Simulation of a Direct Shear Box Experiment using EDEM


AUTHOR(S)

A. D. Orlando, H. H. Shen, J. Favier, S. Ji

PUBLISHER

Minerals Engineering International

SOURCE

Discrete Element Methods (DEM) 07

YEAR

ABSTRACT

EDEM and a Clarkson University in-house DEM code are used to simulate a direct shear box experiment and the results are compared. Spheres with uniform and lognormal mass distributions are used to measure the bulk friction coefficient of the particles. The physical parameters for both programs (i.e. particle size, mass distribution, material properties, box size and shear rate) are the same. The contact model includes a Hertzian elastic term and a viscous damping term. The elastic term is identical in both codes but the damping is not. The in-house code uses a fixed time step of roughly 1% of the Rayleigh time. Two different time steps are used in EDEM to test the effect this has on the results. One is 25% of the Rayleigh time and the other is the same as in the in-house code. This difference in time step affects the results significantly. When all parameters are set to be the same, the results from both codes are very close. Differences between the two codes may be attributed to the damping effect and the numerical treatment of the boundaries in the direct shear box.

KEYWORDS

comparison, shear box, time step