Crush pillars have been extensively applied on the Merensky Reef horizon since the late 1970's. Once in a crushed state, the residual strength of the pillar provides a local support function and must support the hanging-wall to the height of the highest known parting. The design process of crush pillars is mainly limited to specifying a width to height ratio (w:h) of approximately 2:1. It is also required that a pillar crushes close to the face, while the pillar is being formed. On many mines the crush pillar system is problematic owing to the difficulty of controlling pillar sizes. This is mainly caused by poor drilling and blasting practices. As a result, pillar crushing is not always achieved. Crush pillars are implemented at relatively shallow depth, the pillar dimensions have remained essentially unchanged over many years, and the impact of regional pillars and geological losses contributing to the regional behaviour of the rock mass are overlooked. In many cases the pillar system is the source of seismicity. In this paper, the influence of mining losses (potholes) and the use of sidings are discussed as possible contributors impacting on crush pillar behaviour. A limit equilibrium model implemented in a displacement discontinuity boundary element program is used to demonstrate crush pillar behaviour. The results are compared to the pillar behaviour at an underground investigation site, which supports the preliminary findings.
Crush pillars have been extensively used in Merensky Reef stopes. The key function of the pillar system is to prevent the occurrence of back-breaks (large-scale collapses in the back area of a stope) occurring as a result of hangingwall separation along parting planes or fractures. One such problematic parting is the Bastard Merensky Reef, situated between 5–45 m above the Merensky Reef. The pillar dimensions are selected such that the pillars should be fractured while being formed at the mining face. This is typically achieved when a pillar is cut at a width to height (w:h) ratio of approximately 2:1 (Ryder and Jager, 2002). Once crushed, the residual strength of the pillar provides the required support function.
Factors influencing pillar stress (i.e. mining depth, pillar width, mining height, percentage extraction) will impact on crush pillar behaviour. Du Plessis and Malan (2015) demonstrated how oversized pillars could potentially result in unpredictable pillar behaviour. The execution of a mining layout can impact on the size of the pillar cut at the mining face. This is demonstrated by the examples and case study presented in this paper. Similarly, the presence of potholes or blocks of unmined ground will influence pillar crushing. While these are common occurrences on most mines using crush pillars, these factors have historically not been associated with poor pillar crushing or pillar seismicity.