Resume/ Abstract
Since the pioneering works on large triaxial testing conducted in the 1960s in Mexico, USA and Germany, the reliability of sample small-scaling techniques for coarse soils and rockfill materials has remained a subject of great interest. Over the years, numerous experimental methods have been proposed and significant data has been compiled.
However, researchers have often reported conflicting experimental observations, leading to apparent contradictions regarding size effects on the mechanical behavior of coarse materials.
This study aimed at elucidating the factors at the source of size effects. The effects of several material properties and testing conditions are systematically studied based on comprehensive large-scale laboratory and in-situ tests on mine waste rock, rockfills and gravelly soils. The experimental program assesses the effects of grading scaling, sample size and representative elementary volume (REV) conditions.
The findings suggest that a REV of coarse material can only be achieved if the maximum particle size is limited to approximately 1/12 of the triaxial sample diameter, which is more restrictive than current testing standards. It is essential to confirm that the characteristic particle shape of scaled samples remains consistent with that of the original coarse material.
Under these conditions, the critical shear strength of loose samples can be properly characterized using small-scaling techniques.
Understanding these limitations is crucial to avoid optimistic soil property estimates and to ensure reliable geotechnical engineering designs for structures such as rockfill dams and mine waste rock piles.