ABSTRACT

The Hoek-Brown Strength criterion, and associated parameter relationship expressions with GSI, has proven remarkably successful for accurately defining rock mass behaviour for most practical engineering situations, where block size and discontinuity controlled shear failure dominates ground behaviour. However, difficulties have been encountered with application of the strength criterion at the extreme ends of the rock competence scale. In the midrange of this scale, block size and incipient strength is such that rockmass behaviour tends to be controlled by inter-block shear strength rather than by material strength. At the low end of the scale (for very weak rocks with UCSi _10MPa) rock mass strength conforms with matrix strength and structure has minimal impact. At the high end (GSI_65 and m i _15) in situ rock mass strength for a given rock type reaches a maximum controlled by spontaneous crack propagation (spalling) after crack initiation for rocks with high mi and by crack accumulation, interaction and coalescence of cracks (resulting in matrix shearing) for rocks with low or moderate mi values. Transition relationships are introduced on the basis of material behaviour as a basis for better defining the full range of GSI related Hoek-Brown parameters m, s and a.

1 INTRODUCTION

Rock mass classification systems, including the GSI system, for rockmass strength estimation (Marinos&Hoek, 2000), are based on the principle that structure within a rockmass acts to reduce both the cohesion and frictional properties, represented by a degradation in "s" and in "m" respectively, in the Hoek- Brown non-linear criterion (Hoek et al, 2002). The Hoek-Brown criterion admirably handles normal rockmass behaviour for rock excavations, both for underground and surface applications. At the two ends of the rock competence scale however, (for very low strength rocks and for spall-prone, high GSI rockmasses) difficulties have been experienced in using the criterion, largely because outside this range rockmass behaviour becomes less controlled by discontinuities. (Figures 1 and 2). Figure 1. Comparison between (left) shear failure behaviour under squeezing conditions in a low GSI rockmass (courtesy of E. Hoek); and (right) brittle spalling and strain bursting failure behaviour in a high GSI rockmass.

2 EXTENDED APPLICABILITY SUGGESTIONS

This paper builds on two recent publications (Carvalho et al., 2007 and Diederichs et al., 2007) that address the two ends of the rock mass competence scale where current classificationbased strength criteria face limitations in characterizing a rockmass in a way that is consistent with its behavior. The same holds true at the other end of the competency scale, where rockmass behaviour again approaches that of the intact material, rather than that of a jointed rockmass. In a manner analogous to the fundamental behavioural difference recognized in fluid mechanics between flow under laminar and turbulent conditions.

This content is only available via PDF.
You can access this article if you purchase or spend a download.