FIRES IN REAL SCENARIOS

Studies have been carried out to determine the effect of sprinklers on fires typical of a number of occupancies including simulated of office furniture, supermarkets, carpet displays, libraries, video stores and liquor stores. After surveys of actual fire loads, the experiments were conducted in a specially designed sprinklered fire-calorimeter with a collecting hood 6 m x6 m leading to a vertical duct 1 m in diameter. Details of the rig are given elsewhere [1]. Only well ventilated fires were studied. Rate of heat release and production of various toxic chemicals were monitored during the tests. Both sprinklered and unsprinklered fires were used. The results were used to establish the unsprinklered burning behaviour and the fire-control effects of sprinklers. Before sprinkler operation, the rate of fire growth could be modelled as 12_fires as given in NFPA 92B (1991 Edition) [2]. It was found that operation of sprirnklers controlled but did not extinguish the fires. This was expected as parts of the fire load were shielded from the spray. Also there were significant increases in the concentration of carbon monoxide when the sprinklers operated. Sprinklers had little effect on the concentrations of other toxic products measured. The results from the tests were extrapolated to large single storey buildings for the same occupancy classes and the results used to compare the required and the available escape times for different occupancies, particularly whether the use of sprinklers would improve the chances of escape from those premises. It was found that in most of the cases studied, adequate escape times will be available without any special measures. For very rapid fire growths, however, special measures, such as availability of trained staff may be needed. Standard response sprinklers will have little impact.