrade etc. In total, the three critical locations in the
degree of decrease in intensity of stress are corner, edge and interior.
24.8.2 Westergaard’s Stress Equations

Westergaard has postulated a theory based on the condition that the subgrade reaction is
proportional to the deflection and the cement concrete slab is assumed to be homogenous
and to have uniform elastic properties.
He has given expressions for interior, edge and corner loadings.
Stress, Si, for the interior loading is given as:
0.316 P ⎡ ⎛l ⎞ ⎤
Si = ⎢ 4 lo g 10 ⎜⎜ ⎟⎟ + 1.069⎥ (24.4)
h2 ⎢⎣ ⎜⎝b ⎟⎠ ⎥⎦
Stress, Se, for edge loading is given as:
0.572 P ⎡ ⎛l ⎞ ⎤
Se = ⎢ 4 lo g 10 ⎜⎜ ⎟⎟ + 0.359⎥ (24.5)
h2 ⎢⎣ ⎜⎝b ⎟⎠ ⎥⎦
Stress, Sc, for corner loading is given as:
⎡ 0.6 ⎤
3 P ⎢ ⎛⎜ a 2 ⎞⎟ ⎥
S c = 2 ⎢ 1 − ⎜⎜ ⎟⎟ ⎥ (24.6)
h ⎢ ⎜⎝ l ⎟⎠ ⎥
⎣ ⎦
M24_PURU1773_01_SE_C24.indd 829 2/12/2013 8:18:08 AM

where P = wheel load, kg

h = slab thickness, cm
l = radius of relative stiffness, cm
1
⎡ ⎤ 4
⎢ Eh 2 ⎥
l=⎢ ⎥ (24.7)
(
⎢ 12 1 − v 2 k ⎥
⎢⎣ )
⎦⎥
E = modulus of elasticity of cement concrete (kg/cm2)

μ = Poisson’s ratio of cement concrete slab
k = modulus of subgrade reaction (kg/cm2 per cm)
b = equivalent radius of resisting section, cm
b = 1.6 a 2 + h 2 − 0.675 h (24.8)
a = radius of wheel-load distribution, cm

When a > 1.724h, the value of b = a.
In order to compute the slab thickness, h, for a given maximum stress of Si, Se, or Sc, trial
and error approach has to be resorted to.
24.8.3 Indian Road Congress (IRC) Formula (IRC 15, 1981)

Westergaard’s stress equations have been modified by different people. Among the three
locations, the edge and corner loading are generally critical.
Thus, Indian Road Congress has recommended a formula for edge and corner loading.
IRC Edge-loading Formula This equation is based on Westergaard’s equation, but modi-
fied by Teller and Sutherland as:
P ⎛ l ⎞
S e = 0.529 ( 1 + 0.54μ ) ⎜⎜⎜4 lo g 10 + lo g 10 b − 0.4048⎟⎟⎟ (24.9)
h 2 ⎝ b ⎠
IRC Corner Load Formula This equation is based on Westergaard’s equation, but modi-
fied by Kelley as:
⎡ 1 ⎤
3 P ⎢ ⎛⎜ a 2 ⎞⎟ 2 ⎥
Sc = 2 ⎢1 − ⎜⎜ ⎟⎟ ⎥ (24.10)
h ⎢ ⎜⎝ l ⎟⎠ ⎥
⎢⎣ ⎥⎦
The above equations and for finding load stresses at the edge and corner region are pre-
sented in the form of charts by IRC in Figure 24.13.
M24_PURU1773_01_SE_C24.indd 830 2/12/2013 8:18:08 AM

Pavements 831
40 40
K = 30 kg/cm3
35 35
K = 30 kg/cm3 K = 15
K = 10
30 K = 15 K=8
K = 10 30
K=6
Corner load stress, kg/cm2

Edge load stress, kg/cm2
K=8
K=6
25 25
20 20
15 15
10 10
5 5
0 0
14 16 18 20 22 24 14 16 18 20 22 24
Slab thickness, cm Slab thickness, cm
(a) Edge load stress chart (b) Corner load stress chart
Fig. 24.13 IRC edge- and corner-load charts
These charts are applicable for a particular set of design parameters, viz.,
P = 4,100 kg
E = 3 × 105 kg/cm2
v = 0.15
a = 15 cm
and curves for different K values are provided. The design curves are applicable for slab
thickness h, 15 to 25 cm.
WORKED EXAMPLES
Example 24.1 In a highway project, it is decided to provide flexible pavement on the

basis of thickness obtained from group index method. The average group index in that
area is 15. Design the pavement thickness if the anticipated traffic is over 400 commercial
vehicles per day.
Solution
As the average group index is 15, the subgrade may be grouped under very poor category.
From Figure 24.14 (b), the thickness of sub-base only from curve A = 26 cm.
M24_PURU1773_01_SE_C24.indd 831 2/12/2013 8:18:09 AM

Surface course 8 cm
Base course 22 cm
Sub-base course 26 cm
Subgrade, a1 = 15
Fig. 24.14
As the traffic is 400 commercial vehicles per day, the traffic falls under heavy traffic.
∴ The combined thickness for heavy traffic is found from curve D.
i.e., Thickness of surface, base and sub-base = 56 cm
∴ Thickness of surface and base course = 56 − 26 = 30 cm
Assuming a surface of 8 cm, the cross-section is shown in Figure 24.14.
Example 24.2 Well-graded gravel and poorly graded gravel of CBR 86% and 28%, respec-
tively, are to be used in the design of flexible pavement. The soil subgrade has been com-
pacted to have a CBR of 6%. The pavement has to be used for 4,000 commercial vehicles per
day. Design the pavement using IRC curves. The bituminous surfacing is to be of 8 cm thick.
The natural soil has a CBR of 4%.
Solution
As the number of commercial vehicles for which the road is to be used is 4,000, the design
curve (Figure 24.5b) F is to be used.
Bituminous surfacing 8 cm
Well graded gravel – CBR = 86% 10 cm
Poorly graded gravel – CBR = 26% 25 cm
Compacted subgrade – CBR = 6% 12 cm
Natural subgrade – CBR = 4%
Fig. 24.15
Total thickness needed for the natural⎪⎫⎪

⎬ = 55 cm
soil subgrade with a CBR of 4% ⎪⎪⎭
Thickness needed to cover the compacted⎫⎪⎪

⎬ = 43 cm
R 26%
subgrade of with CBR ⎪⎪⎭
M24_PURU1773_01_SE_C24.indd 832 2/12/2013 8:18:10 AM

Pavements 833
Thickness needed to cover the poorly⎪⎫⎪

⎬ = 8 cm
graded gravel with CBR = 28% ⎪⎪⎭
Total thickness over natural soil⎪⎫⎪

⎬ = 55 cm
subgrade⎪⎪⎭
Thickness of compacted soil subgrade = 55 − 43 = 12 cm

Thickness of poorly graded gravel = 43 − 18 = 25 cm
Thickness of well-graded gravel =18 − 8 = 10 cm
Bituminous surfacing = 8 cm
Thicknesses of different layers are shown in Fig. 24.15.
Example 24.3 The flexural strength of concrete is 48 kg/cm2 and modulus of elasticity is
2.8 × 105 kg/cm2.The maximum wheel load is 4,200 kg with an impact factor of 5%. The tyre
pressure is 5.2 kg/cm2 and the modulus of subgrade reaction is 6.0 kg/cm2. Assuming a fac-
tor of safety of 2.5, find the thickness of cement concrete pavement. Use Westergaard’s
corner-stress equation.
Solution
Allowable flexural⎫⎪⎪ Flexural strength
⎬=
strength⎪⎪⎭ Factor of safety
48
= = 19.2 kg/cm2
2.5
Maximum wheel load = 4,200 kg
Impact factor = 5%
5
Design wheel load = 4 , 200 + × 4 , 200
100
= 4,410 kg.
A pavement thickness of 20 cm is assumed and to calculate the radius of relative stiffness.
Thus,
P
Radius of contact area, a =
pu
4 , 410
a= = 16.43 cm
5.2× u
Eh 3
Radius of relative stiffness, l=
( )
4
12 1− v 2 K
2.8 ×10 5 × 20 3
=4
12 (1 − 0.152 )
6
= 75.11 cm
M24_PURU1773_01_SE_C24.indd 833 2/12/2013 8:18:11 AM

⎡ 0.6 ⎤
3 P ⎢ ⎛⎜ a 2 ⎞⎟ ⎥
Corner tensile stress, Sc = ⎢ 1 − ⎜ ⎟⎟ ⎥
h 2 ⎢ ⎜⎜⎝ l ⎟⎠ ⎥
⎣ ⎦
⎡ 0.6 ⎤
3× 4 , 410 ⎢ ⎛⎜ 16.43 2 ⎞⎟ ⎥
= ⎢1 − ⎜ ⎟⎟ ⎥
20 2 ⎢ ⎜⎜⎝ 75.11 ⎟⎠ ⎥
⎣ ⎦
= 33.075 ⎡⎢1 − (0.309) ⎤⎥ .

0.6
⎣ ⎦
2
= 16.74 kg/cm
POINTS TO REMEMBER
24.1 A pavement is a main structure of a road or a runway, which is relatively a stable

layer or a crust constructed over the natural soil or subgrade. The main function of a
pavement is to support and distribute the wheel loads of vehicles through the under-
lying layers to the subgrade.
24.2 Pavements may be classified into two major categories, namely, flexible pavement
and rigid pavement.
24.3 Flexible pavement is the one which distributes the wheel load stresses to the sub-
grade through successive layers of granular materials.
24.4 Cement concrete pavement is the classic example of a rigid pavement, which pro-
vides an ideal traffic service. Cement concrete pavements may be directly placed on
a properly prepared subgrade or a base.
24.5 The design wheel load depends on the (i) maximum wheel load, (ii) contact pres-
sure, (iii) equivalent single-wheel load and (iv) repetition of loads.
24.6 The components of a flexible pavement are a sub-base course, base course and a sur-
face course.
24.7 Stress in the concrete pavement slabs are caused due to wheel loads, cyclic changes
in temperature, changes in moisture and volumetric changes in subgrade or
sub-base.
24.8 Stresses in combined form may cause distress at a particular location of a pavement
due to the resultant stress. The worst condition depends on the time of variation of
temperature and moisture, as loading is a continuous process.
24.9 Critical positions of load are corner, edge and interior location of a pavement.
24.10 Westergaard’s concept of temperature stresses has shown that temperature tends to
produce two types of stresses in a concrete pavement, namely, warping stresses and
frictional stresses.
M24_PURU1773_01_SE_C24.indd 834 2/12/2013 8:18:12 AM

Pavements 835
QUESTIONS
Objective Questions
24.1 Bottom-most layer of pavement is known as

(a) Base course. (b) Sub-base course.
(c) Wearing course. (d) Subgrade.
24.2 Equivalent wheel load of loads of intensity ‘p’each is
(a) 2p. (b) Greater than 2p.
(c) In between 2p. (d) Less than 2p.
24.3 CBR method of design of flexible pavement is
(a) A theoretical method.
(b) An empirical method based on physical properties of subgrade soil.
(c) A semi-empirical method.
(d) An empirical method based on strength characteristics of subgrade soil.
24.4 The IRC-recommended flexible pavement design method is
(a) CBR method. (b) Group index method.
(c) Westergaard’s method. (d) Benkelman Beams method.
24.5 If the Group Index value of subgrade is between 6 and 10, then the subgrade is
treated as
(a) Good. (b) Fair.
(c) Poor. (d) Very poor.
24.6 Flexible pavement distributes the wheel load
(a) Directly to subgrade.
(b) Through base course only.
(c) Through base and sub-base courses.
(d) Through sub-base course only.
24.7 Rigidity factor for a tyre pressure greater than 7 kg/cm2 is
(a) Zero. (b) Greater than 1.
(c) Less than 1. (d) Equal to 1.
24.8 Inflation pressure influences the
(a) Total depth of pavement. (b) Quality of surface course.
(c) Both (a) and (b). (d) None of the above.
24.9 Critical location of wheel load in a rigid pavement design is taken as
(a) Interior-loading. (b) Edge-loading.
(c) Corner-loading. (d) Edge and corner loading.
24.10 Radius of relative stiffness is a function of
(a) E. (b) h and K.
(c) v. (d) All the above.
M24_PURU1773_01_SE_C24.indd 835 2/12/2013 8:18:12 AM

24.11 Differentiate between rigid and flexible pavements.
24.12 What is equivalent single-wheel load? Explain briefly.
24.13 State how variations in climatic conditions affect the design of pavements.
24.14 Explain the Group Index method.
24.15 What are the methods of flexible-pavement design? Write the IRC recommendations
for CBR method.
24.16 Critically appraise the CBR method of flexible-pavement design.
24.17 How Burmister’s two-layer theory is used for the design of highway flexible
pavement?
24.18 Bring out the merits and demerits of cement concrete pavement.
24.19 Describe briefly the factors influencing the design of rigid pavements.
24.20 Discuss the stresses caused in cement concrete pavements.
24.21 What are the critical resultant stresses and their locations?
24.22 Explain Westergaard’s stress equation for cement concrete pavement.
24.23 Explain temperature stresses that develop in cement concrete pavement.
EXERCISE PROBLEMS
24.1 Soil subgrade on a certain area showed the following properties:

Soil passing through 75μ sieve = 78%
Liquid limit = 65%
Plastic limit = 40%
Design the pavement section by Group Index method for heavy traffic with over 3500
commercial vehicles per day.
24.2 Using the following data, design the flexible pavement layers:
CBR of the subgrade soil = 5%.
CBR of poorly graded gravel sub-base = 15%.
CBR of well-graded gravel base = 80%.
Traffic of 3,200 vehicles per day may be considered.
Assume any other data found needed.
24.3 A plate-beating test was conducted with a 30 cm diameter plate on a subgrade soil,
which showed a pressure of 4.2 kg/cm2 at 0.50 cm deflection. Similar test conducted
on a base course of 22 cm thick showed a pressure of 9.4 kg/cm2 at 0.25 cm deflection.
Design the pavement thickness for a wheel load of 5,800 kg with tyre pressure of
6.8 kg/cm2 .Use Burmister’s approach.
24.4 A cement concrete pavement of 20 cm rests over a base course with a modulus of reac-
tion of 15 kg/cm2. Find the load stresses at the edges and corner region under a wheel
load of 5,100 kg. By using IRC stress charts, assume A = 15 cm, E = 3 × 105 kg/cm2
and v = 0.15.
M24_PURU1773_01_SE_C24.indd 836 2/12/2013 8:18:12 AM

List of Symbols
Symbol Definition
A Activity of clay
Aa Cross-section area
A,⎯Α Pore pressure coefficient
A′ Total inter-particle attraction
Ar Area ratio
Av Air content or air void ratio
a Radius
av Coefficient of compressibility
B Buoyancy force
B,⎯Β Pore pressure coefficient
B, Bf Width of footing
b Width
C Clay fines
C Inorganic clay
Cc Compression index
Cα Coefficient of secondary compression
Ce Expansion index
Cf Compressibility of pore fluid
Cs Composite shape factor
Cs Compressibility of soil skeleton
Cu Uniformity coefficient
Cz Coefficient of curvature
c Cohesion intercept
c′ Effective cohesion
c* Modified cohesion intercept
cb Coefficient of consolidation (horizontal drainage)
cm Mobilized cohesion
cu Undrained cohesion
cv Coefficient of consolidation (vertical drainage)
ccu Consolidated undrained cohesion
′
ccu Effective consolidated undrained cohesion
cd′ Effective drained cohesion
cr′ Effective cohesion for residual strength
c Effective contact stress
D, d Diameter
D Particle size
D Depth
D Depth factor
D Length of drainage path
Z01_PURU1773_01_SE_SYM.indd 837 2/12/2013 8:17:39 AM

Dd Design depth of sheet pile

Ds Depth of scour
Df Depth of foundation
DH Height table position
dw, d′w Water table position
E Modulus of elasticity/stress-strain modulus
e Void ratio
ex, ey Eccentricity
ecr Critical void ratio
emax Maximum void ratio
emin Minimum void ratio
eo Original void ratio
F Factor of safety
FB Factor of safety against overturning
FC Factor of safety with respect to cohesion
FH Factor of safety against heave
FOT Factor of safety against overturning
FSL Factor of safety against sliding
Fa Force on anchor rod
Fb Factor of safety with respect to height
Fd Driving forces
Fp Factor of safety against piping
Fr Resisting forces
Fs Factor of safety with respect to strength
Ff Factor of safety with respect to friction
Fp′ Increase in factor of safety against piping
f Frictional resistance
f Frictional resistance between soil to soil
G Gravel
G Specific gravity of soil solids
G Shear modulus
GR Groutability ratio
g Gravitational acceleration
H High compressibility
H, h Head loss
H Height of dam
Hc Unsupported depth of excavation/critical height
Hα Relative height of piling
hc Capillary head
hc Capillary rise
hu Piezometric head
I Current
I Medium compressibility
I Moment of inertia
IB Brittleness index
ID Density index
IL Liquidity index
IQ Quality index
Ic Consistency index
If Flow index

List of Symbols 839
If Influence factor
Ip Plasticity index
It Toughness index
i Hydraulic gradient
i Inclination of slope
ic Critical hydraulic gradient
ie Exit gradient
K Absolute permeability
K Coefficient of lateral pressure
Ka Coefficient of active earth pressure
K0 Earth pressure at rest
Kp Coefficient of passive earth pressure
k Coefficient of permeability
ks Spring constant
kE Coefficient of electrical conductivity
ke Effective coefficient of permeability
kH Effective coefficient of permeability in the horizontal direction
kV Effective coefficient of permeability in the vertical direction
L Length of soil specimen/pile
L Low compressibility
Lr Recovery ratio
M Silt fraction fines
M Total mass
M A coefficient
Ma Mass of air
Md Disturbing moment
Mr Resisting moment
Ms Mass of soil solids
Mw Mass of water
mv Coefficient of volume compressibility
N Percentage finer
N Standard penetration value
Ncd Dynamic cone penetration value
NB Boussinesq’s vertical stress coefficient
NCA Stress coefficient for circular loaded area – stress at any point
NCC Stress coefficient for circular central load
NE Stress coefficient for embankment loading
NF Cousins stability number
NN New Mark’s stress coefficient
NR Stress coefficient for rectangular loaded area
NS Stability factor
NW Westergaard’s vertical stress coefficient
Nc′, Nq′, Nγ′ Bearing capacity factors – deep foundation
Nc , Nq , N γ Bearing capacity factors – general shear
Nc′, Nq′, Nγ′ Bearing capacity factors – local shear
Nd Number of potential drops
Nf Number of flow channels
Nf Flow value
n Porosity
nd Number of potential drops lost

O Organic clays and silts

OCR Over-consolidation ratio
Op Origin of planes
P Poorly graded
P Maximum load on specimen/load at rupture
Pa Active thrust
Pp Passive resistance
pa Unit active pressure
pc Pre-consolidation pressure
po Overburden pressure
pp Unit passive pressure
ps Seepage pressure
Q Vertical load
Q, q Discharge
Qa Allowable load
Qf Resistance due to skin friction
Qn Net ultimate load
Qp Resistance due to point bearing
Qu Ultimate load
Qn Total drag force
Qug Ultimate group capacity
q Stress due to surcharge
q Uniformly distributed load/contact pressure
qa Allowable soil pressure
qc Cone penetration resistance
qf Ultimate bearing capacity
qg Gross foundation pressure
qn Net foundation pressure
qnf Net ultimate bearing capacity
qns Net safe bearing pressure
qs Gross safe bearing pressure
R Radius of influence
R Residual factor
Rb, rb Hydrometer reading
R′ Total inter-particle repulsion
Rn Reynold’s number
Rw, R′w Water table correction factors
r Radius
ru Pore pressure ratio
S Total settlement
S Section modulus of sheet pile wall
SB Settlement of footing
SF Seepage forces on particles
Sc Consolidation settlement
Si Immediate settlement
Sn Stability number
Sp Settlement of plate
Sr Degree of saturation
Ss Settlement due to secondary compression
St Sensitivity of clay

List of Symbols 841
sc, sq, sγ Shape factor corrections

T Coefficient of transmissivity
T Torque
Tr Time factor (radial drainage)
Tv Time factor (vertical drainage)
Ts Surface tension
t Time
Ur Degree of consolidation (radial drainage)
Uv Degree of consolidation (vertical drainage)
ua, ua Pore air pressure
u* Equivalent pore pressure
V Volume
V Voltage
Va Volume of air
Vs Volume of soil solids
Vv Volume of voids
Vw Volume of water
v Velocity
vs Seepage velocity
W Well graded
W Weight
Ws Weight of soil particles
w Water content
wL Liquid limit
wp Plastic limit
ws, wsu Shrinkage limit of soil/undisturbed soil
Z Depth
Z Position head
Z0 Depth of tension zone
Greeks
α Contact angle
β Inclination of retaining wall
γ Total unit weight of soil
γ′ Submerged unit weight of soil
γd Dry unit weight of soil
γs Unit weight of soil solids
γsat Saturated unit weight of soil
γw Unit weight of water
Δe Change in void ratio
ΔH Change in height
Δuw Change in pore water pressure
ΔV Change in volume
Δσ Change in total stress
Δσ ′ Change in effective stress
δ Angle of wall friction
ε Strain
θf Inclination of failure plane
μ Micron

ν Poisson’s ratio
χ A parameter
ρ Total density of soil
ρ Resistivity
ρ Flexibility number
ρ′ Submerged density of soil
ρd Dry density of soil
ρs Density of soil solids
ρsat Saturated density of soil
ρsd Bulk density of soil
ρw Density of water
σ Total stress
σ′ Effective stress
σ ′′ Intrinsic effective stress
σ Mineral–mineral contact stress
σBT Brazilian tensile strength
σfa Allowable flexural stress
σft Flexural tensile strength
σh Lateral stress
σh′ Effective lateral stress
σn Total normal stress
σn′ Effective normal stress
σpt Point load tensile strength
σr Radial stress or radial stress due to surface load
σt Tensile strength
σv Total vertical stress
σ v′ Effective vertical stress
σz Total vertical stress due to surface load
σθ Circumferential stress due to surface load
σ1, σ2 , σ3 Total major, intermediate and minor principal stresses
σ1′ , σ2′ , σ3′ Effective major, intermediate and minor principal stresses
τf Peak shear strength
τr Residual shear strength
τxy , τyz , τzx Complementary shear stresses
φ Angle of shearing resistance
φ Potential function
φ′ Angle of shearing resistance for effective stress condition
φ* Modified friction angle
φcu Consolidated undrained angle of shearing resistance
′
φcu Effective consolidated undrained angle of shearing resistance
φd′ Drained angle of shearing resistance
′
φps Angle of shearing resistance for plain strain
φr Residual angle of shearing resistance
φtr′ Angle of shearing resistance for triaxial test
φu Undrained angle of shearing resistance
ψ Flow function
η Efficiency
η Viscosity of fluid
ηg Efficiency of pile group
ηw Viscosity of water

Bibliography
Abbreviations in References
ASCE American Society of Civil Engineers, USA
ASTM American Society for Testing and Materials, USA
BIS Bureau of Indian Standards, India
ENR Engineering News Record, USA
CGJ Canadian Geotechnical Journal, Canada
GTECH Geotechnique, Institution of Civil Engineers, London, UK
HRB Highway Research Board, USA
ICE Institution of Civil Engineers, London, UK
ICSMEF Proceedings of International Conference on Soil Mechanics and Foundation Engineering
IGJ Indian Geotechnical Journal, India
JGED Journal of Geotechnical Engineering Division, ASCE (1974–2000)
JSMFD Journal of Soil Mechanics and Foundation Division, ASCE (1955–1973, Incl.)
PICE Proceedings of Institution of Civil Engineers, London, UK
STP Special Technical Publication, ASTM
Alam Singh and G. G. Chowdhary, Soil Engineering, Part 1, 3rd Ed., CBS Publishers and Distributors,
New Delhi, 969 pp., 1973.
Anandakrishnan, M. and N. R. Krishnaswamy, “Response of soil-foundation systems,” Proc. 2nd South
East Asian Conference on SMFE, Vol. 1, pp. 107–119, 1969.
Anandakrishnan, M. and N. R. Krishnaswamy, “Response of embedded footings to vertical vibra-
tions,” ASCE, JSMFED, Vol. 99, p. 863, 1973.
AREA, Manual of Recommended Practice, Chicago; Construction and Maintenance Section, Engineering
Division, Association of American Railroads, 1958.
Arthur, H. G., “Selection of type of dam,” In: Design of Small Dams by USBR, Oxford and IBH Publish-
ing Co., New Delhi, pp. 97–105, 1973.
Atterberg, A. “Uber die Physikalische Bodenuntersucshung and uber die plastiziat der Tone,” Int. Mitt.
fur Bodenkund, Vol. 1, pp. 10–43, 1911.
Babu Shankar, N., “Rankine’s earth pressure theory for inclined backfills,” IGJ, Vol. 10,
pp. 361–367, 1980.
Babu Shankar, N., “Coulomb’s theory of earth pressure for c–f soils,” IGJ, Vol. 1, pp. 99–104, 1981.
Baranov, V. A., “On the calculation of excited vibrations of an embedded foundation,” (in Russian),
Voprosy Dynamiki Prochnocti, No. 14, Polytechnical Institute of Riga, pp. 195–209, 1967.
Barkan, D. D., Dynamics of Bases and Foundations, McGraw-Hill, New York, 434 pp., 1962.
Barron, R. A., “Consolidation of fine grained soils by drain wells,” Trans. ASCE, Vol. 113,
pp. 718–742, 1948.
Z02_PURU1773_01_SE_BIB.indd 843 2/12/2013 8:40:01 AM

Barton, N. R., Lien, and J. Lunde, “Engineering classification of rock masses for design of tunnel
support,” Rock Mech., Vol. 6, pp. 189–236, 1975.
Baumann, V. and G. E. A. Bauer, “The performance of foundation on various soils stabilised by vibro-
compaction method,” CGJ, Vol. 11, pp. 509–530, 1974.
Bell, A. L., “The lateral pressure and resistance of clay and the supporting power of clay foundations,”
In: A Century of Soil Mechanics, ICE, London, pp. 93–134, 1915.
Bell, F. G., Fundamentals of Engineering Geology, Butterworth, London, 648 pp., 1983.
Bennett, O. T., “Notes on embankment design,” Fourth Congress on Large Dams, Vol. 1, p. 223, 1951.
Berezantzev, V. G. et al., “Load bearing capacity and deformation of piled foundations,” 5th INCSMFE,
Vol. 2, pp. 11–15, 1961.
Berezantzev, V. G., “Design of deep foundations,” 6th INCSMFE, Vol. 2, pp. 234–237, 1965.
Bernaix, J., “New laboratory methods of studying the mechanical properties of rocks,” Int. J. Rock
Mech., Min. Sci., Vol. 6, pp. 43–90, 1969.
Bertram, G. E., An Experimental Investigation of Protective Filters, Publication of the Graduate School of
Engineering, Harvard University, No. 267, 1940.
Bieniawski, Z. T., “Engineering classification of jointed rock masses,” Trans. S. Afr. Inst. Civil Engineers,
Vol. 15, pp. 335–343, 1973.
Bieniawski, Z. T., “Geomechanics classification of rock masses and its application in tunnelling,” Proc.
3rd Int. Conf. Rock Mech., Denver, Vol. 2, 27–32, 1974a.
Bieniawski, Z. T., “Estimating the strength of rock materials,” Inst. Min. Metall., Vol. 74,
pp. 312–320, 1974b.
Bieniawski, Z. T. (ed.), “Engineering properties of rocks with reference to tunnelling,” In: Tunnelling in
Rock, S. Afr. Inst. Civil Engineers/S. Afr. Nat. Gr. Rock Mech, CSIR, Pretoria, pp. 105–123, 1975.
Bieniawski, Z. T. (ed.), “Rock mass classification in rock engineering,” Proc. Symp. Exploration for Rock
Engineering, A. A. Balkema, Cape Town, Vol. 1, pp. 97–106, 1976.
Blum, H. Einspannungsverhtialtnisse bei Bohlwerken, Diss. Tech. Wilhelm Ernst und Sohn,
Berlin, 1931.
Bjerrum, L., “Geotechnical properties of Norwegian marine clays,” GTECH, Vol. 4, pp. 49–69, 1954.
Bjerrum, L., “Mechanism of progressive failure in slopes of over-consolidated plastic clays and clay
shales,” Third Terzaghi Lecture, ASCE, 1966.
Bishop, A. W., “The use of pore-pressure coefficients in practice,” GTECH, Vol. 4, pp. 148–152, 1954.
Bishop, A. W., “The use of the slip circle in the stability analysis of slopes,” GTECH, Vol. 5, pp. 1–5, 1955.
Bishop, A. W., “Some factors controlling the pore pressure set up during the construction of earth
dams,” Proc. 4th INCSMFE, Vol. 2, pp. 294–300, 1957.
Bishop, A. W., “The principle of effective stress,” Teknisk Ukeblad, Vol. 39, p. 859, 1959.
Bishop, A. W., “The strength of soil as engineering materials,” Sixth Rankine Lecture, Vol. 16,
pp. 91–128, 1966.
Bishop, A. W., “Discussion on shear strength of stiff clays,” Proc. Geotechnical Conf., Oslo, Vol. 2,
p. 142, 1967.
Bishop, A. W., “The influence of progressive failure on the method of stability analysis,” GTECH, Vol.
2, p. 168, 1971.
Bishop, A. W., I. Alpan, E. E. Blight, and I. B. Donald, “Factors controlling the strength of partially
saturated cohesive soils,” Proc. Research Conf. Shear Strength of Cohesive Soils, ASCE, Boulder,
pp. 503–532ff, 1960.
Bishop, A. W. and L. Bjerrum, “The relevance of the triaxial test to the solution of stability problems,”
Proc. Research Conf. Shear Strength of Cohesive Soils, ASCE, Boulder, pp. 437–501, 1960.
Bishop, A. W., G. E. Green, V. K. Garga, A. Andresen, and J. D. Brown, “A new ring shear apparatus
and its application to the measurement of residual strength,” GTECH, Vol. 21, pp. 273–328, 1971.
Bishop, A. W. and D. J. Henkel, The Measurement of Soil Properties in the Triaxial Test, 2nd Ed., Edward
Arnold, London, 228 pp., 1962.

Bibliography 845
Bishop, A. W. and N. R. Morgenstern, “Stability coefficients for earth slopes,” GTECH, Vol. 10,
pp. 129–150, 1960.
Boussinesq, J., Application des Potentials a L’ Etude de L’ Equilibre et du Movement des solides Elastiques,
Ganthier-Villars, Paris, 1885.
Bowles, J. E., Engineering Properties of Soils and Their Measurement, 2nd Ed., McGraw-Hill, New York,
213 pp., 1978.
Bowles, J. E., Foundation Analysis and Design, 3rd Ed., McGraw-Hill, New York, 816 pp., 1982.
Bowles, J. E., Physical and Geotechnical Properties of Soils, 2nd Ed., McGraw-Hill, New York, 478 pp., 1984.
Brace, W. F., “Brittle fracture of rocks,” In: Symp. State of Stress in Earth’s Crest, edited by W. R. Judd,
Santa Monica, Elsevier, pp. 111–180, 1964.
Brace, W. F. and D. Byerlee, Recent Experimental Studies of Brittle Fracture of Rocks, Unpublished Report,
MIT, Cambridge, MA, 1966.
Brooker, E. W. and H. O. Ireland, “Earth pressure at rest related to stress history,” CGJ, Vol. 2,
pp. 1–15, 1965.
Brown, R. E., Vibration Compaction of Granular Hydraulic Fills, Preprint 2657, ASCE National Water
Resources and Ocean Engineering Convention, pp. 1–30, 1976.
Brown, C. B. and I. P. King, “Automatic embankment analysis,” GTECH, Vol. 16, p. 252, 1966.
Brown, E. T. and E. Hoek, “Trends in relationships between measured in situ stresses and depth,” Int.
J. Rock Mech. Min. Sci., Vol. 15, pp. 211–215, 1978.
Burland, J. B., “Shaft friction of piles in clay – A simple fundamental approach,” Ground Engineering,
Vol. 6, pp. 30–42, 1973.
Burland, J. B., et al., “The behaviour and design of large diameter bored piles in stiff clay,” Proc. Conf.
on Large Bored Piles, ICE, London, pp. 51–71, 1966.
Burmister, D. M., “The application of controlled test methods in consolidation testing,” Symposium on
Consolidation Testing of Soils, ASTM, STP No. 126, p. 83, 1951.
Burmister, D. M., “Principles of permeability testing of soils,” Symposium on Permeability of Soils, ASTM,
STP, No. 163, 1954.
Button, S. J., “The bearing capacity of footings on a two-layer cohesive subsoil,” 3rd INCSMFE, Vol. 1,
pp. 332–335, 1953.
Carilo, N., “Simple two and three dimensional cases in the theory of consolidation of soil,” J. Math.
Phys., Vol. 21, pp. 1–5, 1942.
Carroll, W. F., “Dynamic bearing capacity of soils, vertical displacements of spread footings on clay:
Static and impulsive loadings,” Tech. Report No. 3-599, U.S. Army Corps. Engineers, Waterways
Experiment Station, Vicksburg, 1963.
Casagrande, A., “Research on Atterberg limits of soils,” Public Roads, Vol. 13, pp. 121–136, 1932a.
Casagrande, A., “The structure of clay and its importance in foundation engineering,” Contributions to
Soil Mechanics, Boston Society of Civil Engineers, 1925–1940, Boston, pp. 72–112, 1932b.
Casagrande, A., “Characteristics of cohesionless soils affecting the stability of slopes and earthfills,”
Journal of Boston Society of Civil Engineers, Vol. 23, pp. 13–32, 1936a.
Casagrande, A., “The determination of the pre-consolidation load and its practical significance,” 1st
ICSMFE, Vol. 3, pp. 60–64, 1936b.
Casagrande, A., “Seepage through dams,” Journal of New England Water Works Association, Vol. 51,
pp. 295–336, 1937.
Casagrande, A. and N. Carrillo, “Shear failure of anisotropic materials,” In: Contributions to Soil Mech-
anics, Boston Society and Civil Engineers, 1940–1953, Boston, 1944.
Casagrande, A., “Classification and indentification of soils,” Trans. ASCE, Vol. 113, pp. 901–930, 1948.
Casagrande, L., “Electro-osmotic stabilization of soils,” J. Boston Soc. Civil Engrs., Vol. 39, pp. 51–
83, 1952.
Cernica, J. N., Geotechnical Engineering, Holt, Reinhart and Winston, New York, 488 pp., 1982.

Chellis, R. D., “Pile foundations,” In: Foundation Engineering, edited by G. A. Leonards, McGraw-Hill,
New York, 1962.
Chen, W. F., “Soil mechanics and theorems of limit plasticity,” ASCE, JSMFD, Vol. 95, pp. 493–518, 1969.
Converse, F. J., “Foundations subjected to dynamic forces,” In: Foundation Engineering, edited by G. A.
Leonards, McGraw-Hill, New York, pp. 769–825, 1962.
Cook, N. G. W., “The failure of rock,” Int. J. Rock Mech. Min. Sci., Vol. 2, pp. 389–403, 1965.
Cook, N. G. W., E. Hoek, J. P. G. Pretorius, W. D. Ortlepp, and M. D. G. Salamon, “Rock mechanics
applied to the study of rockbursts,” J. S. Afr. Inst. Min. Metall., Vol. 66, pp. 435–528, 1966.
Cooling, L. F. and H. O. Golder, “The analysis of the failure of an earth dam during construction,”
J. ICE, Vol. 19, pp. 38–42, 1942.
Coulomb, C. A., “Essai sur une Application des Regles des Maximis et Minimis a quelques Problems
de Statique Relatifs a l’ Architecture,” Mem. Acad. Roy. Des. Sci., Paris, Vol. 3, p. 38, 1976.
Cousins, B. F. “Stability charts for simple earth slope allowing for tension cracks,” Third Australian–
New Zealand Conf. on Geomechanics, Wellington, Australia, Vol. 2, pp. 1–5, 1980.
Cousins, B. F., “Stability charts for simple earth slopes,” ASCE, JGED, Vol. 104, pp. 267–279, 1978.
Cozens, W. J., “Machinery foundations,” JIEE, Vol. 82, pp. 327–328, 1938.
Craig, R. F., Soil Mechanics, English Language Book Society, Berkshire, 419 pp., 1986.
Cullen, R. M. and I. B. Donald, “Residual strength determination in direct shear,” Proc. First Australia–
New Zealand Conf. on Geomechanics, Vol. 1, p. 1, 1971.
Culmann, C. A., Theorie der Stutz-und Futtermanern. Section 8 in Die Graphische Statik, Meyer und Zeller,
Zurich, 1866.
Culmann, C. A., Die Graphische Statik, Meyer und Zeller, Zurich, 644 pp., 1875.
Darcy, H., Les Fountaines Publiques de la Ville de Dijon, Dalmont, Paris, 674 pp., 1856.
Das, B. M., Fundamentals of Soil Dynamics, Elsevier, Amsterdam, 399 pp., 1990.
Das, B. M., Principles of Foundation Engineering, Cole Engineering Division, California, 595 pp., 1984.
Dastidar, A. G., S. Gupta, and J. K. Gosh, “Applications of sand wick in a housing project,” 7th
INCSMFE, Vol. 2, pp. 59–64, 1969.
Datye, K. R., “Simpler techniques for ground improvements,” Fourth IGS Annual Lecture, IGH, Vol. 12,
pp. 1–82, 1982.
Datye, K. R. and S. S. Nagaraju, “Observation based design approach for preloading,” Proc. 5th Asian
Regional Conference on SMFE, Bangalore, pp. 19–27, 1975.
Datye, K. R. and S. S. Nagaraju, “Practical and economic aspects of the design of ground improvement
system,” IGS Conference on Foundations and Excavations in Weak Soil, Calcutta, p. c-9, 1976.
Datye, K. R. and S. S. Nagaraju, “Design approach and field control for stone columns,” 10th
INCSMFE, Stockholm, Sweden, Vol. 3, pp. 637–644, 1977.
Deere, D. U., “Geological considerations,” In: Rock Mechanics, edited by K. G. Stagg and O. C.
Zienkiewic, John Wiley & Sons, New York, pp. 1–20, 1979.
Deere, D. U. and R. P. Miller, “Engineering classification and index properties for intact rock,” Tech.
Rep. No. AFWL-TR 65-116, Air Force Weapons Lab., Kirtland Air Base, New Mexico, 1966.
De Mello, V. F. B., “Foundations of buildings on clay, state-of-the-art report,” Proc. 7th INCSMFE,
Mexico City, Mexico, pp. 49–136, 1969.
Dhir, R. K. and C. M. Sangha, ‘Relationships between size, deformation and strength for cylindrical
specimens loaded in uniaxial compression,” Int. J. Rock Mech. Min. Sci. Geomech. Abstr., Vol. 10,
pp. 699–712, 1973.
Dobson, T. and B. Slocombe, “Deep densification of granular fills,” Proc. 2nd Geotechnical Conference on
Design and Construction, Las Vegas, 21 pp., 1982.
Drucker, D. C. and W. Prager, “Soil mechanics and plastic analysis or limit design,” Q. Appl. Math.,
Vol. 10, pp. 157–165, 1952.
Dunn, I. S., L. R. Anderson, and F. W. Kiefer, Fundamentals of Geotechnical Analysis, John Wiley & Sons,
New York, 414 pp., 1980.

Bibliography 847
Dupuit, J., Etudes Theoriques et Practques sur le Mouvement des eaux dans les Canaux Decouverts et a travers
les Terrains Permeables, Dunod, Paris, 1863.
Fadum, R. E., “Influence values for estimating stresses in elastic foundation,” 2nd ICSMFE, Vol. 3,
pp. 77–84, 1948.
Fairhurst, C., “Laboratory measurements of some physical properties of rock,” Proc. 4th Symp. on Rock
Mech., Pennsylvania State University, pp. 105–118, 1961.
Farmer, I. W., Engineering Properties of Rocks, Spon, London, 1968.
Fang, H. Y., Introduction to Environmental Geotechnology, CRC Press, Boca Raton, New York, 652 pp., 1997.
Fellenius, W., “Calculation of the stability of earth dams,” Trans. 2nd Cong. Large Dams, Washington,
D.C., Vol. 4, p. 445, 1936.
Fellenius, W., Erdatatische Berechnungen Mit Reibung und Kohasion (Adhasion) und Unter Annabme kreis-
zylindrischer Gleitflachen, Wilhelm Ernst und Sohn, Berlin, 1947.
Foster, C. R., “Field problems: compaction,” In: Foundation Engineering, edited by G. A. Leonards,
McGraw-Hill, New York, pp. 1001–1002, 1962.
Foster, C. R. and R. G. Ahlvin, “Stresses and deflections induced by a uniform circular load,” HRB,
Vol. 33, pp. 467–470, 1954.
Fourmaintraux, D., “Characterization of rocks: Laboratory tests,” In: La Meecanique desroches appliqué
aux auvrages du genie civil, edited by Marc Panet et al., Ecole Nationale des Ponts et Chausses,
Paris, 1976.
Franklin, J. A. and R. Chandra, “The Slake durability index,” Int. J. Rock Mech. Min. Sci., Vol. 9,
pp. 325–342, 1972.
Franklin, J. E. and E. Broch., “The point load strength test,” Int. J. Rock Mech. Mining Sci., Vol. 9,
pp. 669–697, 1972.
Franklin, J. L., Index Properties of Rocks: Part I – Suggested Methods for Determination of Water Content,
Porosity, Density and Related Properties, Imperial College, London, 1970.
Gamble, J. C., Durability–Plasticity Classification of Shales and Other Argillaceous Rocks, Ph.D. thesis,
University of Illinois, 1971.
Goel, M. C., R. P. Singh, and G. Mekonnen, “Design slopes for small homogeneous earth dams,” IGJ,
Vol. 17, pp. 485–498, 1980.
Goodman, R. E., Introduction to Rock Mechanics, John Wiley and Sons, New York, 478 pp., 1980.
Haefeli, R., “Creep and progressive failure in snow, soil, rock, and ice,” 5th INCSMFE, Vol. 3,
p. 134, 1965.
Hall, C. E., “Compacting a dam foundation by blasting,” ASCE, JSMFD, Vol. 88, pp. 33–51, 1962.
Hanna, A. M. and G. G. Meyerhof, “Design charts for ultimate bearing capacity for sands,” Overlying
Clays, CGJ, Vol. 17, pp. 300–303, 1980.
Hansen, J. B., Earth Pressure Calculation, Danish Technical Press, Copenhagen, 1953.
Hansen, J. B., “A revised and extended formula for bearing capacity,” Danish Geotechnical Institute,
Bulletin, Vol. 28, Copenhagen, 21 pp., 1970.
Hazen, A., “Physical properties of sand and gravels with reference to their use in filtration,” Rep. Mass.
State Board of Health, p. 539, 1892.
Hazen, A., Discussion of “Dams on Sand Foundations,” edited by A. C. Koenig, Trans. ASCE, Vol. 73, p.
199, 1911.
Heller, L. W., “Failure modes of impact-loaded footings on dense sand,” Technical Report R-281, U.S.
Naval Civil Eng. Laboratory, Port Hueneme, California, USA, 1964.
Hendron, A. J., “Mechanical properties of rock,” In: Rock Mechanics, edited by K. G. Stagg and O. C.
Zienkiewicz, John Wiley and Sons, New York, pp. 21–53, 1968.
Hoek, E., “Brittle failure of rocks,” In: Rock Mechanics, edited by K. G. Stagg and O. C. Zienkiewicz,
John Wiley and Sons, New York, pp. 99–124, 1968.
Hoek, E. and J. A. Franklin, “Sample triaxial cell for field or laboratory testing of rock,” Trans. Section
A, Inst. Min. Metal., Vol. 77, pp. A22–A26, 1968.

Holtz, W. G. and H. J. Gibbs, “Engineering properties of expansive clays,” Trans. ASCE, Vol. 121,
pp. 641–677, 1956.
Hough, B. K., Basic Soil Engineering, Ronald Press Company, New York, 1969.
Housel, W. S., “A practical method for the selection of foundations based on fundamental research in
soil mechanics,” Univ. Michigan Eng. Res. Bull., Vol. 13, 1929.
Hubert, M. K. and W. Rubey, “Mechanics of fluid-filled porous solids and its application to overthrust
faulting,” Bull. Geol. Soc. Am., Vol. 70, p. 115, 1959.
Hunt, R. E., Geotechnical Engineering Techniques and Practices, McGraw-Hill, New York, 729 pp., 1986.
Hutchinson, J. N., “The free degradation of London clay cliffs,” Proc. Geotechnical Conf., Oslo, Norway,
Vol. 1, p. 113, 1967.
Hvorslev, M. J., “Subsurface exploration and sampling of soils for civil engineering purposes,” Water-
ways Exp. Sta., Vicksburg, Mississippi, 521 pp., 1948.
Hvorslev, M. J., “Physical components of the shear strength on saturated clays,” Proc., Res. Conf. on
Shear Strength of Cohesive Soils, ASCE, Boulder, pp. 169–273, 1960.
IRC: Specification for Road and Bridge Works, Indian Roads Congress, New Delhi, pp. 167–181, 1986.
IRC: Standard Specification and Code of Practice for Road Bridges, Section VII, Indian Roads Congress, New
Delhi, 1987.
IS: 401, Code of Practice for Preservation of Timber, BIS, 1967.
IS: 456, Code of Practice for Plain and Reinforced Concrete, BIS, 1978.
IS: 1080, Design and Construction of Shallow Foundations in Soils (Other than Raft, Ring and Shell), BIS, 1985.
IS: 1498, Classification and Identification of Soils for General Engineering Purposes, BIS, 1970.
IS: 1888, Method of Load Test on Soils, BIS, 1982.
IS: 1892, Code of Practice for Subsurface Investigation for Foundations, BIS, 1979.
IS: 1893, Criteria for Earthquake Resistant Design of Structures, BIS, 1975.
IS: 1904, Design and Construction of Foundations in Soils, General Requirements, BIS, 1986.
IS: 2131, Method for Standard Penetration Test for Soils, BIS, 1981.
IS: 2132, Code of Practice for Thin-Walled Tube Sampling of Soils, BIS, 1986.
IS: 2720, Methods of Test for Soils, BIS.
IS: 2720 – Part 1, Preparation of Dry Soil Samples for Various Tests, BIS, 1983.
IS: 2720 – Part 2, Determination of Water Content, BIS, 1973.
IS: 2720 – Part 3/Sec. 1, Determination of Specific Gravity – Fine Grained Soils, BIS, 1980.
IS: 2720 – Part 3/Sec. 2, Determination of Specific Gravity – Fine, Medium and Coarse Grained Soils, BIS, 1981.
IS: 2720 – Part 4, Grain Size Analysis, BIS, 1975.
IS: 2720 – Part 5, Determination of Liquid and Plastic Limits, BIS, 1970.
IS: 2720 – Part 6, Determination of Shrinkage Factors, BIS, 1972.
IS: 2720 – Part 7, Determination of Water Content – Dry Density Relation Using Light Compaction, BIS, 1974.
IS: 2720 – Part 8, Determination of Water Content – Dry Density Relation Using Heavy Compaction, BIS, 1983.
IS: 2720 – Part 9, Determination of Dry Density – Moisture Content Relation by Constant Weight of Soil
Method, BIS, 1971.
IS: 2720 – Part 10, Determination of Unconfined Compressive Strength, BIS, 1973.
IS: 2720 – Part 11, Determination of the Shear Strength Parameters of a Specimen Tested in Unconsolidated
Undrained Triaxial Compression Without the Measurement of Pore Water Pressure, BIS, 1971.
IS: 2720 – Part 12, Determination of Shear Strength Parameters of Soil from Consolidated Undrained Triaxial
Compression Test with Measurement of Pore Water Pressure, BIS, 1981.
IS: 2720 – Part 13, Direct Shear Test, BIS, 1972.
IS: 2720 – Part 14, Determination of Density Index (Relative Density) of Cohesionless Soils, BIS, 1983.
IS: 2720 – Part 15, Determination of Consolidation Properties, BIS, 1986.
IS: 2720 – Part 16, Laboratory Determination of CBR, BIS, 1987.
IS: 2720 – Part 17, Laboratory Determination of Permeability, BIS, 1986.
IS: 2720 – Part 20, Determination of Linear Shrinkage, BIS, 1966.

Bibliography 849
IS: 2720 – Part 21, Determination of Total Soluble Solids, BIS, 1977.
IS: 2720 – Part 22, Determination of Organic Matter, BIS, 1972.
IS: 2720 – Part 23, Determination of Calcium Carbonate, BIS, 1976.
IS: 2720 – Part 28, Determination of Dry Density of Soils, In-place, by the Sand Replacement Method, BIS, 1974.
IS: 2720 – Part 29, Determination of Dry Density of Soils, In-place by the Core-Cutter Method, BIS, 1975.
IS: 2720 – Part 30, Laboratory Vane Shear Test, BIS, 1980.
IS: 2720 – Part 31, Field Determination of CBR, BIS, 1990.
IS: 2720 – Part 33, Determination of the Density of Soil, In-place, by the Ring and Water Replacement Method,
BIS, 1971.
IS: 2720 – Part 34, Determination of the Density of Soil, In-place, by Rubber-Balloon Method, BIS, 1972.
IS: 2720 – Part 35, Measurement of Negative Pore Water Pressure, BIS, 1974.
IS: 2720 – Part 36, Laboratory Determination of Permeability of Granular Soils (Constant Head), BIS, 1975.
IS: 2720 – Part 38, Compaction Control Test (Hilf Method), BIS, 1977.
IS: 2720 – Part 39/Sec. 1, Direct Shear Test for Soils Containing Gravel-Laboratory Test, BIS, 1977.
IS: 2720 – Part 40, Determination of Free Swell Index of Soils, BIS, 1977.
IS: 2911 – Part 1/Sec. 1, Design and Construction of Pile Foundations – Driven Cast In Situ Concrete Piles,
BIS, 1979.
IS: 2911 – Part 1/Sec. 3, Design and Construction of Pile Foundations – Driven Pre-cast Piles, BIS, 1979.
IS: 2911 – Part 2, Code of Practice for Design and Construction of Pile Foundations – Timber Piles, BIS, 1980.
IS: 2911 – Part 3, Code of Practice for Design and Construction of Pile Foundations – Under Reamed Piles,
BIS, 1980.
IS: 2911 – Part 4, Load Test on Piles, BIS, 1985.
IS: 2968 – Part 1, Dynamic Method Using 50 mm Cone Without Bentonite Slurry, BIS, 1976.
IS: 2974, Code of Practice for Design and Construction of Machine Foundations, BIS.
IS: 2974 – Part 1, Foundations for Reciprocating Type Machines, BIS, 1982.
IS: 2974 – Part 2, Foundations for Impact Type Machines (Bammer Foundations), BIS, 1980.
IS: 2974 – Part 3, Foundations for Rotary Machines (Medium and High Frequency), BIS, 1975.
IS: 2974 – Part 4, Foundations for Rotary Type Machines for Low Frequency, BIS, 1979.
IS: 2974 – Part 5, Foundations for Impact Type Machines other than Hammers (Forging and Stamping Press,
Pig Breaker Elevator and Hoist Tower), BIS, 1970.
IS: 3764, Safety Codes for Excavation Work, BIS, 1966.
IS: 4434, Code for Practice for In Situ Vane Shear Test for Soils, BIS, 1978.
IS: 4453, Code of Practice for Subsurface Exploration by Pits, Trenches, Drifts and Shafts, BIS, 1980.
IS: 4651 – Part 1, Planning and Design of Ports and Harbours, Code of Practice – Site Investigation, BIS, 1974.
IS: 4968 – Part 2, Dynamic Method Using Cone and Bentonite Slurry, BIS, 1976.
IS: 4968 – Part 3, Static Cone Penetration Test, BIS, 1976.
IS: 5529, In Situ Permeability Tests, BIS.
IS: 5529 – Part 1, Code of Practice for In Situ Permeability Test – Tests in Overburden, BIS, 1985.
IS: 6403, Code of Practice for Determination of Bearing Capacity of Shallow Foundations, BIS, 1981.
IS: 6955, Code of Practice for Subsurface Exploration for Earth and Rockfill Dams, BIS, 1973.
IS: 7317, Code of Practice for Uniaxial Jacking Test for Deformation Modulus of Rock, BIS, 1974.
IS: 7894, Code of Practice of Stability Analysis of Earth Dams, BIS, 1975.
IS: 8009 – Part 1, Shallow Foundations subjected to Symmetrical Static Vertical Loads, BIS, 1976.
IS: 8237, Code of Practice for Protection of Slope for Reservoir Embankments, BIS, 1976.
IS: 8414, Guidelines for Design of Under-Seepage Control Measures for Earth and Rockfill Dams, BIS, 1977.
IS: 8763, Guide for Undisturbed Sampling of Sands, BIS, 1978.
IS: 8764, Method for Determination of Point Load Strength Index of Rocks, BIS, 1978.
IS: 8826, Guidelines for Large Earth and Rockfill Dams, BIS, 1978.
IS: 9143, Method for the Determination of Unconfined Compressive Strength of Rock Materials, BIS, 1979.

IS: 9221, Method of Determination of Modulus of Elasticity and Poisson’s Ratio of Rock Materials in “Uniaxial
Compression,” BIS, 1979.
IS: 9429, Code of Practice for Drainage System for Earth and Rockfill Dams, BIS, 1980.
IS: 9640, Specification for Split Spoon Sampler, BIS, 1980.
IS: 9759, Guidelines for Dewatering During Construction, BIS, 1981.
IS: 10060, Code of Practice for Subsurface Investigation for Power Houses Sites, BIS, 1981.
IS: 10082, Method of Test for Determination of Tensile Strength by Indirect Tests on Rock Specimens, BIS, 1981.
IS: 10108, Code of Practice for Sampling for Soils by Thin Wall Sampler with Stationary Piston, BIS, 1982.
IS: 10379, Code of Practice for Field Control of Moisture and Compaction of Soils for Embankment and Sub-
grade, BIS 1982.
IS: 11385, Code of Practice for Subsurface Exploration for Canals and Cross Drainage Works, BIS, 1985.
IS: 11594, Specifications for Mild Steel Thin Walled Sampling Tubes and Sampler Heads, BIS, 1985.
IS: 12070, Code of Practice for Design and Construction of Shallow Foundations on Rocks, BIS, 1987.
Iyer, T. S. R., “Marine deposits,” Proc. 5th Asian Regional Conference, Vol. 1, pp. 35–69, 1975.
Iyer, T. S. R. and S. Padmanabha Pillai, “A study of the in situ stresses in lateritic profile,” Proc. Sympo-
sium on Strength and Deformation Behaviour of Soils, Bangalore, Vol. 1, pp. 159–163, 1972.
Jaeger, J. C. and N. G. W. Cook, Fundamentals of Rock Mechanics, Chapman & Hall, 515 pp., 1969.
Janbu, N., “Application of composite slip surface for stability analysis,” European Conference on Stability
of Earth Slope, Stockholm, Discussion, Vol. 3, pp. 43–49, 1954.
Janbu, N. “Earth pressure and bearing capacity, calculations by generalized procedure of slices,” 4th
INCSMFE, Vol. 2, pp. 207–212, 1957.
Janbu, N., L. Bjerrum, and B. Kjaernsli, Veildring ved losing av fundamentering soppgaver, Norwegian
Geotechnical Institute, Publ. No. 16, Oslo, 93 pp., 1956.
John, K. W., “An approach to rock mechanics,” ASCE, ISMFD, Vol. 88, pp. 1–30, 1962.
Jones, C. J. F. P., Earth Reinforcement and Soil Structures, Butterworth, London, 183 pp., 1985.
Jumikis, A. R., Soil Mechanics, Van Nostrand, New York, 576 pp., 1962.
Kaniraj, S. R., “Limiting of immediate settlement of shallow foundations,” IGJ, Vol. 7, pp. 235–247, 1977.
Kaniraj, S. R. and B. V. Ranganathan, “Limiting of settlement of shallow foundations in normally con-
solidated clay, IGJ, Vol. 7, pp. 159–177, 1977.
Katti, R., K. Kulkarni, V. S. Chandrasekaran, G. Venkatachalan, and D. M. Dewaikar, “Regional soil
deposits of India, state-of-the-art report,” Proc. 5th Asian Regional Conference, Bangalore, Vol. 2,
pp. 35–52, 1975.
Ko, H. Y. and L. W. Davidson, “Bearing capacity of footings in plane strain,” ASCE, JSMFD, Vol. 99,
pp. 1–23, 1973.
Koerner, R. M., Construction and Geotechnical Methods in Foundation Engineering, McGraw-Hill,
New York, 496 pp., 1985.
Kotter, F., Die Bestimmung des bruckes an gekrummten Geeitflachen, Sitzugsber, kgl, Preuss, Akad. der
Wiss, Berlin, 1903.
Kravetz, G. A., “Cement and clay grouting of foundations, the use of clay in pressure grouting,” ASCE,
JSMFD, Vol. 84, pp. 1546–1-30, 1958.
Krishnaswamy, N. R., “Study of surcharge effects on footings vibrations,” Geotechnical Engineering, Vol.
16, pp. 61–64, 1975.
Krishnaswamy, N. R. and M. Anandakrishnan, “Influence on soil moisture on footing vibrations,”
Proc. 5th Asian Regional Conference, Bangalore, Vol. 1, pp. 295–298, 1975.
Ladd, C. C., Strength and Compressibility of Saturated Clays, Pan American Soils Course, Universidad
Catolica Andres Bello, Caracas, Venezuela, 1967.
Lade, P. J. and K. L. Lee, Engineering Properties of Soils, Soil Mech. Laboratory, UCLA-Eng. 7652, 1976.
Lambe, T. W., “How dry is a dry soil,” HRB, p. 491, 1949.
Lambe, T. W., Soil Testing for Engineers, Wiley Eastern, New Delhi, 165 pp., 1951.
Lambe, T. W., The Structure of Inorganic Soils, ASCE, STP No. 315, Vol. 79, 1953.

Bibliography 851
Lambe, T. W., The Permeability of Compacted Fine Grained Soils, ASTM, STP No. 163, 1955.
Lambe, T. W., The Engineering Behaviour of Compacted Clay, ASCE, JSMFD, Vol. 84, pp. 1665-1-35,
1958.
Lambe, T. W., “A mechanistic picture of shear strength in clay,” Proc. Research Conf. on Shear Strength of
Cohesive Soils, ASCE, Boulder, p. 437, 1960.
Lambe, T. W., “Soil stabilization,” In: Foundation Engineering, Edited by G. A. Leonards, McGraw-Hill,
New York, pp. 351–437, 1962.
Lambe, T. W., “Stress path method,” ASCE, JSMFD, Vol. 93, pp. 309–331, 1967.
Lambe, T. W. and R. V. Whitman, Soil Mechanics, Spl. version, Wiley Eastern, New Delhi, 553 pp., 1979.
Lane, K. S. and D. E. Washburn, “Capillarity tests by capillarimeter and by soil filled tubes,” HRB,
p. 1, 1946.
Layman, A. K. B., “Compaction of cohesionless foundation soils by explosives,” Trans. ASCE, Vol. 107,
pp. 1330–1348, 1942.
Leonards, G. A., “Engineering properties of soils,” In: Foundation Engineering, edited by G. A. Leonards,
Leonards, G. A. and B. K. Ramiah, “Time effects in the consolidation of clay,” Symposium on Time Rates
of Loading in Testing Soils, ASTM, STP No. 254, p. 116, 1959.
MacDonald, D. H. and A. W. Skempton, “A survey of comparisons between calculated and observed
stresses and displacements, ICE, London, pp. 318–337, 1955.
Malhotra, B. R. and D. Chandra, “Sub-sieve particle size analysis by different methods,” IGJ, Vol. 12,
pp. 271–280, 1982.
Mansur, C. I. and R. I., Kaufman, “Dewatering,” In: Foundation Engineering, edited by G. A. Leonards,
Marshall, T. J., A Plummet Balance for Measuring the size Distribution of Soil Particles, CSIRO Publication,
pp. 142–147, 1956.
McCarthy, D. F., Essentials of Soil Mechanics and Foundations, Reston Publishing Company, Virginia, 632
pp., 1982.
Menard, L. and Y. Broise, “Theoretical and practical aspects of dynamic consolidation,” GTECH, Vol.
25, pp. 3–18, 1975.
Meyerhof, G. G., “The ultimate bearing capacity of foundations,” GTECH, Vol. 2, pp. 301–331, 1951.
Meyerhof, G. G., “The bearing capacity of foundations under eccentric and inclined loads,”, 3rd
ICSMFE, Vol. 1, pp. 440–445, 1953.
Meyerhof, G. G., “Penetration tests and bearing capacity of cohesionless soils,” ASCE, JSMFD, Vol. 82,
pp. 1–19, 1956.
Meyerhof, G. G., “The ultimate bearing capacity of foundations on slopes,” 4th IMCSMFE, Vol. 1,
pp. 384–386, 1957.
Meyerhof, G. G., “Some recent research on the bearing capacity of foundations,” CGJ, Vol. 1, pp. 16–
26, 1963.
Meyerhof, G. G., “Shallow foundations,” ASCE, JSMFD, Vol. 91, pp. 21–31, 1965.
Meyerhof, G. G., “Ultimate bearing capacity of footings on sand layer overlaying clay,” CGJ, Vol. 11,
pp. 223–229, 1974.
Meyerhof, G. G., “Bearing capacity and settlement of pile foundations,” ASCE, JGED, Vol. 102,
pp. 195–228, 1976.
Meyerhof, G. G. and Adams, J. I., “The ultimate uplift of capacity of foundations,” JSMFD, ASCE,
Vol. 58, SM 6, pp. 1–29, 1968.
Michaels, A. S. and C. S. Lin, “The permeability of kaolinite,” Ind. Eng. Chem., Vol. 46, pp. 1239–
1246, 1954.
Mitchell, J. K., “In-place treatment of foundation soils,” ASCE, JSMFD, Vol. 96, pp. 73–110, 1970.
Mitchell, J. K., “Stabilization of soils for foundation of structures,” Report prepared for U.S. Army
Waterways Experiment Station, pp. 1–34, 1976.

Mohan, D., “A close look at problems of research and its application to pile foundations,” Third IGS
Annual Lecture, IGJ, Vol. 11, pp. 1–41, 1981.
Mohan, D., G. S. Jain, and D. Sharma, “Bearing capacity of multiple under-reamed bored piles,” Proc.
3rd Asian Conf. on SM & FE, Haifa, Vol. 1, pp. 103–106, 1967.
Mohan, D., G. S. Jain, D. P. Sen Gupta, and Devendra Sharma, “Consolidation of ground by vertical
rope drains,” IGJ, Vol. 7, pp. 106–115, 1977.
Mohan, D., V. N. S. Murthy, and G. S. Jain, “Design and construction of multi under-reamed piles,”
Proc. 7th INCSMFE, Vol. 2, pp, 183–186, 1969.
Mohr, O., Uber die Darstellung des spannun-gazustandes und des Deformation-zustandes eines korper-
elements, Zivilingenieur, 1882.
Montgomery, C. W., Physical Geology, WMC Brown Publishers, New York, 2nd Ed., 555 pp., 1990.
Moore, P. J., “Vibration criteria,” In: Analysis and Design of Foundations for Vibrations, edited by
P. J. Moore, Oxford and IBH Publishing Co., New Delhi, pp. 1–20, 1985.
Moore, P. J., “Design of shallow foundations,” In: Analysis and Design of Foundations for Vibrations,
edited by P. J. Moore, Oxford, IBH Publishing Co., New Delhi, pp. 1–20, 1985.
Morgenstern, N. R., “Stability charts for earth slopes during rapid drawdown,” GTECH, Vol. 3,
pp. 121–131, 1963.
Morgenstern, N. R. and V. E. Price, “The analysis of the stability of general slip surface,” GTECH,
Vol. 15, pp. 79–93, 1965.
Morgenstern, N. R. and K. D. Eigenbrod, “Classification of argillaceous soils and rocks,” ASCE, JGED,
Vol. 100, pp. 1137–1158, 1974.
Morgenstern, N. R. and V. E., Price, “A numerical method for solving the equations of stability of gen-
eral slip surface,” Comp. J., Vol. 9, p. 388, 1967.
Nagaraj, T. S., A. Sridharan, and M. V. Paul Alexander, “In situ reinforced earth – An approach for deep
excavation,” IGJ, Vol. 12, pp. 101–111, 1982.
Narain, J., B. Singh, and P. Purushothama Raj, “The strain energy method of stability analysis for gen-
eral slip surface,” Proc. 4th Asian Regional Conference of SMFE, Bangkok, Vol. 1, pp. 151–157, 1971.
Newmark, N. M., Influence Charts for Computation of Stresses in Elastic Foundations, University of Illinois,
Bull. No. 338, 1942.
Nonveiller, E., “The stability of slopes of dams composed of heterogeneous material,” 3rd ICSMFE,
Vol. 2, p. 268, 1953.
Nonveiller, E., “The stability analysis of slopes with a slip surface of general shape,” 6th ICSMFE, Vol.
2, pp. 522–525, 1965.
Osterberg, J. O., “New piston type soil sampler,” ENR, Vol. 148, pp. 77–78, 1952.
Osterberg, J. O., “Influence values for vertical stresses in a semi-infinite mass due to an embankment
loading,” 4th INCSMFE, Vol. 1, pp. 393–394, 1957.
Peck, R. B., “Deep excavations and tunneling in soft ground, state-of-the art report, 7th INCSMFE, Vol.,
pp. 225–290, 1969.
Peck, R. B., W. E. Hanson, and T. H. Thoruburn, Foundation Engineering, 2nd Ed., John Wiley & Sons,
New York, 514 pp., 1947.
Petterson, K. E., “The early history of circular sliding surfaces,” GTECH, Vol. 5, pp. 275, 1955.
Poncelet, J. V., Memoire sur la stabilite des revetments er de loure foundations. Note addition elle sur les relatins
analytiques qui lient entre elles la poussee et la butee de la terre. Memorial de l’officiere de genie, Paris,
Vol. 13, 1840.
Ponnuswamy, N., Bridge Engineering, Tata McGraw-Hill, New Delhi, 544 pp., 1986.
Poulos, H. G. and E. H. Davis, Pile Foundation Analysis and Design, John Wiley & Sons, 397 pp., 1980.
Proctor, R. R., “Design and construction of rolled earth dams (four articles),” ENR, Vol. 3, pp. 245–248,
286–289, 348–351, 372–376, 1933.
Prakash, S., Introductory Soil Testing, Asia Publishing House, New Delhi, 103 pp., 1969.

Bibliography 853
Prakash, S., Soil Dynamics, McGraw-Hill, New York, 426 pp. 1981.
Prandtl., “Uber die Hartc Plastischer Kirper,” Nachr. Kgl. Gass Wiss. Gottingen Math, Ply, K1, 1920.
Purushothama Raj, P., “Stability analysis of slopes with general slip surface,” M.E. Dissertation sub-
mitted to University of Roorkee, Roorkee, 77 pp., 1967.
Purushothama Raj, P., B. K. Ramiah, and K. Basavarajappa, “Bearing capacity of eccentrically loaded
strip footing on two-layer cohesive soils,” Proc. 4th Southeast Asian Conference on Soil Engineering,
Malaysia, Vol. 1, pp. 63–75, 1975.
Purushothama Raj, P., B. K. Ramiah, and K. N. Venkatakrishna Rao, “Limit analysis for bearing ca-
pacity of shallow foundations,” Proc. Symp. on Strength and Deformation Behaviour of soils, Vol. 1,
pp. 191–196, 1972.
Purushothama Raj, P., B. K. Ramiah, and K. N. Venkatakrishna Rao, “Bearing capacity of strip footings
in two-layered cohesive friction soils,” CGJ, Vol. 11, pp. 32–45, 1974.
Ramana, Y. V. and B. Venkatanarayana, “An air porosimeter for the porosity of rocks,” Int. J. Rock Mech.
Min. Sci., Vol. 8, pp. 29–53, 1971.
Ramana Sastry, M. V. B., S. G. Kumar, “Strength characteristics of lime stabilised clay–sand mixes,” IGJ,
Vol. 19, pp. 87–103, 1989.
Ramasamy, G., “Estimation of settlement of footings of sand – A critical reappraisal,” IGJ, Vol. 14,
pp. 112–126, 1984.
Ramasamy, G., “Settlement of footings on normally loaded and preloaded sand beds,” IGJ, Vol. 16, pp.
364–382, 1986.
Ramasamy, G., A. S. R. Rao, and C. Prakash, “Effect of embedment on settlement of footings on sand,”
IGJ, Vol. 12, pp. 112–131, 1982.
Ramasamy, G., A. S. R. Rao, and P. S. Singh, “Influence of capillary zone on settlement of footings on
sand,” IGJ, Vol. 16, pp. 383–393, 1986.
Ramiah, B. K., L. S. Chickanagappa, and T. N. Ramamurthy, “Vertical vibrations of embedded foot-
ings,” Proc. 9th INCSMFE, Tokyo, pp. 343–346, 1977.
Ramiah, B. K., Purushothama Raj, and V. A. Kumaresan, “Stability analysis of slopes subjected to
progressive failure,” Proc. 3rd Southeast Asian Conference on soil Engineering, Hong Kong, Vol. 1,
pp. 171–177, 1972.
Ranjan, G., “Ground treated with granular piles and its response under load,” 11th IGS Annual Lecture,
IGJ, Vol. 19, pp. 1–86, 1989.
Ranjan, G. and B. G. Rao, “Skirted granular piles for ground improvement,” Proc. 8th European Confer-
ence on Soil Mechanics and Foundation Engineering, Helsinki, 1983.
Rankine, W. J. M., “On the stability of loose earth,” Philos. Trans. Royal Soc., London, Vol. 147,
pp. 9–27, 1857.
Rao, K. S. and B. C. Raymahashay, “Influence of clay minerals and iron oxides on selected properties
of two lateritic soils,” IGJ, Vol. 11, pp. 25–266, 1981.
Rassner, H., “Zum Erddruckproblem,” Proc. 1st Int. Conf. of Applied Mechanics, Delft, pp. 295–311, 1924.
Rendulic, L., “Der hydrodynamische Spannungsausgleich in Zentral entwasserten Tonzylinder,” Was-
serwirtschaft und Technik, Vol. 2, pp. 250–253, 269–273, 1935.
Resenqvist, I. Th., “Physico-chemical properties of Soil–water systems,” ASCE, JSMFD, Vol. 85,
p. 31, 1959.
Richart, F. E., “Review of the theories for sand drains,” Trans. ASCE, Vol. 124, pp. 709–736, 1957.
Richart, F. E., Jr., “Foundation vibrations,” Trans. ASCE, Vol. 127, pp. 863–898, 1962.
Rowe, P. W., “Anchored sheet pile walls,” PICE, Vol. 1, pp. 27–70, 1952.
Rowe, P. W., “Sheet pile walls in clay,” PICE, Vol. 7, pp. 629–654, 1957.
Schmertmann, J. M., “The undisturbed consolidation of clay,” Trans. ASCE, Vol. 120, pp. 1201–1233, 1955.
Schmertmann, J. M., “The measurement of in situ shear strength,” Spec. Conf. on In Situ Measurement of
Soil Properties, ASCE, Raleigh, Vol. 2, pp. 57–138, 1975.

Serafim, J. L., “Influence of interstitial water on the behaviour of rock masses,” In: Rock Mechanics,
edited by K. G. Stagg and O. C. Zienkiewicz, John Wiley & Sons, New York, pp. 55–97, 1968.
Serdengecti, S. and G. D. Boozer, “The effects of strain rate and temperature on the behaviour of rocks
subjected to triaxial compression,” Proc. Symp. Rock. Mech. 4th Bull., Mineral Ind. Expr. Sta., Pennsyl-
vania State University, No. 76, 1961.
Sen Gupta, D. P., B. Dey, and D. Rey, “Substrata treatment for a building using rope drains,” IGJ, Vol.
10, pp. 322–331, 1980.
Sherard, J. L., R. J. Woodward, S. P. Gizienzki, and W. A. Clevenger, Earth and Earth-Rock Dams, John
Wiley & Sons, New York, 725 pp., 1963.
Siva Reddy, A. and G. Mogaliah, “Bearing capacity of strip footings on partially saturated soils,” IGJ,
Vol. 6, pp. 220–245, 1976.
Siva Reddy, A. and R. J. Srinivasan, “Bearing capacity of footings on layered soils,” ASCE, JSMFED,
Vol. 93, pp. 83–99, 1967.
Skempton, A. W., “The bearing capacity of clays,” Proc. Building Research Congress, Vol. 1, pp. 180–
189, 1951.
Skempton, A. W., “The colloidal activity of clays,” 3rd ICSMPE, Vol. 1, pp. 57–61, 1953.
Skempton, A. W., “The pore pressure coefficients A and B,” GTECH, Vol. 4, pp. 143–147, 1954.
Skempton, A. W., “Settlement analysis of six structures in Chicago and London,” PICE, Vol. 4,
pp. 525–544, 1955.
Skempton, A. W., “Discussion of the planning and design of the new Hong Kong airport,” PICE,
Vol. 7, pp. 305–307, 1957.
Skempton, A. W., Effective Stress in Soils, Concrete and Rocks, Pore Pressure and Suction in Soils, Butter-
worth, London, p. 4, 1961.
Skempton, A. W., “Long-term stability of clay slopes,” GTECH, Vol. 14, pp. 77–101, 1964.
Skempton, A. W., “Some observation on tectonic shear zones,” Proc. 1st Int. Conf. on Rock Mechanics,
Vol. 1, p. 329, 1966.
Skempton, A. W. and L. Bjerrum, “A contribution to the settlement analysis of foundations on clay,”
GTECH, Vol. 7, pp. 168–178, 1957.
Skempton, A. W. and J. Hutchinson, “Stability of natural slopes and embankment foundation,” 7th
INCSMFE, State of Art Volume, pp. 291–340, 1969.
Skempton, A. W. and D. J. Petley, “The strength along structural discontinuities in striff clays,” Geotech-
nical Conf., Oslo, Vol. 2, pp. 29–46, 1967.
Smith, E. A., “Pile driving analysis by the wave equation,” Trans. ASCE, Vol. 127, pp. 1145–1193, 1962.
Som, N., “Regional deposits,” Proc. 5th Asian Regional Conference, Bangalore, Vol. 2, pp. 70–81, 1975.
Sowers, G. F., “Engineering properties of residual soils derived from igneous and metamorphic rocks,”
Proc. 2nd Pan. Am. Conf. on Soil Mechanics and Foundation Engineering, Brazil, Vol. p. 39, 1963.
Sowers, S. F., Introductory Soil Mechanics and Foundations: Geotechnical Engineering, 4th Ed., Macmillan,
New York, 621 pp., 1979.
Sridharan, A., “Some studies on the strength of partially saturated clays,” Ph.D. Thesis, Purdue
University, Indiana, 1968.
Sridharan, A. and G. Venkatappa Rao, “Mechanics controlling volume change of saturated clays and
the role of the effective stress concept,” GTECH, Vol. 23, pp. 359–382, 1973.
Sridharan, A. and G. Venkatappa Rao, “Shear strength behaviour of saturated clays and the role of the
effective stress concept,” GTECH, Vol. 29, pp. 177–193, 1979.
Stagg, K. G. and O. C. Zienkiweicz, Rock Mechanics in Engineering Practice, 12 contributing authors,
John Wiley & Sons, New York, 442 pp., 1968.
Stamatopoulos, A. S. and P. C. Kotzias, Soil Improvement by Preloadings, John Wiley & Sons, New York,
261 pp., 1985.
Swami Saran, Soil Dynamics and Machine Foundations, Galgotia Publ., New Delhi, 486 pp., 1999.
Taylor D. W., “Stability of earth slopes,” J. Boston Soc. Civil Eng., Vol. 24, pp. 197–246, 1937.

Bibliography 855
Taylor D. W., “Review of pressure distribution theories, earth pressure cell investigations and pressure
distribution data,” Report to U.S. Army Waterways Experimental Station, 1945.
Taylor, D. W., Fundamentals of Soil Mechanics, Asia Publishing House, New Delhi, 700 pp., 1948.
Teng, W. C., Foundation Design, Prentice Hall, New Jersey, 466 pp., 1962.
Terzaghi, K., Erdaumecbanik, Franz Deuticke, Vienna, 1925.
Terzaghi, K., Theoretical Soil Mechanics, John Wiley & Sons, New York, 510 pp., 1943.
Terzaghi, K., “Conditions for the failure of concrete and rock,” Proc. ASTM, Vol. 45, p. 777, 1945.
Terzaghi, K., “Introduction to tunnel geology,” In: Rock Tunneling with Steel Supports, edited by R. Proctor
and T. White, Commercial Shearing and Stamping Co., Youngstown, Ohio, pp. 17–99, 1946.
Terzaghi, K., “Mechanism of landslides,” In: Application of Geology in Engineering Practice, Berkeley Vol.,
Geological Society of America, pp. 83–123, 1950.
Terzaghi, K. and R. B. Peck, Soil Mechanics in Engineering Practice, 2nd Ed., John Wiley & Sons,
New York, 729 pp., 1967.
Tiedemann, B., “Ueber die schabfestigkeit, Bindiger Boden,” Die Bautechnik, Vol. 15, pp. 400, 403, 1937.
Tolia, D. S., “The interpretation of static cone penetration tests,” IGJ, Vol. 8, pp. 152–168, 1978.
Tomlinson, M. J., Foundation Design and Construction, 5th Ed., English Language Book Society, Essex,
842 pp., 1986.
Thorburn, S., “Discussion,” In: Proceedings of the Conference on the Behaviour of Piles, Institution of Civil
Engineers, London, p. 54, 1971.
Tourenq, C. and A. Denis, The Tensile Strength of Rocks (in French), Lab de ponts et Chausses Paris,
Research Rep. No. 4, 1970.
Tschebotarioff, G. P., “Large scale earth pressure tests with model flexible bulkheads,” GTECH, Vol. 1,
pp. 98–111, 1949.
Turnbull, W. J., “Compaction and strength tests on soil,” Annual Meeting, ASCE, January, 1950.
Turnbull, W. J., A. A. Maxwell, and R. G. Ahlvin, “Stresses and deflections on homogeneous soil
masses,” 5th INCSMFE, Vol. 2, 1961.
U.S. Department of Interior, Earth Manual, Bureau of Reclamation, Denver, 783 pp., 1963.
Van Olphen, H., An Introduction to Clay Colloid Chemistry, Interscience Publishers, New York, 301
pp., 1963.
Vesic, A. S., “Bearing capacity of deep foundations in sand,” Highway Research Record No. 39, HRB,
pp. 112–153, 1963.
Vesic, A. S., “Experiments with instruments pile groups in sand,” ASTM, STP No. 444, pp. 177–222,
1969.
Vesic, A. S., “Analysis of ultimate loads of shallow foundations,” ASCE, JSMFD, Vol. 99, pp. 45–73, 1973.
Vesic, A. S., Design of Pile Foundations, National Cooperative Highway Research Program (NCHRP), Synthe-
sis of Practice, No. 42, Transportation Research Board, Washington, D.C., 68 pp., 1977.
Vesic, A. S., D. C. Bands, and J. M. Woodard, “An experimental study of dynamic bearing capacity of
footings on sand,” Proc. 6th INCSMFE, Montreal, Vol. 2, pp., 209–213, 1965.
Vidal, H., The Principle of Reinforced Earth, Highway Research Record No. 282, pp. 1–16, 1969.
Vijayvergiya, V. N. and J. A. Focht, Jr., “A new way to predict capacity of piles in clay,” 4th Offshore
Technology Conference, OTC Paper No. 1718, Houston, 1972.
Vogler, U. V. and K. Kovari, “Suggested methods for determining the strength of rock materials in
triaxial compression, ISRM Commission on Standardization of Laboratory and Field Tests,” Int.
J. Rock Mech. Min. Sci. Geomech. Abstr., Vol. 15, pp. 47–51, 1978.
Walsh, J. B., “The effect of cracks in rocks on Poisson’s ratio,” J. Geophys. Res., Vol. 70, pp. 5249–5257, 1965.
Wawersik, W. R. and C. Fairhurst, “A study of brittle rock fracture in laboratory compression experi-
ments,” Int. J. Rock Mech. Min. Sci., Vol. 7, pp. 561–575, 1970.
Westergaard, H. M., “A problem of elasticity suggested by a problem in soil mechanics: Soft
material reinforced by numerous strong horizontal sheets,” In: Contributions to the Mechanics of Solids,
Stephen Timoshenko 60th Anniversary Volume, New York, 1938.

Whitaker, T., “Experiments with model piles in groups,” GTECH, Vol. 7, pp. 147–167, 1957.
Whitlow, R., Basic Soil Mechanics, Construction Press, London, 439 pp., 1983.
Wickham, G. E., H. R. Tiedemann, and E. H. Skinner, “Support determination based on geologic
predictions,” Proc. 1st N. Am. Tunneling Conf., AIME, New York, pp. 43–64, 1968.
Wiebols, G. A., J. C. Jaeger, and N. G. W. Cook, “Rock property tests in a stiff testing machine,” 10th
Rock Mech. Symp., Rice University, Houston, 1968.
Wineland, J. D., “Borehole shear device,” Special Conf. on In Situ Measurement of Soil Properties, ASCE,
Raleigh, Vol. 1, pp. 511–522, 1975.
Woods, K. B. (ed.), Highway Engineering Handbook, McGraw-Hill, New York, pp. 11-3 to 11-80, 1959.
Wuerker, R. G., “Influence of stress rate on the strength and elasticity of rocks,” Q. Colorado School
Mines, Vol. 54, 1959.
Yamaguchi, H., “Practical formula of bearing value for two-layered ground,” Proc. 2nd Asian
Regional Conference on SMFE, Tokyo, Vol. 1, pp. 176–180, 1963.
Zanten, V. V., Geotextiles and Geomembranes in Civil Engineering, A. A. Balkema, Rotterdam, 658 pp., 1986.

Index
AASTHO soil classification system, 73 types of, 651
acid drainage, 779 uses of, 650–651
acid rain, 779 California bearing ratio (CBR) test, 93, 373–379,
active stress, 388 passive stress, 388 821–823
aeolian deposits, 6 capillary conductivity, 120
air content. See air void ratio capillary head, 119
air void ratio, 27 carbon cycle, 779
allowable bearing pressure, 544–545 cased pile, 602
alluvial deposit, 5, 9 cement stabilization, 719–720
anchor piles, 604–605 chemical stabilization, 719–720
anisotropic soil conditions, 149–151 types of, 719–720
area ratio, 680 chemical weathering, 4
arid lands, 780 circular failure surfaces, 486–498
auger boring, 679 circular wells, 655
clay, 16–19, 223–225
bacteria physio-chemical behaviour of, 16–19
characteristics of, 17 thixotrophy of, 291–292
batter piles, 604 types of, 223–225
bearing capacity, 516–551 clay minerals, 10
criteria of, 516 closed caissons, 651
factors affecting, 516 clusters, 20
belled piers, 642 coarse-grained soils, 66
Bishop’s simplified method, 496–497 coastal margins, 780
black cotton soils, 7 coefficient of passive stress, 388
blasting, 705–706 cohesive soils, 20–21
block foundation, 771–773 sensitivity of, 290–291
methods of analysis for, 771–773 structure of, 20–21
bored compaction pile, 602 combined footings, 568
borehole shear device, 280 design of, 585–589
borings, 794 types of, 585–589
location, spacing and depth of, 694 combined strap footing, 588–589
Boussinesq’s theory, 181–184 compaction
braced excavations, 445–450 effect on soil structure, 92–93
Brinch Hansen’s bearing capacity theory, 537–540 factors affecting, 89–91
brittleness index, 265 principles of, 81–82
brucite sheet, 13 specification and control, 87–89
building cone compaction piles, 707–708
bearing capacity of soils from, 540–542 compactive effort, 82
bulk density, 30–31 complex slides, 474–475
bulk soil, 25 compound slides, 474
buoyant density. See submerged density compressibility, 217
Burmister method, 826–827 characteristics of, 221–223
coefficient of, 222
caissons, 650–655 compression index, 221
bearing capacity and settlement of, 651 cone penetration test, 547–549, 689–690
construction of, 651–655 confining pressure, 271
Z03_PURU1773_01_SE_IND.indd 857 2/12/2013 8:17:16 AM

858 Index
consolidation, 217 eccentric loading, foundations, 526–527

rate of, 226–233 effective stress concept, 171–173
test related to, 219–220 elastic half-space, 179
types of, 217 elastic theory, 198
consolidation settlement, 234–239, 578–579 validity of application, 198
rates related to, 234–239 electrical resistivity method, 673–674
consolidation test, 353–359 electrokinetic phenomenon, 18
constant head permeameter, 106–107 electrolytes, 16–17
contact pressure, 197–198 characteristics of, 16–17
contaminated soils, 781–782 electro-osmosis, 703–705
continuous footings, 568 embankment dams, 729–751
core-cutter method, 315–317 components of, 731–734
in-place dry density of soil by, 315–317 protection against free passage of water
Coulomb’s earth pressure theory, 400–402 through, 744–745
critical hydraulic gradient, 138 selection of site of, 751
Culmann’s graphical method, 402–404 stability of slope of, 745
curvature coefficient, 40 types of, 729–731
embankment loading, 190–191
Darcy’s law, 104–105 end-bearing piles, 603
deep-well drainage, 702–703 environmental geotechnology, 777–783
degree of consolidation, 222–223 erratics, 6
degree of saturation, 27, 117 eskers, 6
density index, 32–34 exchangeable ions, 16
desert soils, 9 extended straight shaft drilled piers, 642
deserts, 780
desiccated soil, foundations on, 530–531 factor of safety, 476–478, 487
design wheel load, 818–819 falling head permeameter, 107–108
deviator stress, 271 Fellenius method, 494–496
diffuse double layer, 16 field compaction, 84–86
dilatancy test, 66 field permeability test, 108–112
direct shear test, 267–269, 362–365 field test, 546–551
dispersed structure, 20 estimation of bearing capacity from, 546–551
dispersion test, 66 evaluation of settlement from, 580–582
domains, 20 filters, 154–155
double-D wells, 656 design of, 154–55
drainage methods, 701–705 fine-grained soils, 66
types of, 701–705 finite slope, 478–480
drifts, 684 analysis of, 483–486
drilled piers, 641–650 fissured clays, 21
bearing capacity of, 642–645 floating foundations, 570
constructive procedures of, 646–650 flocculated structure, 20
settlement of, 645–646 flocculation, 20
types of, 642 flow nets, 142–144
uses of, 641–642 construction of, 144–149
drumlins, 6 properties and applications of, 142–144
dry density, 29 flows, 474
dry soil samples, 310–311 fly-ash stabilization, 720
dry strength test, 66 foundation pressures, 525–526
durability, 791 framework silicate minerals, 15
friction circle method, 488–490
earth dams friction piles, 603
design criteria for, 734–750
earthquakes, 759–761 general flow equation, 139–141
earth-retaining structures, 425–450 general shear failure, 517

Index 859
geological cycle, 3 kelvin model, 216

geomembranes, 721–724
geostatic stresses, 173–179 laboratory compaction, 82–84
geotextiles, 721–724 laboratory permeability test, 106–108
gibbsite sheet, 13 lacustrine deposits, 5
glacial deposits, 5–6 lake deltas, 5
glacio fluvial deposits, 5 land movement, 475–476
grain size distribution curves, 39–40 rate of, 475–476
granular soils, 19, 292–294 land subsidence, 780
shear strength of, 292–294 landslides, 473–476
structure of, 19 types of, 473–476
gravity deposits, 6 Laplace equation, 141–142
gravity-type retaining walls, 425–432 lateral earth pressure, 383–407
ground investigation programme, 672–694 lateral forces, 627–628
ground moraine, 6 lateral stress ratio, 385–386
group index method, 820–821 lateralization, 7
grouting, 715–718 laterities, 7–9
types of, 715–718 leaders, 605
lime stabilization, 719
harmonic vibrations, 762 limit analysis, 384
headings, 685 limit equilibrium methods, 384
heat line loads, 186
characteristics of, 18–19 liquidity index, 792
heavy tamping, 711–712 local shear failure, 518
heavy well, 663–665 loess, 6
stability analysis of, 663–665 long piles, 628
Hookean model, 216
horizontal stress, 796 marine deposits, 5, 7
horizontal under-drainage, 738 mat foundations, 568, 589–591
hydrometer method, 329–334 bearing capacity of, 589–591
grain-size distribution by, 329–334 design methods of, 591
hydrometer method, 36–39 types of, 589
maximum capillary head, 120
immediate settlement, 574–578 Mc Leod method, 823–826
impact machines, 767 metamorphic rocks, 2
inclined load, 528–529 Meyerhof’s bearing capacity theory, 535–537
foundations subjected to, 528–529 modified bearing capacity formulae, 533–534
India modified proctor test, 84
major soil deposits of, 7–9 Mohr’s circle, 169–170
Indian Road Congress (IRC) Formula (IRC 15, 1981), Mohr-Coulomb failure criterion, 262–263
830–831 moisture content. See water content
Indian soil classification system, 70–72 muck, 6
Indian Standard Compaction Test, 84 multi retrogressive slips, 475
infinite slope, 478–480 multiple slides, 474–475
analysis of, 480–483
initial secant modulus, 180 natural cycles, 778–782
initial tangent modulus, 180 negative skin friction, 617–618
injection, 715–718 Newmark’s influence chart, 194–195
inorganic soils, 66 Newtonian element, 216
intact clays. See uniform clays nitrogen cycle, 779
inter-molecular bonds, 12 non-circular failure surfaces, 498–500
inter-particle forces, 12–13 non-cohesive soils
intra-molecular bonds, 12 density index of, 351–353
isomorphous substitution, 14 non-displacement piles, 602

860 Index
non-homogeneous soil conditions, 152 pre-compression, 712–715

normally consolidated clay, 223–225 pressure distribution diagrams, 184–185
pressure–void ratio curves, 221
one-dimensional consolidation, 217–219 pre-wetting, 725
Terzaghi’s theory of, 226–233 primary consolidation, 217
open caissons, 651 primary rock minerals, 10
open-drive sampler, 681 principal plane, 169
organic soils, 66–67 principal stress, 169
oven-drying method, 314–315 proportionating foot size, 584
water content determination of, 314–315 punching shear failure, 518–519
over-consolidated clay, 224–225
over-topping, 734–735 quick clay slides, 475
oxygen cycle, 778
range of validity, 105–106
partially saturated soil, foundations on, 530 Rankine;s earth pressure theory, 389–400
pavements, 815–831 Rankine’s states of plastic equilibrium, 386–389
components of, 815–816 rate of deformation, 277–278
design of flexible, 820–827 reciprocating machines, 766–767
design of rigid, 828–831 rectangular combined footing, 585
design of, 818–820 relief wells, 734
requirements of, 816 re-load modulus, 180
types of, 816 residual factor, 265
peat, 6 residual shear strength, 262
peds, 21 residual soils, 4–5
permeability, 112–116 reversal direct shear test, 270
factors affecting, 115–117 ring shear test, 288
values of, 116 rock, foundations on, 531–533
permeability test, 342–346 rock samples, 672–673
permissible settlements, 542–544 types of, 672–673
physical weathering, 3 rocks
pile cap, 629–630 classification of, 793–796
pile foundations, 599–630 index properties of, 787–793
bearing capacity of, 607–618 mechanical properties of, 800–811
classification of, 599–605 rotary drilling, 677–679
driving equipment, 605–607 rotary machines, 767
lateral resistance of, 627–628 rotational slides, 473–474
uplift resistance of, 626–627 routine field test, 688–692
pile groups, 621–625
pipette method, 35–36, 322–329 samplers, 681–683
grain-size distribution by, 322–329 types of, 681–683
piping, 153–154 sand
piston sampler, 683 compaction behaviour of, 93
plastic flow, 383 sand drains, 239–241
plasticity test, 66 acceleration of consolidation by, 239–241
plate load test, 549–551 compressibility of, 241–242
point load strength index, 790 sand replacement method, 317–320
point-bearing piles. See end-bearing piles in-place dry density of soil by, 317–320
Poisson’s ratio, 180 saturated cohesive soils, 281–286
Poncelet’s graphical method, 404–405 shear strength of, 281–286
pore fluid, 116 saturated density, 30
pore pressure coefficients, 286–290 saturation capillary head, 120
application of, 290 scarp wastes, 780
measurement of, 289–290 secondary compression, 233–234
theory of, 286–289 secondary consolidation, 217
porosity, 26–27 sedimentary rocks, 2

Index 861
sedimentation analysis, 34–38 free swell index of, 346–347

seepage, 137–155 identification and classification of, 65–75
forces related to, 137–139 improvement of, 699–725
seismic refraction method, 674–676 index properties of, 25–59
seismic waves, 760 laboratory measurement of properties of, 309–379
shafts, 685 linear shrinkage of, 341–342
shallow foundations, 567–591 liquid limit of, 334–337
causes of, 573 one-dimensional consolidation of, 217–219
design criteria of, 567–568 origin of, 2–3
designs steps for, 583–584 particle size and shapes, 11–12
dynamic bearing capacity of, 767–768 plastic limit of, 337–338
location and depth of, 571–573 reinforcement of, 720
selection of type of, 570 rheological models of, 216
settlement of, 573–583 shear strength of, 262–294
types of, 568–570 shrinkage and swelling of, 120–121
shear failure, 516 shrinkage factors of, 338–341
modes of, 516–519 stress and stress distribution in, 167–198
shear modulus, 180 structure of, 19–21
shear strength, 262–294 undistributed sampling of, 685–688
field measurement of, 278–280 soil densities, 29–31
laboratory measurement of, 266–278 bulk density, 30–31
of granular soils, 292–294 dry density, 29
of statured cohesive soils, 281–286 saturated density, 30
sheet pipe walls, 432–445 submerged density, 30
shifts, 666–667 soil erosion, 780
short piles, 628 soil minerals, 13–15
shrinkage, 120 soil solids, 311–314
sieve analysis, 34, 320–322 specific gravity of, 311–314
grain-size distribution by, 320–322 liquefaction of, 773–774
Skempton’s bearing capacity theory, 534–535 soil-water system, 16
slaking, 791 solid soils, 27–28
slide in colluviums, 475 specific gravity of, 27–28
slope, 471–502 solution gouts, 718
causes of failures, 471–472 sonic velocity, 791–792
protection measures, 501–502 sounding methods, 677
stability of, 471–502 special loading, 526–530
types of movements, 473–476 specific surface, 15
sloping surface, foundations on, 529–530 split-spoon sampler, 682–683
slump earth flows, 475 spread footings, 568
socketed piers, 642 spreading failures, 475
soil standard penetration test, 546–547, 688–689
arching of, 405–407 standard proctor compaction test, 347–351
bearing capacity of soils, 515–551 standard proctor test, 83
capillary phenomenon in, 118–121 steining, 658
compaction of, 81–99 straight type drilled pier, 642
components of, 9–11 strain hardening. See work hardening
compressibility of, 217 strain softening, 262
consistency of, 40–45 stratified drift. See glacio fluvial deposits
definition of, 1 stratified soils
deposits in India, 7–9 permeability of, 112–115
dynamic loading of, 759–774 stress paths, 170–171
elastic properties of, 180–181 stress–strain curve, 261–262
engineering classification of, 67–75 stress–strain modulus, 180
field identification of, 65–67 subgrade, 817–818
formation of, 4–7 submerged density, 30

862 Index
sub-surface exploration, 673–677 under-reamed piles, 618–620

direct methods of, 677–679 unified soil classification system, 67–70
indirect methods of, 673–677 uniform clays, 21
successive slips, 475 uniformity coefficient, 40
surface compaction, 700–701 unit weights, 31–32
surface loads, 186–197 uplift piles, 604
types of, 186–197 upstream blanket, 734
surface tension, 117–118
suspension grouts, 716–718 vacuum de-watering system, 703
swamp and marsh deposit, 6 Vane shear test, 279–280, 690–691
varved clays, 21
vertical piles, 629
talus, 6 vertical stress, 796
Taylor’s stability chart, 490–493 vibration methods, 705–712
tension piles. See uplift piles types of, 705–712
Terzaghi’s bearing capacity theory, 519–525 vibro-compaction, 705–707
Terzaghi’s theory, 226 vibro-displacement compaction, 707–712
textural soil classification system, 73–75 vibroflotation, 707–711
theory of vibrations, 761–766 void ratio, 26
three-phase system, 25–34
till plain. See ground moraine wash boring, 677
tilts, 666–667 water content, 28–29
total density. See bulk density water flow, 103–104
transition filter, 733 water transported soils, 5
translational slides, 474 weathering, 3
transportation agents, 2 types of, 3–4
transported soils, 5–6 well foundations, 655–667
water transported soils, 5 components of, 656
gravity deposits, 6 construction of, 665–666
swamp and marsh deposit, 6 design of, 656–661
wind transported soils, 6 stability analysis of, 661–663
glacial deposits, 5–6 types of, 655
trapezoidal combined footing, 585–588 well-point systems, 701–702
trenches, 684 Westergaard equation, 185–186
trial pits, 683–684 wetlands, 780
triaxial shear test, 270–274, 365–373 wind transported soils, 6
tunnels, 684 work hardening, 262
twin circular wells, 655–656 work softening. See strain softening
uncased piles, 602 zone of capillary saturation, 119

unconfined compression test, 275–277, 359–362 zone of contact water, 119
under-consolidated clay, 225 zone of partial saturation, 119

Soil Mechanics and Foundation Engineering

Uploaded by

Copyright:

Available Formats

Soil Mechanics and Foundation Engineering

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Soil Mechanics and Foundation Engineering

Uploaded by

Copyright:

Available Formats

SOIL MECHANICS AND

A01_PURU1773_01_SE_PREL.indd 1 2/12/2013 5:21:17 PM

A01_PURU1773_01_SE_PREL.indd 3 2/12/2013 5:21:18 PM

A01_PURU1773_01_SE_PREL.indd 4 2/12/2013 5:21:18 PM

1. Soil Formation and Composition 1

List of Symbols 837

A01_PURU1773_01_SE_PREL.indd 5 2/12/2013 5:21:18 PM

1. Soil Formation and Composition 1

2. Index Properties of Soils 25

3. Identiﬁcation and Classiﬁcation of Soils 65

A01_PURU1773_01_SE_PREL.indd 7 2/12/2013 5:21:18 PM

5. Permeability and Capillarity 103

A01_PURU1773_01_SE_PREL.indd 8 2/12/2013 5:21:18 PM

6.4 Signiﬁcance of Laplace Equation 141

7. Stress and Stress Distribution in Soil 167

8. Consolidation and Consolidation Settlement 215

A01_PURU1773_01_SE_PREL.indd 9 2/12/2013 5:21:18 PM

9. Shear Strength of Soils 261

10. Laboratory Measurement of Soil Properties 309

A01_PURU1773_01_SE_PREL.indd 10 2/12/2013 5:21:18 PM

11. Lateral Earth Pressure 383

12. Earth-Retaining Structures 425

13. Stability of Slopes 471

A01_PURU1773_01_SE_PREL.indd 11 2/12/2013 5:21:18 PM

14. Bearing Capacity of Soils 515

15. Shallow Foundations 567

16. Pile Foundations 599

A01_PURU1773_01_SE_PREL.indd 12 2/12/2013 5:21:18 PM

16.7 Uplift Resistance of Piles 626

17. Drilled Piers and Caisson Foundations 641

18. Ground Investigation 671

19. Soil Improvement 699

A01_PURU1773_01_SE_PREL.indd 13 2/12/2013 5:21:18 PM

19.9 Soil Reinforcement 720

20. Embankment Dams 729

21. Dynamic Loading of Soil 759

22. Environmental Geotechnology 777

A01_PURU1773_01_SE_PREL.indd 14 2/12/2013 5:21:18 PM

23. Introductory Rock Mechanics 787

24. Pavements 815

List of Symbols 837

A01_PURU1773_01_SE_PREL.indd 15 2/12/2013 5:21:18 PM

A01_PURU1773_01_SE_PREL.indd 17 2/12/2013 5:21:18 PM

A01_PURU1773_01_SE_PREL.indd 18 2/12/2013 5:21:18 PM

M01_PURU1773_01_SE_C01.indd 1 2/12/2013 7:50:54 AM

1.2 ORIGIN OF SOILS

M01_PURU1773_01_SE_C01.indd 2 2/12/2013 7:50:54 AM

sible melting of deep rocks

Heat, pressure, and Heat, pressure, and

Fig. 1.1 Geological cycle (Source: Bowles, 1984)

1.3 TYPES OF WEATHERING

1.3.1 Physical Weathering

M01_PURU1773_01_SE_C01.indd 3 2/12/2013 7:50:54 AM

1.3.2 Chemical Weathering

1.4 SOIL FORMATION

1.4.1 Residual Soils

Top Soil – Humus Zone I

Surficial Soil Zone Zone II

Fig. 1.2 Stages of formation of residual soil (Source: McCarthy, 1982)