A series of rock shear tests was conducted on artificial sandstones of uniform and weak nature. The obtained shear strengths appeared to underestimate those by triaxial compression tests slightly in a low stress range. This marginal discrepancy was discussed based on the measured distributions of stresses and strains around the shear plane in the rock shear tests.
Des tests de cisaillement ont ete realises sur des gres artificiels uniformement friables. Les va leurs obtenues dans Ie cas du cisaillement semblent legerernent sous-estimer celles issues des tests en compression triaxiale dans une plage de faibles contraintes. Cette difference marginale fait I'objet d'une discussion basee sur la distribution des contraintes autour du plan de cisaillement au cours des tests de cisaillement.
An gteichformigern.: schwachem kunstlichem Sandstein wurde eine Serie von Felsscherfestigkeitsversuchen durchgefuhrt. Die erhaltenen Scherfestigkeiten waren im niedrigen Spannungsbereich gegenuber den durch dreiachsige Kompressionsversuche erhaltenen Werten erwas unterbewertet. Dieser Unterschied wird auf der Grundlage der in Felsscherfestigkeitsversuchen gemessenen Verteilungen der Spannungen und Verzerrung diskutiert.
As Japan is located in the midst of seismically active areas, geotechnical investigation of foundation ground for nuclear power plants (N PPs) is considered as being of utmost importance for a successful earthquake-proof design. It has been a common practice that large-scale field tests of various kinds, including plate loading tests and rock shear tests, are conducted on the spot in exploratory adits excavated beneath the prospective sites for reactor buildings. Due to limited locations on the land suitable for future plants, Japanese power industry is now seeking possibilities of siting NPPs on offshore man-made islands. This requires new geotechnical investigation technology for foundation ground under the sea, which may preclude the use of exploratory ad its but should at least be comparable qualitatively to the cases on the land. Conceivably, characterization of offshore foundation ground should inevitably resort to conventional large-scale field tests utilizing a huge caisson, smaller-scale field tests conducted in boreholes, and laboratory tests using drilled cores. If sedimentary soft rocks have undergone no weathering and are fairly uniform, the influence of discontinuities, if any, may not be overly significant. Thus it is justified to regard rock mass behavior as not deviating too far from that predicted by a continuum model based on laboratory test data. Recent research has revealed that, pre-failure deformation of soft rock ground can be evaluated appropriately, provided that its non-linear stress ~ strain relationship is accurately characterized by carefully conducted laboratory element tests using high-quality undisturbed samples taken from the ground (Tatsuoka & Kohata, 1995). Crosschecking is recommended by field measurement of shear wave velocities for elastic moduli and also by pressuremeter tests for degraded stiffness at moderate strains. However, there still remains a question whether the same scenario holds good for strength characterization. According to the previous experiences on sedimentary soft rocks and artificial weak rocks, shear strengths evaluated by laboratory tests on small drilled cores are, in most cases, found to be higher than those by large scale field tests. This tendency is more pronounced for fine particle materials, i.e. siltstones ∼ mudstones. Field shear tests, i.e. rock shear tests, were reported to yield strengths as low as one-third to three-quarters of those by laboratory shear tests, i.e. triaxial compression tests (e.g. Miyaike et al., 1993: Sakai et al., 1990). Whereas for large particle materials, i.e. sandstones, the discrepancies were much less. In some cases, somewhat comparable strengths were obtained by both field and laboratory shear tests (e.g. Kudo et aI., 1992). For hard rocks, a well-known scale effect is taken for granted to explain different strength values obtained by tests on different specimens in size. Small rock cores are not considered as good enough to represent mechanical behavior of large rock masses. Joints and other discontinuities are attributed to be responsible for lower strengths obtained by tests of larger scales. This scale effect, however, can hardly be expected to become significant for sedimentary soft rocks, particularly for artificial weak rocks, which rarely include any kinds of local weaknesses.