Case studies illustrate the need for a new concept of coal pillar design

Introduction The most commonly used ground control design in coal mining is the determination of pillar size by various pillar strength formulae for mine layout (Bieniawski, 1983; Holland, 1973; Hustrulid, 1976; Obert and Duvall 1967; Wilson and Ashwin, 1972). However, there have been few documented case histories concerning the validity of those strength formulae. Recently there are indications that those strength formulae tend to overdesign considerably. From a safety point of view, there is nothing wrong with overdesign. But an overdesign with large chain pillars reduces the flexibility of machine movement and decreases production. It also reduces coal recovery and, thus, shortens the reserve's life. This is especially true for small operators who own a limited reserve. Those small mines, through past practices or out of necessity, experiment with smaller pillars than those predicted by the strength formulae. Conclusive ev¬idence has been collected that points to the need of a new concept of coal pillar design. Two case studies will be cited to illustrate this point. 3-D US pillar strength formulae Figure 1 shows a sectional layout of a one-unit mine. The seam was about 1.3 m (50 in.) thick with a uniform cover of 64 m (210 ft). A house was located near the center of the section. The homeowner demanded that no surface subsidence and domestic well dewatering should ever be allowed as a result of underground room-and-pillar mining. Physical properties - strength and Young's modulus - of the rock strata including coal were obtained by testing small samples in the laboratory. The uniaxial compressive strength of the coal was 3660 kPa (823 psi). Using various pillar strength formulae, the homeowner recommended that square chain pillars be at least 11 m (35 ft) wide. The coal company, however, desired to pursue optimum coal recovery in the safest and most economical manner. They engaged in three-dimensional finite element analysis of the section taking into account the complete stratigraphic sequence of the overburden. Results indicated that the section can be developed by 5.5 m (18 ft) wide entries at 21 m (70 ft) center-to-center. In retreat, the pillars can be splitted along both centerlines leaving a stump pillar of 5.5 m (18 ft) square at each of the four corners. The 5.5 m (18 ft) square stump pillars still maintained a safety factor of 1.5, although the safety factors calculated by various pillar strength