Abstract
D.Ing.
The quantification of the surface roughness in order to determine the shear behaviour of
rock joints has been performed using many approaches, amongst which the most
important are the joint roughness coefficient, statistical methods and the fractal approach.
None of these approaches explicitly considers the contribution of grain size to the
creation of surface roughness. An investigation has been undertaken to determine whether
the grain size may be related to the surface roughness characteristics and also to assess its
impact on the shear behaviour of rock joints.
The influence of grain size on the shear behaviour of rock joints has been experimentally
and numerically studied using a profile measurement approach followed by shear tests in
the laboratory. Fresh rock joints in model material made of aggregates of no rite in resin
and two natural rock types, namely norite and sandstone, were used for the investigation.
A numerical analysis of the study has been performed using the Particle Flow Code PFC20
•
Investigation of the model material has resulted in the conceptualisation of grain size
parameters. Of the proposed grain size parameters, the average particle size index As has
been successfully used to define the particle size distribution of a rock joint for both
physical and numerical tests.
From the profile measurements, four roughness parameters, n~mely the average
inclination angle I, the root mean square of the first derivative Z2, the modified root mean
square of the first derivative Z2' and the fractal dimension D, have been computed and a
regression analysis conducted.
Physical and numerical results have demonstrated that grain size distribution, expressed in
terms of average particle size index As shows a good correlation respectively with the root
mean square ofthe first derivative Z2, and its modified version Z2'. The root mean square
of the first derivative Z2, and the modified root mean square of the first derivative Z2'
increase with the average particle size index As. However, no correlation has been found
between the grain size parameter and the fractal dimension D or the average inclination
angle I.
It has also been found that the joint roughness coefficient JRC and the peak friction angle
~P correlate very well with the average particle size index As. The greater the average
particle grain size the higher the joint roughness coefficient JRC and the peak friction
angle ~p· The type of correlation between JRC and ~P with As depends on the rock type.
Exponential shear strength model has been derived where the average particle size index As
is related to the peak friction angle via the residual friction angle ~r·