961
Shear Behaviour of Rock Joints under CNS Boundary Conditions
Comportement en cisaillement de joints rocheux en condition de raideur normale constante
Shrivastava A.K.
Delhi Technological University, Delhi, India
Rao K.S.
Indian Institute of Technology, Delhi, India
ABSTRACT: The shear behaviour of rock joints depends up on many factors, the correct evaluation of this is possible only if these
parameters are properly considered during experimental investigation, constitutive modelling and numerical modelling. Which is
important for safe and economical design of underground openings in jointed rocks, stability analyis of rock slopes, risk assessment of
underground waste disposal, design of foundation on rock and design of rock socketed piles. These concerns invite accurate
quantification of shear strength of unfilled and infilled joints, proper understanding of the basic mechanics of discontinuity and the
principles involved in their shear deformation. This can be done through in-situ or laboratory large scale testing on natural rock or
laboratory testing on physical model. In the present paper the detail account of test results of direct shear tests performed on large size
modeled unfilled and infilled rock joints under different boundary conditions is systematically presented. It is observed that the
constant normal stiffness (CNS) conditions better simulate the field conditions of the loading and shear strength predicted under CNS
condition is more than the constant normal load (CNL) conditions for both unfiled and infilled joints.
RÉSUMÉ : Le comportement en cisaillement de joints rocheux dépend de nombreux facteurs, dont l'identification n'est possible que
par approche expérimentale numérique ou rhéologique. Cela est important pour la conception sécuritaire et économique de cavités
souterraines dans les roches fracturées, l'analyse de stabilité des talus rocheux, l'évaluation du risque d'élimination des déchets
souterraine, la conception de fondation au rocher et la conception pieux. Une quantification précise de la résistance au cisaillement
des joints remplis ou non, ainsi qu'une bonne compréhension des mécanismes de base de la discontinuité et des principes appliqués à
leur déformation en cisaillement sont nécessaires. Ceci est possible grâce à des essais en vraie grandeur in situ ou en laboratoire sur
modèle physique. Dans le présent document, le détail des résultats des essais de cisaillement direct effectués en vraie grandeur avec
ou non un remplissage des joints sous différentes conditions aux limites sont systématiquement présentées. On a observé que les
conditions de la rigidité normale constante (CNS) simulent mieux les conditions sur le terrain et que la résistance au cisaillement
prédite sous condition de CNS est plus grande que pour les conditions de charge normale constante (CNL) pour les deux types de
joint.
KEYWORDS: Shear Behaviour, Rock Joints, CNL, CNS, Direct Shear, Infill, Unfill, Dilation, Shear Strength, Deformation.
1 INTRODUCTION
Rock joints are mechanical discontinuities having geological
origin. These discontinuities are present in the form of joints,
faults, bedding planes or other recurrent planar fractures in the
rock mass. In general, strength and deformability properties of
these discontinuities are quite different from those of intact
rock, and in many cases, the discontinuities completely
dominate the shear and deformation behaviour of the in situ
rock mass in a given stress conditions.
The presence of infill or gouge material in the joints further
reduces the shear strength. The sources of infill material include
products of weathering or overburden washed into open joint,
water conducting in discontinuities, precipitation of minerals
from the ground water, by-products of weathering alterations
along joint walls, crushing of parent rock surfaces due to
tectonic and shears displacements, and thin seams deposited
during formation. In general, infill materials may consist of
partially loose to completely loose cohesionless soil or fine
grained clay. Normally fine-grained clays are more frequently
found as fillers and are more troublesome in terms of structural
stability. Thickness of the infill material varies from
micrometers to several meters and it plays an important role in
shear behaviour. In tectonically crushed zones, the infill
thickness may exceed several meters.
These rock joints unfilled or infilled are the weakest plane
which tries to slide or shear one over the other due to
construction of foundations of a structure and tunnels or
highways and railways on the rock slopes. Hence, for safe and
economical analysis of all the above cases it is important to
understand the strength and deformation behaviour of the rock
joints under direct shearing conditions. Shrivastava and Rao
(2009) discussed in details the influence of factors like (a)
boundary condition (b) shear rate (c) joint roughness (d) size of
joint i.e. scale effect (e) joint condition i.e. unfilled joint/in
filled joint on the direct shear strength of rock joints.
There are two boundary conditions i.e constant normal load
(CNL) and constant normal stiffness (CNS) boundary
conditions under which the shear behaviour of rock joints can
be studied. The planar rock joints can be investigated in the
laboratory by using a conventional direct shear apparatus where
the normal load is kept constant (CNL) during the shearing
process. This particular mode of shearing is suitable for
situations where the surrounding rock freely allows the joint to
shear without restricting the dilation or there is no dilation
during the shearing process, thereby keeping normal stress
constant during shearing process. Shear testing under a constant
normal load (CNL) boundary condition is only beneficial for
cases such as non-rereinforced rock slopes or planar rock joints,
but natural rock joints are seldom planar.
However, for non- planar discontinuities, shearing results in
dilation as one asperity overrides another, and if the surrounding
rock mass is unable to deform sufficiently, then an inevitable
increase in the normal stress occurs during shearing. At any
time t if the normal stress P
n
(t) then increase in normal stress
