Résumé
The critical shear strength characterization of waste rock (WR) is crucial for assessing the stability of waste rock piles. However, laboratory testing is challenging due to the broad grain size distribution (GSD) that ranges from silts to oversized rock clasts. The maximum grain size in a sample is restricted by the size of the equipment and the recommendations by international standards. This implies that WR materials require their GSD to be altered using grading scaling techniques to fit into standard testing cells. The common geotechnical practice for scaling is based on the aspect ratio a = D/dmax, where D is the diameter of the sample and dmax the maximum grain diameter. However, different standards disagree on the minimum a to ensure a representative sample, ranging from a = 5 to 20. The main objective of this paper is to study the effects of a on the critical shear strength of granular materials through numerical triaxial tests in the frame of the discrete-element method (DEM) and laboratory triaxial tests on WR material. The samples used have a values varying from 5 to 20 and GSD of uniformity coefficients ranging from Cu = 1 to 2.2. The results show that the numerical and physical samples exhibit stable critical strength for a ≥ 12, regardless their initial grading. These results demonstrate the advantages of numerically testing materials that are challenging to fit into standard devices and suggest that international standards should be revisited to include the significant effects of a.