Analysis of the stress distribution in a laminar direct simple shear device and implications for test data interpretation

Abstract

Direct simple shear (DSS) testing allows observation of load-deformation response under rotation of the major principal stress plane, which is descriptive of many actual field problems. While the simplicity of the test configuration makes its use popular in research and industry, key uncertainties still remain regarding the interpretation of the laboratory data. This study uses laboratory validated discrete element method (DEM) models to examine the stress transmission in laminar-type direct simple shear devices under drained constant effective stress conditions. The DEM models (comprised of spheres) closely replicate physical specimens of precision chrome steel ball bearings for which the properties (e.g., shape, surface friction, and stiffness) were measured directly. The DEM models were also validated using experimental tests, so that conclusions regarding the system response can be derived with confidence from the available DEM data. The testing program included both loose and dense specimens, allowing for a comparison of the influence of density on stress state which has not been examined in previous simple shear DEM studies. Differences were observed between vertical effective stresses and shear stresses derived from boundary measurements (as commonly carried out in experimental programs) and those derived from force measurements within the DEM specimens. The failure state of the material in simple shear was also examined through Mohr’s circles of stress. The evolution of stresses on both the horizontally and vertically oriented planes were considered so that established methods of direct simple shear interpretation could be critically assessed. For the loose specimens, the angle of shearing resistance can be confidently estimated considering the maximum shear stress acting on the horizontal plane, which is easily inferred from measurements of the shear force during the physical test. This was true considering both internal and boundary calculated stresses. This approach, however, is inaccurate for the dense specimens. Analysis of the particle-scale kinematics of the response illustrates that the deformation field within the central portion of the specimen is in simple shear, although the magnitude of this shearing was significantly larger than what was measured on the boundary. This study and the conclusions derived focus on smooth spherical particle specimens; the objective was to examine the stress distribution within DSS devices and the implications for test interpretation using DEM models that more closely matched the physical laboratory specimens tested than in previous studies. When considered alongside the existing studies, the findings show that there is no broad conclusion that can be applied for all materials and all conditions in simple shear and that interpretation should be carefully tied to the physical conditions simulated.