Abstract
A series of laboratory tests were conducted on matched rock joint samples, no larger than 16 x 16 cm2 of section, which were extracted from an artificial joint, having 4.32 m2 in area, carved in a porphyritic granite block. These tests (1200 pull tests and 200 trials conducted in a sliding machine) involved the systematic levelling of sample middle planes and lead to conclusions that are discrepant with respect to conventional ideas admitted about rock-mass joint mechanics. Those discrepancies are: a) Larger matched samples showed higher strengths in the dilating phase of the sliding tests rather than those from small matched samples; b) Sample shear strengths probably depend on the transverse widths, when JRC, JCS and n (the average normal stress on the rock mass joint) are high, thus inhibiting the use of stability analysis by common slope stability methods such as Fellenius’; c) At the dilating sliding phases, the mechanics of matched joints is essentially different from that of mismatched joints, as the former brings about inverse scale effects (represented by positive exponential regressions) and the latter involves normal scale effects (represented by negative exponential regressions). The results obtained upon those lab tests do not agree with those reported from in situ experiments, as well as the actual behaviour of natural joints. The obtained moderate correlation coefficients do not allow the consideration of these findings as physical laws, nevertheless they do represent certain types of rock mass joint behaviour, or simply useful generic rules. Thus, the subject is full of surprises, as the authors show in text.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have