Shape effect of shear box in large-scale direct shear test for soil-rock mixture

Abstract

The large-scale direct shear test is one of the simplest and most efficient methods for testing the shear strength parameters of soil-rock mixtures (S-RM), with shear boxes available in various shapes such as circular and rectangular. However, it remains unclear whether the shear box shape affects the accuracy of the results from large-scale direct shear tests. Therefore, this study uses a three-dimensional (3D) discrete element method to construct a numerical model for large-scale direct shear tests on S-RM, accounting for realistic rock block shapes. Subsequently, the reliability of this numerical model was validated by comparing it with results from large-scale direct shear tests laboratory. Next, the direct shear test results of the same S-RM were compared between circular and square shear boxes. The results indicate that the test results in the square shear box are more accurate. In the circular shear box, the test results approximate those in the square shear box, but their deformation is not a strict plane strain. Finally, the influence of the length-width ratio (L/W) of the shear box on results of large-scale direct shear tests on S-RM is discussed. Results show that as the L/W decreases, the interlocking effect between rock blocks strengthens, the number of force chains increases, and the shear band thickness decreases; the stress-strain curve transitions from strain-softening to strain-hardening; both cohesion and internal friction angle exhibit two distinct trends within the range before and after the L/W reaches 1:1. Thus, in large-scale direct shear tests of S-RM, square shear boxes with a L/W of 1:1 are preferred, followed by rectangular shear boxes with a L/W greater than 1:1, while shear boxes with a L/W less than 1:1 should be avoided.