The collapse of shallow coal mine workings

Abstract

The present study was -undertaken to investigate the mechanism of void migration and the collapse characteristics of old shallow surface (< 50m), pillar and stall coal mine workings. Simple stereo-photographic techniques have been employed to record these structures where they occur in the high walls of NCB and private opencast coal sites. Several relationships have been identified from this data, and the investigation concludes that the crushing of coal pillars at depth is rare and that the principal mechanism of failure involves the collapse of the roof material into the working. A classification of failure mechanisms based on the frequency and spacing of horizontal and vertical discontinuities relative to the span of the working is proposed. Two distinct situations for analysis are recognised. The first involves the stability of the immediate roof, while the second is concerned with the stability of the 'arch' that develops when the immediate roof beam collapses. Continuous roof beams have been found to be rare in Coal Measures rocks and therefore simple beam analysis is considered to be of little use. Where discontinuities are present Voussoir beam analysis may be appropriate, and Voussoir beam theory has been corrected and extended to overcome some of the problems recognised with the technique. Bulking and arching have been recognised as the 'normal' limiting factors on the height of collapse and are considered as complimentary failure mechanisms. For a 'typical coal mine collapse' situation arching is shown to be the dominant control. However, a review of arching has shown that in general all the theories underestimate the height of collapse. Thus, a statistically derived relationship of (collapse height = 2.68 X span of working) has been proposed as the limiting height for arching situations. Existing bulking relationships have been shown to be rather simplistic and appropriate corrections to the theories are suggested. An analysis of bulking factors derived from colliery discard has shown that a regional variation in this parameter is likely.