ESTIMATION OF ENGINEERING PARAMETERS OF ROCK MASS

A THESIS

submitted in fulfilment of the

requirements for the award of the degree
of
DOCTOR OF PHILOSOPHY
in

CIVIL ENGINEERING -*<«£*>..

>

Bv r<fi£!ll-
V. K. MEHROTRA

DEPARTMENT OF CIVIL ENGINEERING

UNIVERSITY OF ROORKEE
ROORKEE-247667 (INDIA)

DECEMBER, 1992
Dedicated To My Parents
 UNIVERSITY OF ROORKEE

ROORKEE

CANDIDATE'S DECLARATION

I hereby certify that the work which is being presented in

the thesis entitled ESTIMATION OF ENGINEERING PARAMETERS OF
ROCK MASS in fulfilment of the requirement for the award of the
Degree of Doctor of Philosophy and submitted in the Department
of CIVIL ENGINEERING of the University is an authentic record
of my own work carried out during a period from December 1987
to December 1992 under the supervision of Dr. Bhawani Singh.

The matter presented in this thesis has not been submitted

by me for the award of any other degree of this or any other

University.

Signature of the Candidate

This is to certify that the above statement made by the

candidate is correct to the best of my knowledge.

Date : December %S , 1992 (Bhawani Singh)

Professor
Civil Engineering Department
University of Roorkee
Roorkee - 247667

The Ph. D. Viva-Voce examination of . . $v'. . V 1K'* MpfakoVCAs. . . .

Research Scholar, has been held on .£f:S"~: i(t$£

Signature of Signature of Signature of

Supervisor H.O.D. External Examiner
 ABSTRACT

With remarkable increase in civil engineering activities,

availability of sites with favourable geological conditions is
becoming increasingly restricted posing tremendous challenges
to the geotechnical engineers. This is particularly so in India
where the developmental works for harnessing the water resources
need gigantic dams, long tunnels, large underground cavities,
especially in the young Himalaya.
The rock formations in the Himalaya are intricately
folded, faulted or thrusted, comprising different types of
rocks with intrusions and water-charged regions. Besides these
geological complexities, the region is also highly seismic.
Because of the complicated geology, geotechnical problems have
been encountered in a number of river valley projects built in
this region.

For safe and sound design of structures in this complex

geological area, understanding of the strength"and deformation
behaviour of the rocks is of paramount importance. At present
only scanty information is available regarding the engineering
behaviour of the Himalayan rocks. Some empirical relationships
exist in literature but these are based mostly on the
geological data of comparatively better rocks. As a number of
hydroelectric projects are under construction in the Lesser
Himalaya (lower Himalayan region adjacent to river plain),
there is an urgent need that experience in this area is
documented as early as possible.

Rock mass behaviour can be assessed through adequate

number of field and laboratory tests. However, such tests can

(.1)
be carried out only at project sites because of being time

consuming and expensive due to their massive scale of

operation. Moreover, expertise is almost always needed for

performance of these tests which is available for these

projects. The most significant properties of rocks involved

in the design process are as follows:

1. Deformability characteristics - required for the design

of tunnels, underground openings and dam foundations;

2. Shear strength characteristics - required for the design

of rock slopes, foundations and dam abutments;

3. Support pressure - important in the design of tunnels and

underground cavities ;

4. Allowable bearing pressure - necessary for the design of

foundations.

Construction of a number of hydroelectric projects in the

Lesser Himalaya provided an excellent opportunity to author to

conduct the tests to study the properties and rock mass

behaviour of different rock types encountered in the region. An

attempt has been made through this research study, to correlate

the following design parameters with the RMR and Q systems. The

new relationships which have been developed, on the basis of

the data from the Lesser Himalaya, are as follows:

1. Correlation between modulus of deformation and RMR;

2. Correlation between support pressure on tunnels and rock

mass quality Q;

3. Correlation between RMR and shear strength parameters of

rock mass;

4. Failure envelopes for the rock mass; and

5. Correlation between RMR and the allowable bearing

(ii)
 pressure on rocks.

The effect of saturation on the properties of rock mass

has been studied and relationships developed for both, the
naturally moist and the saturated rock mass.

On the basis of field and laboratory test data of six

major hydroelectric project sites of Lesser Himalaya, it has
been possible to examine the applicability of existing
relationships for these Himalayan rocks.

The results of field and laboratory investigations show

that the rocks in the Lesser Himalayan region are generally of

'poor' quality. On the basis of extensive uniaxial jacking
tests, a relationship has been developed between the rock mass
rating (RMR) and the modulus of deformation. The trend of the
relationship shows that it is not in agreement with the
existing relationships of Bieniawski (1978), and Serafim and
Pereira (1983). Bieniawski's relationship is based on the
experience of 117 tunnel sites in hard rock areas. Hence, it is
reasonable to expect that it may not be applicable to Lesser

Himalayan rocks which are generally poor and weak. Moreover,

Bieniawski's curve is valid for RMR values more than 50 showing

that the relationship is not applicable to poor rock masses.

It is further found that the relationship proposed by the

author has a similar trend as that given by Serafim and Pereira

(1983) but the two are not in good agreement. The correlation
given by Serafim and Pereira (1983) would highly overestimate
the modulus of deformation of these rocks. It is thus inferred

that the existing relationships of Bieniawski (1978) and

Serafim and Pereira (1983) are not suitable for application to

(iii)
the Lesser Himalayan rocks.

The effect of saturation on the modulus of deformation

has been found to be significant. On saturation, the reduction

in the modulus values may be as high as 90 per cent for the

poor rock masses and 70 per cent for the fair quality rock

masses.

Modulus of deformation has been found to have a definite

correlation with the support pressure on tunnels. So, if one

knows the modulus of deformation accurately by tests, support

pressure could be predicted objectively. A correlation has,

therefore,been suggested by the author to evaluate the support

pressure for tunnels in the Lesser Himalayan region.

It is found that the estimated support pressure decreases

with increase of modulus of deformation. Initially, the support

pressure decreases rapidly until a modulus value of 3.5 GPa is

reached. Beyond the modulus of 3.5 GPa, the decrease in the

support pressure is relatively at a slower rate. It is noted

that for the modulus values of 10 GPa or more, the support

2
pressures are stabilised more or less at 0.20 kg/cm for the
2
rock masses at natural moisture content and 0.30 kg/cm for the

rock masses at saturation.

Results have led to show that the moisture condition

significantly affects the self supporting capacity of the rock

mass, thereby affecting the support pressures. For the poor

rock masses, as in the Himalayan region, saturated condition

may be considered to be critical for maximum support pressure

because of the possible increase in the pore water pressure and

( iv)
decrease in the angle of internal friction, cohesion and the

modulus of deformation of the rock mass.

In case of saturated rock mass wide scatter of test data

indicates that for modulus values below 3.5 GPa, the pressure

distribution in the rock mass may not remain uniform. This may

be responsible for non-uniform pressure distribution around the

tunnel lining due to non-uniform flow of water through a few

open joints only. Thus proper drainage is essential for

execution of work especially when excavating tunnels in the low

modulus ground (< 3.5 GPa).

The prediction of support pressure through the modulus of

deformation as brought out in this study would prove helpful in

understanding the tunnel-support interaction in the rock mass.

The relationship developed is simple and easy to apply.

Results of extensive in-situ block shear tests performed

under drained condition show that both the shear parameters,

cohesion and angle of internal friction of the rock mass

increase simultaneously. It is observed that initially the

angle of iriternal friction increases rapidly but attains the

maximum value of 57° at a cohesion value of 450 kPa. It is

further noted that shear strength parameters of the rock mass

increase with increase in RMR. However, no appreciable increase

is found in the values beyond RMR value of 60. It is noted

that the angle of internal friction decreases with decrease in

RMR and becomes asymptotic at 32 .

The effect of saturation has been found to be more

striking on cohesion than the angle of internal friction. A

maximum reduction of 70 per cent in cohesion and 38 per cent in

(v)
the angle of internal friction has been observed in poor rock

masses. In case of fair category rock masses a maximum

reduction of 52 per cent in cohesion and 22 per cent in the

angle of internal friction has been observed.

Mohr failure envelopes have been deduced from the data of

large scale block shear tests on different rock types. The

failure envelopes,which have been derived for both dry and
saturated rock masses , show stress dependent behaviour. The
failure envelopes show in many cases similar trend as given by

Hoek and Brown (1980). The envelopes may be used for strength

determination at the desired magnitude of normal stress and may

prove helpful in carrying out stability analyses of rock

masses.

The rock mass rating (RMR) lias also been used to provide

relationship for net allowable bearing pressure on 'poor' and

'fair' category rock masses (RMR 21-60). A non-linear

relationship has been obtained on the basis of the criterion of

permissble settlement. This may be used for estimating the net

allowable bearing pressures for shallow foundations on the

rocks.

The various relationships developed on the basis of rock

mass classification and the results of extensive field and

laboratory tests would provide a rational basis to examine the

geotechnical feasibility of future projects expeditiously in

the Lesser Himalayas.

(vi)
 NOTATIONS

Symbol Description

A a constant in the equation of Mohr's

failure envelope depending upon the
properties of rock mass.

A joint area/or, shearing area

0

a a constant in the Eq. (2.20) depending upon

the normal stress and kind of rock

B width of the opening/or, footing/or,

roadway span

B„ power function in the non-linear failure

0
envelope

B constant of rock mass

m

b a constant in the Eq. (2.20) depending upon

the normal stress and kind of rock

c cohesion of the rock mass

CH chainage

D depth of footing below ground

E modulus of deformation of rock mass (as per authors

" proposal)
* E modulus of elasticity of rock mass
e

E modulus of elasticity of rock material

r

F factor of safety

f correction factor for thickness of overburden

f correction factor for tunnel closure

GSI geological survey of India

H height of overburden

H height of the rock load

P
As per IS:7317-1974, the parameter (Ee) is termed as the
modulus of deformation. The modulus of deformation (Ed)
adopted in the present thesis is with a different concept as
given on page 166.
(vii)
H. height of the opening

I joint frequency

IS Indian standard code of practice

J joint alteration number

a

J joint set number

J joint roughness number

J total number of joints per cubic metre

J joint water reduction factor

JCS joint wall compressive strength

JRC joint roughness coefficient

K a constant in non-linear strength envelope

(Eqs. 2.19 and 2.21) depending upon the
material properties and loading conditions

K joint normal stiffness

LFJ large flat jack

MRF modulus reduction factor

f =
^d 1
—~—
I r J

m rock material constant in Eq. (2.17b);

varies widely - a function of type and
quality of rock

N normal load

n porosity

nmc natural moisture content

P rock load/ or, support load

p. short term support pressure

P or Proof ultimate roof support pressure

p. or p., „,short term roof support pressure

p or p ,, ultimate wall support pressure

rr W3 J- J.

p. or p. . ,, ,. short term wall support pressure

lw n wai ± )

Q Barton's rock mass quality

(viii)
Q, Barton's rock mass quality for short terra
support pressure

Q, or Q., ,,.Barton's rock mass quality for short

lw x (wax j. }
term support pressure for wall

Q or QMaij Barton's rock mass quality for wall

q allowable bearing pressure on rocks
3k

q unconflned compressive strength of the rock

material

R Schmidt rebound on wet unweathered sawn

joint surface

RF reduction factor for saturation

RL reduced level

RMR Bieniawski's rock mass rating

ROD rock quality designation

RSR rock structure rating of Wickhara et al.

r Schmidt rebound on dry unweathered sawn

joint surface

SRF stress reduction factor

s rock material constant in (Eq. 2.17b) [for

intact rock s=l]

sat saturated

T a rock parameter in the equation of Mohr's

failure envelope

T shear strength (or stress)

t shear strength (normalised)

qc
<o normal stress

o major principal stress at failure

(ix)
<y, minor principal stress applied to the
specimen

Cf normal stress (normalised) = —

n
c

uniaxial tensile strength

uniaxial tensile strength (normalised)

tn

unit dry weight of rock

a constant of rock mass

n

ft orientation angle

v Poisson's ratio

strain /or, deformation

0 angle of internal friction of rock mass

basic friction angle

0 residual friction angle

(x)
 SYMBOLS USED IN DRAWINGS

SYMBOL DESCRIPTION

@ JAMRANI SANDSTONE
[j] JAMRANI CLAYSTONE
@ KOTLIBEHL LIMESTONE
£RJ KOTLIBEHL SHALE
LAKHWAR TRAP

LAKHWAR XENOLITH

[Uj LAKHWAR SLATE

A SRINAGAR ftUARTZITE

SRINAGAR METABASIC
W
TEHRI PHYLLITE

A UTYASU aUARTZITE

(*) INDICATES A TEST CONDUCTED IN VERTICAL DIRECTION

-Q- INDICATES A TEST CONDUCTED IN HORIZONTAL DIRECTION
O INDICATES A TEST CONDUCTED AT NATURAL MOISTURE CONTENT
(single line enclosure)

INDICATES A TEST CONDUCTED AT SATURATION

(double line enclosure)

DATA FROM UNIVERSITY OF ROORKEE

(xi)
 ACKNOWLEDGMENT

The author is grateful to Professor Bhawani Singh of the

Department of Civil Engineering for introducing him, to the

real perspective of practical rock mechanics through the

adverse and complex geotechnical problems of the Himalayan

rocks. His persistent supervision, valuable guidance and

imparting a great awareness for the application of rock

mechanics in river valley projects, is deeply acknowledged.

The author wishes to express grateful thanks to Professors

Gopal RanJan, G.C. Nayak and M.N. Viladkar of the Civil

Engineering Department for their valuable suggestions and keen

interest from the beginning of this research programme.

The author expresses his hearty thanks to Professor B.V.K.

Lavania, of the Department of Earthquake Engineering,

University of Roorkee for his suggestions and many stimulating

discussions.

The author is thankful to Dr. A.K. Dube of the Central

Mining Research Station, Roorkee Unit, for his various

suggestions from time to time.

The author is much grateful to Dr. R.Anbalagan of the

Department of Earth Sciences, University of Roorkee, who was

helpful in exchanging ideas on the various matters pertaining

to this work.

Hearty thanks are due, to Sri Chandra Kumar Agrawal, Chief

Engineer, the then Director, U.P. Irrigation Research

Institute, Roorkee for his permission to perform laboratory

tests required for this research work. The author is also

indebted to the senior officers and collegues working at

Jamrani, Kotlibehl, Lakhwar, Srinagar, Tehri and Utyasu dam

(xii)
projects for sponsoring field testing programmes.

The author owes his thanks to Dr. V.M. Sharma, Director,

Central Soils and Materials Research Station, New Delhi for

raany discussions and valuable suggestions for this research.

The author feels grateful to Dr. R.D. Bhargava,

Ex-Professor and Head, Indian Institute of Technology, Bombay,

for numerous suggestions during the preparation of this thesis.

The author gratefully acknowledges Dr. Jaak J.K. Daeraen,

Professor and Head, Mining Engineering, Mackay, School of

Mines, University of Nevada, Professor Raymond Sterling,

Director, Underground Space Center, University of Minnesota,

Mr. Birger Schmidt, Vice President, PB/MK TEAM, U.S.A. for

critical discussions on this work.

4

The author acknowledges Dr. Z.T. Bieniawski, Professor and

Director, Mining and Mineral Resources Research Institute, The

Pennsylvania State University, U.S.A. and Dr. Nick Barton of

the Norwegian Geotechnical Institute, Norway for having

critical discussions with them.

My deep thanks to my son-in-law, Er. Divakar Seth and

daughter Monica who always encouraged and inspired me far-away

from Bristol, U.K. with their numerous enquiries about this

research work.

My deep gratitude always stands for my wife Daya for her

patience and forebearance, and who has been a source of

continuous encouragement during the progress of this work. I

can understand the feelings of my daughter Reetika who had many

a tiraes to part with the company of her father during the

course of this work. ^£^*^f_L__.

(xiii)
 CONTENTS

CHAPTER PAGE

Abstract I

Notations VII

Symbols Used in Drawings XI

Acknowledgment XII

List of Tables XIX

List of Figures XXI

1 INTRODUCTION 1

2 REVIEW OF LITERATURE 6

2.1 Introduction 6

2.2 Deformation Characteristics of Rock Mass 7

2.3 Shear Strength Parameters 22

2.4 Support Pressure for the Design of Tunnels 31

2.4.1 Introduction 31

2.4.2 Rock Mass Classification Approach 37

2.4.2.1 Terzaghi's rock load classification [1946] 37

2.4.2.2 Classification by Lauffer [1958] 38

2.4.2.3 Classification system by Deere et al. [1967] 42

2.4.2.4 Rock structure rating (RSR) concept [1972] 43

2.4.2.5 Georaechanics classification (RMR System) [1973] 43

2.4.2.6 Q-systera of classification by Barton et al.

[1974, 1975] 48

2.5 Allowable Bearing Pressure on Rocks 59

2.6 Concluding Remarks 65

3 GEOTECHNICAL PROBLEMS IN HIMALAYA 68

3.1 General 68

(xiv)
3.2 The Himalaya 68

3.3 Geotechnical Problems in the Himalayan Projects 71

3.3.1 Ramganga Tunnel - Ramganga River Valley Project,

Kalagarh, Uttar Pradesh 73

3.3.2 Chibro-Khodri Tunnel, Yamuna Hydroelectric

Project, Stage II, Uttar Pradesh 79

3.3.3 Tunnel Collapse at Chainage 238 m in Dharasu

Downstream Face of Maneri Bhali Hydel Scheme,
Stage II, Uttarkashi, Uttar Pradesh 83

3.3.4 Ichari-Chibro Tunnel, Yamuna Hydroelectric

Project, Stage II, Uttar Pradesh 88

3.3.5 Other Severe Tunnelling Problems in

Chibro-Khodri Tunnel, Uttar Pradesh 94

3.3.6 Head Race Tunnel - Loktak Hydroelectric

Project, Manipur 96

3.3.7 Head Race Tunnel - Kopli Project, Assam 100

3.3.8 Baira - Siul Head Race Tunnel, Baira-Siul

Project, Chamba, Hiraachal Pradesh 100

3.3.9 Salal Hydel Tunnel - Salal Hydroelectric

Project, Jararau and Kashmir, India 101

3.3.10 Girl Tunnel - Giri Hydroelectric Project,

Hiraachal Pradesh 102

3.3.11 A Project in Highly Folded-Faulted Region -

Charaera Hydroelectric Project, Chamba, Hiraachal
Pradesh 104

3.3.12 Head Race Tunnel, Proposed Pala-Maneri

Hydroelectric Project, Uttarkashi, Uttar
Pradesh 108

(xv)
3.3.13 A Dam on Sandstone - Proposed Jamrani Dam
Project, Uttar Pradesh 109

3.3.14 A Dam in Complex Geology - Proposed Kotlibehl

Dam Project, Uttar Pradesh 112

3.3.15 A Dam on Difficult Foundation - Proposed

Lakhwar Dam Project, Uttar Pradesh 118

3.3.16 A Dam on Metabasic and Quartzite - Proposed

Srinagar Hydel Scheme, Uttar Pradesh 122

3.3.17 A Dam on Phyllites - Proposed Tehri Dam

Project, Uttar Pradesh 125

3.3.18 A Dam on Highly Jointed Quartzite - Proposed

Utyasu Dam Project, Uttar Pradesh 130

3.4 Scope of Present Research 131

C GEOTECHNICAL INVESTIGATIONS 135

4.1 Introduction 136

4.2 Rocks Explored 137

4.3 Drifting and Drilling for Sub-surface

Explorations 139

4.4 Laboratory Tests 140

4.4.1 General 140

4.4.2 Physical Properties 140

4.4.3 Modulus of Elasticity (E ), Poisson's Ratio (u)

and Uniaxial Compressive Strength (qp) 140
4.5 Field Investigations 148

4.5.1 General 148

4.5.2 Modulus of Deformation (E,), and Modulus of

Elasticity (E ) 148

4.5.3 Shear Characteristics - Cohesion (c) and Angle

of Internal Friction (0) 151

(xvi )
4.5.4 Allowable Bearing Pressure on Rocks (q ) 152

4.5.5 Rock Mass Classification 152

4.6 Results of Investigations 156

4.6.1 Laboratory Tests 156

4.6.2 Field Investigations 156

4.7 Discussion 160

5 ANALYSIS AND INTERPRETATION OF DATA 167

5.1 General 167

5.2 Modulus of Deformation (Ed) 167

5.2.1 Estimation of Modulus of Deformation 199

5.2.1.1 Effect of saturation on modulus of deformation 199

5.2.2 Correlation Based on Modulus Reduction Factor

(MRF) 200

5.2.3 Discussion of Results 202

5.3 Short Term Support Pressure in Tunnels 205

5.3.1 Rock Mass Classification Systems 206

5.3.2 Estimation of Short Term Support Pressure in

Tunnels 208

5.3.2.1 Influence of overburden on the support pressure

in tunnels 211

5.3.3 Estimation of Support Pressure Through Various

Approaches 211

5.3.3.1 Support pressure through classification system

of Terzaghi (1946) 213

5.3.3.2 Support pressure through classification system

of Deere et al. (1970) 213

5.3.3.3 Support pressure through RMR system 214

5.3.3.4 Support pressure through Q-system 214

(xvii)
5.3.4 Comparison of Support Pressures by Different
Approaches 215

5.3.5 Support Pressure and Moisture Content 216

5.3.6 Size Effect 219

5.4 Shear Strength Parameters of the Rock Mass 220

5.4.1 RMR and Shear Strength Parameters (c & 0) 221

5.4.2 Discussion and Interpretation 221

5.5 Failure Envelopes for the Rock Masses 228

5.5.1 Trend of the Failure Envelopes 230

5.5.1.1 Mohr envelopes for 'poor' rock masses 230

5.5.1.2 Mohr envelopes for 'fair' rock masses 237

5.5.1.3 Mohr envelopes for 'good' rock masses 240

5.5.2 Recommended criteria of shear strength of rock

masses 240

5.6 Allowable Bearing Pressure on Rocks 245

5.7 Assumptions and Limitations 248

6 CONCLUSIONS 251

7 RECOMMENDATIONS FOR FUTURE RESEARCH 255

REFERENCES 258

(xvi i i)
 LIST OF TABLES

TABLE NO. DESCRIPTION PAGE NO.

2.1 Approximate Equations for Principal

Stress Relationships and Mohr
Envelopes for Intact Rock and
Jointed Rock Masses (after Hoek and
Brown, 1980) 29
2.2 Terzaghi's Rock Load Classification
Currently in Use 41
2.3 Geomechanics Classification of Rock
Masses (after Bieniawski, 1979) 45
2.4 Guidelines for Excavation and
Support of Rock Tunnels in
Accordance with the Rock Mass
Rating System (after Bieniawski,
1984) 46
2.5 Q-Systera of Rock mass(after Barton
et al., 1974) 49
2.6 Classification System based on
Ground Convergence (after Sharma
and Raraamurthy, 1985) 56
2.7 Allowable Pressure on Rocks (after
Peck et al., 1974) 60
2.8 Net Safe Bearing Pressure Based on
Classification (after IS : 12070
- 1987) 64

3.1 Details of Six Major Hydroelectric

Projects in the Lesser Himalaya 134

4.1 Layout Plan of Laboratory Tests on

Rocks 149
4.2 Field Investigations on Rocks 150
4.3- Summary of Laboratory Test Results 157

xix)
4.4 Number of Test Sections
Investigated for Rock Mass
Classification I58
4.5 Sumraary of Classification Results 159
4.6 Summary of Results of- In-Situ
Investigations 161

5.1 Comparison of Support Pressures by

Different Approaches for the
Lakhwar Tunnel 217
5.2 Recommended Mohr Envelopes for
Jointed Rock Masses 244
5.3 Net Safe Bearing Pressures Based on
RMR Classification 247

xx )
 LIST OF FIGURES

Figure No. Description Page No.

2.1 Modulus of Deformation v/s Joint

Frequency (After Cunha, 1991) 12
2.2 Relationship between Geomechanics
Classification Rating and Ratio of
Deformation Modulus of Rock Mass E , to
d
that of Rock Material E (After
r

Bieniawski, 1975) 13
2.3 Relationship between Stress and
Deformation of Clay Rock and Siltstone
with Different Moisture Content (After
Gantang and Uuelan, 1991) 16
2.4 Correlation between the In-situ

Modulus of Deformation and RMR (After

Bieniawski, 1978) 17
2.5 Correlation between the RMR and the

Q-Index (After Bieniawski, 1976 and

Jethwa et al., 1982) 20
2.6 Estimation of In-situ Modulus of

Deformation from Two Classification

methods (After Barton, 1983) 21

2.7 Influence of Foliation Planes on Modes

of Specimen Failure in Compression for

Weakly Foliated Rocks (After Tsidzi,
1991) 26
2.8 Shear Strength of Soft Rock (After
Ramegowda, 1990) 32
2.9 Illustration of Rock Support
Interaction (After Daemen, 1977) 35
2.10 The Tunnel Rock-Load Concept of
Terzaghi (After Terzaghi, 1946) 40
2.11 Relationship between the Stand-Up Time
and Span for Various Rock Mass Classes
According to the Geomechanics

(xxi )
 Classification (After Bieniawski,
1984) 47
2.12 Modified 1988 Lauffer Diagram
Depicting Boundaries of Rock Mass
Classes for TBM Applications (After
Lauffer, 1988) 47
2.13 Equivalent Dimension v/s Rock Mass
Quality [Q-System] (After Barton et
al., 1974) 52
2.14 Seismic Reduction of Q-Value to Obtain
25% Increase in Support Pressure
(After Barton, 1984) 53
2.15 Allowable Contact Pressure on Jointed
Rock (After Peck et al., 1974) 62
2.16 Bearing Capacity Test Result in Case
of Very Soft Shale (After Ramegowda,
1990) 63
2.17 Rock Mass Rating (RMR) v/s Net
Allowable Pressure on Rocks (After
University of Roorkee) 66

3.1 Hydroelectric Schemes in Ganga and

Yamuna Valleys 72
3.2 Geological Cross-Section Along the
Longitudinal Axis of Ramganga Tunnel
No. 1 and 2 77
3.3 Excavation Plan of Ramganga Tunnels 80
3.4 Index Map of the Yamuna Hydroelectric
Scheme, Stage II 81
3.5 Regional Geology of the Yamuna
Hydroelectric Scheme, Stage II 82
3.6 Layout Plan of Yamuna Hydel Scheme,
Stage-II (Part-II) 85
3.7 Geological Sections of Chibro-Khodri
Tunnel 86
3.8 Schematic Representation of Cavity
Development in Head Race Tunnel

xxii )
 Maneri Bhali Hydel Scheme, Stage II 87
3.9 Longitudinal Geological Section and
Cross-Section of Chibro Power Tunnel 90
3.10 Radial Rock Pressure - Time
Relationship at Chainage 1199 ra of the
Chibro-Khodri Head Race Tunnel 97
3.11 Geological Section of Loktak Hydel
Tunnel (After G.S.I., 1979) 98
3.12 Geological Section of Salal Hydel
Tunnel Showing Possibilities of Cavity
Formation and Squeezing Zone (After
Jethwa et al., 1976) 103
3.13(a) Geological Section of Giri Hydel
Tunnel (After Ghosh, 1970) 106
3.13(b) Geological Section of Giri Hydel
Tunnel (After G.S.I., 1977) 106
3.14 Geological Section of Charaera Power
Tunnel 107
3.15 Six Major Hydroelectric Projects
Coraing-Up in Ganga and Yamuna Valleys 110
3.16 Geological Section along the Axis of
the Proposed Jamrani Dara 113
3.17 Geological Section along the Axis of
the Proposed Kotlibehl Dara 116
3.18 Spillway Section of the Proposed
Lakhwar Dara 121
3.19 Cross Section of Deepest Block showing
Foundation Geology of Srinagar Dara 126
3.20 Geological Section along the Tehri Dara
Axis 129
3.21 Cross Section at the Proposed Utyasu
Dara Axis 132

4.1 Location of In-situ Tests Carried-Out

at the Jamrani Dam Project Site 141
4.2 Location of In-situ Tests Carried-Out
at the Kotlibehl Dam Project Site 142

(xxiii)
 43 Location of In-situ Tests Carried-Out
at the Lakhwar Dara Project Site 143
44 Location of In-situ Tests Carried-Out
at the Srinagar Hydel Scheme Site 144
4-5 Location of In-situ Tests Carried-Out
at the Tehri Dara Project Site 145
4-6 Location of In-situ Tests Carried-Out
at the Utyasu Dara Project Site 146
4-7 Set-Up of Equipment for Uniaxial
Jacking Test in Vertical Direction in
Narrow Drifts or Galleries 153
4-8 Arrangement of Block Shear Test in
Drifts 154
49 Typical Pressure-Settlement Curve for
Poor Rock Mass (RMR=25) Obtained from
Uniaxial Jacking Test 162
4-10 Typical Pressure-Settlement Curve for
Poor Rock Mass (RMR=31) Obtained from
Uniaxial Jacking Test 163
4-11 Typical Pressure-Settlement Curve for
Fair Rock Mass (RMR=43) obtained from
Uniaxial Jacking Test 164
4 12 Typical Pressure-Settlement Curve for
Fair Rock Mass (RMR=54) Obtained from
Uniaxial Jacking Test 165
4-13 Method of Calculation of Modulus of
Deformation (Ed) and Modulus of
Elasticity (E ) of the Rock Mass 166

5-1 Correlation Between Rock Mass Rating

(RMR) and Modulus of Deformation (E,) 169
5-2 Modulus of Deformation Data from
Uniaxial Jacking Tests (Rock Mass at
Natural Moisture) 170
5.3 Modulus of Deformation Data from
Uniaxial Jacking Tests (Saturated Rock
Mass) 171

(xxiv)
5.4 Modulus of Deformation - Measured v/s
Predicted Values (Rock Mass at Natural
Moisture) 174
5.5 Modulus of Deformation - Measured v/s
Predicted (Saturated Rock Mass) 175
5.6 Modulus of Deformation - Measured v/s
Predicted (Rock Mass at Natural
Moisture) [Sandstone] 176
5.7 Modulus of Deformation - Measured v/s
Predicted (Rock Mass at Natural
Moisture) [Claystone] 177
5.8 Modulus of Deformation - Measured v/s
Predicted (Rock Mass at Natural
Moisture) [Limestone] 178
5.9 Modulus of Deformation - Measured v/s
Predicted (Rock Mass at Natural
Moisture) [Shale] 179
5.10 Modulus of Deformation - Measured v/s
Predicted (Rock Mass at Natural
Moisture) [Trap] 180
5.11 Modulus of Deformation - Measured v/s
Predicted (Rock Mass at Natural
Moisture) [Xenolith] 181
5.12 Modulus of Deformation - Measured v/s
Predicted (Rock Mass at Natural
Moisture) [Slate] 182
5.13 Modulus of Deformation - Measured v/s
Predicted (Rock Mass at Natural
Moisture) [Quartzite (Srinagar)] 183
5.14 Modulus of Deformation - Measured v/s
Predicted (Rock Mass at Natural
Moisture) [Metabasic] 184
5.15 Modulus of Deformation - Measured v/s
Predicted (Rock Mass at Natural
Moisture) [Phyllite] 185
5.16 Modulus of Deformation - Measured v/s
Predicted (Rock Mass at Natural

( XXV
 Moisture) [Quartzite (Utyasu)] 186

5.17 Modulus of Deformation - Measured v/s

Predicted (Saturated Rock Mass)
[Sandstone] 187

5.18 Modulus of Deformation - Measured v/s

Predicted (Saturated Rock Mass)
[Limestone] 188

5.19 Modulus of Deformation Measured v/s

Predicted (Saturated Rock Mass)
[Shale] 189

5.20 Modulus of Deformation Measured v/s

Predicted (Saturated Rock Mass)
[Trap] 190

5.21 Modulus of Deformation Measured v/s

Predicted (Saturated Rock Mass)
[Xenolith] 191

5.22 Modulus of Deformation Measured v/s

Predicted (Saturated Rock Mass)
[Slate] 192

5.23 Modulus of Deformation Measured v/s

Predicted (Saturated Rock Mass)

[Quartzite] 193

5.24 Modulus of Deformation Measured v/s

Predicted (Saturated Rock Mass)
[Metabasic] 194

5.25 Modulus of Deformation - Measured v/s

Predicted (Saturated Rock Mass)
[Phyllite] 195

5.26 Modulus of Deformation - Measured v/s

Predicted (Saturated Rock Mass)
[Quartzite (Utyasu)] 196

5.27 Influence of Load Direction on the

Modulus of Deformation (Saturated Rock

Mass) 197

5.28 Influence of Load Direction on the

Modulus of Deformation (Rock Mass at

Natural Moisture) 198

(xxvi )
5.29 Effect of Saturation on the Modulus of

Deformation 201

5.30 Relationship between Rock Mass Rating

(RMR) and Ratio of Modulus of

Deformation (E,) to that of Rock

Material (E ) 204
r

5.31 Relationship between Rock Mass Quality

(Q) and Rock Mass Rating (RMR) 209
5.32 Relationship between Modulus of
Deformation (E,) and Short Terra
Support Pressure (P.) for Tunnels
(Rock Mass at Natural Moisture) 212
5.33 Relationship between Modulus of
Deformation (E,) and Short Term
Support Pressure (Pj) for Tunnels
(Saturated Rock Mass) 212
5.34(a) Relationship between Cohesion (c) and
Angle of Internal Friction (0) of the
Rock Mass at Natural Moisture 222

5.34(b) Relationship between Cohesion (c) and

Angle of Internal Friction (0) of the
Rock Mass at Saturation 225

5.35 Relationship between Rock Mass Rating

(RMR) and Shear Strength Parameters,
Cohesion (c) and Angle of Internal
Friction (0) at Saturation and Natural
Moisture 226

5.36 Effect of Saturation on the Shear

Strength Parameters (c & 0) of the

Rock Mass 227

5.37 Failure Envelopes for Jointed

Limestone (RMR = 29-37) 232

5.38 Failure Envelopes for Jointed Slate,

Xenolith and Phyllite (RMR = 23-37) 233
5.39 Failure Envelopes for Jointed
Sandstone and Quartzite (RMR = 22-36) 234

(x x v i i
5.40 Failure Envelopes for Jointed Trap and
Metabasic Rock (RMR = 24-40) 235

5.41 Failure Envelopes for Jointed Slate,

Xenolith and Phyllite (RMR = 43-56) 238
5.42 Failure Envelopes for Jointed
Sandstone and Quartzite (RMR = 41-58) 239
5.43 Failure Envelopes for Jointed Trap and
Metabasic Rock (RMR = 42-59) 242
5.44 Failure Envelopes for Jointed Trap and
Metabasic Rock (RMR = 61-72) 243
5.45 Allowable Bearing Pressure on the
Basis of Rock Mass Rating (RMR =
21-60) 250

(xxviii)
CHAPTER 1
 CHAPTER 1

INTRODUCTION

For safe and sound design of some engineering structures in

rocky region, it is essential to have a comprehensive

knowledge of the mechanical behaviour of the rocks. Designs are
carried out according to their in-situ properties and physical
conditions. Quantitative assessment of the engineering
behaviour of rocks constitutes input information vital for
design. Yet, unlike other engineering materials, rocks
generally present unforeseen problems and thereby pose
tremendous challenges to the designer.

Today, India needs many hydroelectric developmental works

in Himalayan region for harnessing the country's water
resources. Such works include construction of gigantic dams,
large underground cavities, long tunnels, power houses,
spillways; treatment of foundations, slopes and abutments etc.
especially in the young Himalaya; as most of India's water
resources,particularly the hydro-power potential lies in the
Himalaya which has complex geological formations comprising
different types of rocks, faults and fissures, intrusions and
water charged regions. Besides these geological
discontinuities, the region falls within the highly seismic
tectonic belt and is subject to primary stresses whenever any
underground excavation is done. Typical precipitation cycle in
the country and large sunshine hours contribute to high
weathering effects.

Although rock is considered best foundation material, the

designer has also to take into account the dangers associated
with unfavourable rock mass behaviour under induced load. Some
factors controlling rock mass behaviour include the in-situ
*
stress environment, confining pressure, structural defects,
anisotropy, changing moisture content, mineralogical
composition, tirae-dependent deformations and interlocking of
the rough blocks in the rock mass.
Because of the combined effect of factors given above, the
rock and the rock mass exhibit varying amounts of strength and
deformation effects. These aspects are important in relation to
the foundation of dams, underground power houses, stability of >
slopes and tunnels, and excavation in rocks etc. Rock
formations from different geological, environmental and
weathering conditions usually have different rock mass
characteristics. Thus,for safer and efficient designs, it is
important to understand, assess and quantify all the aspects
concerning the engineering behaviour of rock masses.
Engineering rock mass behaviour may be best understood by large
scale in-situ tests. However, such tests are generally
difficult to conduct because of heavy loads involved and v
excessive time required for site preparation. Moreover, the
tests are expensive due to their full scale operation and
therefore, statistically significant number of tests may not be
conducted routinely.

Bieniawski (1978) while apprehending the necessity of

large scale tests commented : "although in-situ tests are
v
expensive and time consuming, they have been employed
throughout the world on most major rock engineering projects.
Yet, there are a number of controversial questions pertinent to
in-situ tests. One can argue, for example, as to whether the
interpretation of the measured in-situ data is at best an

estimation, open to criticisra and thus not justifying the high

expenditure or whether the in-situ tests are truly necessary

and should not be replaced by less costly approaches. Very few

projects to date have featured a sufficient number of different

tests to allow a meaningful comparison of in-situ test data".

In his opinion, "there are uncertainties associated with all

the in-situ tests and the field is not without problems".

Parameters of particular significance commonly used in

the engineering design are as follows :

1. Modulus of deformation/modulus of elasticity; required for

the design of underground cavities, tunnels and dara

foundation;

2. Support pressure; important in the design of tunnels and

underground cavities;

3. Shear strength parameters; required for the design of rock

slopes, foundations and dam abutments; and

4. Allowable bearing pressure on rocks; needed for the design

of foundations.

Estimates of engineering parameters are normally based on

extensive field tests which require great expense and efforts

for their completion. Constructon of a number of hydroelectric

projects in the Lesser Himalaya provided an excellent

opportunity to conduct extensive in-situ tests in different

rock formations.

In this thesis, using the test data, the deformation

characteristics, the shear strength parameters, bearing

pressure and the support pressure for tunnels have been

correlated with comprehensive classification parameters of the

RMR system (Bieniawski, 1973) and the Q-system (Barton et al.,
1974). There were several reasons for selection of these
classifications:

* These have been used for about 2-3 decades throughout the
world. These are well verified and numerous data

pertaining to their use are available;

* These classifications enable evaluation of the
deformability and shear strength parameters, support

pressure for tunnels and bearing pressure on rocks;

* Since these classifications have already been applied
>
successfully to about 400 case histories and have been
generally accepted, these are now regarded as some kind of
convention and thus form a promising link between the

designer arid the field engineer.

The rock mass classification systems are also acknowledged

an important link for the transfer of experience to the future.
Classification systems attained their usefulness on account of
their simplicity and accuracy in prediction.
Investigations were,therefore, carried out using RMR and Q-
systeras for a wide range of rock types, and correlations
between the rock mass quality and the engineering parameters

were obtained as follows :

1. RMR versus deformation characteristics of the rock mass;

2. RMR versus Q;

3. Modulus of deformation versus support pressure for tunnels;

4. RMR versus shear strength parameters of the rock mass;
5. Mohr envelopes for the rock mass; and
6. RMR versus allowable bearing pressure on rocks.
The effect of saturation on the rock mass has also been
studied and the relationships obtained for naturally moist as
well as for the saturated rock mass condition. The effect of
saturation on in-situ moduli, support pressure and 3hear

strength parameters was expected to be significant. Reduction

factors have to be worked out to quantify the effect of
saturation on the rock mass. In case in-situ tests are

conducted under natural moisture condition, the proposed

reduction factors may be applied to predict the engineering

parameters for saturated condition which would prevail in the

rock mass after commissioning of the hydroelectric projects.

Tests were carried out mostly on poor and fair quality

rocks, which are dominant in the Lesser Himalaya. It was

believed that the new correlations would be phenomenally

different and the results may not match with the trends of the

existing correlations. It may be recognised that the existing

RMR and the Q-system correlations are based mainly on good

quality rocks. The inventors of the two classification systems

perhaps did not come across poor and very poor quality rocks
and it is this gap in our present knowledge that really needed

a back-up by field measurements.

The new correlations which are based on tests on fair and

poor quality rocks, would provide not only a rational design

base but ensure a better predictability of performance on the

projects planned in the region. It is hoped that this would be

helpful in reducing the present gap in our knowledge apart from

providing more confidence to the designer and a better safety

of structure besides financial and technical benefits.

CHAPTER 2
 CHAPTER 2

REVIEW OF LITERA1URE

'2.1 Introduction

Knowledge of engineering properties of rocks is essential for

proper design of rock structures like dam foundations, slopes,
tunnels and underground cavities. Three parameters of greater

importance in designing and building engineering structures

over or in the rock are the deformability of the rock mass,

shear characteristics and its excavation behaviour. Current

design procedures for foundations, tunnels and underground

excavations in rock depend heavily on the measurement of

modulus of deformation and in-situ shear strength.

Determination of reliable modulus values is necessary for the

estimation of deformations in the rock mass. For carrying out

the stability analyses of rock masses, it is essential to know

the shear strength along the joints which are potential slip

surfaces. An assessment of the shear strength parameters of the

rock mass is thus necessary. Further, when a structure imposes

large loads on the rock mass, evaluation of the foundation

settlement becomes very important. This involves computation of

safe bearing capacity of foundations on discontinuous rock

masses.

For the design of tunnels and underground cavities,

estimation of support pressure is an important problem. The

analysis of the stability of any underground opening is

dependent on the correct assessment of the rock load. Various

approaches are presently being followed for the estimation of

support pressures. Rock mass classification [Terzaghi (1946),

Wickhara et al. (1972), Bieniawski (1973), Barton et al. (1974)]
is an empirical approach most frequently used for the design of
stable support system. In fact, classification systems have in
recent years developed to a stage where they are used as the
basic design tool in determining support requirements of
tunnels, deformation characteristics for foundations and
underground excavations, shear strength characteristics for an
excavation and allowable bearing capacity for a foundation. By
using classification systems, engineering pararaeters of rock
raas3 may be estimated rapidly and with sufficient accuracy for

site appraisal purposes.

According to Lokin et al. (1983) "classification should

start right from the choice of site and be developed all the
way through to the final completion of work. All the stages of
classification together must constitute an integral system. In

each successive stage the classification is filled-in with new,

raore reliable and specific data, with more and more

quantitative pararaeters, as produced by current investigations

and required by the current stage of design. Thus, its
character changes from the roost general in the initial stage to

become completely specific at the end."

2.2 Deformation Characteristics of Rock Mass

Rock is a discontinuous mass containing cracks, fissures,

joints, faults and bedding planes. The engineering properties

of a rock mass depend far more on these discontinuities. For
foundations on a rock mass, it is generally the deforroability

rather than the strength of rock mass which determines the

design criteria. Modulus of deformation and modulus of
 8

elasticity are essential parameters for the design of tunnels,

underground cavities and dam foundations.

According to Bieniawski (1976), the stability of

structures built in or founded on rock masses depends upon the

strength and deformational behaviour under load of the intact

blocks of rock material as well as upon the properties of

natural discontinuities such as faults, joints, bedding or

schistosity planes. The influence of these discontinuities

depends in turn upon such properties as spacing, orientation,

continuity, tightness, surface roughness, frictional resistance

to sliding and joint water pressures. He further commented that

knowledge of load deformation characteristics of the rock mass

forming the proposed dam foundation is of great importance

since it forms the basis of dam design and determine the type

of dam to be built.

According to Barton (1986), information on rock raa3s

deformation moduli is required in the design of arch dams and

bridge piers, and in all tunnelling projects where analyses of

deformations is required. The most realistic data are obtained

from large scale tests in which the jointing is fully

represented. On many occasions, owing to the prevalence of

horizontally bedded sedimentary rock, the load deformation

curves obtained from large plate loading tests resemble the

normal closure curves for rock joints.

Lo (1987) pointed out that for the design of dams on rock

foundation and underground structures in rocks, the modulus of

deformation of rock is often required. It is well known that

laboratory tests on rock cores seriously over-estimate the

modulus, since the predominant effect of discontinuities can
not be included in the laboratory tests. The common approach is
to employ various field tests including (a) flat jack test in a
slot, (b) plate load test, (c) large scale pressure chamber
test and (d) different bore hole jacking devices such as
dilatoraeters. It has often been found, however, that results
obtained from these tests are widely different from another.

He further opined that empirical correlations between mass

modulus from field tests and some type of rock classification

system, such as Rock Quality Designation (RQD), Commonwealth

Scientific and Industrial Research (CSIR) rating, or Norwegian

Geotechnical Institute (NGI) rating, also have been used.

However, these correlations are more useful in assessing the

variability of the modulus than in determining the value of the

modulus. In addition, there appear to be few systematic

investigations of the variation of modulus from laboratory

tests on small "intact" specimens through increasing size of

specimen to the rock mass.

Londe (1972) believed that in practice it is reasonable to

assume an elastic behaviour for rao3t dam foundations, provided

that the appropriate modulus of elasticity for the stress range

under consideration is used. Comparisons between the results of

analyses and measurements of foundation deformations on raany

dams have shown this approximation to be acceptable.

According to Cunha (1991) the mean deforraability moduli of

*
a rock mass with a single joint set, loaded normally to the

set, are only dependent on the deformability of the rock (Ef),

the joint frequency (I) and the joint normal stiffness (Kn). He
 10

showed that the deformation modulus (E,) is a function of the

mean rock modulus (E ) and decreases as the joint frequency

increases, with the following boundary conditions : 1=0, then

Ed=Er, (intact rock); I • o>, then Ed • 0 (Fig. 2.1).

Commenting on the role of pore water pressure and moisture

content,Okamoto (1981) opined that other properties of soft

rocks which must be considered are the pore water pressure and

the moisture content. The effects of pore water pressure which

extend to the mechanical properties of soft rocks, can be

treated in the same way as for soils, but in case of the

variation of the contained water, it is needless to say that

since the texture of the particles have been solidified by

compaction, once this is broken down by swelling or slaking,

the strength and the modulus of deformation distinctly decline.

Singh and Goel (1975) developed a relationship between

modulus reduction factor and the average number of joints per

metre in the direction of load in plate load test. The

relationship is as follows :
E. 1
Modulus Reduction Factor = " = =—— (2.1)
r 1 + A.n

G/Ed - 1/10 - 1/20

where n = average number of joints per metre in the direction

of load in plate load test.

A = 0.6 m for continuous joints or loose bedding planes

in unweathered rock mass.

= 0.25 m for discontinuous joints in unweathered rock

mass.

= 0.05 m for unweathered cleavage planes but separated.

They emphasized that joints reduce shear modulus (G) much

 11

more than raodulus of deformation (Ed). As such ,rock mass is

anisotropic. The modulus of deformation may be worked out

easily if elastic modulus of rock material is found in the

laboratory for natural confining pressure and degree of

saturation.

Bieniawski (1975) suggested a relationship between

geomechanics classification rating (RMR) and the ratio of

deformation modulus of rock raass Ed to that of rock material Er
(Fig. 2.2).

For rapid site appraisal, the raodulus of deformation (^d)

may be estimated on the basis of geomechanics classification

rating (RMR) and the modulus of elasticity (Er) of the rock

material.

Chaubey and Chaturvedi (1991) highlighted the importance

of scale effects and observed that many scale effects are

evident in rock engineering testing and are caused by

difficulties in obtaining representative samples. Large samples

are more easily damaged and tend to include more flaws. It is

extremely difficult to perform realistic three dimensional

in-situ testing of rock masses. Block size appears to be an

extremely Important parameter for explaining a wide range of

strength values obtained during these geotechnical

investigations.

Hawkins and Macconnell (1991) believed that strength and

deformability of sandstones are controlled primarily by

textural variations and that the mineralogy of framework grains

does not play an important role. Using the values obtained frora

a qualitative microfabric analysis the most important variables

± 4

NJ

Joints per metre IJ Joints per metre

FIG. 2-1 MODULUS OF DEFORMATION v/s JOINT FREQUENCY (After Cunha, 1991)
 4

i-o

0-9.

u 08-
Hi

B Orange fish tunnel

uj 0-7
© Diepsloot bridge
• Nevada test site
o 0-6
u • Dvorshak dam
•

0-5
• Kariba powerhouse
c
o

u
= 0 4-^
*>
L.

5 0-3-1
"3 Px Le Roux
•o
o dam
2 0-2 4

0-1

I? m n
01 t_
10 20 30 UQ 50 60 70 60 90 100

G. C. Rating ( RMR )

FIG. 2-2 RELATIONSHIP BETWEEN GEOMECHANICS CLASSIFICATION

RATING AND RATIO OF DEFORMATION MODULUS OF ROCK
MASS Ed TO THAT OF ROCK MATERIAL Er (After
Bieniawski, 1975 )
 14

appear to be:

1. porosity

2. two dimensional packing density (determined from

computer digitisation of thin sections in the direction

of loading)

3. cement index (based on strength and amount of cement)

4. packing proximity (the number of grain to grain contact

as a percentage of the total number of contacts per

traverse).

5. interlocking index (based on types of . grain to grain

contacts)

6. grains sphericity

He concluded that in case of sandstones , the

inter-relationship between framework grains and the nature of

the inter-granular bonding material appeared to be more

important than framework grain mineralogy.

Gantang and Yuelan (1991) have shown that the presence of

moisture in a rock specimen can reduce the compressive strength

of the rock, e.g., for quartz-diorite, sandstone, clayey

sandstone and shale, the compressive strength of saturated

specimens is reduced to the percentage of 10-40, 40, 50 and 60

of the strength of dry specimens respectively. In addition, the
influence of water on the deformation of rock is also

important. It even changes fundamentally the mode of rock

failure. He also demonstrated that the relations between stress

and strain varies with different moisture content of the clay

rock and siltstone.

It is observed that plastic failure prevails for the

specimens with moisture content higher than about 20-30 per
cent and brittle failure occurs for those with moisture content
less than 15 per cent (Fig. 2.3). Since, in general, water is
always present in the rock mass, it is an important problem on
the investigation of mechanical properties of the wet rock.
Mehrotra et al. (1991) observed that some rocks in the
lower Himalayan region may be significantly affected for their
strength and deformation characteristics once the rock mass is
finally saturated during reservoir filling. The design
V
pararaeters need be evaluated taking into account the long-terra
effects of prolonged saturation.

Based on the results of three case histories involving

three major engineering projects in South Africa, Bieniawski
(1978) obtained a correlation for hard rocks as follows (Fig.
2.4) :

E, = 2 RMR - 100 (2.2)

d

where,

E is the in-situ static raodulus of deformation in GPa and

d
RMR is the rock mass rating In accordance with the geomechanics
classificationj RMR is more than 50.
Bieniawski (1978) believed that Eq. (2.2) may be used for
estimating the modulus of rock masses during the planning stage
of a project with an accuracy better than 20 per cent. Equation
(2.2) may find application under geological conditions similar
to those for which it was developed and may not be applicable
X

for rocks having RMR values less than 50.

Serafim and Pereira (1983) proposed a correlation for full

range of RMR values. The correlation is as follows (Fig. 2.4):
 16

le failure

»
o.

Plastic failure
I

V\
17-9 %>

200 400 600 800 1000

Deformation ( € ) [X 10-4.1

FIG. 2-3 RELATIONSHIP BETWEEN STRESS AND DEFORMATION OF

CLAYROCK AND SILTSTONE WITH DIFFERENT MOISTURE
CONTENT (After Gantang and Yuelan, 1991 )
 *

90

80

m
Ed- 2RMR-100
a.
is
70

HI

60

5 50

•S 40

M
3

I 30
o

^ 20 Case histories
i
+ Bieniawski, 1978
O Serafim & Pereira,1983
10 <o
o o.
.Jl^b'dO"" °
— i t-CT
10 20 30 40 50 60 70 80 90 100

Geomechanics rock mass rating (RMR)

FIG.2-<* CORRELATION BETWEEN THE IN-SITU MODULUS OF DEFORMATION AND RMR (After Bieniawski,J97B)
 18

Ed = 10<RMR " 10'/"0, GPa (2.3)

Based on the results of other researchers, Ramaraurthy
(1985) suggested the following relationships for critically
oriented joints.

(1) For predominantly horizontal joints

E exp (0.0217 RMR - 2.17), GPa (2.4)

r

(2) For predominantly inclined joints at

30 to 45 to vertical
Ed
-g— = exp (0.0564 RMR - 5.64), GPa (2.5)
r

Bieniawski (1976) and Jethwa et al. (1982) established a

correlation between the RMR and the Q. The following
relationship was found for civil engineering tunnels (Fig.
2.5).

RMR = 9. InQ +44 (2.6)

Using the above correlation, Barton (1983) supplemented
the data of Bieniawski (1978) with his own results and plotted
the range of measured values (Fig. 2.6). Barton (1983) found
the following empirical expressions:

Ed max = 40 lo6 Q (2.7)

Ed mean = 25 log Q (2.8)
Ed min ' 10 log Q (2.9)
According to Ramegowda (1990), Eq. (2.3) may be used for a
preliminary assessment of the raodulus of deformation (E ). He
opined that it is inadequate for cases where rock material is
soft and weak.

Rutledge and Preston (1978) found the following

correlation between the rock mass rating (RMR) and Q.
RMR = 5.9 In Q t 43 (2.10)
 19

The above correlation is based on tunnelling projects in

New Zealand. Y
For mining tunnels, Abad et al. (1983) suggested the

following correlation.

RMR = 10.5 In Q + 42 (2.11)

The above correlation is based on 187 coal mine roadways

in Spain.

Cameron and Budavari (1981) developed the following

relationship :

RMR = 5 In Q + 60.80 (2.12)

Moreno Tallon (1982) confirmed the above relationships on

the basis of four tunnelling projects in Spain and suggested

the relationship as follows :

RMR * 5.4 In Q + 55.2 (2.13)

It is seen that these correlations differ significantly.

It would,however,be expected that any of the classification

systems for a given rock mass would lead to similar engineering

properties, if all the relevant factors have been accounted for

and assessed correctly.

According to Kaiser et al. (1986), the RMR system appears

to be less sensitive to minor variations in input parameters

because the sum rather than the product is evaluated. This

insensitivity leads to a flattening of the correlation curve.

To achieve a better correlation between the two classification

systems, the RMR must be evaluated very carefully, particularly ^

in upper ground classes (I & II). On the other hand, in poor

ground conditions, the range of assigned Q-values increases and

produces greater flattening of the correlation curve. Hence,
• y

Ext remely Very poor Poor Fair Good

Veryj Ext. Exc.
Exc. poor
aoor Good|Good Good
100 / •
-D
O
O
at

• N G 1 case studies /% >-

O Geom. case studies o /
80 - A Other case studies o / >

DC
• Indian case studies . ° * / •o
o
X o
K
i
at

.E 60

° / • • •• O

(A U-
Ut

40
• /•*•
E

• •/•• • 1_
o
o

20 -
• ° °°^£° °* o
a.

o
o
Q.

' O RMR»9in a +44

0 ^ • X i i i 1

0.0 01 0-01 0-1 10 100 1000

Rock mass quality -Q.

FIG. 2-5 CORRELATION BETWEEN THE RMR AND THE a-INDEX (After
Bieniawski, 1976 and Jethwa et al., 1982 )
 400 1000

/ EJ-2RMR-100

NJ

60 70 80 90 100

Rock mass rating (RMR)

FIG. 2-6 ESTIMATION OF IN-SITU MODULUS OF DEFORMATION FROM

TWO CLASSIFICATION METHODS: SQUARES REPRESENT
a CASE HISTORIES, OOTS ARE RMR CASES (After Barton,
1983)
 22

extreme care must be taken when the Q-system is applied to poor

ground.

The modulus of elasticity of the rock mass and initial

stress field surrounding a tunnel,may be back analysed by a

method described by Yang and Sterling (1989).

It is understood that the various correlations predicted

by different researchers have emerged from engineering

experience under specific geological conditions. Their

applicability to projects with different geological conditions

needs be verified.

2.3 Shear Strength Parameters

According to Sellers (1974) in-situ shear strength is

notoriously difficult to measure. The problera is mainly a

sampling problem. How is one to isolate a test specimen from

the rock mass without damaging it? How is one to know if the

specimen isolated will be representative? Laboratory tests on

rock specimens obtained by diamond drilling, be they direct

shear tests or triaxially confined test3, yield apparent

shearing strengths which are generally higher than the

strengths of the rock mass in-situ. The reason for this is that

the smaller the specimen, the less likelihood of it including a

major plane of weakness. Furthermore, the weaker part3 of the

rock mass are generally ground up and lost during the drilling
process so that the remaining drill core contains only the

strongest portions.

Hoek and Bray (1981) commented on the importance of

quality of shear tests: "poor quality shear tests are worse
than no shear tests".
 23

According to Oda (1988) the size dependent property is

believed to be caused by a larger rock mass tending to have
larger defects. Therefore, to evaluate the shear strength of
rock masses, in-situ tests are commonly carried out using block
samples containing at least a few joints. On the basis of
numerical experiments supplemented by field measurements he
observed that.to obtain the representative physical constants

in a jointed rock mass, the size must be at least three times

larger than the typical joint trace length. If test blocks were
smaller than this minimum volume, any experiments using thera

may lead to a false prediction about the hydromechanical

properties.

Cunha (1991) believed that the deformation and the failure

modes are dependent not only on the natural block (size,

properties of boundary joints and rock flaws) but also on the

stress levels and on the relation between the prototype

dimensions and the natural block sizes (Barton, 1990). This

fact emphasizes the difficulties in selecting appropriate test

size and extrapolating the measured property values to the

specific engineering problems (Bandis, 1990). The reality of

most projects is such that only a very few large 3cale test3

can be conducted, and even with the greatest care, these are

likely to be unrepresentative of variations in the rock mass.

Nilsen (1985) pointed out that in most rock stability

cases, the normal stress on the potential sliding plane will be

x

relatively low (often in the range of 0.1-0.3 MPa). Because of

the existence of irregularities on joint planes, a non-linear

relationship normally exists between shear 3trength and normal

 24

stress. At low normal stress levels in particular, it i3 of

great importance that the selection of shear strength

parameters is adjusted to the normal stress level in question.

He concluded that serious design mistakes may be the result of

erroneous estimation of shear strength.

Horn and Deere (1962) contended that the influence of

water on frictional strength of rocks, in particular depends

upon the crystal structure of the individual constituent

materials. For the minerals with framework structure (quartz,

feldspar), water acts as an anti-lubricant in which case there

is a little or no wetting whereas it serves as a lubricant in

case of minerals with a layered structure (micas) assuming that

the surfaces are smooth.

Tsidzi (1991) emphasised on the importance of water

saturation on the strength values of foliated rocks. According

to him water saturation has a significant influence on the

strength values. On the whole,there have been reductions upto

50 per cent although increments of a few per cent (&%) were

also observed in some rocks. Patrographic peculiarities seem to

govern the nature and degree of strength variation with water

saturation. For the mica-rich strongly foliated rocks the

change is that of a reduction, whereas it is an increase for the

quartz and feldspar rich weakly foliated rocks.

Tsidzi (1991) on the basis of experiments concluded that

for the weakly foliated rocks, the influence of foliation

planes on specimen failure was found to be practically

insignificant so that the dominant mode of failure was by shear

fracture through the intact rock meterial for all orientations

 25

(Fig. 2.7).

It may be seen that for weakly foliated rocks the

V

influence of foliation planes on specimen failure is

insignificant. Water saturation had a significant reduction in
the strength values of moderately foliated rocks. For weakly
foliated rocks the strength tended to increase on account of
anti-lubricant effect of water.

According to Barton and Chaubey (1977), joints that are

smooth and planer or those that exhibit signs of earlier
-V
movement can not be relied upon to have any asperity component
of shear strength and in such cases design (i.e. rock slope
design) will have to be based on the minimum shear strength,
namely the residual friction angle (0r). The residual friction
angle (& ) may be determined using the following relationship
of Barton and Chaubey (1977)

0 = (0, - 20) + 20 (r/R) (2.14)

r b

where &,_ - basic friction angle estimated from residual tilt

D

tests on dry unweathered sawn surfaces.

r and Ft = Schmidt rebound on dry unweathered sawn and wet

joint surface respectively.

The value of & is scale independent and relevant to

r

saturated conditions.

Based on the results of a back analysis of an artificial

slope-cut, Grasso et al. (1991) showed that allowances for

scale effects were always necessary when using the JRC - JCS
model of Barton and Chaubey (1977) to predict the shear

strength of rock joints. He suggested that such adjustments may

be made using the following relationship (Barton and Bandis,
 26

^ 0 /9 =45° 0=90°
(a) MECHANICALLY SIGNIFICANT FOLIATIONS
(DEGREE OF FOLIATION; GREATER THAN MODERATE )

r / ?
/
/ /

'
' /
/ / i' /
•'
/ / /
/

/
/J/ ' /
/
/ /
/ /
/ s

p=o 0=45 B=9 0

(b) MECHANICALLY INSIGNIFICANT FOLIATIONS

(DEGREE OF FOLIATION: LESS THAN MODERATE)

[THICK WAVY LINES DEFINE FAILURE SURFACE]

FIG. 2-7 INFLUENCE OF FOLIATION PLANES ON

THE MODES OF SPECIMEN FAILURE FOR
WEAKLY FOLIATED ROCKS ( After Tsidzi ,
1991 )
 27

1990)

-0.02 JRC,,
L ••
n
JRC = JRC\a (2.15)
n 0
L0

-0.03 JRC,,
L -"n
0
JCS = JCS~ (2.16)
n 0
L0
where the subscripts (0) and (n) refer to the laboratory

size samples (100 mm) and in-situ natural blocks respectively.

Davachi et al. (1991) considered the following factors to-

determine the design angle of shearing resistance :

(i) residual angle of shearing resistance; (ii) in-situ state

component of angle of shearing resistance; (ill) angle of

shearing resistance attributable to shear plane roughness; and

(iv) angle of shearing resistance attributable to shear plane

thickness.

It follows from the above that component (1) is the

primary factor determining the design angle of shearing

resistance. Component (ii) accounted for a small additional

component of the angle of shearing resistance in excess of the

residual value. This seemed true because the in-situ angle of

shearing resistance indicated by direct 3hear testing often

exceeded the residual value. Component (iii) recognizes the

fact that shear-plane roughness contributes to the angle of

shearing resistance. Component (iv) recognizes that all other

conditions being equal, thin shear planes with rock-to-rock

contact have higher angles of shearing resistance than thick

shear planes with fillings.

 28

Hoek and Brown (1980) presented Mohr envelopes defined by

equations having the form :
tn =A(on -a.tn )B° (2.17a)

and, a - o^ + Vmq^ + s. qQ (2.17b)

whfrp t and a are normalised shear and normal stresses r/q
n n

and <Vq , and a is the normalised uniaxial tensile strength

c n
defined by

a - o"t/q =1/2 [m -/m2+4s J 4*

n

q = uniaxial compressive strength of rock material

c

Table 2.1 shows approximate equations for principal stress

relationships and Mohr envelopes for intact rock and jointed
rock masses (Hoek and Brown, 1980).

More recently Hoek et al. (1992) modified the original

Hoek - Brown failure criterion and on the basis of ten years
practical experience, developed a modified rock mass failure
criterion. The most significant changes include :
*
1. A re-forraulation of the criterion for Jointed rock
masses to eliminate the tensile strength predicted by

the original criterion.

2. The introduction of a simplified qualitative rock mass

classification for the estimation of the parameters in

the modified criterion.

3. The presentation of a procedure for calculating

J*

the parameters defining the Mohr failure envelope for

the modified criterion, and for determining the
instantaneous friction angle and cohesive strength for
 4

>

TABLE 2.1
APPROXIMATE EQUATIONS FOR PRINCIPAL STRESS RELATIONSHIPS AND
MOHR ENVELOPES FOR INTACT ROCK AND JOINTED ROCK MASSES
(After Hoek and Brown, 1980)

zrsisys! r~™M HESSs,r' ssrsssss: sss^s?;

POORLY DEVELOPED TALLINE ROCKS METAMDRPHIC
TAL CLEAVAGE
CRYSTAL CLEAVAGE

mudstone,siltstone, sandstone and andesite,dolerite, amphibolite,gabbro,

dolcmite,limestone diabase & rhyolite gneiss,granite,norite &
and marble shale and slate quartzite
quartzdiorite
(normal to cleavage)

INTACT ROCK SAMPLES 'ln=°3n+ '7°3n+1-° in- 3n

+/I0o3n+1.0 otf 3n+ /lSo^+l.O V= 3nWl7o3n+1-° ° ln= 3n+ ^So^+l.O
Laboratory size rock
specimens free from Tn=0.816(an+0.140) •658 Tn=0.918(an+0.099)0-677 x„=1.044<ch+0.067)0-692 xn=1.086(on+0.059)0-696 xn=1.220(on+0.040) 0.705

structural defects
CSIR rating 100+,
NGI rating 500
VERY GOOD QUALITY
0l=Cn+/3^30
'ln= <3n+ VJ-Sa3n+
*lrro3n+ /^-^l 'tf^KsW* -m-V^V0 "in-3n^l2.5^+0.1
ROCK MASS
Tightly interlocking 0.712
undisturbed rock with x=0.651(o +0.028)°-679 ,=0.739(a +0.020)0-692 xn=0.848(^+0.013)°"702 ,n=0.883(on+0.012)0-7°5 xn=0.998(on+0. 008)
unweathered joints
spaced at + 3 metres
CSIR rating 85, NGI
rating 100
GOOD QUALITY ROCK MASS
^0T oln=a3n+ /l.0a3n+0.004 V= °3n+ ^5^0M ^^ /l.7a3n+0.004 .tf V /2.5a 3n
+0.004

Fresh to slightly
weathered rock,slightly
disturbed with joints
xn=0.369C, -n+0.006)0-669 xn=0.427(an+0.004)°-683 x^O.SOK^+0.003)0-695 *n=0.525< on+0.002)°-698 xn=0.603( c^+0.002 0.707

spaced at 1 to 3
metres, CSIR rating 65,
NGI rating 10.

FAIR QUALITY ROCK MASS o^o_Wo.l4 ^+0.0001 °ln=°,n+fo.20V0.0001 °ln= V V°-30 V0"0001 °ln=a3n+ ^34°3n+0-°°01 °ln= V '0.50^+0.0001
in 3n 3n' In 3n
Several sets of
moderately weathered
joints spaced at 0.3 to
xn=0.198(an+0-0007)0-662 xn=0.234(on+0.0005)°-^ xn=0.280<on+0.0003)°-^ ^=0.295(^.0003,0-691 ^346(^.0002) 0.700

1 metre. CSIR rating

44, NGI rating 1.0
POOR QUALITY ROCK MASS
an=%+ /Q.04^+0.00001 .„,- a^O.OSo^+O.OOOOl o^Wo.OSo^+O.OOOOl "ln=^0.09^+0.00001 °ln= V^" V0"00001
Numerous weathered

xn=0.115(on+0.0002) 0.646 xn=n.129Con+0.0002)°-655 xn=0.162(on+0.0001)0-672 *n=0.172(o n+0.0001)0-676 V<).203( V0.0001) 0.686

joints spaced at 30 to
500nm with some gouge
filling/clean waste
rock CSIR rating 23,
NGI rating 0.1
VERY POOR gUALTTY
ROCK MASS
oln=3n+V'0.007"3n+0 jz
°ln=»3n+/0.010
in" 3nT'«•«•"' °3n+°
3nT "« ln=o3n+'0-0l;>u3n'
1„=a,n+V0.015»,n+0 oln=o1nW0-017°3n+0 °ln=°3n+M3-025 V°
Numerous heavily wea
0.556
thered joints spaced
,n=0.050(on)°-"9 xn=0.061Cn)
0.546 on=0.065(on) 0.548
less than 50nm with -h=0-042(on) 0.534
gouge filling/waste
rock with fines
CSIR rating 3, NGI
rating 0.01
 30

a given normal stress value.

The new criterion is

"l = °3 + qc C mb V^c3"
where m, and a are constants of rock mass.

The modified failure criterion provides a means of

estimating the strength of jointed rock masses during

feasibility studies of a project. However, more detailed

studies, including laboratory or in-situ tests, should be

carried out for detailed engineering designs.

Ramaraurthy (1985) proposed a similar criterion of shear

strength of rock mass,

(*! " &3)/&3 = Bmtqc/°'3;,0'ln (2.18)

where Bm
m
and a m are constants of rock mass,

Yoshinaka et al. (1991), on the basis of direct shear test

for flat surface, concluded that there is no influence against

the angle of internal friction due to the size effect.

Regarding the size effect on shear strength, following equation

concerning the failure envelope was suggested :
.,
t =K aa . AJft (2.19)
where, t = shear strength

& - normal stress

A = Joint Area

K, a and ft are the constants depending upon the material

properties and loading conditions. In a particular case, test
results showed K = 7.73, a=0.83 and ft = 0.175.

According to the above relationship the shear strength

decreases in the form of power function with increment of the
shearing area.
 31

Yoshinaka et al. (1991) also suggested a relationship

between the shearing area and peak shear strength as follows :

^r

log10r =a +b.log10(Ao)06 (2.20)

where, a and b are the parameters depending on the normal

stress and kind of rock or the joint roughness respectively.

Ramegowda (1990) pointed out that in case of soft rock

masses and filled discontinuity planes containing soft

material, the intensity of normal stress influences the shear

strength significantly. The Mohr-Coulomb straight line may not

be the best fit to the experimental data. He suggested an

empirical power law relation of the form (Fig. 2.8) :

B
t = Ko ° (2.21)
2.4 Support Pressure for the Design of Tunnels

2.4.1 Introduction

When dealing with underground structures constructed as part of

civil engineering projects the fundamental problem is whether

the support of the excavation has been properly designed? The

stability of an underground excavation depends, on one hand,

the geological, hydrogeological, and mechanical properties of

the rock mass and on the other hand, the manner in which the

support for the excavation has been constructed. When an

underground excavation is made in a rock mass the state of

equilibrium of the rock mass gets changed. The initial state of

equilibrium passes through a series of transition states to

attain a final equilibriura after the excavation is completed. j^

The initial 3tate and the final state of equilibrium of the

rock mass constitute the 'short term' and the 'long terra'

support pressures respectively.

w 4

10-0 r-

(<r ) Normal stress in kg /cm^

FIG. 2-8 SHEAR STRENGTH OF SOFT ROCK (After

Ramegowda,1990 )
 33

The short term and the long term support pressures are

influenced by factors such as the geology of the rock mass and

its in-situ strength parameters, pre and post excavation stress

distribution, stand-up time of the rock mass, sequence and

method of excavation, type of support measures used and its

installation time, rate of advance and ground water conditions.

These factors influence the stability of the opening to a large

extent. Keeping all the factors in view, the predicted support

requirements should be safe, practical and economical.

Before any support system is designed and excavations

started, it is necessary to estimate the expected support

pressures. Estimation of support pressure for the design of

tunnel supports may be grouped into three categories namely

analytical, observational and empirical approaches. Any of the

three approaches may be used but it would be better to start

simultaneously with atleast two different approaches 30 that

design assumptions are compared with the recoraraendations of

other design methods.

Analytical methods utilize the analysis of stresses and

deformations involving certain assumptions. The accuracy of

such solutions depends on the correctness of such assumptions.

The methods include techniques like finite element, finite

difference and boundary element methods, closed form solutions,

photoelastic and physical modelling methods.

Observational methods involve interpretation of monitored

data during underground excavation. In this approach, in-situ

measurements are essential to predict the stress distribution

process. The observational approach is thus a process of

 34

continuous "design as you dig" and is distinct from other

approaches. Use of this method needs practical experience with

a number of points to be taken into consideration. Some of the

points needing attention are as follows:

1. Continuous control of the rock and support elements

through measurements;

2. Suitable support installed in a correct sequence at a

correct time;

3. Creating or conserving triaxial stress condition

compatible with the rock strength; ~?

4. Direct contact between the rock and support elements

for load transfer;

5. Avoidance of detrimental loosening;

6. Careful excavation to avoid disturbing the rock; and

/ 7. Adjustment In construction procedures to the changing

rock conditions.

An example of observational approach 13 New Austrian

Tunnelling Method i.e., NATM (Rabcewicz, 1964). The NATM

involves a combination of excavation and tunnelling with

continual monitoring of rock movement. The results of

raeasurements are interpreted in a scientific manner. The method

raay be used effectively only with highly qualified contractors

and experienced engineers.

Highlighting the problems in tunnel support mechanics

Daemen (1977) illustrated stepwise mechanism of the developraent

-v

of pressure on a steel support as the excavation in a tunnel

advances. Load gets transferred to the support in five stages

(Fig. 2.9).
 35

X —,

*J- -V-^-V STEP

i i

STEP b

STEP

K^~H—it

STEP d

*—i^

A_rLkJk_d.s J* .m wv ^ hi, y>, yy ,", j STEP e

X —I Section X-X

Radial deformation

Excavated profile
Linear-elastic
deformation Tunnel profile
Start of failure of
rock surrounding tunnel

Support reaction
Load deformation curve
for tunnel roof

y- Load deformation curve

4f for tunnel sidewall
Radial deformation

FIG. 2-9 STEPWISE DEVELOPMENT OF PRESSURE ON

A STEEL SUPPORT AS THE TUNNEL
ADVANCES (After Daemen,1977)
 36

1. Stress changes begin to affect the rock about one

tunnel diameter ahead of the face;

2. Immediately after excavation, the rock mass remains

unsupported because of the influence of face and the
adjacent supports;

3. The supports are now installed in place, usually with

no initial load;

4. The face is further excavated. The support which was

available upto now by the excavated rock is removed.
The rock at the section, deforms further and the
steel sets get compressed and loaded; and

5. As the tunnel advances, the load increases until the

support load is fully developed at about two
diameters behind the face.

Final equilibrium is reached when the primary stress, the

stresses caused by excavation and the stresses caused by the
support reaction forces are balanced.

The empirical design methods relate practical experience

gained on previous projects to the conditions anticipated at a
proposed site. These are considered to be the best suited
approaches and are still considered to be the only alternative
available for assessing the stability of underground
excavations in rocks. Analytical methods are supposed to be
still in developmental stage and are currently in use in India
on experimental basis as well as for research and developraent.
Thus use of empirical approach seems to suit comparatively
better over the other design approaches for underground
excavation.
 37

2.4.2 Rock Mass Classification Approach

Today there are many different rock mass classification systeras
in existence which have been successfully applied throughout
the world viz. in the United States (Deere et al., 1967;
Wickharaet al., 1972; Terzaghi, 1946; Bieniawski, 1979), Canada
(Coates, 1964; Franklin, 1976), Western Europe (Lauffer, 1958;
Pacheret al., 1974; Barton et al,, 1974), South Africa
(Bieniawski 1973; Laubscher, 1977; Olivier, 1979), Australia
(Baczynski, 1980), New Zealand (Rutledge, 1978), Japan (Nakao,
1983), India (Ghose and Raju, 1981; Bieniawski, 1973; Barton et
al., 1974, 1975; Sharma, 1985), Russia (Protodyakonov, 1974),
some East European countries (Kidybinski, 1979) and France
(AFTES, 1981).
The rock mass classifictaion systeras by different
researchers have emerged from engineering experience under
specific geological conditions and each has its own limitations
and drawbacks. Of the many different rock mass classification
systeras in existence today, the systeras proposed by Terzaghi
(1946), Lauffer (1958), Deere et al. (1967), Wickhara et al.
(1972), Bieniawski (1973) and Barton et al. (1974) shall be
discussed in detail because they are raost coraraon, siraple and
practical.
2.4.2.1 Terzaghi s rock load classification (1946)
Terzaghi (1946) was the first to evolve a practical
classification for estiraating rock loads for steel arch
supported tunnels. In this classification he recognised the
significance of structural discontinuities and influence of
weathering but the properties of the rock reraained overlooked.
 38

Cecil (1970) concluded that the classification was too general

to permit an objective evaluation of rock quality and that it
provided no quantitative information on the properties of rock
mass. Terzaghi's classification is appropriate for estimating
rock loads for steel-arch supported tunnels and is not suitable
for modern tunnelling methods using shotcrete and rockbolts.
The main features of Terzaghi's tunnel rock-load concept are
shown in Fig. 2.10.

In his classification rock mass was divided qualitatively

into nine categories, viz (i) hard and intact (ii) hard,
stratified or schistose (iii) massive, moderately Jointed (iv)
moderately blocky and seamy (v) very blocky and seamy (vi)
completely crushed but chemically intact (vii) squeezing rock,
moderate depth (viii) squeezing rock, great depth (ix) swelling
rock.

Terzaghi's rock load classification was later modified by

Deere et al. (1970) and Rose (1982) (Table 2.2). It may be seen
that Terzaghi's rock load pressure increases with tunnel size.
On the basis of more than 25 case records of large cavities,
Cording and Deere (1972) concluded that increase in support
pressure with increasing dimensions of cavity was not
necessary. It would therefore be better if rock load factors
are replaced by constant values of support pressures in Table
2.2.

2.4.2.2 Classification by Lauffer (1958)

Stini (1950) emphasized the importance of structural defects in
rock masses. Lauffer (1958) based his classification on Stini"s
work and introduced the concept of stand-up time for any active
 39

unsupported rock span in various rock raass classes. The

stand-up time is the period of time that a tunnel will stand
unsupported after excavation. An active unsupported span is the
width of tunnel or the distance from the face to the support if
this is less than the tunnel width. He illustrated the
Influence of orientation of tunnel axis, shape of opening,
method of excavation and support system on the unsupported
span.

Lauffer's classification was later improved by Pacher et

al. (1974). The main significance of the improvement lies in
the fact that an increase in tunnel span leads to a large
reduction in the stand-up time. Use of this system requires
considerable experience to judge accurately both the stand-up
time and the unsupported span.
According to Ward (1978) there is nothing like the rock
load on a tunnel support in rock and it all depends upon when
and where the support was installed.
On the basis of tunnelling experience, Mussger (1986)
expressed that the very short stand-up time, in some areas less
than one hour, required an immediate sealing of all freshly
excavated areas not only in the top heading but also face,
bench and occasionally the supporting body. All support
elements had to be installed as soon as possible. He viewed
that a number of damages are wrongly allocated to geological
hazards. In such cases, the excavation and support works were
not adjusted within stand-up time according to the geological
conditions which were either predictable or could not be seen
in advance.
 40

SURFACE

FIG. 2-10 THE TUNNEL ROCK-LOAD CONCEPT OF TERZAGHI

(After Terzaghi, 19^6)
 41

TABLE 2.2
•,b
TIRZAGH'S ROCI LOAD CLASSIFICATIOI CORRRITLT II BSI

Rock Condition ROD Rock Load B, (ft) Reiarks

1. Bard and Intact 95-100 Zero Saie as Terzaghi (1946)

2. Hard stratified or schistose 90- 99 0-0.5 B Saie as Terzaghi (1946)

3. Massive Moderately jointed 85- 95 0-0.25 8 Sate as Terzaghi (1946)

4. Hoderately blocky and seaiy 75- 85 0.25 B-0.20(B+Bt) Types 4,5 a.nd 6 reduced
by about 50% froi
5. Very blocky and seaiy 30- 75 (0.20-0.60)(B+Bt) Terzaghi values because
sater table has little
6. Coipletely crushed but cheiically 3- 30 (0.60-1.10)(B+Bt) effect on rock load
intact (Terzaghi, 1946:Brekke,
6a. Sand and gravel 0- 3 (l.l0-1.40)(B*Bt) 1968)

7. Squeezing rock, loderate depth HA' (1.10-2.10)(Btfit) Saie as Terzaghi (1946)

8. Squeezing rock, great depth BA' (2.l0-4.50)(B*Bt) Saie as Terzaghi (1946)

9. Swelling rock RAe Op to 250 ft Saie as Terzaghi (1946)

irrespectlre of
value of (BtBt)

a As lodlfied by Deere et al. (1970) and Rose (1982)

b Rock load B, in feet of rock on roof of suuport in tunnel Blth Bidth B(ft) and height Bt
(ft) at depth of lore than 1.5 (BtBt)
c Hot applicable
 42

2.4.2.3 Classification system by Deere et al. (1967)

Deere (1964) developed a quantitative index called the rock
quality designation (RQD). Deere et al. (1967) and Deere (1968)
proposed a relationship between RQD and the core quality as
follows:

RQD, per cent Rock Quality

< 25 Very poor
^.5-50 poor
50-75 falr

75-90 goocj

90-100 excellent
Palmstrom (1982) suggested that when core is not
available, the RQD may be estimated from the number of Joints
per unit volume, in which the number of joints per metre for
each joint set is added. The conversion for clay free rock
masses is:

RQD = 115 - 3.3 Jv (2.22)

where, Jy represents the total number of joints per cubic
metre.

Later, Deere et al. (1970) modified Terzaghi's rock load

classification and proposed a relationship between RQD and
support requirements both for the conventional and the machine
tunnelling methods. Deere's RQD classification though simple,
quick and inexpensive disregards orientation, tightness,
roughness and gouge material of joints and is not suitable to
provide adequate description of rock mass. Roof support
pressure for a tunnel may be estimated using support
recommendations by Deere et al. (1970).
 43

Merrit (1972) pointed out the limitation of RQD for clay

filled joints or where the rock material is weathered. Brekke ^
and Howard (1973) discussed Influence of clay seams and fault
gouge on tunnel stability.
2.4.2.4 Rock structure rating (RSR) concept
Wickham et al. (1972, 1974) proposed a method to describe the
quality of a rock mass and then to select an appropriate
support. Based on the concept of rating system, RSR method
considered two categories of factors, namely (1) geologic
pararaeters and (ii) construction parameters. The geologic
pararaeters include (a) rock type, (b) average spacing of
joints, (c) orientation of joints, (d) condition of
discontinuities (e) major faults, shears and folds, (f) rock
material properties, and (g) weathering or alteration. The
construction parameters included (a) size of tunnel, (b)
direction of drive, and (c) method of excavation. All the above
factors were grouped into three basic pararaeters A, B and C.
Summing the weighted numerical values for each parameter gives
a
the RSR value. Thus, RSR = A + B + C with a maximum value of
100. The RSR values were correlated with actual support
installations.

The system appears to be useful for selecting steel rib

supports for tunnels and is not used for selection of rock bolt
or shotcrete supports.

2.4.2.5 Geomechanics classification (RMR System) y

Bieniawski (1973) developed a quantitative system for
engineering classification of rock masses. The following
parameters are used to classify a rock mass using the RMR
 '

44

system:

1. Uniaxial compressive strength of rock material (q );

2. Rock quality designation (RQD);

3. Spacing of discontinuities;
4. Condition of discontinuities;
5. Ground water conditions; and

6. Orientation of discontinuities.

The first five of these parameters are grouped into five

ranges of values. Importance ratings have been allocated
according to the value of the parameters, a higher rating
indicating better rock mass condition. After the importance
rating of the classification parameters are established, the
ratings of the five parameters are summed to yield the basic
RMR for the structural region under consideration. The rating
of the sixth parameter is obtained depending on the nature of
construction. After the adjustment of the sixth parameter, the
final rock mass rating is estimated. The importance ratings
given for discontinuity spacings apply to rock masses having
three sets of discontinuities.

Table 2.3 shows the results of classification and

practical meaning of each rock mass class by relating it to
specific engineering problems.

The RMR system has been modified over the years as and
when more case histories were available (Bieniawski, 1979). It
may be mentioned that the angle of internal friction is
unlikely to decrease due to joints beyond a certain limiting
value. Thus angle of internal friction will not tend to be zero
when RMR tends to be aero as per last row of Table 2.3.
 45

TABLE 2.3

Geomechanics Classification of Rock Masses after

Bieniawski (1979)

41-60 21-40 <20

Rating 81-100 61-80

III IV V
Class No. I II

Fair Poor Very Poor

Description Very Good Good
Rock Rock Rock Rock
Rock
1 yr 1 wk 10 h 30 min
Average 20 yr
for 15-m for 10-m for 5-m for 2.5-m for 1-m
stand-up
span span span
time span span

300-400 200-300 100-200 <100

Cohesion of > 400
the rock
mass (kPa)
25 -35 15 -25 <15
Angle of >45° 35 -45
internal
friction of
the rock
raass (deg)

The RMR system initially developed for tunnels is also

applicable to rock foundations (Bieniawski and Orr, 1976),
slopes (Steffen, 1976 and Romana, 1985), ground rippability
assessment (Hardy, 1986) as well as to mining problems. The
system provides guidelines for the selection of roof support in
tunnels for long-term stability of various rock mass classes
keeping in view factors such as tunnel size and shape, in-situ
stress and method of excavation (Table 2.4).
Most recently, Lauffer (1988) presented a revised stand-up
time diagram specifically for tunnel boring machine (TBM)
excavation and superimposed it on the RMR diagram between the
y
stand-up time and span for various rock mass classes (Fig.
2.11). The superimposed modified diagram is shown in Fig. 2.12.
This diagram is useful for making an RMR adjustment for machine
 46

TABLI 2.4

Galdeliies for Iicavatlon aid Support of lock Tunnels it Accordaice litk the lock lass latlig Systei
[After Bieniawski, 1984]
Shape : horseshoe; width : 10 •; Vertical Stress :<2S IPa.'

Support

Excavation Rock Bolts (20 •• Dia,

Rock lass Class (drilling and blasting) Fully Grouted) Shotcrete Steel Sets

Very good rock Full face

I 3-1 advance Generally, no support required except for occasional spot bolting
RHR:81-100
Good rock Full face Locally, bolts in crosn 50 •• in crown where lone
II 1.0-1.5 • advance 3 • long, spaced 2.5 • required
RIR:61-80 Coiplete support 20 • Blth ocassional sire
froi face lesh
Fair rock Top heading and bench Systematic bolts 4 i long, 50-100 ii in crown and lone
III 1.5-3 i advance in top spaced 1.5-2 i in 30 ii in sides
KM8:41-60 heading crofln and Balls with
Coiience support after wire lesh in crown
each blast
Coiience support 10 i
froi face
Poor rock Top heading and bench Systeiatic bolts 4-5 • 100-150 ii in cron Light to
IV 1.0-1.5 • advance in top long, spaced 1-1.5 • and 100 ii in sides •ediui ribs
RMR:21-40 heading. Install support in crown and Ball with spaced 1.5 •
concurrently Blth wire iesh where
excavation 10 • froi required
face
Very poor rock Multiple drifts Systeiatic bolts 5-6 • 150-200 n in crown, lediui to
V 0.5-1.5 • advance in top long, spaced 1-1.5 i 150 •• In sides, and heavy ribs
RMR:<20 heading. Install support in crown and walls Blth 50 ii on face spaced 0.75 •
concurrently sith wire iesh. Bolt invert with steel
excavation. Shotcrete lagging and
as soon as possible forepoling if
after blasting required.
Close invert

A
c
M
Q.

H 10* 10 10 10" 10'

id 10 10
Stand-up time , hrs

FIG. 2-11 RELATIONSHIP BETWEEN THE STAND-UP TIME AND SPAN FOR
VARIOUS ROCK MASS CLASSES ACCORDING TO THE GEOMECHANICS
CLASSIFICATION (After Bieniawski, 1984 )
1day 1 week 1 month 1 year 10 years

o
o

10* 10" 10 10"

10 1 10* 10
Stand- up time , hrs
FIG 2-12 MODIFIED 1988 LAUFFER DIAGRAM DEPICTING BOUNDARIES OF ROCK
MASS CLASSES FOR TBM APPLICATIONS (After Lauffer, 1988 )
 48

excavated rock masses. Support load may also be determined from

the RMR system as proposed by Unal (1983):

p. Mfc 8MB $B (2 23)
2
where P = support pressure, kg/m ;

B = tunnel width, m;
3
#• = rock density, kg/m .

2.4.2.6 Q-system of classification (Barton et al., 1974 and

1975)

Q-system of rock mass classification was developed by Barton et

al. (1974) basically considering the tunnels and large

underground cavities under a variety of rock conditions and

support systems. The system derives its importance from the

fact that it was proposed on the basis of an analysis of 212

tunnel case histories from Scandinavia. Further it is a

quantitative classification system facilitating the design of

tunnel su^orts.

Barton's Q-sy3tem takes into consideration six different

parameters to determine the rock mass quality, namely (1) RQD,

(ii) number of joint sets, (iii) roughness of the most

unfavourable joint or discontinuity, (iv) degree of alteration

of filling along the weakest Joint, (v) water inflow and (vi)

stress condition.

The rock mass quality (Q) is calculated from the

relationship:

«= PK-» -£-* -iSrY

where,

FQD = rock quality designation,

Jn = joint set number,

 49

Jr = joint roughness number,

Ja = joint alteration number,

Jw = Joint water reduction number, and

SRF= stress reduction factor.

The rock mass quality can range from Q=0.001 to Q=1000 on

a logarithmic scale. The three quotients appearing in the above

relationship happen to be a rough measure of block size with

extreme values of 200 and 0.5, inter-block shear strength (of

the joints) and "active stress" respectively. Table 2.5 shows

the description of classification in Q-system.

TABLE 2.5

Q-system of Rock Mass after Barton et al. (1974)

0.001-0.01 0.01-0.1 0.1-1 1-4 4-10 10-40 40-100 100-400 400-
1000

exception extreme very poor fair good very Extre Excep

ally poor ly poor poor good mely tiona
good lly
good

Barton et al. (1974, 1975) plotted the long-term support

pressure against the rock mass quality (Q) and proposed an

empirical relationship as follows :

2.0 ^-1/3
kg/cm* (2.25)
roof * Q
wall Jr
Qw 1/3 kg/cm2 (2.26)

(number of joint sets is three or more)

where, P = ultimate rock pressure on roof

roof

P .,, - ultimate rock pressure on wall

wall

Qw = wall factor
>-

The wall factor (Q ) is obtained after multiplying Q by a

w

factor which depends upon Q as follows :

.- UC**-'-
 50

Range of Q Wall Factor Q

w

> 10 5.0 Q

0.10 - 10 2.5 Q

< 0.10 1.0 Q

If the number of joint sets is less than three, the above

equations have been expressed as follows :

Proof =2/3 Jn1/2 V' Qroof" ^™ (2.27)

Pwall '- 2/3 Jn1/2 V* QwaU ^GV? ^-^
From the above empirical approach, the short term support
pressure may be estimated by increasing "Q" to "5 Q". It may be
seen that the long-term support pressure is predicted to be
approximately 1.7 times the short-term pressure according to
Eq. 2.25. It may be recalled that Eqs. 2.25 and 2.26 are based
on capacities of support systems and not observed values.
The Q-system emphasizes that the support pressure is
independent of the tunnel size. The classification is thus more
general because of being non size-specific in character.

According to Jethwa (1984), the ultimate rock pressure

would reach upto 3 times the short-term elastoplastic support
pressure. He observed that the ultimate rock pressure predicted
by Q-system appears to be on lower side and may be applicable
to non-squeezing or elastic rock conditions. Barton (1988)
agreed that the Q-system is generally more effective and useful
for the elastic rock conditions.

Barton et al. (1974) related 'Q' to tunnel support

requirements by defining equivalent dimension (ED) which may be
obtained by dividing the span, diameter, or the wall height of
excavation by a quantity called excavation support ratio (ESR)

SOOKES I I!
 51

Thus,
excavation span or diameter or height (m)
Equivalent dimension - ESR
ESR is related to the use for which excavation is intended
and the degree of safety demanded. For permanent mine openings,
water tunnels for hydropower, pilot tunnels. drifts and
headings for large excavations ESR may be taken as 1.6. They
provided 38 support categories which gave estimates of
permanent support (Fig. 2.13).
For temporary supports ESR is increased to 1.5 ESR.
The maximum unsupported span raay be obtained as follows :
Maximum unsupported excavation span = 2(KSR) Q ' (2-29)
On the basis of empirical results, Barton (1984) proposed
a thumb rule for tunnels or excavations requiring seismic
design. This is to increase the term SRF of a particular rock
mass. Consequently, he found that Q (Seismic) is approximately
half the Q(Static). This provides for approximately 25 per cent
increase in support pressure (Fig. 2.14). This however needs
verification in the field.
It may be inferred from the above that Barton's Q-systera
may provide a design aid regarding the estimation of support
pressure in tunnels.
On the basis of their experience in two tunnels in China,
Zhen et al. (1984) commented that in the designing of rock
engineering projects many factors which are difficult to
define, are to be encountered, such as the orientation and
distribution of joints (discontinuities) in rock mass, in-situ
stress field, the mechanical properties of rock mass and
discontinuities and water flow in them.
 y

0001 100 1000

Rock mass quality Ol

FIG. 2-13 EQUIVALENT DIMENSION v/s ROCK MASS QUALITY (Q-SYSTEM ) ( After Barton et al., 197^ )
 *

Very poor Very good

OJ
E
o

2C

m tn
m
01

o
a
a

tn

-n ->a SRF
FIG. 2-1*» SEISMIC REDUCTION OF Q-VALUE TO OBTAIN 25 7. INCREASE IN SUPPORT PRESSURE
(After Barton, 198*f)
 54

He further commented that for large permanent underground

openings, it is much raore difficult to conduct these geological

investigations and in-situ rock mechanics tests. Furthermore,

in the case of deeply buried long tunnel it is sometimes

impossible to estimate the geological conditions for every part

of it. The design of permanent linings and temporary supports

have to be based on only preliminary investigations. It is

necessary to revise the original design for every part of the

tunnel according to the newly revealed geological data and the

performance of rock mass monitored in the process of

excavation.

On the basis of experience from 44 case studies of coal

mine roadways, Sheorey (1991) observed that the Q-systera is

inadequate for certain geological features not covered by it

and also when joint orientation is unfavourable. According to

him some modifications are necessary for coal mines. These were

suggested for horizontol stratifications, joint orientation and

some special geological structures which may be encountered in

these functions. For horizontally stratified formations it was

recommended to double the original value of rock quality which

may then be reduced further for unfavourable joint orientation

(or horizontal stress direction) or special structures. The

classification known as CMRS classification (Venkateshwarlu,

1986) applicable mainly to coal measures gives ratings to five

geomechanical factors, namely (i) layer thickness (ii)

structural features (iii) slake durability (iv) intact rock

strength and (v) ground water. Rock load or mean support load
density is then calculated from :
 55

P =£.B.F. (1.7-0.037 R+0.0002 R2) (2.30)

where, £ = unit dry wt. of rock
B = roadway span

F - safety factor

R = total adjusted rating

Sharma and Ramaraurthy (1985) presented a rock mass

classification system for underground openings based on ground
convergence and recommended that the method of convergence
confinement or the method of characteristic lines as it is
sometimes called, is probably the best method available today >
for explaining the complex phenomenon of restabilization of
ground subsequent to tunnelling and supporting. It has the
specific advantage that it can provide a rational explanation
to various aspects of the interaction between the ground and
the support to some extent, taking the element of time into
account (Ladanyi, 1974, 1984).
He reported that the basic requirement of a tunnel support
system is that it should control the movement of the strata
surrounding the tunnel in such a way that the function of the A
tunnel is not impaired. Unless the tunnel is passing through an
urban area where very small deformations can be permitted, for
tunnels in weak rock subject to high stresses, it is more
realistic to accomodate the ground movement rather than resist
it. The support system may be so chosen, as to permit the
required amount of deformation. The support systeras for
A

different categories were obtained as follows (Table 2.6).

 56

TABLE 2.6

Classification System Based on Ground Convergence

after Sharma and Ramaraurthy (1985)

Total Rating > 80 60-80 40-60 20-40 <20

Ground Conver Behaviour Less 5- 20 20-100 More

gence (u1/ri)X10 J is elastic than 5 than 100

Category Very Good Good Fair . Poor Very

Poor
Suggested No support Rock Rock Shot Yielding
Support bolts bolts crete arch alo-
System ngwith
rockbolts
and shot
crete
,

Sharma (1990) emphasised the importance of main factors

which affect the performance of the support system and should

therefore be considered in support design. These are as

follows:

(1) the displacements occuring after support

installation;

(ii) the stiffness of the rock raass;

(iii) the strength of the rock mass;

(iv) the stiffness of the support system; and

(v) the strength of the support system.
The presence of water is a factor which not only affects
the above parameters but also has a very significant additional
action. One over-riding consideration that deserves special
mention is the influence of construction procedures on support
loads.

Sharma (1990) re-emphasised on the following points

regarding the performance of support system.
 57

(i)
the time of support installation; since it would
determine the extent of displacements which have
occured prior to installation.

(ii)
the stiffness and the strength of rock mass are
affected differently depending on whether the tunnel
was bored, blasted, pre-splitted, smooth-wall blasted
etc.

(iii)
the stiffness and the strength of the support system
depend on many factors, for example,

Steel sets block spacing, amount of over-break, footing

conditions etc.

Concrete voids behind the lining, quality of back-fill,

Rockbolts anchoring, bearing plate,

Shotcrete composition, application.

The above facts indicate that it may not be possible to

define "strength" or "load bearing capacity" for support
systeras in an absolute sense.
According to Singh et al. (1992), the correlation between
rock mass quality and support pressure proposed by Barton et
al. (1974) proved useful except in cases of squeezing ground
conditions. Field data collected by Central Mining Research
Station, Dhanbad from 20 tunnel sections indicated a clear need
for correction factors to account for height of overburden and
tunnel closure which do not seem to be adequately accounted for
by the stress reduction factor. On the basis of available field A

data Singh et al. (1992) showed that;

1. Squeezing is likely to occur, where the height of
1/3
overburden in metre exceeds 350 Q
 58

2. The shot-term roof support pressure is given by the

following correlation :

Pi(roof) =-IT" ^Q)"1/3 .«.*'

r
(2-31)
in which f is the correction factor for thickness of
overburden (H) in metres, and £' is the correction factor for
tunnel closure (equal to 1 for non-squeezing ground
conditions). The value of correction factor f is given as
f = 1 + (H - 320)/800 > 1 (2.32)
3. In squeezing ground conditions, the support pressure
was significantly influenced by tunnel closures. The
correction factor f" for tunnel closure varied from 0.7

to 1.8 in the case of a single tunnel. The minimum

support pressure occured when the tunnel closure was

about 5 per cent of the tunnel diameter. The support

pressure increased rapidly beyond this limiting
closure.

4. The actual wall support pressure for non-squeezing rock

conditions is likely to be negligible. The short-terra
values of PlR/Plr depend on Qx (=5 Q), as given below :
Piw/Pir \
10 5Q < 0.1
1.0-0.0 5 < 5 Q < 0.1
00 5Q > 5
Piw and pir are predicted short-term wall and roof
support pressures respectively. Q± is the short term
rock mass quality.

5. The ultimate support pressure may be 1.75 times the

short terra support pressure for tunnel sections under
 59

non-squeezing ground conditions, except for cases of

soluble and erodible joint fillings with seepage. *

6. Support pressure is independent of the tunnel size,

provided that Q is obtained from a full sized opening.
2.5 Allowable Bearing Pressure on Rocks

Rock is usually an excellent foundation material, but

near surface rock raay be significantly fractured and it is

always necessary to establish the competence of the rock to

bear the required load at acceptable levels of deformation or

settlement. When a structure imposes a large loading on the

rock rasss, the selection of a design bearing pressure and the

computation of the foundation settlement become very important.
However, there i3 some difficulty in estimating the bearing

capacity of rocks. The estimation of a design bearing pressure

can be accomplished using any of four different procedures :

(1) full scale load tests; (2) standard codes of practice; (3)
rational methods based on bearing capacity and settlement

analysis; or (4) empirical rules. Full scale load tests are the
surest but these are expensive and much time consuming and

therefore raay not be conducted in routine. Moreover, it is not

practicable to test at full scale and, even at reduced scales,

as the representative volume may be too large to test to

failure. Codes of practice provide conservative design

pressures and also vary considerably in their recommendations.

Rational methods are quite lengthy and involve a number of ^

assumptions. For the purpose of preliminary designs, allowable

bearing pressure may be quickly estimated through an "empirical

approach".
 60

Peck et al. (1974) suggested an empirical correlation

between the allowable contact pressure and the RQD (Fig. 2.15).

The correlation was suggested for a rock mass in which

discontinuities are tight or are not open wider than a few mm.

The authors contend that with above considerations, the

foundation settlement raay not exceed 12.5 mra.

Table 2.7 presents an abstract from various building

codes, for assessing allowable pressure on different quality

rock masses.

TABLE 2.7
2
Allowable Pressure on Rocks (kg/cm )

Abstract from various Building Codesa (After Peck et al., 1974)

Material Codeb
A B C D

Massive crystalline bed 110.00 110.00 0.2qc 11.0

rock including granite,
diorite, gneiss, traprock,
hard limestone and dolomite

Foliated rocks such as 44.0 44.0 0.2qc 4.4

schist or slate in sound
condition

Bedded limestone in sound 44.0 16.5 0.2q 4.4

c
condition

Sedimentary rocks, 27.5 16.5 0.2qc 3.3

including hard shales and
sandstone

Soft or broken bedrock 11.0 0.2qc

(excluding shale and soft
limestone)

Soft Shale 4.4 0.2qc

a. values do not include increase allowed for embedment,

b. A=B0CA (1968), B= National Building Code (1967),

C= Uniform Building Code (1964), D= Los Angeles (1959)

 ' '.. ' ;•• ••: "2 '^^"

•.'••'; .• . _ •'•• ' • •.;...- '•_••' • ••-.....

61

c. Uniaxial Compressive Strength

Limitations

1. Settlement is unlikely to exceed 12.50 mm even for large

loaded areas.

2. Bearing pressure of jointed rocks is governed by

settlement rather than by bearing capacity (failure) of

rock mass.

3. Correction of submergence under ground water-table and

correction for unfavourable dip of joints must also be

applied.

IS:12070-1987 recommends that plate load test must be

conducted on very poor rocks where allowable bearing pressure

is suspected to be less than 100 %/ tn . For rocks of better

quality, Table 2.7 and classification no. 'A' may be used.

Ramegowda (1990) believed that bearing capacity of the

rock mass is not considered to be a major problem in large

structures like darns. Measurement of ultimate bearing capacity

of rock mass by in-situ test is difficult since it is seldom

possible to reach the failure load due to experimental

constraints. Using a 20 cm diameter rigid pad, tests were

carried out (Ramegowda, 1990) at several dam sites. It was

possible to reach failure load only in the case of very soft

rock masses (Fig. 2.16). In situations where it was not

possible to reach the failure load, the approach was to stress

the rock mass to at least 5 to 6 times the maximum stress that

is likely to be imposed after structure is built.

Standard practice as adopted in India suggests estimates

of net safe bearing pressure on the basis of classification

•
 62

300

1*0 60

R Q D (•/•)

FIG. 2-15 ALLOWABLE CONTACT PRESSURE ON JOINTED

ROCK (After Peck et al., 1974 )
• • •'

GATEWAY *

40h
118 kg/cm2
Rigid pad
35 315 cm2

30

S 25
c
o

.£ 2 0

8 10 t2 U
Displacement in mm

FIG.2-16 BEARING CAPACrTY TEST RESULT

IN CASE OF VERY SOFT SHALE
(After Ramegowda * 1990 )
 64

approach (Table 2.8).

TABLE 2.8

Net Safe Bearing Pressure Based on Classification

(After IS:12070 - 1987)

Material Net Safe Bearing Pressure (t/m2)

Massive crystalline bed rock 1000
including granite, diorite,
gneiss, trap rock

Foliated rocks such as schist 400

or slate in sound condition

Bedded limestone in sound 400

condition

Sedimentary rock, including 250

hard shales and sandstones

Soft or broken bedrock 100

(excluding shale) and soft
limestone

Soft shale 40

Experimental work (unpublished) carried out at University

of Roorkee demonstrated that rock mass rating (RMR) may also be
used to give net allowable pressure on rocks (Fig. 2.17). The
curve has been developed so as to ensure settlement of raft
foundation upto 6 m width to be less than 12 mm. The RMR in the
Fig. 2.17 is the average within a depth below foundation level
equal to the width of the foundation, and the RMR is fairly
uniform within the depth. If the upper part of the rock, within
a depth of about one fourth of the width of foundation, is of
lower quality the value of this part should be used, or the
inferior rock should be removed. The values in Fig. 2.17 are
based on limiting value of the settlement. They should not be
increased, if the foundation is embedded into the rock.
 65

2.6 Concluding Remarks

Review of literature reveals the following major gaps of

immediate importance :

1. There is no unique classification system which could

cater for all types of rock masses and support systeras.

However, the RMR and the Q-system seem to be useful for

most of the situations.

2. The existing empirical approaches for estimating

modulus of deformation have been basically developed in
hard rock areas and these may be more applicable to
hard rock situations. The rock raass under investigation
at the six hydroelectric projects under study fall3
under poor to fair quality rocks. It is necessary to
test the validity of existing empirical approaches and
recalibrate the same for poor and fair quality rock

masses.

3. The existing empirical relationship between the support

pressure and the Q-system developed by Barton et al.

(1974) originated for tunnels in hard rock situations.
The extent of validity of this relationship when

applied to poor and fair quality rock masses is not

known. Strong need is felt to suggest a suitable
relationship to predict short term support pressure for
the design of tunnels in poor and fair quality rock

masses.

4. The shear strength pararaeters, cohesion (c) and the

angle of internal friction (0) of the rock mass, being
the output from the geomechanics classification have
 66

30 40 50 60 70 80 90

Rock mats rating (RMR)

FIG. 2-17 ROCK MASS RATING (RMR) v/s NET ALLOWABLE PRESSURE ON ROCKS
[After University of Roorkee, 1991 (unpublished )]
 67

emerged from the rock raass under specific geological

conditions. It is necessary to investigate whether or

not, the shear strength parameters (c & &) predicted

through geomechanics classification are applicable to

rock masses of poorer quality.

5. The non-linear Mohr envelopes developed by Hoek and

Brown (1980) are based on a few tests and need to be

re-evaluated for application to Lesser Himalayan rocks.

6. It is necessary to'investigate the effect of saturation

on the engineering properties of poor rock masses to

predict the behaviour of pressure tunnels and dam

foundations.

7. Investigation is also required on the use of

engineering rock raass classification approach for rapid

estimation of allowable bearing pressures on the rocks

on the basis of large size plate load tests.

8. Civil engineers have unfortunately ignored the

anisotropic behaviour of rock mass. Proper field

testing is thus required.

In view of the foregoing gaps, an attempt has been made in

the subsequent chapters to fill a few of these gaps in our

understanding by field raeasurements and careful interpretation

of data from 3ix major hydroelectric projects located in the

Lesser Himalaya in India.

CHAPTER 3
 CHAPTER 3

GEOTECHNICAL PROBLEMS IN HIMALAYA

3.1 General

India's sixty per cent hydro-power potential lies in the

Himalaya. Out of this huge potential hardly ten per cent has

been developed so far. Therefore, the development of river

valley projects for harnessing the untapped water resources is

of vital necessity in the Hiraalayan region. The Himalaya owing

to their unique tectonic history present a complex

heterogeneous mass. The rock formations in Himalaya, being

youngest in the world, are intricately folded, faulted or

thrusted. Because of weak rocks and the complicated geology,

variety of geotechnical problems have been encountered in a

number of river valley projects built in the region.

3.2 The Himalaya

The Himalayan range extends from Kashmir to Assam having a

length of about 2400 km and width ranging between 160 and 400

km. Geomorphologlcally the Himalayan region is divided into

three belts from North to South.

(i) The Great Himalaya

The inner line of high mountain ranges rises above the limit of

perpetual snow with average height 6100 m. On it are situated

peaks like Mount Everest, Kanchenjunga, Dhulagiri, Nandadevi

etc.

(ii) The Lesser Himalaya

This is the middle range which i3 a series of ranges closely

related to the former but at lower elevation. The average width

 69

is about 80 km and average height'3200 to 4500 metre.

(iii) The Outer Himalaya or The Siwalik Range

It is the mountain ranges which intervene between the Lesser

Himalaya and the plains. The width of these ranges varies from

8 km to 50 km and average height 900-1500 m.

Geological Structure of the Himalaya

The geological structure of the Himalaya falls into three broad

zones :

(i) The Northern or Tibetan zone - This zone is composed

of a continuous series of highly fossiliferous marine

I

sedimentary rocks, ranging in age from the earliest

Palaeozoic to the Eocene age. Except near the

North-Western extremity, (in Hazra and Kashmir) rocks

belong to this zone but rarely occur South of the

line of snow peaks.

(ii) The Central or Himalayan Zone - It comprises most of

the Lesser or Middle Himalaya together with the Great

Himalaya. It is mostly composed of crystalline and

metaraorphic rocks, gneisses and schist with

unfossiliferous sedimentary Purana and Mesozoic

deposits.

(iii) The Outer or Sub-Himalayan Zone - This zone

corresponds to the Siwalik ranges and is composed

almost entirely of tertiary or principally of upper

tertiary, sedimentary river deposits.

The rock formations in the Himalaya are intricately

folded, faulted or thrusted because of tectonic deformations.

 70

The slopes of the Himalayan valleys are quite steep and mostly

unstable and comprise of rocks which are generally weak due to

tectonic uplift followed by river cutting and are therefore

susceptible to weathering and erosion. The younger rocks are

made up of sandstone/sandrock with bands of steep dipping shear

zones of claystone/siltstone. These rocks are soft, highly

weathered and fractured.

Seismic History of the Region

The Himalayan region is seismically very active as is evidenced

from the past earthquakes of Delhi 1720, Calcutta 1737, Bengal

1762 and 1885, Kashmir 1885, Assam 1897, Kangra 1905, Bihar

1934, Baluchistan 1935, North East Assam 1950 and Uttarkashi

1991. The entire Himalayan belt along the Ganga and Yamuna

valleys is also seismically very active and has been marked by

frequent occurrence of moderate size earthquakes. In a recent

earthquake in the region on October 21, 1991 registering 6.1 on

the Richter scale, more than 500 people were killed and over

1000injured in house collapses and landslides in Uttarkashi

district alone. The tremors which began at 2.53 A.M. lasted for

43 seconds. The epicentre was located at 30.7-5° N(Lat), 78.86°E

(Long) with its focal depth at 12 km.

Seismicity and its Influence on Local Geotechnical Conditions

It is known that each seismic activity brings about some

changes in the physical and mechanical properties of rocks. For

example, a seismic activity raay create problems of water

leakage and increased inflow due to joint slip around a tunnel

in the neighbourhood of faults and thrusts. Cracking of

 71

concrete or shotcrete lining and associated deformations may

have several adverse effects on the overall performance of a

project. The new joints after seismic shock, may cause renewed

movements along the existing fractures. Such movements may

further be responsible for more support pressures on the tunnel

linings. Besides, the dynamic loading may change the overall

deformability of rock mass, its strength characteristics and

the load bearing capacity of the foundations. This may in turn

reduce the resistance of engineering structures to

after-shocks.

3.3 Geotechnical Problems In the Himalayan Projects

Several multipurpose and hydroelectric schemes are located in

the Lesser Himalayan region of India. The raost notable of these

are mainly in the States of Jamrau and Kashmir, Punjab, Hiraachal

Pradesh, Uttar Pradesh, Manipur and Assam. A couple of projects

in these States have already been commissioned. A number of

similar projects are proposed to be constructed in the Lesser

Himalaya confronting the same complex geology. Some of these

projects are under investigation and planning stage while

others are under design and construction stage. Fig. 3.1 shows

the location of multipurpose and hydroelectric schemes in Ganga

and Yamuna valleys. '

It has by now been common experience that those

hydroelectric projects which do not involve underground works

are generally completed expeditiously but those projects which

involve underground works have to face time over-runs due to

various uncertainties about the rock conditions in the

 *r

FIG. 3-1 HYDROELECTRIC SCHEMES GANGA AND YAMUNA VALLEYS

 73

Himalaya. In future constructions, almost all civil engineering

projects in Ganga and Yamuna valleys will involve construction
of tunnels and underground power houses due to unstable nature
of steep valley slopes.

In almost all the major underground works in the Himalaya,

surprises were encountered during execution. It is felt that no
amount of intensive investigation can completely prevent such
surprises. Apart from the known faults, thrusts, shears,
synclines etc. the rock mass behaviour depends not only on the
type of rock, but also on its state of fragmentation, Joint
pattern, dip, strike etc. which vary rapidly, suddenly and
unexpectedly in the Himalaya.
The case records of major river valley projects in the
Lesser Himalaya are given below :

3.3.1 Ramganga Tunnel - Ramganga River Valley Project,

Kalagarh, Uttar Pradesh

General

Ramganga project comprised completion of a 127.5 ra high earth

and rock fill dara, a power house, irrigation outlet system and
a vast canal system. The project was commissioned in 1975. It
is one of the first few high earth and rock fill dams in India
constructed on weak rocks close to areas prone to seismic
activity. It was the first venture in the Himalaya which laid
the foundation for still greater ventures coming up today in

these lofty mountains.

Regional Geology

The regional geology consists of Siwalik formations which have

 74

been folded into an asymmetrical anticline pitching at an angle

of 20° towards the South-East direction. The repeated

downstream dipping Siwalik sequence forming the Southern limb

of anticline is not fully exposed and is either lying concealed

below the Indo-Gangetic alluvium truncated by a possible thrust

fault located approximately 3.2 km downstream of the main dara.

The formations continue Northwards and Eastwards from the main

dara and at the saddle dam site. The rocks in the area consist

of sandrock with horizons of calcerous nodules, clayshales,

large boulder gravels, cobble gravels, sands and silts. The

area has been subjected to intense compression in the

geological past resulting in the development of a number of

thrust planes. All these thrust planes are of regional type and

possess the NWSE structural trend of the Himalaya and dip in a

North-Easterly direction.

The geology in the projet area consisted of alternate

bands of poorly consolidated and cemented sandstone and

clayshale,the latter constituting about 25% of the total rock

sequence in the dam area and the sandstone predominates the

clayshale. The general strike of the rock was observed N 40

W/S 40°E to N 80°W/S 80°E with dip varying between 35° to 45°
towards upstream direction.

Two diversion tunnels (each of 9.45 m finished diameter)

were provided to cater for the diversion requirements during

construction period. Both the tunnels passed through soft and

complex geology of Lesser Himalaya. There were intermittent

rock falls and damages which occurred during excavation of the

 75

above tunnels. The inlet portals of both the tunnels were

located in sandstone band dipping 30 towards North-East.

Outlet portals of tunnels were also located in sandstone. The

sandstone was found to be massive and coarsely jointed in

general but showed disintegrating property under water. The

clayshales were found to be either greenish or chocolate in

colour and at many sections the rock mass was found to be

extremely soft. Fig. 3.2 shows geological cross-sections along

the longitudinal axis of Ramganga tunnels no. 1 and 2.

Geological investigations in pilot tunnel (2.5 m X 2.7 m)

driven along the alignment of tunnel No. 2 showed the following

typical features :

1. The sandstone band frequently showed thinly bedded and

highly laminated zones which were found to "give way and

fall easily in lumps on saturation. The over-break in one

case rose to 100 per cent of the excavated height of the

pilot tunnel, the progressive failure taking place along

the bedding planes in the micaceous rock through a series

of zig-zag pattern of vertical shear cracks.

2. Sorae of the thinly bedded and cemented zones were found to

be intersected by well developed sets of joints.

3. Most of the sandstone bands were found to be saturated.

4. Active flow of groundwater with ferric oxide and sulphate

was recorded along certain joints in a few bands of

sandstone.

5. Short stand-up time were noticed during excavation of

pilot tunnel.
 76

6. Large scale flowing conditions encountered in the

sandstone bands.

7. The clayshale bands were found to be dry in general, but a

Ii 1 \ , .
few bands of interbedded siltstone showed swelling
I'M

property with no active flow.

Pilot tunnel was excavated to ascertain the rock

characteristics in tunnel no. 2. The compressive strength of

2
clay-rocks and sand-rocks was found to be 2.1 kg/cm and 7.0

kg/era respectively, whereas modulus of elasticity ranged

between 0.85 X 104 kg/cm2 to 3.5 X 104 kg/era showing very poor
rock mass condition. The field permeability ranged between 3.0

X 10~5 era/sec to 9.45 X 10 cm/sec in clay and sandstone band.

Due,to their generally poor cementation, the sandrocks were
found to be very soft and friable, sorae of the bands completely

disintegrated when soaked under water.

Twenty nine years ago when construction work of the

diversion tunnels was undertaken (1963), no advanced technology

of tunnel excavation was available. Therefore, old conventional

raethod of drilling and blasting was adopted for excavation.

Rock mass in the area being soft, full face excavation was

considered unsafe. Installation of steel rib support in the top

heading was considered necessary before taking next blast. The

excavation in the lower half portion was proposed after

installing proper steel supports.

In order to continue tunnelling operation during monsoon,

masonry protection walls were constructed at the inlet and

outlet portals of tunnel no. 1. The excavation programme was
 it-

-Crest structure
Chute spillway
Ground profile
Intake

Inlet portal D/S Portal

EL 268-836
Stilling basin
Reach
-Valve chamber ^Drainage tunnel El. 262-712
K-A-*kC-»K D- --hBK>HBk 0 •++*- -B
K- •H*A*
713 m >J

L-SECTION OF TUNNEL NO. 2

Control chamber u_Oam axit •j

Access tunnel. \Tb. 372-0 -j

Intake structure **^°" a^r¥*t

Natural ground

Cover section

L- SECTION OF TUNNEL NO-1

LEGEND
Reach A-Normal ribs at 305 mm Over burden f°H3
Reach B- Normal ribs at 610 mm Terrace gravel j*2]
Reach C- Larger ribs Sand rock p^
Reach D-Ribs at 610mmto 1220mm
Clay shale

FIG. 3-2 GEOLOGICAL CROSS SECTION ALONG THE LONGITUDINAL AXIS OF

RAMGANGA TUNNEL NCM AND NO-2
 78

made in three stages viz. (i) top heading (ii) side segments
and (iii) bottom portion. After about 35 m excavation of
heading, removal of side segments was started for the erection
of ribs. It was only after the erection of four ribs @ 1.2 m
spacing, when a portion of rock raass slipped (after 'seven
weeks) in a length of about 1.5 ra resulting in twisting of
ribs. Further rockfall continued in about 6 ra length. Spacing
of ribs was thereafter reduced from 1.2 m to 0.6 ro and
excavation continued with blasting in 1.5 ra deep drill holes.
After about seven weeks, the blasting holes were increased upto
2.13 ra to obtain better progress which caused excessive
over-break in the arch portion. A chimney of several metres was
observed inside the tunnel and rocks continued to fall. After a
lapse of about a week, there was again a heavy rock-fall. The
roof was then supported by fixing perfobolts. But a series of
rock failure continued to take place during subsequent month
before the tunnel collapsed eventually with a thundering sound.
Finally, the modified multiple drift heading and bench
method was adopted for tunnel excavation. In this method, the
upper half section of the tunnel was excavated first. Wall
plate drifts were made in advance and they were placed to
correct alignment, grade and angles. The upper segments of
circular steel ribs were then installed resting on wall plates.
After this, lower half section was excavated ensuring that
adequate bearing was maintained under the wall plates and lower
segments of the full circular ribs were installed In the first
instance. Fig. 3.3 shows tunnel excavation plan of Ramganga
 79

tunnels.

The tunnel failure may be attributed to the extremely poor

rock mass condition in the diversion tunnel.

3.3.2 Chibro-Khodri Tunnel, Yamuna Hydroelectric Project, Stage

II, Uttar Pradesh

General

A 55 ra high, 155 ro long concrete diversion dam has been

constructed at Ichari about 32 km upstream of Kalsi on river

Tons a tributary of river Yamuna. Through an intake on the

*
right bank, the water is fed to a sedimentation chamber in the

river and then to a 7 ra diameter lined 6.3 km long tunnel

taking water to Chibro power house. On the downstream is an

underground collection gallery which leads water through a

syphon tunnel, and 5.9 km long, 7.5 ra diameter tunnel to Khodri

powerhouse (Fig. 3.4). ,, , . ..,,.,

Geological Appraisal of the Problem

,The 5.9 km long Chibro-Khodri head race tunnel passes through

two major tectonic features - Krol and Nahan thrusts (Fig.

3.5). It was anticipated, that tunnelling through this

intra-thrust zone would be a problera. To enable study of rock

raass behaviour in the intra-thrust zone in advance and to

collect data for design of tunnel lining, an intermediate

gallery was provided. However, when the intra-thrust zone was

,,actually encountered, tunnelling became extremely difficult due

to highly squeezing ground condition in the crushed rock

charged with water and work came to a standstill. The

excavation of tunnel through this most difficult squeezing

 80

Beam segments

FIG. 3-3 EXCAVATION PLAN OF RAMGANGA TUNNELS

 81

Kishau dam.

Yamuna hydel sch[e.me

Stage-ll, part-l

Dakpathar

Dhakrani
branch

Jiwangarh
grant
Jelpura
branch

Dakpathar barrage &

head regulator

FIG- 3^ INDEX MAP- YAMUNA HYDROELECTRIC SCHEME STAGE-II

 82

FIG. 3-5 REGIONAL GEOLOGY OF THE YAMUNA HYDROELECTRIC

SCHEME STAGE-I
 83

ground was made possible through ingenious engineering. Instead

of a single 7.5m diameter tunnel, three smaller tunnels of 4.8

m diameter were excavated and one km length of tunnel of this

reach was designed a3 three smaller tunnels (Fig. 3.6). The

work was successfully completed in 1983 having no precedence in

the world. Fig. 3.7 shows the geological sections of the

Chibro-Khodri tunnel.

Another major problem in the Chibro-Khodri head race

tunnel had been the heavy seepage and building of excessive

external pressure behind the lining. Thick lining was required

to take-up the excessive external pressures. Instead, high

pressure grouting of upto 6 to 7 times the internal pressure

was resorted to. Very high closures of 10 ram to 150 ram and rock
2
pressures upto 10 kg/cm were recorded.

At the Khodri power house site the main problem was that

of instability of power house back slope which could have

damaged the power house and also reduced the lateral cover for

surge shaft and adit portal. The problem was. indetified and

duly tackled by installing rock anchors and shotcreting the

rock slope surface.

3.3.3 Tunnel Collapse at Chainage 238 m in Dharasu, Downstream

face of Maneri Bhali Hydel Scheme, Stage-II, Uttarkashi,

Uttar Pradesh

General

The 6 m diameter, 16 km long head race tunnel of Maneri Bhali

Hydel Scheme, Stage II faced many unforeseen problems due to

the geologically complicated nature of rock formations. On 22nd

 84

July 1982 a portion of the tunnel collapsed at chainage 238 ra.

The studies revealed that the collapse had taken place due to

the mobilisation of the brecciated greywacke, overlying a rock

o

wedge, created by an adversely oriented joint dipping at 20

towards North, above the foliation shear zone. The rock mass

overlying the shear zone was saturated as this was acting as a

barrier, which in turn, facilitated failure even at low angle
of intersection line getting daylighted on the left side face

of the tunnel in these reaches. This insight led to realignment

of the tunnel in this distress reach.

Reasons for tunnel collapse

The successive development of cavity formed at chaniages 238 ra,

242 m and 245 ra has been schematically shown in Fig. 3.8. The

stereographic projection of the discontinuities revealed that

the rock wedges were formed by the intersection of joint sets

1-2, 2-3, 2-4 and 3-4. The plunge of the last three rock wedges
was towards the heading, so no failure had taken place. The

rock wedge,formed by the intersection of joint sets 1 and 2 ^

plunged at 20° towards portal,failed when foliation shear along
joint set no. 1 was punctured at the crown. Dripping was
continuously observed from chainage 221 ra but no arrangements

were ready for immediate removal of water. With the gradual

removal of muck resting at 30°, the toe support was disturbed
and cavity went on enlarging and reached to the maxiraura of 7 m
at chainage 245 m (Fig. 3.8). It then gradually reduced when
foliation shear acting as the barrier for water ingress shifted
to the left of the invert and brecciated greywacks overlying
 ¥• 4

/Expansion chamber
/ Syphon tunnel gate Kalawar trifurcation
/ / operating chamber
Parallel observation gallery
Kalawar inspection gallery

Spilling
channel

oo

FIG. 3-6 LAYOUT PLAN OF YAMUNA HYDEL SCHEME STAGE-II (PART-II )

 i

Quartzites, slates, limestone Talus

Intra-thrust zone consisting of eJ
Fault
crushed red shales,slltstone,clay V
Sandstone, siltstone, claystone *• Thru;

m
1200
rrom underground power \\
at ChW -

y?PH^. 7-Om to 7»5m dTatTTni

Head race tunnel
'^0?* o>

»\* Jo be excavated •*!*-

a- ORIGINAL GEOLOGICAL SECTION (After Auden,1942)
0 200 400 600 m
Scale

1000

800-

600 lr '1292m. XZZ3, -*" V /»um-/-am. •••••••••

=*- ESSp^T <<r'r'-ll590m.-* . Head race tunnel " fT Khod ri
400 Chhibro
/' m
200 F R» Fo
Excavated LJ3^tobeexcavate**+» Excavated

5600m

b- REVISED GEOLOGICAL SECTION (After Jain et al.,1975)

FIG. 3-7 GEOLOGICAL SECTION OF CHHIBRO - KHODRI TUNNEL

 87

S47°W

N43°W

SECTION AT CH 2 38 m
SECTION AT CH 242 m

SECTION AT CH 245 m

FIG. 3-8 SCHEMATIC REPRESENTATION OF CAVITY DEVELOPMENT IN

HEADRACE TUNNEL - MANERI BHALI HYDEL SCHEME,
STAGE-E
 88

the rock wedge had already slipped down. The tunnel was later

diverted through a side drift made at right spring level

commencing from chainage 232 m and it bypassed the collapsed

reach.

3.3.4 Ichari-Chibro Tunnel, Yamuna Hydroelectric Project, Stage

II, Uttar Pradesh

General

The scheme provided construction of;

(1) 55 ra high concrete gravity dara for diversion of river

Tons into power tunnel at Ichari; and

(ii) 7 m diameter, 6.3 km long circular concrete lined

tunnel to conduct water to the underground power

house at Chibro (Fig. 3.9).

The project site existed in an area which is geologically

much disturbed. The tunnel alignment runs from North to South

and is wholly located in geological formations which are folded

into a syncline called "Jaunsar Syncline". It is bounded on

either side by thrusts dipping in opposite directions on the

North by Tons thrust and on the South by the Krol thrust. The

axis of the syncline is in East-West direction i.e., nearly at

right angles to the direction of the tunnel.

Geology

The tunnel passes through main formations occuring successively

from the intake (i) Chandpurs comprising mainly of slates with

minor quartzites (ii) Nagthats consisting mainly of quartz!tes

with slates and (iii) Mandhalis comprising of mainly slates and

limestones. The entire formation along the tunnel alignment was

 89

interspersed with minor and medium shear zones, minor folds and
faults. In raany reaches the rocks were almost paper thin. The
bedding planes of the rock in some reaches were graphitic and
showed a tendency to slide along the bedding planes. They

generally had a dip of about 60o-60°. In some' ireaches the

Joints were filled with clay gouge. In general, the rock raass
conditions were poor and unfavourable for tunnelling.

Throughout the tunnel, seepage of water and dripping conditions

were met.

The alignment of the tunnel was kept at right angle to the

; axis of the Jaunsar Syncline supposed to be the most favourable

direction for tunnelling through the bedded rock. No serious

problems were expected during construction of the tunnel.

However, in practice the frequent existence of faults, seams

and slip planes running nearly parallel to the axis of the

tunnel presented many problems. Some of the problems

encountered during construction are as follows :

1. Considerable loosening of rock and sizeable rock falls

took place progressively after blasting. The rock

behaviour was closely watched and it soon became clear

that the tunnel support had essentially to be such that

radial rock restraints are restored as quickly as possible

after blasting and over as much area as possible of the

exposed rock.

Shotcreting seemed the only and obvious answer to the

problem and was used in conjunction with rock bolts and

perfo-anchors followed by concurrent lining of the arch.
 90

(chart
U-Axis of jaunsar syncline Chhibro power Khodri power
InJake- house house

1000 r,ver
500

(Slate) 1 (Quartzites with lime- t Sand-stone and clay

stone stone
Nagthat quartzltes L_oogshal subbathu

GEOLOGICAL SECTION

Minimum excavation Shuttered concrete

line lining

Kerb

SECTION OF H;R.T.

FIG. 3-9 LONGITUDINAL GEOLOGICAL SECTION AND CROSS SECTION

OF CHIBRO POWER TUNNEL
 91

Difficulties however arose because over-breaks extending 3 to 4

m took place almost simultaneously with the blast followed by

auxiliary rock falls due to subsequent loosening of the rock.

These rock falls, even from reaches where shotcrete was

complete, did not allow continuous application of shotcrete.

Backfill concreting became necessary at many places and a very

peculiar situation arose. Controlled blasting and shotcreting

reinforced with chain link fabric and rock bolts were tried in

such zones with no success.

After a careful study of the persistent rock behaviour the

rock raass was reinforced by perfo-anchors in advance of

blasting and thus it tended to become self supporting and

permitting sufficient bridging period for application of

shotcrete or steel ribs. The following methodology was adopted:

1. A pilot tunnel 4 ra X 4.5 ra was driven ahead of the face

in a length of 15 to 20 ra keeping the same invert level

as that of the raain tunnel.

2. Radial holes were driven along the roof from the pilot

tunnel to a depth of 3 to 4 ra beyond the final

excavation profile of the main tunnel.

3. Perfo-anchors 3 to 4 m long were installed through

these holes in a manner that the part of the holes

between the pilot and the final excavated profile of

the tunnel was blank.

4. A minimum of 48 hours were allowed for the grout in the

perfo-anchors to get set before next widening blast in

the tunnel.
 92

5. Immediately after the widening blast, shotcreting was

applied to*the exposed rock surface.

The method proved very successful in arresting overbreaks

and rock falls and attaining a reasonable cycle of advance

inspite of continued adverse geological conditions. A

consistent cycle of 20 hrs was achieved with this technique

with a progress of 5 to 6 m per day.

»

Another factor which created much problem in sorae reaches

was sizeable rock fall involving secondary mucking operations

during drilling process. Simultaneous drilling of top and

bottom holes was thus not possible. Various shapes and profiles

of face cutting were tried to control the rockfalls but without

success. The problem was solved by (i) changing the pattern of

holes from time to time to suit the rock response (ii) reducing

the specific charge (iii) reducing the charge density and (iv)

careful observations of rock behaviour.

In some reaches, loosening of surface rock took place

during blasting operations. Shear strength between the adjacent

laminations of the stratified rock was completely lost and

rockfalls continued to take place progressively after blasting.

Careful and continuous shotcreting and re-shotcreting provided

an effective solution of this problem.

From chainage 3138 m onwards the Chibro heading passed

through an exceptionally unfavourable geological structure

comprising quartzitic slates polished with graphite and had a

repetitive pattern of joints running parallel to the tunnel.

The dip of this strata was getting steeper which was all the
 93

more disturbing as it would further add to the tunnelling

problera being faced due to major rock falls. There was •a fast

and progressive destressing phenomenon which allowed no time

even for application of any support system.

As a result of destressing, progressive rock falls started

travelling backwards also. Thick shotcrete with interlayers of

chain link fabric were used in some locations but it appeared

that the time element surpassed all attempts at any

satisfactory rock control and supporting system. A new

arrangement was introduced for reducing the interval between

blasting and application of initial shotcrete layer so as to

maintain the intactness of the rock mass to the maximum

possible extent. In this technique, a special arrangement was

introduced behind the main jumbo where the entire shotcrete

train was kept in advance and shotcreting operation over the

roof arch was started even from the top floor of jumbo as soon

as the mucking on the face had reached a stage that the Jumbo

could be moved forward to an extent that the five metre

cantilever platform of the jumbo easily reached the excavated

face. This technique helped in minimising the time interval for

application of initial shotcrete and also in reducing the cycle

time of advance by partial overlapping of operations. The work

could progress under these conditions upto chainage 3000 ra.

Beyond chainage 3000 m the rocks were thickly

laminated slates with graphitic joints and quartz veins at

short interval interspersed with numerous joints and shear

planes. In addition, the rock cover was high and there was
 94

ingress of water. That resulted in high overbreak. Shotcreting

was not possible and continuous process of rib and concrete

placement had to be resorted to. The conditions were acute and

unsurraountable. The overbreak which was 1 ra at chainage 3110 ra

increased to 15 m at chainage 3161 m and there was a sudden

rock collapse at chainage 3161 m. Further mucking was

discontinued when even after mucking out 250 cum, the process

of rauch replenishment continued by further loose fall.

The troublesome zone was treated by umbrella grouting and

multidrift raethod. It took about four months to cross the zone

which was about 22 m in length. The process proved entirely

satisfactory. In umbrella grouting, the zone of overbreak is

grouted by drilling radial drill holes and pumping cement grout

at very low pressure.

The conditions continued to be bad even after crossing the

above shear zone and overbreaks in the rock appeared almost

uncontrollable. There were chimney formations in some reaches.

A major chimney formation occured at 3517 m and the entire

tunnel section was filled up with muck. The chimney formation

was at an angle of about 60 and in a width of about 3 m. The

problera was solved by urabrella grouting of the muck as done at

chainage 3161.

3.3.5 Other Severe Tunnelling Problems in Chibro-Khodri Tunnel,

Uttar Pradesh

Apart from occurrence of rainor rock falls, there were also raost

severe rock falls in the Chibro-Khodri tunnel. A few are listed

below :
 95

1. Rock fall at Chibro heading on 8.12.1971 between

chainage 693.5 to 697 m

The tunnel failure in that reach was mainly due to badly

jointed rocks, full of local faults and bedding shears. The

fault and bedding shears converged and met near the crown level

at approximate chainage of 704 ra. Seepage of water had further

aggravated the situation. These resulted in more severe

conditions and caused complete collapse of tunnel. The work was

completed by high pressure grouting of the surrounding rock

mass through deep holes. This converted the crushed rock mass

into a consolidated tunnelable rock media. The reaches of major

rock falls were continuously completed by urabrella grouting and

multidrift raethod without resorting to blasting.

2. Probleras faced in the excavation and supporting of the

Chibro-Khodri tunnel in the red shale zone from

chainage 1135 ra to 1310 ra approximately

There were hazardous conditions in the whole red shale

reach (1135 m to 1310 m). Heavy rock falls made the things

still worse in the above reach. Probleras of tunnel closure, rib

deformation and support closure were much severe. The support

2 2
pressures were too high and varied from 10 kg/cm to 18 kg/cm .

The pressures were so high that the horizontal diameter of the

supports reduced to a large extent. The crowns of the ribs were

pushed-up towards the rock, resulting in negative closures or

extension in the vertical direction. Fig. 3.10 shows the radial

rock pressure-time relationship at chainage 1199 ra of the head

race tunnel. It can be observed that after 6 months the

 96

2
pressure on the support increased to 6-13 kg/era .

3.3.6 Head Race Tunnel - Loktak Hydroelectric Project, Manipur y

General Features

Loktak hydroelectric project comprised diversion of 58.8 curaecs

of water from Loktak lake, out of which 16.8 cumecs have been

used for lift irrigation. The balance of 42 curaecs of water

with a gross head of 312 ra have been used to generate 105 MW of

power by three units. The 10.265 km long water conductor system

consists of a 2.27 km long open channel, a 1.22 km long and 5.0

ra diameter horse-shoe shaped cut and cover section, a 6.505 km

long and 3.81 m diameter horse-shoe shaped head race tunnel and

a 0.27 km long and 3.658 m diaraeter circular pipe tunnel. The

construction of the head race tunnel was taken up in 1970 frora

seven faces applying a conventional steel support raethod.

Geology

The tunnel was driven through the bands of shale, sandstone,

silty clay and highly jointed shale. The tunnel crosses a N-S

trending syncline. Thick layers of sandstone and siltstone

occur at the trough of this syncline whereas splintary shales

with thin bands of sandstone and siltstone occupy the flanks at

the tunnel grade. The axial portion of the fold has been

refolded into several N-W trending cross folds. Both limbs of

the fold have been affected by a number of faults. Several

vertical or steeply dipping joints had also developed. Fig.

3.11 shows the geological section of Loktak tunnel. y

Tunnelling Problems

Abnormal support pressures, methane gas explosion and squeezing

 97

5
X.
Q.

u
o
CE

100 150 200 300

Time (days) -♦-

FIG. 3-10 RADIAL ROCK PRESSURE-TIME RELATIONSHIP

AT CHAINAGE 1199m OF THE CHIBRO-KHODRI
HEAD RACE TUNNEL
 *

SO
00

(Face 0 iFace 2
r*~ Facef1^" Face T
540 1930 4550 6505 m

Lake deposits Silt zones with Tunnel

£> O 0 ! subordinate shale
Terrace deposits
P7^ Shale with subordinate - Fault
f§3j Shale siltstone /sandstone
Thin bands of sandstone and
Sandstones siltstone
Sandstone with shale parting

FIG. 3-11 GEOLOGICAL SECTION OF LOKTAK HYDEL TUNNEL (After G.S.I.,1979)

 99

ground conditions were found to be the main problems during

tunnel excavation. High support pressures and heavy fractured

and water bearing rocks created unexpected probleras in the

tunnel. The tunnel became unstable, steel supports buckled and

re-excavation became necessary on a considerable length of

tunnel. This failure resulted in very costly and time consuming

reconstruction works.

Reasons for Tunnel Failure

The reasons for tunnel failure have been attributed to the

following facts;

1. The tunnel section was situated in intensely folded and

fractured shales under high overburden.

2. Rare system of discontinuities consisted of steep fault

planes, mostly slickensided or filled with gougy

material with strike parallel to the tunnel axis. The

joint system and the local geology were most

unfavourable and rare in character. The above

conditions were responsible for highly reduced natural

cohesion (c) and joint friction angle (0) of the rock

mass.

3. There existed squeezing and swelling rock raass

conditions in the tunnel. The swelling of shales was

further aggravated by insufficient water drainage in

the area.

A switch over to the New Austrian Tunnelling Method (NATM)

improved the conditions and the tunnel could be completed in

1981.
 100

3.3.7 Head Race Tunnel - Kopli Project, Assam

General ^

The project comprised construction of a 65 m high concrete

gravity dam across Kopli river along a dyke system. The water

from the reservoir was taken to Khandong power house through

4.5 m diameter, 2.7 km long tunnel which passed through granite

and shale.

The rocks were highly fractured and severe water seepage

problems were encountered throughout the construction activity

of project. Another seepage problem was encountered in Khandong

power house tail race channel which passed through carbonaceous

shales.

In another similar problem during the initial filling and

pre-coraraissioning tests in Sept. 1986, a portion of the tunnel

developed cracks and the water spilled out with consequent

erosion of the overburden of hill slope. The cracking could be

due to low thickness of lining and low overburden.

Solution channels, overbreak and flowing ground conditions

>
surcharged with water were the other probleras encountered

during construction of the project.

3.3.8 Baira-Siul Head Race Tunnel, Baira-Siul Project, Chamba,

Himachal Pradesh

General

Water of Baira and Bhaledh have been stored in reservoir formed

behind Baira rock fill dam and the same is carried through head

race tunnel. Water of Siul is led into head race tunnel through

drop shaft connected by feeder tunnel emanating through the

 10

reservoir formed behind Siul weir in the river Siul.

Geology

The head race tunnel passes through geologically complex

phyllite and schist formations. The area around the power house

is prone to landslides and one such landslide downstream of the

main dam had created a lot of construction difficulties right

from the start of project in 1978.

Problems Encountered

High seepage conditions, strong schistocity, overbreak, cavity

formation, flowing ground, poor shear strength characteristics

and unfavourable geotechnical conditions were the problems

encountered during construction of the project.

3.3.9 Salal Hydel Tunnel - Salal Hydroelectric Project, Jammu

and Kashmir, India

General

The construction work on the project started in 1978.

Excavation of 2.5 km long, 11 m diameter tail race tunnel was

started in 1980.

Geology

The tunnel has been constructed through jointed dolomites of

the great limestone series of Pre-Carabrian age. It passes below

a hill under a maxiraura cover of 630 ro. The geological section

of the tunnel is shown in Fig. 3.12. The dolomites have been

intersected by three sets of closely spaced joints. The bedding

joints dipped at 45°-55° due North. The transverse joints were

dipping at 70 -80 due West and the cross joints were dipping

at 30°-40° due South. The tunnel makes an angle of 20°-65° with

 102

the strike of these formations.

The jointed dolomite was intersected by several shear ^

zones varying in thickness from a few centimetres to a couple

of metres. The shears are filled with crushed calcareous raatter

formed during shearing movements. The joints were generally

tight except at few locations where hill water found access.

Problems Encountered '

Dripping water and sudden flooding conditions were encountered

during construction of the project. Support problems were also

encountered in some structural regions of the tunnel. The

tunnel was excavated by conventional methods and was supported

by steel ribs.

""" During excavation of the foundation of spillway, the badly

sheared rock was met with and the same condition of. the bed

rock persisted to appreciably greater depth. This was possibly

due to the intense shearing to which the bed rock was subjected

during Hiraalayan tectonic activity when the rocks were affected

upto greater depths. The sheared rock condition at foundation

levels caused considerable design problems and costly remedial

treatment had to be adopted for safety of the spillway

structure against shear failure of the foundation.

3.3.10 Giri Tunnel-Giri Hydroelectric Project, Himachal Pradesh

General

A 7.4 km long, 3.66 m diameter pressure tunnel has been

excavated through slates with boulder beds, phyllite, shale, f

clay and sandstone. The tunnel passes through an area which is

geologically very complex.

 I*

1080

920 I

•*—»—»•-
Shear zones Debris material Blocky dolomites Crumbly and sheared
dolomites

Jofnts Highly jointed dolomites

I StraumatoOtic bands
uJ Possibility of cavity
formation

• •• • River borne material Higly jointed dolomites O O o

• • • ^ with slaty bands c-

O
o
O
o
o
Cherty dolomites

FIG.3-,2 eEOWGICAL SECTION OF SALAL HYOEL TUNNEL SH0WIN3 POSSIBILITIES OF CAVITY FORMATION AND
bUttziNG ZONE (After Jethwa et al., 1976 )
 08

The support was provided by tensioned rock bolts and occasional

steel sets.

3.3.12 Head Race Tunnel, Proposed Pala-Maneri Hydroelectric

Project, Uttarkashi, Uttar Pradesh

General

The project envisages the utilization of huge drop available in

river Bhagirathi between Pala and Maneri. It > involves
construction of a 74 m high concrete gravity dara, intake and
underground sedimentation chamber on' right bank, a head race
tunnel of 6 m diameter and 12.70 km long and a power house
having installed capacity of 400 MW.

Geology

The tunnel is proposed to be driven in gneiss quartzite and

basic rocks.

Problems Anticipated

The project is under investigation and planning stage.

Geotechnical investigations have revealed difficult ground
conditions due to main Hiraalayan thrust with heavy ingress of
water and squeezing ground conditions.

Recent Hydroelectric Development in the Lesser Himalaya

Apart from a number of river valley projects already complete,
raany new schemes have been proposed for construction in the
Himalaya. Six major hydroelectric schemes proposed for
construction have been selected under the present research work
(Fig. 3.15). These projects have complex geological formations
comprising different types of rocks inter-spersed with thrust
zones, shear zones, folds, faults, fissures and water charged
 109

horizons. These projects may have intricate geotechnical

problems on account of geological surprises anticipated during >
construction. Geological details of the projects and problems

anticipated during construction are discussed in the following

paragraphs.

3.3.13 A Dam on Sandstone - Proposed Jamrani Dam Project, Uttar

Pradesh

General

The proposed Jamrani dara project envisages construction of a

150 ra high roller compacted concrete dam across river Gola at

Jamrani village about 10 km upstream of Kathgodara in Uttar

Pradesh. The dam would provide additional irrigation to 62,114

hectare of land and 52.53 MCM of drinking water annually.

Besides, it will also provide 15 MW of peak power generating

capacity.

Geology of the dam site area

The dara site is located on the lower Siwalik rocks of Kuraaun

Himalaya which consists of sandstones with alternate bands of

siltstone and claystone. These generally trend NW-SE direction

with dip of 30° to 50° towards NE direction. The rock raass is

traversed by a number of joints. The major structural

discontinuities such as thrust, and shear zones are

conspicuously absent in the area, except for minor shearing

along sandstone/claystone contact at places. There are five

prominent joint sets out of which four intersect resulting in

the formation of rock wedges which are likely to cause problems

in abutment stripping.
<JAI tWAT

OAltvvMi

HIMACHAL
PRADESH

^ "-JAMRANI DAM
X (IRRIGATIONS, WATER
SUPPLY SCHEME)

«»« S-X »AJ0R HY0R0ELECT,C PR0JECTS C0M1NC-UP „0ANGA AND YAMUNA VALL.YS
 111

On the right bank, rock is exposed all along while the

entire left bank is covered by debris and hill wash material

having thick forest growth. The bedding and joints control the

drainage and topography on either bank. The depth of weathering

on the right and left bank is approximately- 5 m and 12 ra

respectively. Fig. 3.16 shows the geological cross-section

along the dara axis. The profile of the section indicates

moderate slope on left abutment controlled by bedding and

moderately steep slope on the right abutment controlled by

South-dipping joints.

Geotechnical Properties of Rocks

Geological logs of the subsurface drilling indicate thick beds

of siltstone and sandstone. The core recovery is generally good

below 9 m depth and varies from 40 to 95 per cent and in most

cases 80-90 per cent. The RQD improves considerably below the

depth of 18 m. The permeability decreases with depth and

becomes less than 10 Lugeons below 35 ra. The compressive

strength of sandstone which is the dominant foundation rock

2
varies from 300 to 750 kg/cm under dry condition and the

modulus of deformation (Ed) lies in the range 0.10 X 105 -0.20

5 2 2
X 10 kg/cm at a stress level of 20-30 kg/era . The cohesion
2
(c) of the rock mass lies between 4.3 to 6.3 kg/cm and the

angle of internal friction (0) varies from 39° to 50°.

Special features in the dam area

The dara site lies in an active seisraic belt and was affected

by earthquakes in various parts of Kuraaun and adjacent seismic

belt at frequent intervals and that a number of earthquakes

 112

occured in this region, sorae of which are Dharchula earthquake

of 1916 (M-7.5), Kapkote earthquake of 1956 (M-6.25) and West

Nepal earthquakes of 1966 (M-6) and 1980 (M-6.5).

i i I l i ' '

Rock slides have been observed on both the abutments.

However, major slope slides are seen on left abutment due to

undermining of the toe at river bed level.

At sorae locations, the sandstone is very soft and

weathered. On being immersed in water it disintegrated rapidly

into small pieces and lost its strength.

V
Problems Anticipated

1. Advance rock anchoring of suitable length raay be needed at

the time of slope excavation.

2. Thin claystone bands (1-3 m) should be given treatment to

provide uniform dispersal of stresses in the dam

foundation.

3. The modulus of deformation (E,) of the rocks is very low;

so, foundation settlements of high order are expected.

4. Rocks being soft, weathered and of poor quality, may

require special treatment in the dam foundation and

abutments.

3.3.14 A Dam in Complex Geology - Proposed Kotlibehl Dam

Project, Uttar Pradesh

General

The project envisages construction of a 210 m high earth-rock

fill dara on river Ganga with an installed capacity of 1000 MW.

The dam derives its importance for the fact that it will serve

as a balancing reservoir for a number of projects planned in

 N2 W

DJ+-39

•DL-2( projected;
Jogofjiam_ EL 7_65i0m. ^

DH-35 „_
DL-3 / QB*

LEGEND it'
Ul
j°f^| Rtver borne material (RBM)/Debris
?~j] Finegrained sandstone
YJTF] Medium-grained sandstone
jy^| Coarse-grained sandstone/sandrock
550
Siltstone /vanegated siltstone 5 •>-.-•;
l=z~t] Dryshale/clayseam
j^^l Sedimentary breccia
500 53 Drift

470

FIG. 3-16 GEOLOGICAL SECTION ALONG THE AXIS OF THE PROPOSED JAMRANI DAM PROJECT
 1 14

upper reaches of Ganga valley together with partial flood

protection works in this region.

Geology at the dam site

The proposed dam site area falls in the South Western limb of

Narendra Nagar major syncline belonging to poor sedimentary

rocks mainly consisting of shales, highly jointed limestones

with pockets and veins of gypsum, highly disintegrated

gypsyferrous limestone, all belonging to Krol thrust in the

region. The four 13 ra diameter diversion tunnels are to be

established in gypsyferrous limestone reach. Fig. 3-. 17 shows

the geological section along the axis at the proposed Kotlibehl

dam.

Geological features at the dam site

The dara site area is located in quite a complex geology with

repeated folding and faulting. There are alternate bands of

soft and hard quality rocks in the dam foundation area. A major

fault has been mapped on the left bank which is exposed on the

upstream side passing through the left abutment on the river

bed towards downstream. This fault has been further offset by

three cross faults. These faults specially i the cross-faults

stated to be the youngest, have shifted towards left abutment

fault. The possibility of further movement of these faults

can not however be ruled out. This aspect is therefore

considered of added significance as one of the twin epicentres

of the Kangra earthquake of 1905 was located at Dehradun. The

matter needs further rethinking in view of the recent

earthquake in this region on October 21, 1991 registering 6.1

on the Richter scale.

 115

Presence of soft pockets or local shear zones is evident

from the fact that wherever exploratory drifts passed through

*
weak bands of gypsyferrous limestone, gradual formation of

inverted chimneys has been observed. In raany structural

regions, it was found that either the support system had caved

in or displaced excessively from its original position. In raany

cases the wooden sleepers had bulged out. Such conditions are

suggestive of high support pressures during tunnel excavation.

Patrographic studies indicated the presence of limestone,

breccia and slate as basic rocks. In a few samples pressure

solution seams and pitted contacts between the grains were

observed. In a few places calcium carbonate was found to have

been deposited within the joints of the cavities. A few samples

indicated fault zone features and fracturing.

From the foregoing details, it appears that a few

structural regions of the proposed dam site lie in major fault

zone. Voids, pressure solution cavities and seams as indicated

in the patrographic analysis show the presence of sorae minerals

>
which seem to be chemically active with moisture.

Geotechnical Properties of Rocks

Laboratory and field test results indicate that the material as

well as rock mass is of very poor quality. No core recovery

could be possible at raany locations because of

disintegrated-jointed gypsyferrous limestone and charred

shales. In few cases core recovery was not possible due to

thinly bedded slates dipping at 70° • to 80°. Explorations

revealed that most of the dam area is covered with deeply
 4

600

500

a\

- 400
oooi River borne
fi-e-°l material
Gypsiferrous
lime stone
300 -
- 300
Shale of various
colours

Mass ive/com pact

lime stone
200
200

Note -Elevations are Tn metre

FIG.3-t7 GEOLOGICAL SECTION ALONG THE AXIS AT THE PROPOSED KOTLIBEHL DAM
 1 17

seated shale and limestone mixed with different minerals. The

modulus of deformation (E,) in general was found to be of lower

5
order with maximum and minimum values ranging from 0.30 X 10
o 5 2
kg/cm to 0.03 X 10 kg/cm . The modulus decreased drastically

under saturation. Plate load tests on shale under saturated

condition showed settlements of the order of 4.1 cm against

2
pressure intensity of 24.5 kg/cm . Under moist condition,

settlement was found 0.48 cm against pressure intensity of

2
25.50 kg/cm . In another test conducted on the sarae rock mass

under saturated condition, settlement of the order of 18 cm was

2
found against a small pressure intensity of 10.20 kg/cm . These

results indicate presence of some soft pockets behind the test

surface. Results of block shear test on shales and gypsyferrous

limestones under saturated condition show cohesion (c) ranging

2 2
from 0.25 kg/cm to 1.95 kg/cm . The value of angle of internal

friction (0) varied from 36° to 45°.

Laboratory tests on rocks collected from different

locations show compressive strength of the rock material

2 2
ranging from 160 kg/cm to 500 kg/era for limestone.

Problems Anticipated

1. Severe rock falls with chimney formation raay be expected

from the bands of gypsyferrous limestone during excavation

of tunnels.

2. Heavy rock loads may demand unique support system during

tunnelling.

3. Extensive foundation treatment raay be necessary on account

of poor rock raass behaviour and low raodulus of

 1 18

deformation.

4. The four diversion tunnel inlet portals and sorae length of Y

tunnels is to be built in gypsyferrous limestone a weak

and disintegrated rock which is likely to pose probleras

during tunnelling.

5. Extensive instrumentation would be necessary so that a

comparison is made between the parameters assumed in the

design and observed at site.

3.3.15 A Dam on Difficult Foundation - Proposed Lakhwar Dam

Project, Uttar Pradesh

Genaral

Lakhwar dam project envisages construction of a 204 ra high arch

- gravity dam across river Yamuna and a 3X100 MW installed

capacity underground power house near village Lakhwar. At the

proposed dam site, the river flows through a narrow gorge and

the side hills are fairly high.

Geology

The rock formations in the region where Lakhwar dam is proposed a.

to be located comprise phyllite, slates, quartzites and

limestones which form the Southern limb of a major syncline.

These rocks are intruded by a number of minor basic trap rock

bodies. The Lakhwar dara is located across one such intruded

body of basic rock, the maximum width of which is about 300 m

in the river section and is Just sufficient to accomodate the

base width of a gravity dara section and the spillway bucket.

The trap is generally coarse grained and highly jointed. On the

left abutment an outcrop of slates and quartzites is enclosed

 119

on three sides by the traps forming a rock body called the

"Xenolith". The quartzites of the xenolith are mediura to coarse

grained and are massive but have profuse Jointing and iron
oxide stains. On the right abutment, the band of trap rock
contains a number of joints in different directions with
numerous calcite veins. The joint surfaces are mostly fresh,
but a few of them show slight alteration also. Some joint
planes are stained due to iron oxide coating. They are mostly
tight and wherever open, the opening is of the order of 1 to 3
mm. Most of them are devoid of filling whereas a few contain
rock powder and a few contain white gougy clay also.
The lithological contact between xenolith and trap is
often sheared. The foliation planes in the slates have dip
ranging between 65°-85° N:25°-45° E (upstream). The trap/slate
contact is dipping at angles ranging from 30o-70° N : 30°-45° E
direction (upstream). In the foundation of overflow section the
trap is underlain by crushed and thinly foliated slates of
indefinite thickness at a depth of 30-40 ra below the toe
region. The contact having shear zone of about 1.5 ra is dipping
upstream at an angle of 45° such that the depth of trap rock in
the heel region becomes about 200 m. Fig. 3.18 shows the
overflow section of the proposed Lakhwar dara.
Geotechnical Properties of Rock Mass

There are 8-9 sets of joints in the trap rock raany of which are
quite persistent and repeated at closed spacing. The trap rock
comprises dolerite to hornblend rhyolite with following
geotechnical properties :
 20

2
Compressive Strength (q^) 690-1950 kg/cm
Modulus of deformation (Ed) 0.16-1.03 X 10 5 kg/era 2
2
Cohesion (c) 1.01-3.66 kg/cm
o o

Angle of internal friction (0) 37 -57

Permeability 3-47 Lugeons

The slate body on the left abutment consists of thinly

foliated, soft and friable Chandpur slates. Core recovery could

not be possible in these rocks. The following properties were

•.

obtained on the two specimens of thinly foliated slates :

2
Unconflned compressive cube strength (Qc> 10-11 kg/cm
Modulus of deformation (Ed) (0.01-0.1) X 105 kg/cm2
Geotechnical Problems Identified at the Dam Site

The main geotechnical problems identified at the dam site

include :

(i) Occurrence of the large slate body (Xenolith) within

the basic rock falling in the dam blocks on the left abutment.

The xenolith has been explored in detail and as a result of

that its configuration and depthwise persistency has been

established.

(ii) The shape and extent of the intrusive basic rock body

in the dara area has been delineated. The properties and the

extent of intrusion of the slate body raay play vital role in

proper placing of the various appurtenant engineering

structures of the Lakhwar dam complex.

(iii) It was thought to raise the height of dam to have a

larger storage but on geological considerations, especially due

to the presence of unsound rock on left abutment, it was not

121
Z
<
Q
or
<
i
<
o
Ol
to
o
a
o
a.
a
u.
o
»-
o
UJ
>-
<
a.
in
tn
 122

found feasible to raise the height of the dara.

(iv) The slate body and the gougy material at the shear

zone contact have very low values of raodulus of deformation

(E,). Excessive deformations may therefore be expected in the

dara foundation which raay require some unique measures to take

care of it.

(v) The work of dara construction may be taken up only

after realistic geotechnical appraisal of thinly foliated

slates under saturated condition. Extensive foundation

treatment to improve the properties of foundation rock raay have

to be carried out. If no solution is sought feasible to improve

the competence of left abutment, it would be necessary to

change the type of dam that suits the geotechnical aspects of

the foundation rocks. Shifting of dam on upstream side may also

be necessary from geotechnical considerations. Thus arch cum

gravity dara would be a better choice.

3.3.16 A Dam on Metabasic and Quartzite - Proposed Srinagar

Hydel Scheme, Uttar Pradesh

General

Srinagar hydroelectric project comprises of a 90 ra high

concrete gravity dam on river Alakhnanda near Srinagar town and

a surface power house with an installed capacity of 330 MW

(6x55 MW). The spillway has been accomodated in the dam body
itself due to non-availability of sufficient suitable space on
the left bank, fhe intake and sedimentation chamber have been
proposed on the left bank. The water from the sedimentation

tank would be led through a 4.3 km long open power channel on

 123

right bank after crossing the river through a 180 ra long

aqueduct.

Geology of Dam Area

The project is located within the rocks of Dudatoli and Garhwal

groups separated by a major tectonic lineament called Srinagar
thrust. The proposed dara and appurtenant works are located in a
complexly folded, sheared and fractured sequence of quartzite
and basic rock belonging to Garhwal group. The proposed dara

site area exposes a sequence of complexly folded quartzite and

metabasic rock. The local trend of the formation varies from N

40° E - S 40° W to N 60° E - S 60° W with moderate dips on

either side. The generation of folding is identifiable in field

in both the rock units. Folds of both the generation are

coaxial in nature and are steeply plunging towards North.

Geotechnical Appraisal

Quartzite and metabasic rocks in the dara area have been further
divided according to their physical competence, degree of
jointing, shearing and morphological expressions'. The
quartzites have been divided into three types (1)
recrystallised quartzite (ii) massive quartzite and (iii)

thinly bedded and fractured quartzite. The metabasics have been

classified into three types, namely (i) massive, (ii) foliated

and (iii) thinly foliated and sheeted.

Massive quartzite and massive metabasic in general have

two sets of joints with one random set and are exposed in 25
per cent of the dara area. The recrystallised quartzite and
foliated metabasic have three sets of joints with one random
 124

set and are exposed in 65 per cent of the dam area. Thinly
bedded and sheared quartzite, and thinly foliated and sheared
metabasic have more than three well defined Joint sets and
occupy 10 per cent of the dam area. Most of the joint sets in
all rock types have moderate to steep dips (50° to 85 ) with a
very few joints having low to moderate dips (less than 30 ).
The steeply dipping joints have a strike continuity of 2-6 ra,
whereas the low dipping joints had a strike continuity of less

than 3 m.

It was seen that in most of the cases, the core recovery

was 60 to 70 per cent. The permeability of the rock raass ranged

between 2.00 and 11.50 Lugeon. The raodulus of deformation (Ed)
c p
lies in the range (0.10-1.30) X 10 kg/cm . The cohesion (c) of
2
the rock mass lies between 2.2 to 3.16 kg/cm and angle of
o o

internal friction (0) varies from 31 to 57 . •

Problems Anticipated

The proposed dara and its appurtenant structures are located

within a coraplexly folded and fractured sequence of quartzite
and metabasic rocks of poor quality. It will therefore be

necessary to undertake thorough geotechnical evaluation of the

project site.

The open sedimentation chamber has been proposed to be

located on the left bank immediately after the intake

structure. The hill side cuts have been anticipated to be of

the order of 50 ra or more. For slopes to be stable, the cuts

may require support by way of benching or rock bolting. Careful

determination of strength characteristics (c & 0) of rock raass

 125

would be necessary.

Two third of the plunge pool area would be accomodated in

A
quartzite while the remaining one third would be in metabasic
rock. Unequal settlements in the foundation can not be ruled
out. Careful study of the characteristics of foundation rock

raass is essential especially at the junction of two rocks.

The power channel would cross the river through a 180 ra

long aqueduct with five spans of 36 m each. The aqueduct has

been designed for a discharge capacity of 550 curaecs. The

height of the piers supporting the aqueduct would be about 65 ro

from the deepest foundation. The area lies in the zone of

seismic activity. Study of rock mass behaviour under dynamic

conditions would be necessary. Fig. 3.19 shows the foundation

geology of the deepest block of the dam.

3.3.17 A Dam on Phyllites - proposed Tehri Dam Project, Uttar

Pradesh

General

Tehri dam project has been taken up for exploiting the vast

potential of river Bhagirathi, a tributary of Ganga, for

irrigation and generation of power. The project envisages

construction of a 260.5 m high earth and rock fill dara across

river Bhagirathi near Tehri in Uttar Pradesh. The project area

lies in a tectonically disturbed belt and is characterised by

high relief.

Geology of the Tehri Dam Area

Tehri dara project is located in the Lesser Himalaya. The rock

formations at the dara site are phyllites of Chandpur series
 K 4

590

FIG. 3-19 CROSS SECTION OF THE DEEPEST BLOCK SHOWING FOUNDATION GEOLOGY IN SRINAGAR DAM
 27

which are banded in appearance. Tha bands are constituted of

argillaceous and arenaceous material which have been classified

into three categories.

Grade I Phyllite - Grade I is phyllitic-quartzitic in

composition and predominantly arenaceous in nature, massive in

character and distinctly jointed.

Grade II Phyllite - Grade II is quartzitic phyllite in

composition with alternate bands of argillaceous and arenaceous

materials.
•

Grade III Phyllite - Grade III is composed of mainly

argillaceous material and is schistose in nature with foliation

planes, cleavages and joints very closely spaced.

Grade I and Grade II units which form about 70 per cent of

the foundation area are nearly of the same quality so far as

deformability characteristics and permeability are concerned.

The formation of phyllites exposed in the Tehri gorge generally

strike N 55° W - S 55° E to N 80° W - S 80° E with a South -

Westerly (downstream) dip of 35 -55 . Seven prominent joint

sets have been analysed on the basis of stereographic plots.

The average RQD values for phyllite Grade I, II and III were

found to be of the order of 30, 22 and 10% respectively. The

core recovery of phyllite grade I and II was found to increase

with depth. After a depth of 10 m the core recovery was

generally above 50 per cent.

The raodulus of deformation for philite varies from

5 2
(0.04-0.40) X 10 kg/cm . The permeability values vary from

0.50 to 30 Lugeon,

•
 1 • »" ! •

,/'V.-^v.v^".^ .-.'-v-.-. •••••.:^- <f+.:. /••.-,;- -:-•••.'' •-• -•> ....-v-..-.- .. * '••'•••• •-•••-•• i* '--W...- •.-;-.

128

There are a number of shear zones in Tehri gorge which

have crushed rock and are expected to be relatively more
pervious. There are four sets of shear zones along the
foliation planes, two across the foliation planes and two other
sets of steeply dipping features trending N 60° W - S 60° E and
N 30° E - S 30° W direction. In many cases the shear zones are

clay filled. Fig. 3.20 shows the geological section at the dam
axis.

Problems Anticipated f
In a high dam like Tehri, importance of geotechnical
exploration can not be over emphasized in view of the
unprecedented size of the structures, seismically active region
and complex geological conditions. From the foregoing details,
it appears that the permeability of the rock is comparatively
high, the core recovery is poor and the modulus values are low.
Therefore, in order to maintain the integrity of foundation, it
may require extensive foundation treatment to improve the
permeability and modulus values of the rock mass. So far , there >
have been some problems of rock slope stability and land

slides.

It raay be visualised that study of the following

geological aspects is essential :
1. Estimation of support pressure and requirement of
supporting system for underground excavations.
2. Estimation of modulus of deformation and shear strength
characteristics of the foundation rock raass.

3. Estimation of allowable bearing pressure in view of

 900r

850>
Right bank

800'

750' INDEX

M 1 Phyllitic quarzite
E T Phyllitic quartzite thinly bedded
c 700|
• *' . " Over-burden surcharge

(m/t Lithological contact

~~ 1 Over-burden rock
650-

1 Acceptable rock

— XX—
Destressed zone

[^"^j Shear zone

600-

River bed

550
50 50 100 150 200 250 300 350 400 450 500 550 600

Chainage in metre

FIG. 3-20 GEOLOGICAL SECTION ALONG THE TEHRI DAM AXIS

 130

heavy hydrostatic load imposed by the reservoir and

mass of the dam body itself.

3.3.18 A Dam on Highly Jointed Quartzite - Proposed Utyasu Dam

Project, Uttar Pradesh

General

A 246 m high concrete gravity dara is proposed on Alakhnanda, a

tributary of river Ganga near Rudraprayag. The power station of

the Utyuasu dam complex will have an installed capacity of 1000

MW. The project is under investigation and planning stage.

Geology of the Project Area

The proposed dam and the appurtenant works are located in

highly jointed and fractured quartzite. The rocks are highly

charged with water. Five prominent sets of joints have been

identified on the basis of stereographic plots. Two faults have

been located in the area. Fig. 3.21 shows the cross section at

the proposed Utyasu dam axis.

Geotechnical Properties of the Rocks

The rock formation in the region where dara is located comprises

of thinly bedded and fractured quartzite. The extent of

weathering of these rocks is not very deep laterally into the

abutment or vertically below the overburden. The limited

explorations by drifting and boring have shown that the zone of

weathering below the overburden extends 2m-3ra both laterally

and vertically. The strength of the weathered rock is not very

markedly different from the fresh rock. It has been experienced

that in most of the cases the core recovery seldom exceeded 60

to 70 per cent.
 131

Problems Anticipated

Low compressive strength, low shear strength parameters, and

raodulus of deformation being of lower order indicate that
extensive foundation and slope treatment might be necessary to
improve the rock mass quality. Proper evaluation of engineering
pararaeters of rock mass is thus necessary.
3.4 Scope of Present Research
Case studies of the major hydroelectric projects show the
magnitude of varied geotechnical probleras in the Himalaya. The y
problems encountered in the Himalayan rocks basically pertain
to supporting of tunnels and underground caverns, stability of
slopes and settlement of foundations. It raay be inferred that
the Hiraalayan rocks do not generally qualify as good
foundations and abutment and the behaviour of rocks during
tunnelling raay not be free from many hazards and risks.
From the case histories of the completed projects, it is
observed that at each project - (i) the local geology and the
rock formations were found to be untypical, (ii) the joint sets ^
were large in number with unfavourable dips, (iii) weathering
potential of rocks appeared to be serious, (iv) rock material
properties showed high scatter, (v) moisture conditions were
found to be extremely heterogeneous and affected rock behaviour
drastically. It is thus understood that engineering
correlations of the rock mass would be different at each
*
construction project which may be planned in future.
Presently there are a number of ongoing projects in the
Lesser Himalaya. Under the present investigation it was not
 M

900-

Drift at
r.l. aco-o J\p L-4 Top of dam 840-0 m T t J)rift at R.L. 840 0
Normal reservoir level 830-0m g

Left bank Right bank

800-1

Drift at R.L. 770-0 j Drift at R.L-7700

c
o

o
R-2 *-{ Drift at Rl. 700-0
-* 700
UJ

Drift at R.L. 6800

Road - -

Drift at R4.. 640-0

Drift at R.L. 6300 > c
Not to scale

600

580
400 300 200 100 100 200 300 400 500

Distances (m >

FIG. 3-21 CROSS SECTION AT THE PROPOSED UTYASU DAM AXIS

 133

possible to explore and assess geotechnical probleras of all the

projects in the region. Therefore, six untypical hydroelectric

schemes to be started soon have been selected for the present

research study. During construction of these projects the

geotechnical problems raay prove to be challenging. Table 3.1

shows the details of six projects selected for geotechnical

investigation.

It may further be seen from the foregoing case histories

that each rock engineering project has its site specific

problem; yet through extensive field and laboratory

investigations, application of improved engineering skills and

a sound engineering judgment, it would be possible to overcome

such problems.

On the basis of laboratory and field tests, comprehensive

data were gathered at normal moisture content and saturated

states from these six major hydroelectric project sites.

Approximate empirical relationships have been developed to

estimate the various engineering parameters generally involved

in the designs in the feasibility stage.

 134

TABLE 3.1

DETAILS OF SII MAJOR HTDBOELECTRIC PBOJECTS ID THE

LESSEB HIMALAYA

S.No. PBOJECT DOMINANT ROCK(S) DAM LOCATION GENERAL

ELEVATION
TTPE HEIGHT
(•) (•)

1. JAMRANI Sandstone, Clays tone Concrete 150 Loner Siwalik 615.00

Gravity

2. KOTLIBEHL Shale, Lliestone Earth and 210 Ganga Valley 370.00

Bock Fill

3. LAKHHAR Trap, Slate, Xenolith Arch 204 Taiuna Valley 620.00

Gravity

4. SBIHAGAB Quartzite, Metabasic Concrete 090 Ganga Valley 545.00

Gravity

5. TEHBI Phrllite Earth and 260 Ganga Valley 610.00

Bock Fill

6. UTYASU Quartzite Concrete 246 Ganga Valley 605.00

Gravity
 104

Geology ir ' ,

Auden (1942) mapped the regional, geology of the project area

and established the existence of three low angle reverse faults

called locally as Renuka, Krol and Nahan thrusts lying in close

proximity of one another. The geological section of the Giri

hydel tunnel is shown in Fig. 3.13.

Problems Encountered

The various problems which were encountered during construction

of the project include overbreak, excessive support pressure,

tunnel closures, twisting of ribs and occuring of gases.

Adopted remedial measures include shotcreting with

perfobolting, flexible lining, excavating larger diaraeter

tunnel to allow closure below supports and use of gas

detectors.

3.3.11 A Project in Highly Folded-Faulted Region - Chamera

Hydroelectric Project, Chamba, Himachal Pradesh

General

A 540 MW Chamera hydroelectric, project, stage-I envisages

utilization of waters of Ravi river by construction of 140 m

high arch gravity dara across Ravi river. The -water conductor

system comprises of 6.4 km long horse shoe shaped power tunnel

of 9.5 ra diameter. Other works include 25 m diaraeter circular

pressure shaft 160 m deep, trifurcating into 3 X 5ra diaraeter

horizontal penstock branches to utilise a gross head of 207 m;

a 112x24.9x37 ra high underground power house of 540 MW
installed capacity; and a tail race system 2.65 kra long
consisting of 9.5 m diaraeter horse-shoe shaped 2.405 kra long
 105

tunnel and 245 ra long open channel. The other features of the

project are 10.7 m diameter diversion tunnel 500 m long, and a

network of 1.2 kra access tunnels. The work on diversion tunnel

started in May 1985 by heading and benching raethod. The work on

this project is in progress.

Geology

The project is located within the geological zone of the Lesser

Hiraalaya which is forraed mainly of successive steeply dipping

beds of phyllitic quartzite slates, carbonaceous phyllites, „

quartzites and volcanics. The entire sequence is highly folded

and faulted. Two major regional thrusts namely Jatogh thrust

and Shall thrust are exposed in the project area. The Jatogh

thrust is inside the reservoir and Shall thrust cuts across

power tunnel. The geological longitudinal section is shown in

Fig. 3.14.

Geotechnical Problems in the Tunnels

Due to poor rock conditions and intersection of Shall thrust

there were raany holdups in the tunnel construction. The work at .

r

chainage 54 ra of adit-I for power tunnel was abandoned due to

geological considerations. A large chimney formation had

occured which proved uncontrollable. Therefore, a new site was

selected for adit-I which was comparatively in better rock.

Similar problem was faced at chainage 150 m of adit-II but the

problera of this worst geological reach was solved through

extensive treatment. *"

In most cases, the tunnels passed through thrusted

phyllites which proved to be exerting high support pressures.

1500 Tunnel inlet Tanlog inclined ?ile branch
approach,
1200 Nahan thrust Marar adit

Tunnel outlet

3000 4000 6000 7000

Distance in metres

a-ORIGINAL GEOLOGICAL SECTION (After Ghosh, 1970 )

o
•3\

Tunnel Inlet Tanlog inclined! ^Krolthrust^ r- Nahan thrust Marar adit Tunnel outlet-,
approach

1000 2000 3000 4000 5000 6000 7000

Distance in metres

b-REVISED GEOLOGICAL SECTION (After G.S.I., 1977)

1 ZL'~ Clay stone /siltstone 3 ; '. Sandstone 5 Z ~ Shale 7 Phyllites/slate 9i *j Shear zones
2 *"
s e n
Psedoconglomerates 4 X?Z\ Basic rock 6 Sc £ Shale with quartzites 8 Blainfc slates

FIG. 3-13 GEOLOGICAL SECTIONS OF GIRI HYDEL TUNNEL

 ^ GATEWAY A

^^Construction access no-2

~row«r
w«r tunnirsuiqe
runnetsun shaft
.Power house
gTTransformer
ITMnsformex gallery

iuiae shaft tnv. EL 702-50

LONGITUDINAL PROFILE
©

Quartzitic phyllite
Limestone with bands 'Wl Moderately sheared L^j shear zone
W of phyllite upto 20 •/• £EL 1 phyllite
phyUfti p-^—

Close jointed and

& Quartzitic calcareous phyllite ;®; Carboneceous phyllite & fractured quartzite

jfefl Phyllite with bands of limestone Kg^J Highly sheared phyllite VjT^l Moderately jointed to
J&H] upto 20 V. EsSlj with band of limestone y-fju closely jointed to volcanics
upto 20°/.

FIG.3-11* GEOLOGICAL SECTION OF CHAMERA POWER TUNNEL

CHAPTER 4
 CHAPTER 4-

GEOTECHNICAL INVESTIGATIONS

4.1 Introduction

Study of twelve case studies of major hydroelectric projects in

the Lesser Himalaya reveals the need of proper identification

of geological problems and assessment of engineering parameters

of rock mass. For proper evaluation of engineering parameters

of rock mass, the working philosophy should be such that it

relates the experience of rock conditions at one project to the

conditions and experience encountered at others. An ideal

approach must fulfil the following two tasks :

1. it should combine the past experience with the results of

laboratory and field observations; and

2. it should yield the raost realistic values of engineering

design parameters of the rock mass:

With the above aims in view, extensive geotechnical

investigations were planned and carried out at six major

hydroelectric projects upcoming in the Lesser Himalaya. The

test programme was accompanied by detailed geotechnical

explorations which aimed at providing various correlations

between the laboratory and the in-situ data. The predominant

rock types which were encountered during investigation at the

dam sites included : sandstone, claystone, limestone, shale,

trap, slate, xenolith, metabasic, quartzite and phyllite. The

geotechnical assessment of the various rock types has been done

by dividing the px-ojects into a number of structural zones

 136

having more or less uniform geological features.

Laboratory investigations have been classified into four

main categories (i) visual examination of rocks, (ii)

determination of physical properties, (iii) assessment of

uniaxial compressive strength (q,,), and (iv) static modulus of

elasticity (E ) and the Poisson's. ratio (v). Large scale

in-situ tests have been carried out carefully in order to study

the deformability and shear characteristics of the different

rock types which were encountered at the project site. Plate

load tests have been performed to assess the allowable bearing

pressures on rocks.

Rock mass classification approach has been applied and

used on the six projects so as to compare and evaluate the use

of the different classification systems and to develop

correlations between RMR, Q and the various data which have

been obtained on the projects in the Lesser Himalaya. The

classification systems of Terzaghi (1946), Deere et al. (1970),

A
Bieniawski (1973) and Barton et al. (1974) have been used to

estimate the short term support pressure for the tunnelling

works at the proposed hydroelectric projects. On the basis of

classification approach, the extent of applicability of various

existing correlations to the Lesser Himalayan rocks has been

examined. The in-situ and laboratory tests have been carried

out as per the standard procedures including the

recommendations by the International Society for Rock Mechanics

(ISRM).
 137

4.2 Rocks Explored

The geological formations encountered during site investigation
included a wide variety of rocks. Brief geological details of
these rocks are as follows :

(i) Sandstone and Claystone

Sandstone occupies eighty per cent of the Jamrani dam

foundation area. The rock is generally stratified with
alternate bands of siltstone and claystone. The yellowish grey
to brown sandstone is micaceous (biotite), fine to coarse
grained, generally thinly bedded having 5 to 10 mm thick
laminations. The sandstone is generally flaggy and friable at
places due to poor cementation. Cross bedding has also been
locally observed in sandstone. Thick encrustations of
calcareous materia] wer* noticed on the 1nn«r surface of the
drifts, indicating that some of the sandstone have calcareous
cementing material. The major geological discontinuities like
thrusts, faults and shear zones are absent in the area, except
for minor shearing along sandstone/claystone contact at places.
The claystone is weak and highly weathered at surface. The
weaker claystone beds occupy 15 per cent to 20 per cent of the
foundation in the form of 1 m to 3 ra thick beds.

(ii) Limestone and Shale

Limestone and shale are the raain lithological units in the

foundation and inlet portion of the diversion tunnels of the
proposed Kotlibehl dam.

Limestone has been found to be highly jointed with pockets

 138

and veins of gypsum and few lenses of black/grey shales. At

places, the limestone has been found to be highly disintegrated
and fragile. Shear zones of varying width traverse the
lithological units at a few places.

Shale is another major lithological unit exposed at the

dam site. The shales are red (purple) and green varying to

black and grey in colour. The shales have been found with
gypsum lenses and calcite veins and are highly jointed.

(iii) Trap, Xenolith and Slate

A major part of the Lakhwar dam seat is to be founded on

the basic rock (called 'trap'). However, a small part of the

dam seat on the left abutment would be founded on xenolith - a

huge slate body entrapped within the basic rock. The trap is
coarse grained and highly to moderately jointed and at sorae

places assimilation of the slate and trap is evident. It has

some slickensided joint planes also. A few joint planes are

filled with quarts/calcite veins of varying thickness. The

joints are mostly tight but the presence of rock flour and
gouge fillings have been conspicuously noticed.

The xenolith rock body consists mainly of the mediura to

coarse grained quartzites having considerable jointing and iron

oxide stains.

The slate is very soft, compact but fragile and mostly

paper thin. It disintegrates and loses its strength completely

when immersed in water. The contact between the basic rock and

slate is defined by a highly plastic gougy zone of about 1.20 ra

thickness. The slate contains a few quartz veins which are

 139

sometimes fractured.

X (iv) Quartzite and Metabasic

The two main lithological units over which Srinagar dam is

proposed to be founded are (i) Quartzite and (ii) Metabasic.

The rocks are highly fractured. Reraobilisation of free silica
in the form of quartz crystals along the Joint planes have been

noticed at places.

(v) Phyllite

Phyllite is the common rock type encountered at various

project sites in the Himalayan region and is serving as the
essential foundation rock. The phyllites of the proposed Tehri

dam have been investigated. The rocks which have been broadly

grouped in Grade I, II and III are banded in appearance, the

bands are of argillaceous and arenaceous materials. The rock is

impregnated with quartz veins both along and across the

foliation planes.

4.3 Drifting and Drilling for Sub-Surface Explorations

The sub-surface explorations were carried out by excavating

exploratory drifts and through core drilling. Twenty eight

exploratory drifts totalling 550 m in length were excavated in

the area under investigation. This covered all the six

hydroelectric projects meant for investigation. The section

size of the drift generally was 2.1 ra X 2.1 ra. In all 175 test

sections were identified and investigated for the purpose. NX

size diamond core drilling was undertaken at 44 different

locations for collecting cores from different rock formations.

 140

Figures 4.1 to 4.6 show the location of the exploratory

drifts, and drill holes investigated at the project sites.

4.4 Laboratory Tests

Laboratory tests were carried out in three parts namely visual

examination, determination of physical properties and

estimation of engineering properties on rock cores/cubes/luraps.

4.4.1 General

Rock samples collected from various test sites were examined

visually for properties like colour, luster, hardness and

cleavage. Degree of weathering, extent of micro-fracturing and

presence of bedding in the rock samples were carefully noted.

Behaviour in the rocks , when put under water, was closely
observed for any change. Features like mechanical flaws,

micro-fractures and layers of apparently weaker material were

carefully noted.

4.4.2 Physical Properties

Elaborate experimental work has been carried out to determine

properties like natural water content, density, specific

gravity, porosity, resistance to abrasion, swelling
characteristics and the slake durability of different rocks.

4.4.3 Modulus of Elasticity (Kr), Poisson's Ratio (v) and

Dniaxial Compressive Strength (qc)

Core samples were tested to determine the modulus of elasticity

(E ) and uniaxial compressive strength (qQ) of the rocks. In

case where rock cores could not be available the compressive

strength was determined on rock cubes prepared in the

 141

730

LEGEND

I 3 Drift

Drill hole

Block shear test

D Plate load test

o Uniaxial jacking test

790

FIG.A.1_LOCATION OF IN-SITU TESTS CARRIED OUT AT THE

PROPOSED JAMRANI DAM PROJECT SITE
 > *•

LE GEND

==» Drift

• Block shear test

D Plate load test

Uniaxial jacking test

to

FIG.A.2.LOCATION OF IN-SITU TESTS CARRIED OUT AT THE PROPOSED KOTLIBEHL

DAM PROJECT SITE

Open
bench

LEGEND

Drill hole

Block shear test

Plate load test

Uniaxial jacking
' ft

FIG- U.3 LOCATION OF IN-SITU TESTS CARRIED OUT AT THE LAKHWAR DAM
PROJECT SITE
 ± * M

LEGEND

i Drift

• Drill hole

• Block shear test

D Plate load test
Uniaxial jacking test

FIG.4.A.LOCATION OF IN-SITU TESTS CARRIED OUT AT THE SRINAGAR HYDEL SCHEME

 V

MAIN DAM AXIS A

LEGEND

Drift 1
i
Drill hole •
900
Plate load test D
Block shear test •
Uniaxial jacking test o

FIG. 4.5 LOCATION OF IN-SITU TESTS CARRIED OUT AT THE TEHRI DAM PROJECT
 146

Drift 1
Drill hole •
Block shear test •
Plate load test D
Uniaxial jacking test o

FIG.4.6 LOCATION OF IN-SITU TESTS CARRIED OUT AT THE

PROPOSED UTYASU DAM PROJECT SITE
 147

laboratory, or by the standard point load strength index (Ir)

L

on rock lumps. In order to study the effect of saturation on

different rocks, tests were conducted on both saturated and

naturally moist specimens. The specimens of slates tended to

become more fragile during moisture loss and disintegrated

under water, losing its strength completely. All laboratory

experiments were carried out nearly under similar laboratory

test conditions. The following informations were noted for each

test :

1. Number of Specimens tested.

2. Orientation of bedding planes, foliations etc. with

respect to the loading axis.

3. Location, depth and method of sampling.

4. Specimen diameter and height.

5. Moisture content,

6. Duration of test and rate of loading.

7. Smoothness of specimen surfaces.

8. Specimen features, if any.

It was seen that the tests performed on shale, claystone

and slates were not totally successful because the specimens

fractured during initial stages of loading. Under saturation

most of the specimens of slate, shale, claystone and sandstone

crumbled and could not be tested. This seems to be one of the

biggest limitations of the weak rocks.

Table 4.1 gives the test plan of laboratory tests on

rocks.
 48

4.5 Field Investigations

Field investigations include determination of mechanical

properties and classification parameters of rock masses.

4.5.1 General

Extensive field investigations were carried out to 3tudy the

engineering behaviour of different rock masses encountered at
the project sites. Investigations have been carried out with
the object to study the following aspects concerning the rock

mass behaviour :

1. Deformation characteristics namely, modulus of deformation

(E,) and modulus of elasticity (E );

a e

2. Shear strength characteristics that is cohesion (c) and

angle of internal friction (0) ;

3. Allowable bearing pressure (q ) on rock masses; and

a,

4. Rock mass classification using the following approaches;

(i) RMR system of Bieniawski (1973)

(ii) Q-systera of Barton et al. (1974)

Table 4.2 presents the type and number of in-situ tests

carried out at different projects. Data were collected at the

natural moisture condition as well as the saturated states of

the rock mass.

4.5.2 Modulus of Deformation (Ed), and Modulus of Elasticity

Uniaxial jacking tests on 60 cm diameter plates were conducted

 49

TABLE 4.1

LATODT FLAI OF LABOBATOBT TESTS Ol BOCKS

DAM SITES
Total
Jairani Kotlibehl Lakhwar Srinagar Tehri Utyasu nuiber
of
laie of Bock Saiples
Tested
Test Sand Clajr Liie Shale Slate lenolith Trap Meta Quart - Phyl Quar
stone stone stone basic site lite tzite

Muiber of Saiples fested

1. Visual Exaiination 12 2 8 7 8 3 15 7 8 10 8 88

2. Mater Content 12 2 8 7 8 3 15 7 8 10 8 88

3. Density 12 2 8 7 8 3 15 7 8 10 8 88

4. Specific Gravity 12 2 8 7 8 3 15 7 8 10 8 88

5. Abrasion Value 12 2 8 7 8 3 15 7 8 •6 8 84

6. Slake Durability 4 1 8 7 8 3 15 4 4 6 3 63

7. Swelling Pressure 5 1 3 2 2 2 5 4 4 3 5 36
Index

8. Uniaxial Compress 28 -
8 7 6 8 27 7 8 12 7 118
ive Strength

9. Foisson's Ratio 12 -
1 1 -
2 1 1 4 3 25

10.Static Modulus of 3 -
1 1 2 2 9 7 8 6 4 43
Elasticity

Total Mo. of Saiples 112 12 61 53 58 30 133 58 65 77 62 721

Tested
 150

TABLE 4.2

FIELD IIVESTIGATIOIS 01 BOCKS

DAM SITES

Jairani Kotlibehl Lakhwar Srinagar Tehri Utyasu Total

nuiber
BOCK TYPE of
tests
laie of Test Sand Clay Liie- Shale Slate Ieno Trap Meta Quart Phyl Quar
stone stone stone lith basic zite lite tzite
(Srinagar) (Utyasu)

luiber of Tests

Dniaxial Jacking 9 4 10 3 10 8 29 7 5 22 13 120

Block Shear 11 1 8 1 8 13 12 8 5 6 4 77

Plate Load 2 5 4 4 2 11 4 6 1 39

A
 151

in exploratory drifts excavated at each project site to

determine how the rock mass reacts to controlled loading and

unloading cycles at the desired stress. The test provided data

on the modulus of deformation (E.) and the modulus of

elasticity (Ee). In order to assess modulus anisotropy, the

tests were conducted in horizontal as well as in vertical

directions. Figure 4.7 shows the test set up of equipment for

uniaxial jacking test in drift or gallery. It was assumed that

the plate settlement is about half of the total deflection

between the top and the bottom plate3.

4.5.3 Shear Characteristics - Cohesion (c) and Angle of

Internal Friction (O)

In-situ cohesion (c) and the angle of internal friction (0) of

the rock mass were estimated on rock to rock interface. The

location of the test blocks (700mra x 700mm x 300mm) and the

system of applied loads were selected such that the plane to be
sheared generally coincided with the plane of weakness in the
rock mass. For each test generally five blocks were prepared on
the same test horizon with each specimen tested at a different
but constant normal stress. Due care was taken to allow pore
water pressure in the rock adjacent to the shear plane to
disscipate under full normal stress before shearing. Type of
failure and the manner in which additional cracks developed in
the specimen consequence to the loading were observed.
Temperature, humidity in the test area at the start and at the
end of the test, water content of the rock at the test
interface and the description of the failure surface were
recorded. Figure 4.8 shows the arrangement of the block shear
 152

test in the drift.

4.5.4 Allowable Bearing Pressure (q&) on Rocks

For the assessment of bearing capacity of the rock strata,
plate load tests on 60 cm diameter plates were conducted at
different locations of the project sites. Curves between
pressure versus settlement have been obtained for different
rock formations. Using pressure - settlement curve, the
allowable bearing pressure corresponding to 12.5 mm maximum
permissible settlement of foundation of 6 m width have been
evaluated. A relationship has been developed between the rock
mass rating (RMR) and the allowable bearing pressure (qfl). The
plate load test set up is the same as shown in Fig. 4.7.
4.5.5 Rock Mass Classification
(i) RMR system of Bieniawski (1973)
The RMR system was applied and used by dividing the projects
area into a number of geostructural zones so that the type of
rock and discontinuity spacings in a zone were more or less the
same throughout. The zones were further divided into a number
of structural regions such that certain features were more or
less uniform within each region. The test sections were
selected such that these were representative of the typical
conditions likely to be encountered during tunnel excavation.
In a number of structural regions, mixed quality conditions
were encountered at the excavated rock face. For example, good
quality and poor quality gypsyferrous limestone were present in
sorae drifts of the Kotlibehl dara project. In such cases, the
most critical condition for the assessment of rock strata were
 153

2-5 cm thick steel Rock mass _..

/ Plates with legs at
plate 60 cm dia 120 for fixing dial
__^p^__guages

Rock mass

FIG. k-7 SETUP OF EQUIPMENT FOR UNIAXIAL

JACKING TEST IN VERTICAL DIRECTION
IN NARROW DRIFTS OR GALLERIES
 15<>

Rock, mass
/I
Cement -sand mortar
/ 0:2)
. \\\Y///
Steel plate
y%&^#
^^;
Packing
^

<4<i
\\

-Hydraulic jack

Cement-
mortar (

- Cement-concrete Roller arrangement

(1:2:4 ) M-Odia)

£
3r x x x n Steel plate
jTY\fY^YYSV^SVrYV\ VJ - (50X50X2-5cm)
Test block Steel casing
70X70': 3D cm

Rock mass

FIG. 4-8 ARRANGEMENT OF BLOCK SHEAR TEST IN DRIFTS

 155

identified.

After the importance ratings of the classification

parameters were assessed, the ratings of the first five

parameters were summed-up to obtain the basic RMR value for the

structural region under consideration. The influence of the

strike and dip of the discontinuities was included at this

stage and flnaJ RMR value was obtained. The rock mass was

classified into one of the five classes. No adjustment was made

for primary (in situ) stress field or a change in stress. All

pertinent information regarding the location, size and special

features in the exploratory drifts were recorded. The shear

zones in the rock mass were taken as the boundaries of the

structural regions and these have been treated as a special
structural region for the purpose of rock mass classification.

In a few structural regions two, sometimes three or even more,

clearly different zones were observed to be present. In such

cases, overall weighted value of the RMR was computed.

(ii) The Q-system of Barton et al. (1974)

The Q-system has been applied on the same structural regions as

those used for the RMR system. The following guidelines were
observed while estimating Q.

(a) Same RQD values have been used as those used for the
estimation of RMR values.

(b) In case of strongly foliated slates of Lakhwar dara

project, the parallel joints were counted one joint set.

(c) The parameters Ja and Jf have been taken for the weakest
 156

joint set anticipated in the te3t section.

(d) Primary stress field has not been considered while

evaluating SRF and Q.

4.6 Results of Investigations

The results of laboratory and field investigations are

given as follows :

4.6.1 Laboratory Tests

The standard laboratory tests for specific gravity, dry

density, water content, abrasion value, slake durability,
swelling pressure, uniaxial compressive strength, raodulus of
elasticity and Poisson's ratio were carried out mostly in the
test laboratory of the Irrigation Research Institute at
Roorkee. The laboratory test results are summarised in Table
4.3. Number of test sections investigated and summary of
classification results are given in Tables 4.4 and 4.5
respectively.

4.6.2 Field Investigations

*

Extensive field tests were carried out at varied locations of the

projects as follows :

1. Uniaxial jacking tests

2. In-situ block shear tests

3. Plate load tests

Table 4.4 gives the number of test sections investigated ^

for RMR and Q-systerns of rock mass classifications carried out
at the six hydroelectric projects under study. Table 4.5 shows
the summary of classification results. Summary of the results
 157

TABLE 4.3

SOMA!! Of LABOBATOBT TIST IKSDLTS

Dniaxiai Natural Dry Porosity Abrasion Slake Saelling

Coipressire Moisture Density Value Durability Pressure
Bock Type Strength Content Index Index
[««] [mc] m [a]

(MPa) (X) (gi/cc) (X) (X) (XI (kPa)

D 32.00- 75.00
Sandstone 1.25- 1.50 2.28- 2.50 5.66- 14.49 25.00- 45.00 34.5- 46.5 0.0- 0.1
S 24.00- 48.00

D
Claystone 2.25- 4.50 2.38- 2.42 7.75- 12.00 29.00- 35.00 49.5 1.0- 2.5
S .

D 1.00- 38.00
Slates 0.30- 0.94 2.64- 2.79 1.80 - 4.28 36.00- 90.30 19.7-100.0 9.0-25.0
S 0.00- 20.50

D 21.00
lenolith 0.25 2.69- 2.74 2.54 - 3.52 44.00- 70.30 27.1- 33.5 2.0
S 10.00- 21.00

D 98.00-196.50
Trap 0.28- 0.30 2.63- 2.84 1.43 - 5.73 19.75- 39.00 14.2- 39.0 0.0
S 71.50-163.00

D 16.80- 37.00
Shale 1.00- 1.95 2.06- 2.42 5.22 - 8.44 42.00- 49.75 27.3- 36.0 2.5- 4.5
S 12.50- 30.50

D 21.00- 49.00
Liiestone 0.30- 2.50 2.08- 2.43 4.40- 17.03 28.00- 39.55 13.7- 33.3 -

S 16.00- 40.00

D 70.90-104.00
Metabasic 0.30- 0.60 2.72- 2.88 0.31 - 2.98 2.35- 13.10 7.2- 19.7 0.0
S 63.00- 88.50

D 67.00-128.00
Quartzite 0.40- 0.45 2.51- 2.87 1.37 - 5.70 9.30- 23.40 13.5- 30.3 0.0
S 54.50-112.00

D 38.00-133.00
Phyllite 0.35- 0.90 2.61- 2.82 2.60- 8.74 25.00- 38.00 17.0- 37.5 0.10
S 25.50- 95.50

D indicates natural loisture content

S indicates saturated condition

 158

TABU 4.4

ID8BII Of TIST SICTIOIS IITISTIGATID 101 BOCI IASS CLASSIflCATIOl

STBDCTDBAL ZONE AND NDMBEB Of TEST SECTIONS

Total
Classification Systei Jairani lotlibehl Lakhuar Srinagar Tehri Dai Dtyasu Dai Nuiber of
Dai Site Dai Site Dai Site Dai Site Site Site test sections
inrestigated

8MB 20 28 53 35 29 10 175

q 18 19 47 21 23 8 136
 159

TABLE 4.5

SUMMARY OF CLASSIFICATION RESULTS

CLASSIFICATION SYSTEM

ROCK RMR Q

Sandstone Class IV : Poor rock mass Poor rock mass

Claystone Class IV : Poor rock mass Poor rock mass

Limestone Class IV : Poor rock mass Poor rock raass

Shale Class IV : Poor rock mass Poor rock mass

Slate Class IV : Poor rock mass Poor rock mass

Xenolith Class IV : Poor rock mass Poor rock mass

Traprock Class III: Fair rock mass Fair rock mass

Quartzite Class III: Fair rock mass Fair rock mass

Metabasic Class III: Fair rock mass Poor rock mass

Phyllite Class IV : Poor rock mass Poor rock mass

 160

of in-situ investigations is given in Table 4.6. Figs. 4.9 -

4.12 show typical pressure-settlement curves of poor and fair

rock masses, obtained from uniaxial Jacking tests. Fig. 4.13

illustrates the method of calculation of E,, and E .

d e
4.7 Discussion

The laboratory and insitu tests carried out to determine the

geotechnical properties of rocks have a number of limitations.

Although extensive care was taken while carrying out the tests,
yet it is possible that values obtained are not always
representative of the actual rock mass condition existing at

the site. The nature of the rock mass varies from place to
place, and from project to project because of the varying
moisture conditions, effects of anisotropy, the stress

environment in the rocks, the effects of temperature and other

unknown geological factors. The above mentioned factors seem to

have been responsible in the data scatter which is seen in some

test results. Looking into the overall perspective, the wealth

of geotechnical data so obtained requires proper interpretation

so that the most representative values are estimated. This

objective has been achieved in chapter-5 titled "ANALYSIS AND

INTERPRETATION OF DATA".
 161

%M TABLI 4.6

SDHBABT Of BISDLTS Of II-SITO IITISTIGATIOI

Modulus of
deforiation elasticity rock Q m Poisson' s Pen
Bock Type [Ed] [Be]* [Er] Batlo eabi
litjr
(GPa) (GPa) (GPa) (Lugeon)
I 0.44- 2.10 1.75- 2.90 6.76 0.7 - 2.0 20- 4! 0.30- 0.3( 7- 19
Sandstone
£ 0.35 1.20 17.50 0.2 27 - -

1 0.74- 1.55 1.43- 3.67 8.41 0.7 - 1.8 24- 3! -

Claystone
S

D 0.49- 4.04 0.98- 7.80 20.00 0.3 - 3.9 18- 45 0.39 2.5- 5.0
Slates
S 0.50- 1.30 1.09- 3.60 14.74 0.1 - 2.1 23- 42 - -

D 1.58 2.95 14.74 1.5 - 1.6 28- 33 0.32 -

lenolith
S 0.41- 4.60 0.98- 5.07 12.35 0.4 - 2.7 30- 60 -
-

D 1.60- 9.12 1.98-13.00 12.35-36.43 1.7 -11.7 30- 61 0.24- 0.27 7- 21

Trap
S 0.86- 8.47 1.60- 9.70 21.50-42.80 1.2 -13.2 43- 64 -
-

D 0.90- 1.57 2.22- 2.95 10.80 0.9 - 1.5 25- 30 0.36 -

Shale
S 0.30- 0.50 1.09- 1.15 -
0.1 - 0.2 23- 25 -
9

D 0.26- 3.08 0.55- 4.80 11.90 0.1 - 4.0 11- 53 0.32 14- 21
Liiestone
S

D 1.45- 3.08 4.38- 7.11 21.00-22.40 3.3 - 4.7 37- 60 0.29 2- 11

letabasic
S 1.20- 1.25 3.39- 5.73 22.40 2.5 - 3.0 49 • - -

D J.84-13.70 0.98-14.37 28.25-49.80 J.3 -19.0 27- 71 0.33 8

( luartzite
S 1J.78- 5.55 1.78- 6.52 30.00-40.16 1.2 - 4.9 37- 58 -
-

Dl 1.54- 3.48 0.73- 4.13 6.68- 7.07 11.3 - 4.1 18- 50 0.24- 0.32 1- 28
[ hyllite
S I1.38- 4.08 1.25- 5.14 ]6.27 11.3 - 6.5 31- 61 -

'

modulus
modulus 'of73^974;- thS The
of deformation. Parameter
modulus(E*)of is deformation
termed as r™T
the

D - indicates the value at the natural moisture

s - indicates the value at saturated condition
 162

70-0-1

60-0

I 50-0- Maximum stress 50-60 kg/cm2

Ed - 11850 kg/cm2
40-0 E-e =• 52.000 Ks/Cvm"

V)

30-0-

20-0-

10«0-

0-0
2-0 4-0 6-0 80 10-0 12-0 K-0

Deformation , mm —♦-

FIG. k 9 TYPICAL PRESSURE - SETTLEMENT CURVE FOR POOR ROCK

MASS (RMR =25) OBTAINED FROM UNIAXIAL JACKING
TEST ( DIAMETER OF PLATE 60 cm )
* As per IS:7317-1974, the parameter (Ee) is termed as the
modulus of deformation. The modulus of deformation (Ed)
adopted in the present thesis is with a different concept as
given on page 166.
 163

450-1

Max. stress = 42- 44 kg/cm2

400- Ed =r 13 840 kg /cm2

35.0.

300-

Deformation ^ mm

FIG. if-10 TYPICAL PRESSURE-SETTLEMENT CURVE FOR POOR ROCK

MASS (RMR-31) OBTAINED FROM UNIAXIAL JACKING TEST
(DIAMETER OF PLATE 60 cm)
As per IS:7317-1974, the parameter (Ee) is termed as the
modulus of deformation. The modulus of deformation (Ed)
adopted in the present thesis is with a different concept as
given on page 166.
 164

70-0-1

60-0-

Max. stress 50*60 kg/cm2

500

Ed • 27950 kg/cm'
E E.e =» 4-3800 Kg/Cm
u
400-

^300-
V)

20-0

10-0

0-0
14-0

Deformation, mm

FIG.it-11 TYPICAL PRESSURE-SETTLEMENT CURVE FOR FAIR ROCK

MASS (RMR* i*3) OBTAINED FROM UNIAXIAL JACKING TEST
(DIAMETER OF PLATE 60cm)

* As per IS:7317-1974, the parameter (Ee) is termed as the

modulus of deformation. The modulus of deformation (Ed)
adopted in the present thesis is with a different concept as
given on page 166.
 165

70-0-1

1-5 2-0 2-5 3.655 4-0

Deformation , mm _

FIG. ^-12 TYPICAL PRESSURE - SETTLEMENT CURVE FOR FAIR ROCK MASS
(RMR = 5<*) OBTAINED FROM UNIAXIAL JACKING TEST (DIAMETER
OF PLATE 60 cm )

As per IS=7317-1974, the parameter (Ee) is termed as the

modulus of deformation. The modulus of deformation (Ed)
adopted in the present thesis is with a different concept as
given on page 166.
 1G6

Stress kZ'Uk kg/cm

Ed = 13870 kg/cm2
Ee • 28>90kg/cm2
35-0

300

250

CM
E 200-
u

J*

in

150-

10-0

r T"
2-0 30 i»-0 5-0 60 70 8-0

Deformation, mm-»-
Ed _ mCl-M2) x P (Bousslnesq's rigid punch equation)
^A~- S/2
-Where, E _ 0»96(1-0-0U)xa2-t»ttx2828
P • Normal load in kg ^2 828 x (0-3/2 )
V • Poisson's ratio («• 0-2,assumed) = 13,870 kg/cm2
m » Constant =0-96 for circular plate
A - Area of plate in sq. cm (2828 cm2
m 0-96(1-0-Qt»)xfr2-^x2828
for 60cm diameter plate)
o • Deformation corresponding to load
e" /2828 x (0MU6/2 )
P in cm
= 28,U90 kg/cm2

FIG. if-13 METHOD OF CALCULATION OF MODULUS OF DEFORMATION (Ed) AND

MODULUS OF ELASTICITY (Ee)

* As per IS:7317-1974, the parameter (Ee) is termed as the

modulus of deformation.
CHAPTER 5
 CHAPTER 5

ANALYSIS AND INTERPRETATION OF DATA

5.1 General

Study of the construction problems encountered due to the

complex Himalayan geology reveals an urgent need to understand

the rock mass behaviour. Accordingly, the field and laboratory

test data obtained for over 4-5 years period in the Lesser

Himalaya have been analysed, interpreted and suitably

correlated through RMR and Q system of the rock mass

classification.

As mentioned in Chapter 4, the project site3 under

investigation involved ten different kinds of rocks, which have

been categorised from poor to fair class rock masses (Table
4.5). The results as obtained through field and laboratory
tests depict varying range of the engineering parameters. For

clarity and convenience, each parametric study as analysed has

been discussed separately.

5.2 Modulus of Deformation (E,)

d

Figure 5.1 shows the data obtained from seventy uniaxial

jacking tests carried out at the natural moisture content and

another fifty tests carried out on saturated rock raass.

Basically, the test involved applying load to rock surface by

means of hydraulic jack and measuring the resulting
deformation. The experimental technique and "other conditions
during test performance were kept the same throughout (Fig.
4.7).

For the estimation of modulus of deformation (E,), the

 168

area of the six hydroelectric project sites was divided into

eleven geostructural zones involving ten different kinds of
rocks (Table 4.2). The results have been plotted in Fig. 5.1
and show wide range of modulus values obtained during the

tests.

In order to check the appropriateness of the method and to

examine consistency in test results, 14 sets of uniaxial
jacking tests were carried out on the same rock raass at the
natural moisture content,and at saturation. Each set consisted
of repeating the test on the same horizon, under similar test
conditions but separated by a distance of 4-5 m. Results of
tests conducted at natural moisture content,and at saturation
are plotted in Figs. 5.2 and 5.3 respectively. As is seen in
Fig. 5.2 there is not much scatter in the results. It is worth
noting that out of fourteen sets of tests performed, there is
exact repeatability in four tests, quite close comparison in
eight tests; and two tests correspond to the ratios of more
than 2 between the values of deformation moduli obtained at the
two locations (separated by 4-5 m).

Similarly, no appreciable scatter is seen in the test

results obtained on the saturated rock mass. In one case only,
the data point corresponds to the ratio of more than 2. Sorae
data scatter seen in the results may be due to the influence of
certain unknown geological factors present at the two test
locations, which can not be considered in the field test data.
The results of these pilot tests thus indicate that consistency
of data in the test method is appropriate and it may be
inferred that the data have not been influenced by the test
 169

SATURATED ROCK MASS

Data at nmc
Data at saturation

30 40 50 80

ROCK MASS RATING (RMR)

FIG. 5-1 CORRELATION BETWEEN ROCK MASS RATING CRMR) AND MODULUS
OF DEFORMATION (Ed)
 w « v-.w/' aM££ ,-;.'. •N'v,',' v.v.',' ~-x>y// ^-nv>

Foce m—<t-5ni-^ Portal

(Ed)M2 (Ed)M1

Exploratory drift (2-1m x 2-1 m )

d?
o

*^

T3
111

C
o

o
o
E

•o

1 2 3 U ~5
Modulus of deformation (Ed)M1 , GPa—»•
FKS.5-2 MODULUS OF DEFORMATtON DATA FROM UNIAXIAL JACKING TESTS [ROCK MASS AT nmc]
 f- <

£ Face ^.ii-5m _^ Portal

I"•«y (Ed>M2
jw y—777-\7—pny wy wry
tEd>M1
»y m"'\l **\J—" vi wy w*r**&

Exploratory drift (2-1mx2-1m)

5i

a
Q.
O

Ui

4-.

•
E

•a

3
•o
o

2 3 i» 5
Modulus of deformation ( Ed )M1 gp —»-

FIG.5-3 MOOULUS OF. DEFORMATION OATA FROM UNIAXIAL JACKING TESTS (SATURATED ROCK MASS)
 172

raethod. It is also seen in Figs. 5.4 and 5.5 that the mean

values of the moduli compare well with the predicted values

from curves 1 and 2 in Fig. 5.1 for both moist and saturated

rock masses.
'•

Figures 5.6 to 5.16 also show the plots of results for

each rock type tested at natural moisture content. It is seen

again that in all the cases except Tehri phyllite and Utyasu

quartzite there is close agreement between the measured and

predicted values (from curve 1 of Fig. 5.1). In case of Tehri

phyllite (Fig. 5.15) and Utyasu quartzite (Fig. 5.16) the

values are seen to be in fair agreement. In case of Tehri

phyllite the prediction error is 13 per cent.

The plots of the moduli values obtained on the saturated

rock mass are shown in Figs. 5.17 to 5.26. It is seen that

there is close agreement between the measured and predicted

values according to curve 2 in Fig. 5.1 in cases of sandstone,

xenolith, quartzite (Srinagar) and phyllite rocks (Figs. 5.17,

5.21, 5.23 and 5.25). The values are in fairly good agreement

in case of shale, trap, metabasic and quartzite (Utyasu) rocks

(Figs. 5.19, 5.20, 5.24 and 5.26). It may be seen from Fig.

5.24 that in case of metabasic rock the prediction error is

24.8 per cent. In case of shale, trap and Utyasu quartzite, the

ratio (E, predicted/E, measured) varies from 0.80 - 0.85. In

case of trap the prediction error is 18 per cent while in case

of shale and Utyasu quartzite it is 25 per cent.

The modulus of deformation data presented under this study

includes results of 120 uniaxial jacking tests. Of the total

 173

number of tests performed, 92 tests were carried out in

vertical and 28 tests in horizontal direction.

In order to study modulus anisotropy, an independent

series of tests was performed covering different kinds of rock

masses. The series consisted of performing 17 uniaxial jacking

tests in vertical as well as in horizontal direction. Each set

of test (i.e. one horizontal and one vertical) was performed on

the same rock type but at locations which were separated by a

distance of 4-5 m. It was ensured that the properties of rock

raass apparently did not differ at the two locations. Of the 17

sets of tests, 10 were performed at the natural moisture

content and 7 on the saturated rock mass using the peak stress

level of about 4 MPa. The plot of the results are shown in

Figs. 5.27 and 5.28.

Figure 5.27 indicates that anisotropy is more marked in

rocks having bedded structure (^(Vertical/SdCHorlaontair1-85

for phyllite and 0.59 for zenolithl. It is also seen that
anisotropy is more predominant in fair rocks than in poor

quality rocks becuase of too many joint sets in the latter. It

is also noted that anisotropy is more pronounced in saturated

rock mass than at natural moisture content (nine).

Ej,„ ,,/Ej,,, . -,v varies from 1.39 to 0.76 for

d(Vertical)/ d(Horizontal)
saturated rock mass as against 1.27 to 0.65 for rock mass at

natural moisture.
>
The results indicate that the rocks show moderately

anisotropic behaviour in their deformation characteristics.

 174

2 3 k 5
Modulus of deformationCy measured (mean), GPq

FIG.5* MODULUS OF DEFORMATION-MEASURED v/s PREDICTED VALUES(ROCK MASS AT nmc )

4 *-

U1

2 3 k 9 S
Modulus of deformation [Edl measured (mean), GPa —-

FIG. 5-5 MODULUS OF DEFORMATION - MEASURED v/s PREDICTED [SATURATED ROCK MASS]
4 fr

>£ (Ed) Predicted

o (Ed) Measured
a
o

•a
<u

c
o

a
e

3
•a
o
2

2 3 4

Modulus of deformation (E<j) measured , GPa

FIG. 5-6 MODULUS OF DEFORMATION-MEASURED v/i PREDICTED [ROCK MASS AT nmcj

4 ii
>~

a -a
Q. -3

•o
at

•a
41

^6
m

c
e 2-

a
E

*»
T3

in

O
1 -

2 3 <»

Modulus of deformation (E<j) measured , GPa—*"

FIG 5-7 MODULUS OF DEFORMATION-MEASURED v/s PREDICTED [ROCK MASS AT nmc]

 -J
CD

2 3

Modulus of deformation(Ea) measured , GPa

FIG. 5-8 MODULUS OF DEFORMATION-MEASURED v/s PREDICTED [ROCK MASS AT nmc]

 -J

2 3 4
Modulus of deformation (Ed) measured, GPa

FIG. 5-9 MODULUS OF DEFORMATION - MEASURED v/s PREDICTED CROCK MASS AT nmc]
 9-

8 -

Moduls t 7.
0?
o

•o
6

2 6-
tl

T3

— 5 -
c
o

a
E
u
o

« u.
•o

CO
o o

2-
<&

1 -
/%

5 6 7 8
Modulus of deformation (Ed ) measured , GPa —-
FIG.5-10
MODULUS OF DEFORMATION-MEASURED v/% PREDICTS D[ROCK MASS AT nmc]
 0-87

U -

3 -
o

•o
0)

•o
01

.§ 2
E

Modulus of deformation(Ed) measured GPa

FIG. 5-11 MOOULUS OF DEFORMATION- MEASURED v/s PREDICTED [ROCK MASS AT nmc]
 «

o
a.
o

-a
v

•a

•o
LU 2 -

c
o CO

a
E

at
•a

1 -
3

2 3
Modulus of deformation (Ed ) measured ,GPa-

FIG.5-12 MODULUS OF 0EF0RMAT10N- MEASURED vs PREDICTED [ROCK MASS AT nmcj

 o
a.
o

T3
01

aii
u o
00
c

a
E

•a

•o
o

2 3 4
Modulus of deformation (Ed) measuredt GPa""*"

FIG.5-13 MODULUS OF DEFORMATION-MEASURED v/s PREDICTED [ROCK MASS AT nmc]

*> *

9 -

t
8-

7-1
/\(Ed ) predicted
>r (Ed) measured
•a
at

6-
•D
4>

5-
C
O

a
E
CO

•a
o

/**
¥ /

1 -

_ 1 • • •

2 3 4 5 6 10

Modulus of deformation (Eg) measured, GPq

FIG. 5-14 MODULUS OF DEFORMATION-MEASURED v/s PREDICTED [ROCK MASS AT nmc]

fc

00

1 2 3 U
Modulus of deformation (Ed) measured , Gpa—*•
FIG.5-15 MODULUS OF DEFORMATION-MEASURED v/*i PREDICTEO [ROCK MASS AT nmc]
 15

a
D.
O

0 -

Ui

c
o

a CO
E 31

Ol
T3

5-
3

O
2

5 10 15
Modulus of deformation (Ej) measured , GPa *"

FIG.5-16 MODULUS OF DEFORMATION-MEASURED v/s PREDICTED [ROCK MASS AT nmc]

^ ft

i
20

/C^ (Ed) predicted

(Ed) measured

? l.fl

01
wk
•o
u

Ui

2 10
*•*
a
E CO

m
3

"3
n

| 0-5

1 1

0-5 1 T-5 2 2-5

Modulus of deformation (Ed ) measured , GPa

FIG. 5-17 MODULUS OF DEFORMATION - MEASURED v/s PREDICTED VALUE [SATURATED ROCK MASS]
 CO
CO

0-5 1-0 1-5 2-0 2-5

Modulus of deformation (Ed) measured ,GPa

FIG. 5-18 MODULUS OF OEFORMATION-MEASURED v/s PREDICTED [SATURATED ROCK MASS]

 ft

2 -I

a
o.
o

•D
*«

Ol

•o

01
•o

= 0-5
T3
O

0-5 1 1-8 2-5

Modulus of deformation(Ed) measured, GPa-

FIG.5-19 MODULUS OF OEFORMATION - MEASURED v/s PREDICTED VALUES [SATURATED ROCK MASS]
1

0-85
(Ed) predictedf
( Ed) measured"'

3 a 5 6

Modulus of deformation l&d) measured ,GPa

FIG. 5-20 MODULUS OF DEFORMATION- MEASURED v/s PREDICTED VALUES [SATURATED ROCK MASS!
 a
c 3 .
(3

01

T3
Ol

•a
UJ

.2 2
o
E

•o

1 2 3
Modulus of deformation( Ed) measured ,GPa
FIG. 5-21 MODULUS OF DEFORMATION —MEASURED v/s PREDICTED [SATURATED ROCK MASS]
 ft

0-6

10 1-5 20 2-5
Modulus of deformation ( Ed ) measured , GPa —»-

FlG.5-22 MODULUS OF DEFORMATION - MEASURED v/s PREDICTED [SATURATED ROCK MASS]

 2-0-1

(Ed) predicted
(Ed) measured
of 1-5

•o
2!
Q.

-o
Ul
~ 10.

a
E

«i
T3

1 0-5.
•o
o
X

1-0 1-8 2-0 2-8

Modulus of deformation (Ed) measured , GPa —»•

FIG.5-23 MODULUS OF DEFORMATION-MEASURED v/s PREDICTED [SATURATED ROCK MASS]

 ¥

1-33 1

/£ (Ed) predicted
(Ed) measured

0-3 10 1-5 20 2-5

Modulus of deformation (Ed) measured , GPa

FIG. 5-24 MOOULUS OF DEFORMATION-MEASURED v/s PREDICTED [SATURATED ROCK MASS]

 195

Modulus of deformation (Ed) measured, GPa

FIG.525 MODULUS OF DEFORMATION-MEASURED v/% PREDICTED

[SATURATED ROCK MASS]
 *

U-5-
0-8
/
2&
(Ed) predicted
X
(Ed) measured
4

S 3"
n
Ol

•a
UJ

a
en
S 2
o

Ol
T3

3
-o
O
Z

1-

2 3 «. '
Modulus of deformation (Ed) measured , GPa •"

FIG.5-26
MODULUS OF DEFORMATION'MEASURED v/s PREDICTED VALUES [SATURATED ROCK MASS]
 197

0-59

Modulus of deformation [Ed]h , Gfb (horizontal loading)

FIG.5-27 INFLUENCE OF LOAD DIRECTION ON THE MODULUS OF DEFORMATION
[SATURATED ROCK MASS]
 198

1 2 3 U 5

Modulus of deformation CEd]n , GPa (Horizontal loading)

FIG5-28 INFLUENCE OF LOAD DIRECTION ON THE MODULUS OF DEFORMATION (ROCK MASSAT nmc)
 99

5.2.1 Estimation of Modulus of Deformation

The geology in the Lesser Himalaya where the six hydroelectric

projects are located, varies from place to place and project to

project. The rock mass in the region varies considerably in

quality, having RMR ranging from 17 - 64-, the moduli values vary

from 0.5 GPa to 12 GPa. The following relationship has been

observed,

RMR - 25

Ed = 10 40 GPa (5.1)
The relationship yields the prediction error of 20.8 per

cent as shown in Fig. 5.1.

The modulus data obtained on the saturated rock raass have

been presented in curve-2 of Fig. 5.1. For the rock masses

having natural moisture content of the order of 5 per cent, the

RMR has been estimated by taking into account the condition of

groundwater having the rating 10 (for damp rock mass). For rock

masses having moisture content at saturation, the rating has

been taken as 7 (for wet rock mass). Thus for the same type of

rock mass, the RMR values of dry (at nmc) and saturated states

differ by a rating of 3 (Bieniawski, 1978).

5.2.1.1 Effect of Saturation on Modulus of Deformation

It is seen that the effect of saturation is more predominant in

poor rocks than in fair quality rocks. It is worth noting that

with increase in the RMR values the two curves in Fig. 5.1

(curve 1 and curve 2) approach to come closer which may be

expected also. As is seen,the trend of the curve is not clearly

defined for RMR values less than 30 in case of saturated rock

 200

mass.

The effect of saturation has been studied in detail

through a series of separate tests performed on poor as well as
fair quality rocks (Fig. 5.29). Nine cases have been studied on
rocks having RMR values between 21 - 40 (poor rock) and eleven
cases on rocks having RMR values between 41 - 60 (fair rock).
Reduction factors have been worked out for both the cases. It
is seen that for poor rocks, the saturation reduction factor
(RF) lies in the range 0.1 to 0.3, and for fair quality rocks,
the RF lies between 0.3 and 0.6. The results of two cases

(denoted by question mark in Fig. 5.29), one each for poor rock
and fair rock have not been included as these are not

consistent with the other results., because of testing errors. It

is found from Fig. 5.29 that RMR bears a bi-linear relationship

with the reduction factor for saturation as follows :

E,, i% /E., , = 0-01 RMR - 0-1 (for RMR 21—• 40) (5.2)
d(sat) d(nmc)

EJ, ^ /EJ/ x = 0.016 RMR - 0.385 (for RMR 41—• 60) (5.3)
d(sat) ' d(nmc)

>

It is observed that the effect of saturation is more

striking in poor rocks (average saturation reduction factor *

0.2) than in fair quality rocks (average saturation reduction
factor * 0.45). This is expected because RMR was generally less

in argillaceous rocks with water sensitive minerals in Lesser

Himalaya.

5.2.2 Correlation based on Modulus Reduction Factor (MRF)

The raodulus of elasticity (Er) values obtained in the

laboratory on rock cores have been used to develop the
 0-6-

saturated

0.5

u
E
c

Ol 0-V

3
*-•
CO
a
o
(A

•a
0-3-
ai

c
o 0-2-
u
[Edsaturated/Ednmc] =0-01RMR-0-1
3
•a
[3
Ol

c
o

0-1
a

10 20 30 t*Q 50 60 70

Rock mass rating (RMR) *•

FIG.5-29 EFFECT OF SATURATION ON THE MODULUS OF DEFORMATION (ed)
 202

relationship between modulus reduction factor (E,/E ) and the

RMR. It is seen that a definite relationship exists between RMR

and the MRF (Fig. 5.30). It may be seen from curve 1 of Fig.

5.30 that for rock mass at natural moisture content, for RMR 21

-60, the ratio (Ed/Er) is 0.1 - 0.2 which is quite close to the
correlation of Bieniawski (1975). For the saturated rock mass-,

the ratio (Ed/Er) is in the range 0.012 - 0.12 (curve 2; E is

at nmc). Unfortunately, E for saturated condition could not be
r

obtained because of disintegration of rock cores in raany cases.

5.2.3 Discussion of Results

Figure 5.1 shows the RMR values plotted against the raodulus of

deformation (E,) for RMR values from 17 to 64. Curve 1 is based

on the data obtained on rock raass tested at natural moisture

content and curve 2 is based on the data on the saturated rock

raass. Bieniawski (1978) correlated the raodulus of deformation

(E,) through the geomechanics classification for hard rocks as

shown by curve 3, and for poor quality rock masses, Serafim and

Pereira (1983) proposed an overall correlation which is shown

by curve 4 (Fig. 5.1).

It is observed that Bieniawski's correlation (Eq. 2.2) is

applicable for rock masses having RMR more than 50. The

relationship suggests that when RMR is 50, the raodulus of

deformation is zero. However, it is found that even the

extremely poor rocks and soils have values of modulus greater

than zero. Use of this equation is thus not justified at least

for poor rocks. The relationship proposed by Serafim and

Pereira (1983) is applicable for RMR values lower than 50 also.

 203

It is worth noting that curves 1 and 2 have similar trend as

that of curve 4 (Fig. 5.1) but the two are not in good

agreement. The difference may be attributed to various

geological factors such as weak rock material (Er), in-situ

stress field, mineralogical composition of the rocks especially

water sensitive minerals, water flow in rock mass and different

test procedure. These factors may be altogether different in

the Lesser Himalayan rocks than those on which Serafim and

Pereira's Eq. 2.3 is based. Doubts have been expressed that

tests are conducted in drifts after several months of

excavation by blasting so that rock mass becomes weaker due to

blasting vibrations, stress relief, creep and weathering

particularly in argillaceous rocks. The damage to rock mass

will be more in the roof of the drift. If borehole extensometer

had been used, higher values of E, would have been obtained. In

fact, there is close comparison of modulus reduction factors

from proposed tests at natural moisture content and

Bieniawski's (1975) correlation in Fig. 5.30. As such, the

testing procedure is found to be satisfactory for purpose of

correlating E, and RMR which is found on the spot at the time

of test.

It is seen that under saturation, the reduction in the

raodulus of deformation (E,) is as high as 90 per cent for the

poor rocks and 70 per cent for fair quality rocks. If in-situ

tests are conducted in dry condition, the proposed correction

factors (Eqs. 5.2 and 5.3) raay be applied to predict the design

parameters for saturated condition which may prevail after

commissioning of projects.
V 4

0-3

0-275-

u
(After Bieniawski, 1975).
f 0-2-
UJ I

Q£

is:
o

0-1-
Rock mass at natural
moisture

T3
O
2 rock mass

10 20 30 dO 50 60 70

Rock mass rating (RMR )

FIG.5-30 RELATIONSHIP BETWEEN ROCK MASS RATING (RMR) AN0 RATIO OF MODULUS OF DEFORMATION
(Ed) TO THAT OF ROCK MATERIAL (Er)
 205

The modulus of deformation (E,) may also be estimated

rapidly and with sufficient accuracy frora the relationship

between RMR and the MRF (=E,/E ), as the raodulus of elasticity

a r

(E ) of the rock material may be easily obtained from small

laboratory specimens. It may be seen that the RMR bears a

non-linear relationship with the MRF.

It is noted that the curve proposed for saturated rock

raass is not in agreement with the Bieniawski's relationship,

whereas the curve proposed for the rock mass at natural

moisture content seems to be in good agreement. However, in the

higher RMR range (for RMR values > 60), the two curves appear

to follow the Bieniawski's relationship. The difference in the

trend, in the poor and fair quality rock masses raay be due to

the variation in minerals and uncertainties of geological

conditions in the field from where the data have been obtained.

It is observed that the average modulus reduction factors

varied as 0.05 - 0.09 for poor quality rocks and 0.09 - 0.18

for fair quality rocks. The relationship has thus provided a

useful link between the laboratory measurements and the insitu

data for estimating modulus of deformation (E,) values of the

poor and fair quality rock masses.

5.3 Short Term Support Pressure in Tunnels

The basic problem that needs proper attention in an underground

excavation is the temporary as well as permanent stabilisation

of the ground. In almost all cases, the stabilisation problem

is basically the problem of design of a proper support system.

An ideal approach must yield reliable values of support

 206

pressure for the design of a suitable support system before the

commencement of an underground excavation. Among the various

approaches available today, the empirical approach to tunnel

design is based on experience gained on previous projects of

similar nature. The approach is based on engineering rock raass

classifications that enable use of previous experience with

respect to the relationship between the rock raass quality,

excavation method and the required tunnel support.

5.3.1 Rock Mass Classification Systems

Among the major rock mass classification systems for the

estimation of support pressure, Q-system devised by Barton et

al. (1974) has been applied and used for implementation at the

six hydroelectric projects. The Q-system is based mainly for

hard rock situations as the RMR system does and the method is

especially appropriate for establishing the support pressure

for tunnels in hard rocks. In order to compare the values

predicted by the two classification systems, namely, the RMR

and the Q-systems, the classification approach was

simultaneously applied at the six hydroelectric projects. It is

to be expected that the application of different classification

systems to the same ground condition would lead to similar

conclusions.

The Q-system has been used as the basis for establishing

the support pressure whereas the Terzaghi (1946) rock load

system as modified by Deere et al. (1970) and the RMR system by

Bieniawski (1973) were used for checking and comparison of

results. The RMR and Q as obtained at the project sites have

been plotted and the relationships that have been obtained are
 207

as follows :

RMR = 13 In Q + 31 (rock raass at nrac) (5.4)

RMR = 10 In Q + 40 (saturated rock mass) (5.5)

Figure 5.31 shows that the data is scattered about the

mean and both the relationships (lines 1 and 2) lie within the

range of 90 per cent confidence limits (lines 4 and 5) of the

Bieniawski's (1976) relationship (line 3).

An interesting aspect is seen in the three relationships

(lines 1, 2 and 3) that at Q2s40, all the three correlations

predict nearly the same value of RMR which is equal to 78. For

other values of Q, the three relationships predict different

RMR values. It is being felt that a unique relationship raay not

be possible between Q and RMR because the result depends on the

mean value of classification parameters and their respective

coefficients of variation. Moreover, the difference in the

trend in relationships 1, 2 and Bieniawski's correlation may

also be due to difference in the geology of the Lesser

Hiraalayan rock masses and those on which Bieniawski's (1976)

correlation is based.

It follows that the correlation obtained between RMR and Q

system is reasonable within the acceptable limits. Such a

comparison deemed desirable because most quantitative rock raass

classfications were developed and calibrated for geological

conditions other than which exist in the Lesser Himalayan

rocks. Moreover, having applied at least two rock raass

classification systems on the six major hydroelectric projects,

the data-base of soft rock case histories has been strengthened
 208

for application to other projects which may be planned in

future in the Lesser Himalaya.

5.3.2 Estimation of Short Term Support Pressure in Tunnels

Barton's Q-system has been used for the determination of short

term support pressure in tunnels. The empirical relationships
have already been given by Eqs. 2.25 and 2.26 in chapter 2.
Although Q-system provides reliable results for the long terra
support system, the recommendations for temporary supports are
based on the meagre data and may therefore be considered only
as a rough approximation. Therefore, it was necessary to obtain
sufficient data and use the recommendation of Barton et al.

(1975) for temporary support so that a more reliable estimate

of support pressure is obtained in poor rock conditions.

The support pressures have been worked out on the basis of

Barton's Q values estimated at 97 test sections identified in

the Lesser Hiraalaya. In order to see the effect of water on the

support pressure, estimates of Q have been worked out on the

rock raass at saturation. Out of a total of 97 case studies, 55

cases pertain to the rock masses at natural moisture content

(nmc) and 42 pertain to the rock masses at saturation. For the

estimation of Q values, all the test sections were identified

in exploratory drifts which were approximately of size 2.1m X

2.1 m.

In order to permit correct designs, the rock support

measures roust be able to accomodate the expected deformations

without failure. It is on this account that deformability

characteristics of rock mass play an important role in the

 209

80
78

<
70 90V. Confidence limit

60

50-

a.

oe

<

S3tH
o
or

20H

10

Data at nmc

Data at saturation
—»
0-0001 0001 001 0-1 10 <t0 100
ROCK MASS QUALITY (ft

FIG.5.31 RELATIOSHIP BETWEEN ROCK MASS QUALITY (0.) AND ROCK MASS RATING (RMR}

<
 210

design of support measures for tunnels and other underground

structures. It has been found that a correlation exists between

the modulus of deformation (Ed) and the estimated short terra

support pressure tPi(roof>]. Correlation (Eq. 2.25) of Barton
et al. (1975) for short term supports has been used to develop
a relationship between raodulus of deformation (Ed) and the
support pressure on tunnels.

Figure 5.32 shows the plot of data from 55 test sections

at natural moisture content and Fig. 5.33 shows the plot of

data from 42 test sections at saturation. It will be seen from

Figs. 5,32 and 5.33 that inspite of high scatter in the data, a
definite correlation seems to exist between the modulus of
deformation (E,) and the estimated short terra support pressure

rp. , ,»]. Review of Figs. 5.32 and 5.33 further show that
Ltl(roof)
there is no decrease in the support pressure beyond the raodulus

value of 10 GPa and the short terra support pressures seem to be

stabilised. The two relationships also show that between
modulus range of 10 GPa to 3GPa, there is not much increase in
the support pressure values, whereas between 3GPa to 0.25 GPa
there is fast rise in the support pressure. It may be inferred

from the results that for rock masses having modulus of

deformation of 10 GPa or more, the support pressures seem to be

about 0.2 kg/cm for rock mass at natural moisture content and
p
about 0.3 kg/cm for rock raass at saturation. The fact however

needs to be verified through instrumented tunnels.

 21 1

5.3.2.1 Influence of Overburden on the Short Term Support

Pressure in Tunnels

The raeasurements of wall-displacements in the exploratory

drifts have shown that apparently no squeezing or swelling

conditions existed in the drifts. Out of 97 cases explored,

closures measured in the drifts were found to be within one per

cent of the drift size. According to Singh et al. (1992), a

rock mass may undergo squeezing when the depth of tunnel

1/3
section exceeds 350 Q . The depth of the tunnel sections

proposed at the six project sites varies between 250 ra to 530

ra. Application of Barton's Q - system to the Lesser Himalayan

rock masses shows that the rock mass quality (Q) in the region

generally varied from 0.1 to 19. From the foregoing details, it

is clear that the estimated values of short term support

pressure are free from the effects of squeezing ground, usually

the result of high overburden. The rock mass is thus supposed

to be under elastic conditions only.

5.3.3 Estimation of Support Pressure Through Various Approaches

Support pressures for the present study have been estimated

from the empirical approach of Terzaghi (1946), Deere et al.

(1970), Bieniawski (1979), and Barton et al. (1974). The

predicted support pressures from the above theories have been

compared with observed pressures. The observations were taken

at two different locations i.e. chainage 9.0 m and 16.0 ro of

the 5.0 ro diaraeter drainage tunnel of the Lakhwar power house

cavity.
 r

@ G) ©
®i? © IT H~
7-0 7-5 8-0 8-5 9-0 10-0
£♦-0 5-0 6-0
0 0-25 0-75 10 20 2-5 3-0 RMR:=60
[RMRi=20l RMRi=<*0 MODULUS OF DEFORMATION (Ed)-GPa ——

FIG.5-32 RELATIONSHIP BETWEEN MODULUS OF DEFORMATION (Erf) AND SHORT TERM SUPPORT PRESSURE (W)
FOR TUNNELS ,
[ROCK MASSAT NATURAL MOISTURE CONTENT]

70 7-5 80 9-0 10-0

0 0-25 0-75 10 20 2-5 3-0 <*'0 5-0 6-0
MODULUS OF DEFORMATION (Ed)- GPa •»

FIG. 5-33 RELATIONSHIP BETWEEN MODULUS OF DEFORMATION (E<j) AND SHORT TERM SUPPORT PRESSURE (P,)
FOR TUNNELS
(SATURATED ROCK MASS)
 213

The unit weight (r) of the rock is 2.76 gra/cc. The support

pressures are now estimated.

5.3.3.1 Support Pressure Through Classification System of

Terzaghi (1946)

In accordance with Terzaghi s rock load classification criteria

the rock mass condition can be described as "moderately blocky

and seamy" with no side pressure (category 4 of Terzaghi's

table) for which height of rock load is given as,

H = 0.25B to 0.35 (B +• Ht)

Since the tunnel under consideration is circular having

the diameter 5 m, the height H is given by,

H = 1.25 m to 3.5 n
P

Thus, the support pressures in the tunnel roof work out to

be:

pi(roof) =0'36 to 0-97 kS/cm2

p
An average value of 0.66 kg/era can be taken for the

purpose of comparison.

5.3.3.2 Support Pressure Through Classification System of Deere

et al. (1970)

The RQD of the rock mass in the tunnel has been estimated to be

varying from 50 - 60 per cent (fair rock mass). The height of

rock load by Deere's approach is given as :

H = (0.20 - 0.60) (B + H.)

IT KJ

Since the diameter of the tunnel is 5 m,

 214

H = 2 m to 6 m
P

Thus, the support pressures in the tunnel roof work out to X

be

Pj,
*i(roof)
„, = 0.55 to 1.66 kg/cm2

An average value of 1.10 kg/cm2 can be taken for the

purpose of comparison.

5.3.3.3 Support Pressure Through RMR System

The RMR value of the rock mass in the drainage tunnel has been ^
estimated as 59. The support pressure can be determined from
the RMR system as proposed by Unal (1983) :

u - 100 -RMR n
HP " 100~~ ti

Thus, the support pressure in the tunnel works out to be:

2
p - =0.57 kg/cm
*roof

5.3.3.4 Support Pressure Through Q-system

Like other empirical approaches, Q-system also provides only

final support measures. However, Barton et al. (1974) provided
approximate suggestions by increasing the value of Q and
temporarily decreasing the significance of the project, as

follows:

Q, = 5Q
l

The Q-value estimated in the tunnel section is of the

order, Q = 5 to 9. Taking a root mean square value of Q equal
to 6.7, joint roughness number (Jr) equal to 2 for smooth,
 215

undulating joints, the short term support pressure on the basis

of Q-value works out to be :

2 n-l/3 . . 2
pi(roof)=-^- Q kg/cm

•I—(33.5) 1/3 kg/cm2

=0.31 kg/cm2

Thus, short term support pressure on the basis of Q-system

2
is estimated as 0.31 kg/cm .

5.3.4 Comparison of Support Pressures by Different Approaches

The predicted support pressures (Fig. 5.32) have been compared

with the pressures predicted from different empirical

approaches.

Table 5.1 gives a comparison of results from the various

empirical approaches. A comparison of different results shows

that:

(i) The empirical approach of Terzaghi (1946) provides support

2 2
pressure in the range from 0.35 kg/cm to 0.97 kg/cm ;

whereas the support pressure by Deere et al. (1970) are

2 2
between 0.55 kg/cm to 1.66 kg/cm . It is seen that the

pressures predicted by Deere et al. (1970) are about 1.7

times of the pressures predicted by Terzaghi (1946).

(ii) Q-systern of Barton et al. (1974, 1975) provides short term

support pressure which is nearly equal to the observed

 216

value. It is worth noting that the measured pressure

matches with the predicted values on the basis of Q as
well as Ej.

Thus proposed relationship between modulus of deformation

and the support pressure provides a convenient method for
estimating short term support pressures. The relationship can
be used for preliminary designs to be carried out during
feasibility and planning stage of a project. Further refinement
of the correlation is possible when the data from more
instrumented tunnels are available.

5.3.5 Support Procure and Moisture Content

Laboratory studies have shown that strength and deformation

modulus of rocks tend to decrease with increase in moisture
content. The effect of water is more significant In rocks which
have a layered structure and which are rich in clay content. In
these rocks, the mineralogy of the rock material seems to
directly affect the support pressure. Depending upon the
mineralogical composition, some rocks lose while others may
gain weight even in short periods of saturation (Mehrotra et
al., 1991). With the action of water, soft rocks try to
disintegrate which involves further softening, dispersion and
movement of colloidal matter within the saturated rock raass.
Thus, under increased moisture, the soft rock mass tends to
become disintegrated. Moreover flow of water through a few open
joints only is very non-uniform.
The situation as explained above may be responsible for
the development of non-uniform pressure distribution on the
 217

TABLE 5.1

COIPABISOI OF SDPPOiT PBESSOBES BT DIFFEBIIT HITIODS FOI TIE DEAIIAGI TIIIEL

AT LAIIHAB DAI PBOJECT

Short ten support pressure by Predicted support Obserred

Bock Hass Details different classification srsteas pressure kg/ci' support
(kg/ca«) on the basis of pressure
•odnlos of
Terzaghi Deere BMB Q- detonation (kg/ci')
et al. Systei Sfstei
(lit) (Pit) (Pit) (Pit) (Pit) (pUtitrrtO

Trap rock - a composite of 0.35 to 0.55 to 0.57 0.31 0.26 0.28

dolerite to homblend - 0.97 1.66 (after 85
rhfolite, coarse grained days)
and jointed, density 2.76 (Av.0.66) (Arl.10
gi/cc, BQD:50 - 60X,Q:6.7,
iU-i9, diaieter of tunnel 0.287
5 a, natural loisture (after 180
content 2.6X, Ed-5.05 GPa days)
 218

tunnel supports. Loosening of rock mass may also start at

adjoining places. Too much loosening may be responsible for

unequal distribution of pressure before installation of support

in the tunnel. Although loosening before installation of

support may not be avoided completely, yet it should be

accounted for, because it may give non-uniforra distribution of

pressure on the tunnel support apart from diminishing self

supporting capacity of the rock mass. A comparison of the

curves shown in Figs. 5.32 and 5.33 clearly shows two aspects

which are as follows :

(i) Overall increase of support pressure under the

effect of saturation; and

(ii) Comparatively more data scatter when the rock mass

is saturated.

Increase in support pressure is due to the fact that under

saturation the modulus of deformation of the rock mass has

decreased substantially. Further, a strong scatter in the data

points indicates the non-uniform pressure distribution on the

supports. Under such situation, the support raay either fall

without giving any warning or the tunnel lining raay crack at

such locations.

It is understood that under saturated conditions, the

empirical approach of Terzaghi would lead to safer designs. The

Q-system of Barton et al. (1974, 1975) does not seem to be

applicable to rock mass under saturation as it is difficult to

consider all effects of water in an empirical approach at least

 219

in poor rocks.

5.3.6 Size Effect

According to Terzaghi (1946), Wickham et al. (1972, 1974),

Deere et al. (1970) and Unal (1983) the support pressure
increases directly with tunnel size. However, a number of case
histories show that the support pressure is independent of the
tunnel size. The fact has been recently verified and confirmed
by Singh et al. (1992) who cited detailed studies at the
underground power house complex of Lakhwar dam project in India
and found that the observed roof support pressures were nearly
the same for the 6 ra wide approach adit, 14 m wide expansion
surge tank and 21 m wide power house cavern, all excavated
through tightly jointed traps. According to Barton et al.
(1974) the support pressure is independent of tunnel size.

However, it is worth noting that the rock mass quality (Q)

estimated from a larger tunnel would be smaller than that
obtained from drifts in a similar rock mass. It would therefore
be unsafe to obtain Q from smaller drifts and use it to
estimate support requirements for large excavations.
Relationship (Fig. 5.32) between modulus of deformation and
support pressure is based on the data obtained from the
exploratory drifts. In order that the results are applicable to
tunnels of larger diameter, it is considered necessary to use a
correction factor for the support pressure values obtained in
the drifts. The correction factors for the support pressure
should be based on the reduced values of Q which raay be
encountered in the larger diameter tunnels. Thus the correction
 220

factors for Q should be found out.

5.4 Shear Strength Parameters of the Rock Mass

Rock is a heterogeneous, anisotropic and discontinuous raass.

When civil engineering structures like daras are founded on

rock, they transmit normal and shear stresses on the

discontinuities of the rock mass. Failure raay be initiated by

sliding along a joint plane near or along the foundation or

along the abutments of dam. For a realistic assessment of the

stability of structure, estimation of the shear resistance of

rock mass along any desired plane of potential shear or along

the weakest discontinuity becomes essential. This involves an

assessment of shear strength parameters i.e., cohesion (c) and

angle of internal friction (0) of the rock mass. Estimates of

these pararaeters are usually not based on extensive field

tests. However, in this research study, attempt has been made

to develop correlations between the RMR and the shear strength

parameters (c & 0) on the basis of large number of tests. The

relationship may be found useful for the purpose of preliminary

designs of a project.

Large size (700 mmX 700 mraX 300 ram) block shear tests were

conducted to estimate the shear strength parameters of the rock

masses. The tests were performed in exploratory drifts under

anticipated stress range which varied from 10-15 kg/era . All

the tests were performed under consolidated drained condition.

Twenty two tests were carried out on rock raass at natural

moisture content and another twenty one on the saturated rock

mass.
 22 1

By plotting the normal stresses and the corresponding peak

shear stresses on the Coulomb plot, cohesion (c) and angle of

shearing resistance (0) in terms of effective stress have been

estimated.

5.4.1 RMR and Shear Strength Parameters (c & 0)

Accurate representation of the real field conditions is

difficult because cohesion (c) and angle of internal friction

(0) are not constant throughout the rock raass but vary from

place to place and from project to project. Yet it is believed

that an engineering rock mass classification approach may be

used to predict shear strength parameters (c & 0). In the

present study RMR system has been used to predict the shear

strength pararaeters (c & 0) of the rock masses.

Figures 5.34(a) and 5.34(b) show variation of cohesion (c)

with the angle of internal friction (0) of the rock mass at

natural moisture, and at saturated conditions respectively.

Figure 5.35 shows the correlation between RMR and the cohesion

(c), angle of internal friction (0) at the natural moisture as

well as at the saturated rock mass conditions. Figure 5.36

shows the effect of saturation on the shear strength pararaeters

(c & 0) of the rock raass. Reduction factors for saturation

(Fig. 5.36) are based on the data for the rocks which have been

classified as poor to fair quality rock masses (RMR= 21-60).

5.4.2 Discussion and Interpretation

For the rock masses at natural moisture content, wide scatter

is seen in the results especially in the lower portion of the

Fig. 5.34(a).It is observed that angle of internal friction (0)

 *•

60 -

57

50- ©
m
Ol
u ®?
»

^ 30

I 20

c
<
10-

100 200 300 aoo ^50 500

Cohesion (c)- KPa »-

FIG.5-3t»(a) COHESION (c) AND ANGLE OF INTERNAL FRICTION 10) OF THE ROCK MASS AT nmc (0-15 V.-2-5 %)
 223

increases with increase in cohesion (c) in rocks unlike in

soils. It is worth noting that the increase in angle of
internal friction (0) is at a faster rate untill a cohesion
value of about 300 kPa. Within the range of data obtained, the
maximum value of angle of internal friction (0) [=57 ] is
observed at the cohesion (c) value of 450 kPa.

In case of saturated rock mass, the trend in Fig. 5.34(b)

is the same as for rock mass at natural moisture content.
However, it is observed that except for two results> data is
comparatively more consistent in case of saturated rock mass.
The scatter in the results reflects the variation in mechanical
characteristics along the joints of the various rocks tested
Including the type of fillings in the joints, the index
properties of the fillings, insitu stresses and the joint
matrix of the rock raass tested in each case. On account of
above limitations, a realistic representation of the field
condition seemed difficult, yet it is believed that the average
value of shear strength parameters as predicted are
representative of the rock mass strength for preliminary
designs.

The relevant c-0 curve should be superimposed on c-0 curve

obtained from back analysis of rock slopes. The point of

intersection of two curves will give more realistic values of

shear strength pararaeters.

 224

It can be seen from Fig. 5.35 that even with a

considerable scatter, a definite trend is marked in the

results. Figure 5.35 suggests that irrespective of the moisture

state, both,the cohesion (c) and the angle of internal friction

(0) have an increasing tendency with the increase in RMR. It is

observed that at natural moisture content the angle of internal

friction (0) lies between 42 and 54 for poor rocks and from

54 to 57 for fair quality rock mass. The maximum value of the

angle of internal friction (0) is 57° at RMR 60. For the

saturated rock mass the values are in the range, 0 = 31
53t>
o o o

46 for poor rock masses and 0 = 46 - 53 for fair rock

S 3.T*

masses. The average cohesion for saturated rock mass is found

to vary from 75 kPa to 270 kPa for RMR 20-60 and is in

agreement with values suggested by Bieniawski (1979).

Figure 5.36 shows that the effect of saturation is more

marked on cohesion (c) than on friction angle (0). Figure 5.36

further shows that the reduction factors for cohesion vary from

0.30 to 0.48, and 0.48 to 0.60 for 'poor' and 'fair' quality

rock masses respectively. In case of angle of internal friction

(0), the reduction factors vary from 0.65 to 0.80 for poor rock

masses, and from 0.80 to 0.90 for fair rock masses. It is

observed that the angle of internal friction (0) decreases

with decrease in RMR and becomes asymptotic to 32 in

o

case of naturally moist rock masses and 20 for saturated

rock masses (Fig. 5.35). The curves may be used to obtain

the values of cohesion (c) and angle of internal

friction (0) of the poor and fair quality rock masses.

In case of rock slopes, factor of safety may be

 60 -

50-
Ol
a«
i_

01
T3

~ 40-

on

- 30-
a
c

15 20

C
<

10-

200 300 U00 425 500

75 100

Cohesion (c )— kPa —

FIG.5-3*»(b) COHESION(c) AND ANGLE OF INTERNAL FRICTION (rfj) OF THE SATURATED ROCK MASS
 226

60

g 50
P
U

£j 30
z
LL
'Bicfu'aswki' (1979)

* 10
20 30 W 50 -tT -*r —»
eo

600
| 500 ©
1 MX)
u
z 300 0]
flZ.
__ 2. — ^ vBfenia
niaswkK 1979)
o
, ®
10 20 30 40 50 60 70 80
ROCK MASS RATING (RMR)
[ROCK MASSAT nmc]

I 70

± 60

rj w

<*&

a:

30
Bieniowski(1979)

uj 20

<
10
20 30 <*0 50 60 70 60

rBien'tawskl(1979)

60 70 80
ROCK MASS RATING (RMR)
[SATURATED ROCK MASS)

FIG.5-35 RELATIONSHIP BETWEEN ROCK MASS RATING (RMR) AND SHEAR STRENGTH
PARAMETERS COHESION (C) AND ANGLE OF INTERNAL FRICTION (0)
 227

1-0-

0-9-

<(> saturated

0-8-

o
C 0-7-1

- 0-6H

c
o c saturated
•» 0-5H

vi
2
0-k-

§ 0^
c
o

£ 0-2 ^
c
o
4-»

s
=3

<5? o-i-

10 20 30 K0 50 60 70 80

Rock mass rating (RMR) »•

FIG.5-36 EFFECT OF SATURATION ON THE SHEAR STRENGTH PARAMETERS

(c and 0)OF THE ROCK MASS
 228

calculated after obtaining the values of c and 0, and using the

standard design procedures.

The following inferences may be drawn frora the foregoing

discussion :

1. RMR system may be used to estimate the shear strength

parameters (c & 0) of the rock mass. It is observed that

cohesion (c) and angle of internal friction (0) generally

increase with the increase in RMR.

2. The effect of saturation on the shear strength parameters

has been found to be significant. If in-situ tests are

conducted in dry condition, the proposed reduction

factors may be applied to predict the design parameters

for saturated condition which normally prevails after

commissioning of the hydroelectric projects. For poor rock

masses, a maximum reduction of 70 per cent has been

observed in cohesion (c) while the reduction in angle of

internal friction (0) is of the order of 35 per cent.

5.5 Failure Envelopes for the Rock Masses

Usually failure of a rock mass occurs partially along the

joints and partially in solid rock, but in extreme cases it may

occur entirely along a joint or entirely in solid rock.

Therefore, the failure of a rock mass lies within the area

bounded by the failure envelope for a single joint and the

failure envelope for solid rock. The mode of failure depends on

the quality and the type of rock mass under investigation.

In case of poor rock masses, the magnitude of normal

 229

stress (o) influences the shear strength significantly. A

straight line envelope is therefore not a proper fit for such .

data and is likely lead to over-estimation of angle of internal

friction (0) at higher normal stresses.

The failure envelopes for the rock masses show generally a

non-linear trend. Straight line criterion may be valid only

when loads are small (c << q ) which is generally not the case

in civil engineering hydroelectric projects where the intensity

of stresses is comparatively high. The failure envelopes based £

on generalised empirical power law may be expressed as follows

(Hoek and Brown, 1980):

t = A (cr + T)B (5.1)

= 0 if & < 0 (tensile)

It is assumed that shear strength is zero when normal

stress is tensile in nature as suggested by Hoek et al. (1992).

In other words tensile strength of rock raass js 0 as generally

assumed in "No Tension Anslysis" by FEM. Thus, parameter T is

not tensile strength but a rock parameter. The cohesion >

intercept (c) is equal to AT .

For known values of power function (B), constants A and T

have been worked out from a series of block shear test data.

Consequently, empirical equations for the rock masses both at

natural moisture and at saturation, have been calculated for

defining failure envelopes. The values of the power function B ^

have been assumed to be the same as in the equations proposed

by Hoek and Brown (1980) for heavily jointed rock masses.

 230

5.5.1 Trend of the Failure Envelopes

Figures 5.37 to 5.44 show the plots of failure envelopes for

four different category of rock masses namely (i) limestone

(ii) slate, xenolith, phyllite (iii) metabasic, trap and

(iv)sandstone, quartzite. The constants A and T have been

estimated using the results obtained from the in-situ block

shear tests carried out on the Lesser Himalayan rocks.

5.5.1.1 Mohr's Envelopesfor 'Poor' Rock Masses

Figure 5.37 shows the failure envelopes for naturally moist and

saturated limestones. The rock has been categorised as "poor

rock mass" (RMR = 29-37). Figure 5.37 also shows tan0 plotted

as a function of normal stress (o>) , 0 being the angle of

internal friction of the rock mass at the failure plane. It is

noted that the failure envelopes are stress dependent with

2
cohesion intercepts of 2.16 kg/cm for the naturally moist rock

mass and 1.25 kg/cm for the rock mass under saturation. The
2
parameter T of the rock mass is estimated as 0.80 kg/cm for

the rock mass at the natural moisture content. It is observed

that there is no significant change in the values of tan 0

2
beyond a normal stress {&) value of 20 kg/cm' at which tan 0

equals 0.55 (0=29°) for tha naturally moist and 0.33 (0=18°)
for the saturated rock mass.

Figure 5.38 shows the failure envelopes for naturally

moist and saturated slates, xenoliths, phyllites. The rocks

have been categorised as "poor rock mass" (RMR=23-37). Figure

5.38 also shows tan 0 plotted as a function of normal stress

(<=") . It is seen that the failure envelopes are non-linear with

 231

cohesion intercepts of 2.20 kg/cm2 and 1.38 kg/cm" for

naturally moist and saturated rock masses respectively. The y
parameter T for the rock mass at natural moisture content is
0.75 kg/cm2. It is observed that there Is no significant change
in the values of tan 0 beyond a normal stress (c) value of 20
kg/cm2 at which tan 0 equals 0.61 (0=31°) and 0.40 (0=22°) for
the naturally moist and saturated rock masses respectively.
Compared with the results of limestone, it is seen that the
cohesion intercept is comparatively higher in the case of
2 X
saturated slates, xenoliths and phyllites (1.45 kg/era for
2
slates, xenoliths and phyllites, as against 1.30 kg/cm in case
of limestones). The values of tan0 are also comparatively
higher in case of slates, xenoliths and phyllites.
Figure 5.39 shows the failure envelopes for naturally
moist and saturated sandstone and quartzite. The rocks have
been classified as "poor rock mass" (RMR=22 - 36). The failure
envelopes are non-linear and show the cohesion intercepts of
2.20 kg/cm2 and 1.50 kg/cm2 for the naturally moist and
saturated rock masses respectively. The parameter T for the
2
rock mass at natural moisture content is 0.70 kg/cm . It is
observed that there is no significant change in the values of
2
tan 0 beyond a normal stress values of 20 kg/cm at which tan0
o

equals 0.70 (0=35°) for the naturally moist, and 0.50 (0=26 )
for the saturated rock masses.

Figure 5.40 shows the failure envelopes for naturally

moist and saturated metabasic and trap rocks. The rocks have
been classified as "poor rock mass" (RMR=24-40). It is observed
that the failure envelopes are stress dependent and show the
 V

IV)

6 8 10 12
Normal stress (<r ) , kg/cm2 —*-

FIG.5-37 FAILURE ENVELOPES FOR JOINTED LIMESTONE AT NATURAL MOISTURE CONTENT (0-3%~ 2-8 % ) AND
SATURATED CONDITIONS [RMR-29-37, a-0-<»-1-5, qc(nmc)-300-500 kg/cm2, qc(sot)- 200-<fOO kg/cm2]
^ r*

to
Ul
Ul

6 8 10 12
Normal stress (a-), kg/cm2 —

FIG.5-38 FAILURE ENVELOPES FOR JOINTED SUTE, XENOLITH AND PHYLLITE AT NATURAL MOISTURE CONTENT (0-75%-l-85V.)
AND SATURATED CONDITIONS [RMR =23-37, a= 0-3-1-0, qc(nmc)=350-850 kg/cm2, qc(sat)«200-600 kg/cm2 ]
 16
^nmc«2-8C(T+0.7> 0-672

Ik

12.

CM
1
E
u I
*>-
o>
X
13-
"\

k
*^

c
+-t

0» 1
c
8-
*-.

</%

L. i

6-

-•—

6 8 10 12 Ik 16 18 20
Normal strtsff (o~),kg/cm2

FIG.5-39 FAILURE ENVELOPES FOR JOINTED SANDSTONE AND Q.UARTZITE AT NATURAL MOISTURE CONTENT (0-<+0 V.-I-5V.)
AND SATURATED CONDITIONS CRMR=22-36, a-0-1-2-5, qc(nmc)« 000-1000 kg/cm2, qc(satr 350-700 kg/cm2!
t f

to

6 8 10 12
Normal stress ( cr) , kg/cm2

FIG.5-if0 FAILURE ENVELOPES FOR JOINTED TRAP AND METABASIC ROCK AT NATURAL MOISTURE CONTENT (0-28V.-0-6 V.)
AND SATURATED CONDITIONS f/RMR= 2<t-<+0, a-0-3-2-5, qc(nmc] =750-1500 kgfcm2, qc(sat)= 600-1200 kg/cm J
 236

2 2
cohesion intercepts of 2.24 kg/cm and 1.59 kg/cm for the

naturally moist and saturated rock masses respectively. The

parameter T for the rock mass at natural moisture content is

2
0.65 kg/cm . It is seen that beyond a normal stress (a) value
2
of 20 kg/cm there is no significant change in the value of
p
tan 0. At normal stress (a) of 20 kg/cm , tan 0 equals 0.76

(0=37 ) for the rock mass at natural moisture and 0.57 (0=30°)

for the rock mass under saturation.

It is evident from the Figs. 5.37, 5.38, 5.39 and 5.40

that the results of in-situ block shear tests performed on both

naturally moist and saturated rocks, namely (i) limestone (ii)

slate, xenolith, phyllite (iii) sandstone, quartzite and (iv)

metabasic, trap can be suitably used to represent the failure

envelopes. The following inferences may be drawn :

1. It is possible to estimate the approximate shear

strengths from the data obtained from the in-situ

block shear tests.

2. Shear strength of the rock mass is stress dependent.

The cohesion intercept of the rock mass varies from

2 2
1.25 kg/cm to 1.59 kg/cm for saturated and 2.16
2 2
kg/cm. to 2.24 kg/cm for naturally moist rock

masses.

3. Beyond the normal stress (o) value of 20 kg/cm ,

there is no significant change in the values of

tan 0. It is observed that the angle of interna.1

friction (0) of the rock mass is asymptotic at 20°

 237

4. The effect of saturation on the shear strength is

found to be significant. When saturated the reduction

in the shear strength is about 30 per cent at the

2
normal stress (c) of 20 kg/crn .

5.5.1.2 Mohr's Envelopes for 'Fair' Rock Masses

Figure 5.41 shows the failure envelopes for naturally moist and
saturated slates, xenoliths and phyllites. The rocks have been

categorised as "fair rock mass" (RMR = 43-56). Figure 5.41 also

shows tan0 plotted as a function of normal stress (c). It is
x

seen that the failure envelopes are non-linear with cohesion

intercepts of 3.02 kg/cm for the rock mass at natural moisture

and 2.29 kg/cm for the rock mass under saturation. The

parameter T of the rock mass at natural moisture content is

1.15 kg/cm2. It is observed that there is no significant change
in the values of tan 0 beyond a normal stress (c) value of 20

kg/cm2 at which tan 0 equals 0.64 (0=35°) for the naturally

moist, and 0.54 (0=28°) for the saturated rock masses.
Figure 5.42 shows the failure envelopes for naturally

moist and saturated sandstones and quartaite3. The rocks have

been classified as "fair rock mass" (RMR=41-58). The failure

envelopes are stress dependent and show the cohesion intercepts

of 3.04 kg/cm for the rock mass at natural moisture and 2.33

kg/cm for the. rock mass under saturation. The parameter T of

2
the rock mass at natural moisture is estimated as 1.10 kg/cm .

It is observed that beyond a normal stress (&) value of 20

kg/crn there is no significant change in the values of ban 0.

2 o
At normal stress (cr) of 20 kg/crn , tan 0 equals 0.76 (0=37 )

for the naturally moist, and 0.60 (0=31°) for the saturated
 k V-

00

Normal stress (cr ), kg/cm

FIG.5-M FAILURE ENVELOPES FOR JOINTED SLATE, XENOLITH AND PHYLLITE AT NATURAL MOISTURE CONTENT (0-25%-0-9^V.)
ANO SATURATED CONDITIONS [RMR-^3-56, a-2-0-3-5, qc(nmc)«350-850 kg/cm2, Qc(sat)= 200-600 kg/ cm2}
k y

to
Ul

Normal stress (cr), kg/cm

FIG. 5-^2 FAILURE ENVELOPES FOR JOINTED SANDSTONE AND aUARTZITE AT NATURAL MOISTURE CONTENT (t>l»VMW
AND SATURATED CONDITIONS [RMR=<+1-58; a-1-9-5-1., qc(nmc) = WO-1000 kg/cm2, qc(sQt ,=350-700 kg/cm2]
 240

rock masses.

Figure 5.43 shows the failure envelopes for naturally

moist and saturated trap and metabasic rocks. The rocks have

been classified as "fair rock mass" (RMR=42-59). It is seen

that the failure envelopes are stress dependent and show the
2
cohesion intercepts of 3.05 kg/cm for the rock mass at normal
2
moisture and 2.36 kg/cm for the rock mass under saturation.

The parameter T for the rock mass at natural moisture content

is estimated as 1.00 kg/cm . It is observed that there is no

significant change in the values of tan 0 beyond a normal

2
stress (ey) value of 20 kg/cm at which tan 0 equals 0.82

(0=39°) for naturally moist, and 0.66 (0=33°) for the saturated
rock masses.

5.5.1.3 Mohr's Envelopes for 'Good' Rock Masses

Figure 5.44 shows the failure envelopes for jointed trap

and metabasic rocks. The rocks have been classified as "good
rock mass" (RMR=61-72). It is seen that the failure envelopes
are stress dependent. The cohesion intercepts for naturally
moist, and saturated rock masses have been estimated as 7.6
2 2
kg/cm and 3.9 kg/cm respectively. The parameter T have been
estimated as 2.7 kg/cm for the naturally moist and 1.2 kg/cm2
for the saturated rock masses. It is observed that there is no
significant change in the values of tan 0 beyond a normal
stress (a) value of 22 kg/cm at which tan 0 equals 1.00
(0=45°) for naturally moist, and 0.88 (0=42°) for the saturated
rock masses.

5.5.2 Recommended Criteria of Shear Strength of Rock Masses

Table 5.2 summarises the shear strength equations for various
 241

rock mass ratings, degree of saturation and rock types. The

recommended criteria are based on the test results from Figs.
4
5,37 - 5.44, experience and judgment of the author. It has been
realised that for highly jointed rock masses shear strength
will not be governed by the strength of the rock material as
suggested by Hoek and Brown (1980). Moreover their parameters
ara based on statistically insignificant data. Further, the
cohesion intercept will not be negligible as proposed by Hoek
et al. (1992). The results show that saturation does affect
shear strength of rock mass. Table 5.2 does not suffer from

these limitations.

The results obtained from the study of failure envelopes

of fair quality rock masses show the following trends :

1. It is possible to estimate the aproxiroate shear strength
from the data obtained from the in-situ block shear tests.

2. Shear strength of the rock mass is stress dependent. At

natural moisture content the cohesion intercept of the
2 2
rock mass varies from 3.02 kg/cm to 3.05 kg/cm . Under
2
saturation the cohesion intercept varies from 2.29 kg/cm

to 2.36 kg/cm .

2
3. Beyond a normal stress (c) value of 20 kg/cm , there is no
significant change in the values of tan 0.It is observed
that the angle of internal friction (0) of the rock mass

is asymptotic at 27 .

4. The effect of saturation on the shear strength is found to

be significant. When saturated, the reduction in the shear
strength is about 25 per cent at the normal stress (a) of
 •J^
L.
fcw- >• V

0»

c
0)

i/>

6 8 10 12
Normal stress (cr ), kg/cm2 »»
FIG.5-43 FAILURE ENVELOPES FOR JOINTED TRAP AND METABASIC ROCKS AT NATURAL MOISTURE C0NTENT(Q-28%-0-60 7.)
AND SATURATED CONDrTIONS [RMR- J»2-59, 9=1-5-4-0, qc(nmc)=750~1500 kg/cm2, qc(sat)=600-1200 kg/cm2]
 <K

Normal stress (cr), kg/cm *

16 18 20 22
6 8 10 12 ]U

0-030

0-044
0005
Normal stress ( a~n )

FIG.5-44
FAILURE ENVELOPES FOR JOINTED TRAP AND METABASIC ROCK AT NATURAL MOISTURE <^NT
(0-40 V.-0-80V.) AND SATURATED CONDITIONS [RMR =61-72, O-8-0-14-5, qc(nmc)-750 -1500kg/crrf,
Pccsat)=600'1200 k9/cm2]
 ^
-r- 14-

TABLE 5.2

Recommended Mohr Envelopes for Jointed Rock Masses

n I
T , „in kg/cm2,- t= 0 if a < 0)
(• qc/ O
n —
qc
S = degree of saturation [average value of degree of saturation is shown by %v ]
1, for completely saturated rock mass

Slate,Xenolith,Phyllite Sandstone, Quartzite Trap,Metabasic

S. Rock Limestone

NO- Quality
1. 'Good'Rock Mass Vnmc)=°-38(<V°-°05)U-^ ^n(nmc)=0-42(%+0.004)0-683 Tn(nmc)=°-44(%+0-003)U-6yb tn(nmc)
, ,=0.50(o n+0.003)0-698
[Sav=0.30]
RMR = 61-80

Q = 10-40
*n(satr^<V°-004>°-669 *n(satr°-38<On+0-003)°-683 • Tn(sat)=O-43(V0-°°2>°-695 ^n(sat)=0-49(V°-0°2)0-698
[S=l] [S=l] Fig.5.44
[S=l] [S=l]

Tnmc=2.85(a+1.10)0-688 tnmc=3.05(o+l.00)0-691
2. 'Fair' Rock Mass Tnmc=2-6°(0+1-25>0-662 "nmc^-75'^1-15'0-675
RMR = 41-60 [Sav=0.25] [Sav=0.15] [Sav=0.35]
Q = 2-10
^,.=1.95(0+1.20)0-662 ^,.=2.15(0+1.10) °-675 tsat=2.2S(a+1.05)0-688 i^^^sto+o.gs)0-691
[S=l] Fig. 5.41 [S=l] Fig.5.42 [S=l] Fig.5.43
[S=l]

^nmc=2.50(o+0.80)0-646 w=2.65(o+0.75)0-655 Tnmc=2.80(o+0.70)0-672 W=3.00(a+0.65)0-675

3. 'Poor' Rock Mass
[Sav=0.40] [Sav=0-25] [Sav=0.15]
RMR = 21-40 [Sav=0.20]
g = 0.5-2

Tsat=l.50(o+0.75)0-646 xsac
^=2.00(o+0.65)0-672 ^at=2.25(a+0.60)0-675
Tsat=1-75(o+0-70,°'655
[S=l] Fig.5.38 [S=l] Fig.5.39 [S=l] Fig. 5.40
[S=l] Fig.5.37

t nmc=2.90(o+0.50)0-548
4. 'Very Poor1 -nmc=2.25(a+0.65)0-534 xnm==2-45(o+0-60)°-539 ^nmc^-65'^0-55'0'546
Rock Mass
RMR < 21 Tg-^O^CO)0-539 tnmc=1.05(o)0-546 ,nmc
=1.25(0)0-548
g < 0.5
[S=l] [S=l] [S=l]
[S=l]
 245

20 kg/cm2.

The equations which have been developed on the Lesser

Himalayan data may be used only for preliminary designs of

underground excavation or rock slope. There is further scope
for refinement if the present data are supplemented with
triaxial test data. For RMR > 60, shear strength will be
governed by strength of rock material, Eq. 2.17 of Hoek and
Brown (1980) may be used.

5.6 Allowable Bearing Pressure on Rocks

Rock is usually recognised as the best foundation material.

Major civil engineering structures such as dams, bridges, tall
buildings, towers etc. are generally founded on a rock mass
which nearly always contains discontinuities and many other
unfavourable rock conditions. The designer has to foresee the
dangers associated with the heterogeneity and unfavourable rock
conditions since overstressing a foundation can give rise to
settlements leading to the failure of structure. For important
structures such as mentioned above, the settlement of
foundations becomes a governing parameter in design since
excessive total and differential settlements endanger the
safety or utility of structure. Thus the ultimate bearing
capacity at which the foundation rock will fail has not much
engineering significance. Instead, the differential settlement
becomes the controlling factor for the safety of structure.
Thus, there is need for rationalising the approach to estimate
the allowable bearing pressure on rocks so that settlement does
not exceed a permissible limit.
 246

Thirty nine plate load tests were conducted in different

rock types which have been classified from poor to fair rock
4-
masses (RMR = 21-60). The tests were performed at the natural

moisture content which varied from 0.6% to 6.5%. The

permissible settlement for calculation of safe bearing pressure

from the plate load test has been taken as 12 mm even for large

loaded areas. For 12 mm settlement of footing, the settlement

of plate has been obtained by using the following relationship

for jointed rock masses as given in Indian Standard Code of

Practice (IS:12070-1987); X

B (Bf + 30)
(5.2)
B, (Bp + 30)

Where,
S = settlement of plate (mm)

Sf = settlement of footing (mm) [^12 mm]

B = width of plate (=60 cm)

B„ = width of footing (taken as 600 cm)

>

Equation 5.2 is based on the assumption that behaviour of

jointed rock ma3s will tend to be similar to that of the sand.

With this criterion, the settlement of the plate has been

worked out as 5.88 mm. From the pressure settlement curve, the

safe bearing pressures have been estimated for the plate

settlement of 5.88 mm.

Average RMR values have been determined at the actual

locations at which the plate load tests were carried out.

Figure 5.45 shows the relationship developed between RMR and

 247

the allowable bearing pressure on rocks. The relationship is

based on the assumption that the rock mass is relatively

uniform beneath the foundation. In some cases, layered rocks of

inferior quality were encountered below the stratum

investigated. In sucli cases, the RMR value of the inferior rock

has been used in the correlation proposed. Table 5.3 gives net

bearing pressures estimated on the basis of RMR.

TABLE 5.3

Net Safe Bearing Pressures Based on RMR Classification

Rock Quality & Pressures III IV

Description Fair Rock Mass Poor Rock Mass

RMR 41-60 21-40

q (MPa) 1.4 - 3.0 0.4 - 1.4

a

Because the numerical values of bear inn vrvuisuro (q ) a re

based on the criterion of permissible value of total

settlement, these need not be increased, even if the foundation

is embedded into the rock. Furthermore, no correction factors

such as for conditions like submergence, presence of cavities,

loose stratum and slopes etc., need be applied because the

effects of these are already covered in the RMR values.

Moreover, as the predicted bearing pressures are based on total

settlement criterion, these are supposed to be inclusive of the

effect, of eccentricity and the effect of interference of

adjoining foundations, if any. However, for designs the bearing

pressure may be taken as 75 - 80 per cent of the predicted

value because of many unforeseen geological conditions or other

uncertain factors. The values of allowable bearing pressures in
 248

Fig. 5.45 are in good agreement with the values obtained by the
University of Roorkee from plate load tests on 30 cm diameter
4
plate (unpublished). It is suggested that the values may be
taken only as a guide and if thought necessary, the permissible
settlement in each case may be decided as per requirement of

the designer depending upon the geological environment and

importance of structure.

From the above discussions, it is brought out that the

allowable bearing pressures based on the RMR values can be used

for the preliminary design of structures proposed to be founded

on poor and fair quality rock masses.

5.7 Assumptions and Limitations

The work presented under this research study is the result of

extensive laboratory and in-situ investigations performed over

a period of 4 years. The data obtained from the tests carried

out at the six project sites sliow that there is a considerable

degree of scatter in the results, especially those obtained for

>
poor rock masses. The scatter of the results indicates that
some unknown geological conditions or other uncertain factors

have shown their influence on the performance of rock mass

during testing. This is ' true, because it is difficult to

visualise and assess the geological conditions in every part of

the x'ock mass under exploration.

\
The laboratory and field methods adopted to obtain the

properties of rocks usually always incorporate certain

assumptions. The results of the present study are also under
 249

certain assumptions which may be given as follows :

1. The rock mass is isotx'opic, homogeneous and semi-infinite

with ideal elastic properties.

2. The rock mass is free from any damage while preparing rock

bed for the test and the results are free from blasting

effects.

The foregoing assumptions have been taken care of during

performance of tests in the field and in the laboratory as far

as possible. For example, equipments and test procedures for

various laboratory and in-situ tests have been mostly as per

ISRM suggested test methods for classification and

characterisation of rock materials and rock masses. However,

the settlement has been measured on the surface and not inside

the rock mass.

Apart from the assumptions made during deriving results

from experimental work, the deformation and shear

characteristics of the intact as well as the rock in-situ are

affected by many other factors, the most important being

in-situ stress condition, varying moisture content, structural

defects, specimen size and shape, mineralogical composition,

interlocking of the rock blocks. Consequently, it seems

desirable to include the effect of all these factors while

estimating the engineering parameters of the rock mass by

conducting statistically significant number of field tests on

representative volumes of undisturbed rock masses soon after

excavation.
 r

on
O

25 30 35 WJ 45 50 55 60
15 20

ROCK MASS RATING ( RMR ) —•-

FIG. 5-^5 ALLOWABLE BEARING PRESSURE ON THE BASIS OF ROCK MASS RATING [RMR =21~60]
( NATURAL MOISTURE CONTENT = 0-60 % - 6«50 •/. )
X-

CHAPTER 6

J
 CHAPTER 6

CONCLUSIONS

Study of engineering behaviour of various rock formations in

the Lesser Himalaya at construction sites of a number of

hydroelectric projects in the complex Himalayan geology has

been brought out. The results obtained provide a basis for

understanding the characteristics of different rocks found in

this region. The RMR and the Q systems have been applied and

used to classify the rocks encountered at the six hydroelectric

project sites. The data obtained from the rock raass

classification approach, and the field and laboratory

investigations,have been used to evolve correlations between

the commonly used engineering parameters of the rock mass. It

has been observed that the rocks in the Lesser Himalaya are

generally poor and weak having inferior engineering properties.

Presently, very scanty information is available x-egarding

the engineering behavioux" of Himalayan rocks. The repox'ted case

histories (Chapter 3) reveal that the problems encountered in

the Lesser Himalayan rocks are basically the problems of

foundations, supporting of tunnels and underground cavities

i.e. the problems related to rock raass deformability and shear

characteristics.

The RMR and the Q systems have been correlated with the

parameters commonly used in civil engineering designs. The

effect of saturation on the modulus of defox-mation, support

pressure and shear strength parameters, lias been found to be

significant. In case in-situ tests are conducted in dry

condition, the correction factox-s developed can be used to

 252

predict the design parameters for saturated condition.

The study has led to the following main conclusions:

1. Inspite of some scatter of results, mostly at low RMR

values, a definite correlation (Eq. 5.1) has been

found to exist between the RMR and the modulus of

deformation. It is found that the rock formations in

the Lesser Himalaya are comparatively of low modulus

values ranging from 0.75 GPa to 2.4 GPa for poor rock

masses and from 2.4 GPa to 7.5 GPa for fair rock

masses (Fig. 5.1). A

2. Equations as suggested by Bieniawski (1978) and by

Serafim and Pereira (1983) to estimate the modulus

values of rock mass are inadequate for application to

the Lesser Hiraalayan rocks. The relationship given by

Serafira and Pereira (1983) highly overestimates the

raodulus of deformation of these rocks. The

relationship given by Bieniawski (1978) is not

applicable to poor rock masses.

3. The effect of saturation on the modulus of

deformation ha3 been found to be significant. When

saturated, the reduction in the raodulus values from

Eq. 5.2 can be as high as 90 per cent for 'poor' and

from Eq. 5.3, 75 per cent for 'fair' rock masses.

4. The relationship between RMR and raodulus reduction

factor (MRF) is non-linear (Fig. 5.30). It is found

that the relationship suggested by Bieniawski (1975)

is not applicable in estimating MRF for Lesser

 253

Himalayan rocks because comparatively higher values

are predicted through this relationship.

5. The proposed relationships (Figs. 5.32, 5.33) between

modulus of deformation (E,) and the estimated support

pressure [p., ^,] show a non-linear trend. The

* i(roof)
results indicate that the support pressure has a

decreasing trend with increase in the modulus. For

values of modulus between 0.7 5 GPa and 2.5 GPa the

estimated support pressux-es lie in the range 0.90

2 2
kg/cm to 0.33 kg/cm , whereas for rock masses having

modulus values between 2.5 GPa and 7.5 GPa the

2
support pressures lie between 0.33 kg/cm and 0.21
o
kg/cm . It is observed that beyond the modulus of 7.5

GPa, the support pressures attain nearly a constant

2
value of 0.2 kg/cm .

It is seen that under saturation the support

pressures rise by about 100 pex* cent fox* x'ock masses

in the modulus range of 0.75 GPa to 2.5 GPa.

6. Cox-relations (Fig. 5.35) established between RMR

system and shear s ti-ength parameters c and 0 on the

basis of extensive large size block shear tests have

shown that both, cohesion (c) and the angle of

internal fx'iction (0) increase with the RMR. Maximum

values of shear strength parameters (c & 0) within

the range of the tests are predicted at RMR of 60, at

which cohesion (c) is 450 kPa and angle of internal

o

friction (0) is 57 .

The effect of saturation (Fig. 5.36) is found to

 254

be significant on the shear strength parameters. It

has been observed that the effect of saturation is

'4
more striking on cohesion (c) than the angle of

internal friction (0). It is found that under

saturation,the reduction in cohesion (c) is as high

as 70 per cent, whereas,reduction in the angle of

internal friction (0) is only 35 per cent.

7. The Mohr envelopes (Figs. 5.37-5.44) deduced from the

data of large size block shear tests,are non-linear.

A
The approximate shear strength of the rock mass for

triaxial compression, and developed friction (tan 0)

can be estimated at desired magnitude of normal

stress from these envelopes (Table 5.2). The data of

large size block shear tests have been used to

represent the stress dependent failure envelopes for

the rocks.

8. There is relationship between the rock mass rating

k
(RMR)and the allowable bearing pressure on Lesser

Himalayan rocks , as developed in the study (Fig.

5.45). This relationship would be useful for

estimating allowable bearing pressures for

foundations on the rock3.

S<
A-

CHAPTER 7

1
 CHAPTER 7

RECOMMENDATIONS FOR FUTURE RESEARCH

Provision of reliable input data for engineering design of

structures in rocks is one of the most difficult jobs facing

the design engineers. The present study is an attempt to

provide an assessment of engineering properties of rock mass.

The parameters of particular significance have been correlated

with the RMR and Q systems. The correlations would find useful

application in obtaining estimates of engineering parameters of

Lesser Himalayan rocks without involving time consuming costly

field tests.

The study gives a rational approach fox- rapid estimation

of engineering parameters of rock mass fox- planning stage of a

project. Detailed investigations can be carried out later on as

there ax-e uncex-tainties which are always associated with the

obtained results and therefore, there is scope for further

refinement of these correlations.

On the basis of present study, the following suggestions

may be helpful for futux-e research.

1. In case of underground excavations the behaviour of

the rock mass around the cavity is dominated by

initial state of stress. It is thus recommended to

study the effect of initial state of stress on the

engineering parameters of the rock mass. The

parameters estimated aftex- taking into account the

effect of in-situ stress shall be more compatible

with the rock mass conditions existing in the region.

 256

2. In near surface excavations, the effect of stress

distribution in the rocks in the vicinity of river

valley should be studied and taken care of while

estimating engineering parameters of rock raass.

3. Influence of rock anisotropy upon the distribution of

stresses in rocks need to be studied so that the

effect of anisotropy is incorporated in the various

engineering parameters of the rock raass.

4. It has been found that the saturation of the rock

affects the mechanical properties of rocks ^

significantly. Moreover, data obtained from in-situ
investigation is sometimes difficult to interpret
because the values of parameters change in different

seasons. It is thus necessary to study the effect of

moisture change on the mechanical properties of

rocks, especially the poor rocks.

5. Rock masses are essentially inhomogeneous and

discontinuous media. Hence the results of the

laboratory and in-situ tests for the determination of (

their mechanical properties show a dependency on the

specimen size (scale effect).

Scale effect has been found to be a function of

the type of rock and the amount of moisture present.

It is thus recommended to investigate into the scale

effects in the determination of mechanical properties

V
of different types of rocks. For this purpose

uniaxial jacking tests/plate load tests on different

sizes of plates should be conducted.

 257

6. In seismic regions, the effect of dynamic loading on

the various engineering parameters need to be studied

in details. In case any live thrust or fault crosses

a prospective dam site, it is necessary to account

for the effect of dynamic loading on the engineering

parameters of rock mass. Detailed dynamic studies are

thus necessary for any prospective dam planned in

seismic region.

7. It is also possible to refine the new correlations by

using additional pararaeters such as overburden

height, time etc. so that the results will be more

realistic.

8. Some improvement in uniaxial jacking test as per 1SRM

procedure is essential. The test should be conducted

soon after excavation of drift. Simultaneously

dilatometex- tests may be conducted to find out

increase in E, with depth.

9. The effect of confining pressure on modulus of

deformation should be investigated on rock masses.

This has tremendous beneficial effect on the

stability of underground openings and dam foundations

because favourable redistribution of stresses takes

place.
 REFERENCES

Abad, J., Celada, B., Chacon, E., Gutierrez, V. and Hidalgo,

E.(1983), "Application of Geomechanical Classification to
Predict the Convergence of Coal Mine Galleries and to
Design Their Supports," proc. 5th Int. Cong. Rock Mech.,
ISRM, Melbourne, Vol.2, pp. E15-E19-

Association Francaise des Travaux en Souterrain [AFTES] (1981),

"Textes des Recommendations Relatives au Choix d'un Type
de Soutenement en Galerie," Tunnels et Ouvrages
Souterrains, Avril.

Auden,J.B. (1942), "A geological Investigation of Tunnel

Alignment for Yamuna Hydroelectric Scheme", Rec, Geol.
Surv. of India, Vol. 78, pt II.

Baczynski, N.(1980), "Rock Mass Characterization and Its

Application to Assessment of Unsupported Underground
Openings", Ph. D. Thesis, University of Melbourne, 233
PP.

Bandis.S.C. (1990), "Scale Effects in the Strength and

Deformability of Rocks and Rock Joints", Special Lecture,
Int. Workshop on Scale Effects in Rock Masses, Loen,
Balkema.

Barton,N., Lien, R. and Lunde, J. (1975), "Engineering

Classification of Rock Masses for the Design of Tunnel
Supports," Rock Mechanics, Vol.6, No. 4, pp. 189-236.

Barton,N., Lien, R. and Lunde, J. (1974), "Estimation of

Support Requirements for Underground Excavations, "Proc.
XVI Symp. on Rock Mechanics, University of Minnesota,
y Minneapolis, U.S.A., pp. 163-177.

Barton, N. and Chaubey, V. (1977), "The Shear Strength of Rock

Joints in Theory and Practice, Rock Mechanics, Vol. 10,
pp. 1-54.

Barton, N.(1983), "Application of Q-Syetem and Index Tests to

Estimate Shear Strength and Deformability of Rock
Masses," Proc. Int. Symp. Eng. Geol. Underground Constr.,
A.A. Balkema, Boston, pp. 51-70.

Barton,N. (1984), "Effects of Rock Mass Deformation on Tunnel

Performance in Seismic Regions," Adv. Tunnel Technol. and
Subsurf. Use. Vol.4, No.3, pp. 89-99.
J
Barton,N. R. (1986), "Deformation Phenomena in Jointed Rock"
Geotechnique 36, No. 2, pp. 147-167.

Barton,N. (1988), "Personal Discussion".

Barton,N. (1990), "Scale Effects or Sampling Bias", Closing

Lecture, Int. Workshop on Scale Effects in Rock Masses,
Loen, Balkema.
 259

Barton,N. and Bandis, S. (1990), "Review of Predictive

Capabilities of JRC-JCS Model in Engineering Practice,
Proc. Int. Symp. on Rock Joints, pp. 603-610.
Bieniawski, Z.T.(1973), "Engineering Classification of Jointed
Rock Masses," Trans. S. Afr. Inst. Civ. Eng. 15, pp. 33a-
344.

Bieniawski, Z.T.(1975), "Case Studies: Prediction of Rock

Behaviour by the Geomechanics Classification Proc. znd
Austr. - N.Z. Conference on Geomechanics, Brisbane, pp.
36-41.

Bieniawski, Z.T.(1976), "Rock Mass Classifications in Rock

Engineering; Proc. Symp. Exploration for Rock
Engineering. ed. Z. T. Bieniawski, A.A. Balkema,
Rotterdam, pp. 1-106.

Bieniawski, Z.T. and Orr, CM. (1976), "Rapid Site Appraisal

for Dara Foundations by the Geomechanics Classification,
Proc. 12th Congr. Large Dams, ICOLD, Mexico City, pp.
483-501.

Bieniawski, Z.T. (1978), "Determining Rock Mass Deformability:

Experience from Case Histories", Int. J. Rock Mech. Mm.
Sci. Geom. Abst., Vol. 15, pp. 237-247.
Bieniawski, Z.T. (1979), "The Geomechanics Classification in
Rock Engineering Applications", Proc. 4th Int. Congr.
Rock Mech., ISRM, Montreux, Vol. 2, pp. 41-48.
Bieniawski, Z.T. (1984), "Rock Mechanics Design in Mining and
Tunneling," A.A. Balkema, Rotterdam, The Netherlands.
Bieniawski, Z.T. (1988), "Personal Discussion".
Bieniawski, Z.T. (1989), "Engineering Rock Mass
Classifications", John Wiley and Sons, Inc., New York,
U.S.A.

Brekke, T.L. and Howard, T.R. (1973), "Functional

Classification of Gouge Materials from Seams and Faults
in Relation to Suitability," Problems in Underground
Openings, U.S. Bureau of Mines, Denver, Contract No. HO
22022.

Cameron-Clark, I.S. and Budavari, S. (1981), " Correlation of

Rock Mass Classification Parameters Obtained from
Borehole and In-Situ Observations, "Eng. Geol. 17, pp.
^
19-53.

Cecil 0 S (1970), "Correlation of Rock Bolts-Shotcrete Support

'and Rock Quality parameters in Scandinavian Tunnels; Ph.
D. Thesis, University of Illinois, Urbana, U.S.A. pp. 1-
414.

Chaubey, Vishnu. D. (1988), "Personal Discussion."

 260

Chaubey, Vishnu. D. and Chaturvedi Shailendra (1991),

"Geotechnical Appraisal of the Foundation Rock Mass
Behaviour of Narmada Sagar Dam Project, Central India : A
Case Study." Can. Geotech. J., 28, pp. 148-159.

Coates.D. F. (1964), "Classification of Rock for Rock

Mechanics", Int. J. Rock Mech. Min. Sci., Vol. 1, pp.
421-429.

Cording, E.J. and Deere, D.U.(1972), "Rock Tunnel Supports and

Field Measurements," Proc. Rapid Excav. Tunnelling Conf.,
AIME, New York, pp. 601-622.

Cunha Pinto Da, A.(1991), "Scale Effects in the Determination

of Mechanical Properties of Joints and Rock Masses,"
Proc. 7th Int. Cong, on Rock Mechanics, Aachen, Germany,
Vol. 1, pp. 311-318.

Daemen,J.J.K. (1977), "Problems in Tunnel Support Mechanics",

Underground Space, Vol. 1, pp. 163-172.

Daemen,J.J.K. (1988), "Personal Discussion."

Daemen,J.J.K. (1991), "Personal Discussion".

Davachi, M.M., Sinclair, B.J., Hartraaier, H.H., Baggott, B.L.

and Peters, J.E. (1991), "Determination of the Oldman
River Dam Foundation Shear Strength," Can. Geotech. J.
Vol. 28, pp. 698-707.

Deere, D.U. (1964), "Technical Description of Rock Cores for

Engineering Purposes", Rock Mech. and Engg. Geol., Vol.
1, pp. 17-22.

Deere, D.U., Hendron, A. J., Patton, F. D. and Cording, E.J.

(1967), "Design of surface and Near surface Construction
in Rock," Proc. 8th U.S. Symp. Rock Mech., AIME, New
York, pp. 237-302.

Deere, D.U., (1968), "Geological Consideration", Rock Mechanics

in Engineering Practice, Ed. R.G. Stagg and D. C.
Zienkiewicz, Wiley, New York, pp. 1-20.

Deere,D. U., Peck, R.B., Parker, H., Mouses, J.E. and Schmidt,
B.(1970), "Design of Tunnel Support Systems", Highway
Research Record, No. 339, pp. 26-33.

Franklin, J.A. (1976), "An Observational Approach to the

Selection and Control of Rock Tunnel Linings", Proc.
Conf. Shotcrete Ground Control, ASCE, Easton, MA, pp.
556-596.

Gantang, L. and Yuelan, H. (1991), "Influence of Rock Moisture

on the Longitudinal Wave Velocity," Selected papers from
Chinese J. of Geotech. Engg. Ed. Haunz, ASCE, pp. 190-
196.
 261

Ghose.A.K. andRaju, N.M. (1981), "Characterization of Rock

Mass vis-a-vis Application of Rock Bolting in Indian Coal
Measures", Proc. 22nd U.S. Symp. Rock Mech., MIT,
Cambridge, MA, pp. 422-427.

Ghosh, Amitava and Daemen, Jaak J.J. (1987), "An Expert System
Approach to Geomechanics Design Problems," Proc. The
Third Int. Conf. on Innovative Mining Systems, Univ. of
Missouri-Rolla, pp. 1-14.

Grasso, P., Xu, S.L. and Carrier!, G. (1991), "Stability

Analysis of an Artificial Slope Out in a Weak Siltstone,"
Proc. 7th Int. Cong, on Rock Mechanics, Aachen, Germany,
Vol. 2, pp. 881-886.

Hardy,J.S. (1986), "Estimating Rippability by Rock Mass

Classification," Proc. 27th U.S. Symposium, University of
Alabama, pp. 443-448.
A
Hawkins, A. and Mcconnell, B.J. (1991), "Influence of Geology
on Geomechanical Properties of Sandstones," Proc. 7th
Int. Cong, on Rock Mechanics, Aachen, Germany, Vol. 1,
pp. 257-260.

Hoek, E. and Brown, E.T.(1980), "Underground Excavations in

Rock", The Institution of Mining and Metallurgy, London,
pp. 131-177.

Hoek, E. and Bray, J.W. (1974/1981), "Rock Slope Engineering",

Institution of Mining and Metallurgy, London, I.Ed. 1974/
2.Ed. 1981.

Hoek, E., Wood, D. and Shah, S. (1992) "A Modified Hoek-Brown

Failure Criterion for Jointed Rock Masses," 1SRM
Symposium: Eurock' 92 on Rock Characterisation, ed. J.A.
Hudson, Thomas Telford, London.

Horn, H.M. and Deere, D.U. (1962), "Frictional Characteristics

of Minerals, Geotechnique 12: pp. 319-335.

IS:7317-1974, Indian Standard Code of Practice for Deformation

Modulus of Rock, Bureau of Indian Standards,New Delhi.
IS:7746-1975, Indian Standard Code of Practice for In-Situ
Shear Test on Rock, Bureau of Indian Standards, New
Delhi.
•. i

IS:8764-1978, Indian Standard Code of Practice on the Method

for Determination of Point Load Strength Index on Rocks,
Bureau of Indian Standards, New Delhi.
^

IS:9143-1979, Indian Standard Code of Practice on the Method

for Determination of Unconfined Compressive Strength of
Rock Materials, Bureau of Indian Standards, New Delhi.

IS:9179-1979, Indian Standard Code of Practice on the Method

for Preparing of Rock Specimen for Laboratory Testing,
Bureau of Indian Standards, New Delhi.
 262

IS
9221-1979, Indian Standard Code of Practice on the Method
for Determination of Modulus of Elasticity and Poisson s
Ratio of Rock Materials in Uniaxial Compression, Bureau
of Indian Standards, New Delhi.

1988-1982, Indian Standard Code of Practice on the Method of

IS
Load Test on Soils, Bureau of Indian Standards, New
Delhi.

IS:11315 (PT 1-10), Indian Standard Methods for Qualitative

Description of Discontinuities in Rock Mass Bureau of
Indian Standards, New Delhi.

IS:12070-1987, Code of Practice for Design and construction of

shallow Foundations on Rocks, Bureau of Indian Standards,
New Delhi.

International Society for Rock Mechanica (1981) "B^sif

y Geotechnical Description of Rock Masses" lint. J.Rock
Mech. Min. Sci., Vol. 18, pp. 85-110.

International Society for Rock Mechanics (1982), "ISRM

Suggested Methods ; Rock Characterization, testing and
Monitoring, ed. E.T. Brown, Pergamon, London, 211 pp.
Jethwa, J.L., Dube, A.K. and Singh, B. (1976), "Assessment of
Rock Pressure and Support Requirements for Tail Race
Tunnel of Salal Hydroelectric Project, Central Mining
Research Station, Dhanbad, Report Dec. 1976.

Jethwa, J.L., Dube, A.K., Singh, B., Singh Bhawani and Mithai,
R.S. (1982), "Evaluation of Methods for Tunnel Support
Design in Squeezing Rock Conditions," Proc. 4th Int.
Cong., Int. Assoc, of Engg. Geol., New Delhi, Vol. 5, pp.
125-134.

c Jethwa, J.L. (1984), "Estimation of Ultimate Rock Pressure for

Tunnel Linings Under Squeezing Rock Conditions - A New
Approach," Proc. ISRM Symp. on Design and Performance of
Underground Excavations, Cambridge, U.K., pp. 231-238.

John, K.W. (1962), "An Approach to Rock Mechanics," J.Soil

Mech. Found. Div., Am. Soc Civ. Eng., Vol. 88, SM-4,
Ang. , pp. "1-30.

Kaiser,Peter K., Mackay, C and Gale, A.D. (1986) "Evaluation of

Rock Classifications at B.C. Rail Tumbler Ridge Tunnels",
Rock Mech. Rock Eng., 19, pp. 205-234.

J Kidybinski, A. (1979), "Experience with Rock Penetrometers for

Mine Rock Stability Predictions", Proc. 4th Int. Congr.
Rock Mech., ISRM, Montreux, pp. 293-301.

Kulhawy, Fred. H. and Goodman, Richard. E. (1930), e s i gn of

Foundations on Discontinous Rock", Proc. Int. Con ii. on

Strcl. Fdns. on Rock, Ed. P.J.N. Pells, Vol. 1 PP- 209-

220.
 263

Ladanyi, B. (1974), "Use of Long Term Strength Concept in the

Determination of Ground Pressure on Tunnel Lining",
Advances in Rock Mechanics, Proceedings of the Third V
Congress of the International Society.

Laubscher, D.H. (1977), "Geomechanics Classification of Jointed

Rock Masses - Mining Applications", Trans. Inst. Min.
Metall. 86, pp. A1-A7.

Lauffer, H.(1958), "Gebirgs Klassifi Ziering fur den

Stollenbau," Geol Bauwesen, Vol. 74, No. 1, pp. 46-51.

Lauffer, H.(1988), "Zur Gebirgsklassifizierung bei

Frasvortrieben", Felsbau 6(3), pp. 137-149.

Lo, K.Y., Yung, T. C. B. and Lukajic, B(1987), "A Field Method

for the Determination of Rock Mass Modulus" Can. Geotech.
J. Vol. 24, pp. 406-413.

Lokin,P., Nijajilovic, R. and Vasic, M. (1983), "An Approach to

Rock Mass classification for Undergx'ound Works", Proc.
'"" ' 5th Int. Congr. Rock Mech., ISRM, Melbourne, Vol. 1, pp.
• B87-B92.

Londe.P. (1972), "The Mechanisms of Rock Slopes and

Foundations", Imperial College Rock Mech. Rep., No. 17,
pp. 1-23.

Mehrotra.V.K., Mitra Subhash and Agrawal, C.K.(1991), "Need of

Long Terra Evaluation of Rock Parameters in the Himalaya",
Proc. 7th Int. Cong, on Rock Mechanics, Aachen, Gex-many,
Vol. 1, pp. 289-292.

Merritt,A.H. (1972), "Geologic Prediction for Underground

Excavations", Proc. Rapid Excav. Tunneling Conf., AIME,
New York, pp. 115-132.

Moreno Tallon, E.(1982), "Comparison and Application of the

Geomechanics Classification Schemes in Tunnel
Construction," Proc. Tunnelling ' 82, Institution of
Mining and Metallurgy, London pp. 241-246.

Muller.L. (1963), "Der Felsbau, Stuttgart, Ferdinand Enke-

Verlag, pp. 1-624.

Nakao.K., Iihoshi, S. and Koyama, S. (1983), "Statistical

Reconsiderations on the Parameters for the Geomechnics
Classification", Proc. 5th Int. Congr. Rock Mech., ISRM,
Melbourne, pp. B13-B16.

Nguyen Minh, D., Habib, P. and Guerpillon, Y. (1984), "Time

Dependent Behaviour of a Pilot Tunnel Driven in Hard
Marls," Proc. ISRM Symp. on Design and Performance of
Underground Excavations, Cambridge, U.K., pp. 453-459.

Nilsen.B. (1985), "Shear Strength of Rock Joints at Low Normal

Stresses - a Key Parameter for Evaluating Rock Slope
 264

Stability," Proc. Int. Symp. on Fundamentals of Rock

Joints, Ed. Bjorkliden, pp. 15-20.

Oda Masanobu (1988), "A Method for Evaluating the

Representative Elementary Volume Based on Joxnt Survey of
Rock Masses," Canadian Geotech. J., Vol. 25, pp. 440-447.
Okamoto Ryuichi, Kojima Keiji and Yoshinaka Ryunoshin (1981),
"Distribution and Engineering Properties of Weak Rocks in
Japan," Proc. Int. Symp. on Weak Rock, Tokyo, pp. 1269-
1281.

Olivier, H.J. (1979), "Applicability of the Geomechanics

classification to the Orange - Fish Tunnel Rock Masses",
Civil Engg. S. Afr. 21, pp. 179-185.

Pacher.F., Rabcewicz, L. and Gosler, J. (1974), "Zum der

Seitigen Stand der Gebirgs - Klassfizienung in Stollen-
und Tunnelbau," Proc. XXII Geomech. Colloq., Salzburg,
Austria, pp. 51-58.

Palrastrom, A.(1982), "The Volumetric Joint Count -A Useful and

Simple Measure of the Degree of Rock Jointing," Proc. 4th
Int. Congr. Int. Assoc. Eng. Geol., Delhi, Vol. 5, pp.
221-228.

Palrastrom, A.(1985), "Application of the Volumetric Joint Count

as Measure of Rock Mass Jointing. Proceedings of the
International Symposium on Fundamentals of Rock Joints,"
Bjorkliden, Norway, pp. 103-110.

Peck.R.B., Hanson, W.E. and Thornburn, T.H. (1974), "Foundation

Engineering, 2nd Ed., New York, Wiley.

Pinto Da Cunha, A.(1991), "Scale Effects in the Determination

of Mechanical Properties of Joints and Rock Masses,"
Proc. 7th Int. Cong, on Rock Mechanics, Aachen, Germany,
Vol. 1, PP. 311-318.

Protodyakonov, M.M. (1963), "Firmness Coefficient for

Estimation of Rock Loads", Personal Communication to Beas
Organisation, N. Delhi.

Protodyakonov, M.M. (1974), "Klassifikacija Gornych Porod,

Tunnels Overages Sovterrains", 1, pp. 31-34. (English
Version).

Rabcewicz, L.(1964)., "The New Austrian Tunnelling Method,"

Water Power, Vol. 16, No. 11, pp. 453-457., No. 12, pp.
511-516.

Raraamurthy, T.(1985), "Stability of Rock Mass," Eighth Indian

Geotechnical Society Annual Lecture, Roorkee, Dec. 1985,
PP. 1-74.

Raraamurthy, T.(1992), "Strength and Modulus for Classification

of Jointed Rocks", Asian Reg. Symp. on Rock Slopes, N.
Delhi, pp. 1-10.
 265

Ramegowda, B.M. (1990), "Insitu testing of Rock Masses," Proc.

Workshop on Rock Mechanics, Central Board of Irrigation
and Power, New Delhi, pp. 28-36.

Roraana, M. (1985), "New Adjustment Ratings for Application of

i'M Bieniawski Classification to Slopes," Proc. Int. Symp.
Rock Mech. Min. Civ. Works, ISRM, Zacatecas, Mexico, pp.
.'< '•'• ' 59-63.

Rose, D.(1982), "Revising Terzaghi's Tunnel Rock Load

Coefficients," Proc. 23rd U.S. Symp. Rock Mech., AIME,
New York, U.S.A., pp. 953-960.

Rutledge, J.C. and Preston, R.L. (1973), "Experience with

Engineering Classifications of Rock," Px-oc. Int.
Tunnelling Symp., Tokyo, pp. A3.1-A3.7.

Schmidt, Birger (1991), "Personal Discussion".

H
Sellers,J.B. (1974), "Measurement of In-Situ Shear Strength
Using the Torsional Shear Method, "Field Testing and
Instrumentation of Rock, ASTM STP 554, American Society
for Testing and Materials, pp. 147-155.

Serafim,J.L. and Pereira, J.P. (1983), "Considerations of the

Georaechanics Classification of Bieniawski, Proc. Int.
Syrap. Eng. Geol. Underground Constr., LNEC, Lisbon, Vol.
1, pp. II.33-11.42.

Sharp,J.C, Enderbee, L.A. and Mellors, T.W.(1984), "Design and

Observed Performance of Permanent Cavex-n Excavations in
Weak, Bedded Strata," Proc. ISRM Symp. on Design and
Performance of Undex-ground Excavations, Cambridge, U.K.,
pp. 493-507.

Sharraa.V.M. and Raraamurthy, T. (1985), "A Rock Mass

Classification System for Underground Openings Based on
Ground Convergence," Workshop on Engineering
Classification of Rocks, Centx-al Board of Ix'x-igation and
Power, N. Delhi, pp. 185-196.

Sharma,V.M.(1990), "Rock Support Interaction; Workshop on Rock

Mechanics, Central Board of Ix*x-igation and Power, New
Delhi, pp. 77-84. '

Sheorey.P.R. (1991), "Experience with Application of the NGI

Classification to Coal Measures," Int. J.Rock Mech. Min.
Sci. & Geomech. Abstr. Vol. 28, No. 1, pp. 27-33.

Singh, Bhawani. and Goel, P.K. (1975), "Estimation of Elastic

Modulus of Jointed Rock Masses - Lecture Notes on Rock
Mechanics in Engg. Px-actice, Indian Geotechnical Society,
Local Chapter, Roox'kee, Chaptex- No. 3.

Singh, Bhawani.(1979), "Geological and Geophysical

Investigation in Rocks fox- Engineering Projects", Proc.
Int. Conf. on In-Situ Testing of Soils and Rocks and
Performance of Stx-uctux-es, Roorkee, Vol. 1, pp. 486-497.
 266

Singh,Bhawani, Jethwa, J.L., Dube, A.K. and Singh, B. (1992),

"Correlation Between Observed Support Pressure and Rock
Mass Quality," Tunnelling and Underground Space
Technology, Vol. 7, No. 1, pp. 59-74.

Steffen, O.K.H. (1976), "Research and Development Needs in Data

Collection for Rock Engineering", ed. Z.T. Bieniawski,
A.A. Balkema Press, Rotterdam Vol 2., pp. 93-104.

Sterling, Raymond. (1991), "Personal Discussion".

Stini,I.(1950), "Tunnulbaugeologie," Springer-Verlag, Vienna,

Asutria, pp. 1-336.

Terzaghi, K.(1946), "Rock Defects and Loads on Tunnel

Supports," Rock Tunnelling with Steel Supports Ed.-R.V.
Proctor and T.L. White, Commercial Sliearing and Stamping
Co. Youngstown, Ohio, U.S.A., pp. 15-99.

Tsidzi,K.E.N. (1991), "Compressive Strength Anisotropy of

Foliated Rocks," Mechanics of Jointed and Faulted Rock,
Rossmanith (ed), Balkema, Rotterdam, The Netherlands.

Unal.E. (1983), "Design Guidelines, and Roof Control Standards

for Coal Mine Roofs, Ph.D. Thesis, Pennsylvania State
University, University Park, 355 pp.

Venkateswarlu, V., Sinha, A. and Raju, N.M. (1985),

"Classification of Coal Mine Roof Rocks in India,"
Workshop on Engineering Classification of Rocks, Central
Board of Irrigation and Power, N. Delhi, pp. 173-181.

Venkateswarlu,V.(1986), "Geomechanics Classification of Coal

Measure Rocks vis-a-vis Roof Supports," Ph.D. Thesis,
Indian School of Mines, Dhanbad, pp. 1-251.

Ward,W.H. (1978), "Eighteenth Rankine Lecture Ground Supports

for Tunnels in Weak Rocks," Geotechnique Vol. 28, No. 2,
London, England, pp. 133-170.

Whittaker.B.N., Bonsall, C.J. and Kapusniak, S.S. (1984)

Stability Aspects of Major Coal Mining Tunnel Projects ;
Proc. ISRM Symp. on Design and Performance of Underground
Excavations, Cambridge, U.K., pp. 461-470.

Wickham, G.E., Tiedmann, H.R. and Skinner, E.H. (1972),

"Support Determination Based on Geologic Predictions",
Proc. Rapid Excav. Tunnelling Conf., AIME, New York, pp.
43-64.

Wickham,G.E., Tiedmann, H.R. and Skinner, E.H. (1974), "Ground

Support Prediction Model, RSR Concept, Proc. Rapid Excav.
Tunnelling Conf., AIME, New York, pp. 691-707.

Yang.L. and Sterling, R.L., (1989), "Back Analysis of Rock

Tunnel Using B.E.M.," ASCE, J. of Geotech. Engg. pp.
1163-1169.
 267

Yoshinaka, R., Yoshida, J., Shirnizu, T., Aral, H. and Arisaka,

S.(1991), "Scale Effect in Shear Strength and
Deformability of Rock Joints," Proc. 7th Int. Cong. on
Rock Mechanics, Aachen, Germany, Vol. 1, pp. 371-374.

Zhen.CX., Cheng, D.X., Min, Z. and Ling, H.(1984), "Monitoring

and Back Analysis of the Deformations in Two Tunnels in
China", Proc. ISRM Symp. on Design and Performance of
Underground Excavations, Cambridge, U.K., pp. 445-451.

<

Sv