A research and testing facility has been developed at Simtars to allow the spontaneous combustion testing of coal on a large scale. In this adiabatic reactor, the coal particle sizing is closer to that which could be found in the goaf and therefore the reaction is more likely to simulate the underground situation. This allows aspects of the spontaneous combustion process and methods for detecting a heating to be investigated which cannot be done in an operational mine. Up to 18 tonnes of run of mine or crushed coal is loaded into the reactor and allowed to self-heat while the temperature and gas profiles within the coal pile are monitored. The gases evolved from the coal are analysed using a fixed gas monitor system and gas chromatography to determine which of the common mine fire indicators and ratios can be used to predict the onset and progress of coal heatings. Benchmarking of the inherent propensity of a coal to spontaneously combust at both the large- and laboratory-scale is also possible by comparing the sample against previously tested samples._x000D_
It has been found that the coal particle size affects both the time it takes for the coal to react and the location of the developing heating. The heatings generate approximately one metre from the inlet but in the crushed coal case do not progress to an advanced heating stage until the hot spot' migrates towards the coal surface where there is an increased oxygen supply. Physical properties such as the compressive strength of the coal also affect the heating size and the time it takes for a heating to develop. Higher strength coals appear to oxidise more slowly and therefore take longer to self-heat as oxidation of the coal is the heat-generating step in the spontaneous combustion process._x000D_
The best indicators of the state of the heating were found to be CO make and Graham's ratio which are independent of air flow. Investigation of a potential spontaneous combustion should begin as soon as the CO make or Graham's ratio exceed the background levels. The results from the tests undertaken on a number of Queensland and NSW coals are reported here together with the implications for the mining industry.