This paper describes observed crush pillar failure which occurred in a stope 550 metres below surface two years after a portion of a regional stability pillar was partially extracted.
Two different modes of failure were identified, one being a stable crushing mechanism, the other an unstable explosive mechanism. Seismic data recorded from this area supports the observations that some pillars failed violently. Investigations undertaken suggest that the presence and intensity of jointing is one of the controlling factors influencing pillar behaviour.
Impala workings are supported with a pillar system consisting of two components. Firstly a series of dip orientated barriers with a secondary system of small pillars between them. Barrier pillars are left to provide regional stability whereas the small pillars provide support for the immediate hangingwall up to a distinct parting approximately 15 metres above the reef.
Since tremors have sporadically been felt on Impala, the recent installation of seismic monitoring devices has made the location of these events possible. Soon after commissioning a triaxial geophone, the clustering of events in a stope in which mining ceased a few years ago initiated a visit to the stope to assess whether damage was occurring. The underground vi- Sit revealed that all pillars had failed, some in a stable manner and others in a violent manner.
In this paper the investigations which were undertaken, to offer some explanations for the happenings in this particular stope, are described. Based on the findings some tentative conclusions are put forward.
The Rustenburg and adjacent area is sub- Ject to natural earth tremors. Between September 1920 and January 1988 twenty five tremors were recorded. Magnitudes ranged between 1.8 and 4.8 on the Richter scale. At Impala itself, tremors have been felt on surface since the late seventies. With the absence of underground damage, it was impossible to determine where the events were taking place or what was causing them. In 1980 an unsuccessful attempt was made to monitor and locate the source of this seismicity. In April 1989 a triaxial geophone was installed and commissioned close to number 1 shaft at the southern end of Impala's lease area where most of the tremors were felt.
After the first month of operation, thirty events had been recorded and located in the immediate area of the geophone. Their magnitudes ranged from -1.2 to 1.0 on the Richter scale. The most striking feature of that initial data was the clustering of fourteen events in a specific area of the mine that had been worked out a number of years previously.
In the months that followed, further seismic events were recorded in this same area, but never with the same frequency or distinct clustering of that first month. Having identified this specific area, an underground visit was made to inspect the workings to try and establish possible causes of the seismicity, bearing in mind that mining ceased a number of years ago.
