Evaluation of the Onset of Flashover in Room Fire Experiments

Two series of full scale room fire tests comprising 16 experiments are used for a study of the onset of flashover. The fire loads were varied and represented seven different commercial applications and two non-combustible linings with significantly different thermal inertia were used. The test results showed that by lowering the thermal inertia and thereby lowering the heat loss from the room and at the same time increasing the thermal feedback, a thermal runaway occurred before significant fire spread; but only for objects composed of a mixture of plastic/rubber/textiles and wood/celluloses. In these cases the onset of thermal runaway was found to occur at room temperatures in the range 300C to 420C, supporting that the room temperature at the onset of thermal runaway is strongly dependent on the thermal inertia. It also shows that the onset of thermal runaway cannot in all cases implicitly be predicted by the traditional flashover temperature criterion of 500C to 600C. For fire loads composed of pure wood/celluloses the onset of flashover occurred about the same time as fire spread irrespectively of linings and at significantly higher room temperatures (725C). This can be explained by flammability parameters making wood/celluloses less sensitive to thermal feedback.