ABSTRACT
A common assumption in determining the shear strength of a rock joint using a direct shear test is that the surfaces undergoing relative shear are in complete contact. Furthermore, the limiting displacement associated with the residual strength values is obtained based on engineering judgement through visual inspection of shear force versus displacement curves. This study proposes a systematic way of determining the residual shear strength of rock joints by evaluating the limiting displacement leading to a minimum cohesive strength value and taking the true area of contact into account. The results of the study indicate that there is a limiting displacement associated with the residual shear strength, and some rocks may have an inherent interface residual cohesive strength different from zero. When the cohesion is different from zero, the residual shear strength is likely to be underestimated if the true area of contact is not considered. In the case where zero calculated cohesion is obtained, the area of contact has no bearing on the shear strength equation because the frictional force depends only on the normal load.
INTRODUCTION
When determining the shear strength of a 'rock joint using a direct shear test, it is common to assume that both surfaces undergoing shear are in complete contact with each other. This conventional approach to direct shear test analysis, particularly in the case of hard rock joints, neglects the fact that the two surfaces are only in partial contact with each other at various localized peaks. In addition, the conventional approach does not take into consideration the increase in contact area with increasing normal loads. Engineering judgement has to be exercised when estimating the correct displacements at which the residual strength values are reached. The selection of these displacements, which are based on visual inspection of shear force-displacement curves, are frequently highly arbitrary. Due to the arbitrary nature of limiting displacement selection (i.e., displacement at which the residual shear strength is attained), erroneous calculated cohesion and angle of friction are likely to result. This is particularly true in a test which yields a non-zero calculated cohesion. This study proposes a systematic way of determining the true residual value of the interface cohesion and the angle of friction of rock joints from laboratory shear testing by evaluating the limiting displacement. The laboratory shear testing is subdivided into two phases: 1) in the first phase, the normal load is applied in increasing stages to show the change in configuration of the sheared area with increasing normal loads, and 2) in the second phase, the normal load is applied in decreasing stages starting with a normal load simulating field normal stress. The joint residual shear strength is determined from the data obtained in this phase. A comparison of the residual shear strength obtained by using both the true area of contact and the area of the superimposed surfaces is included. An analysis of the deformational characteristics of the tested joints based on the works of Byeflee (1968), Roberds and Einstein 1978), and Farmer (1983) is presented. This analysis emphasizes the brittle-ductile transition behavior encountered during successive shearings.
