Fire Engineering Research Report 95/01

This project focused on the effects of scale on the backdraft phenomena. A full residential-scale compartment was constructed using a shipping container. An extensive amount of time and effort was taken, in order to design and construct a compartment in which backdrafts may be safely simulated and observed. To determine the effects of scale, results were compared with previous work (Fleischmann et al. 1993) which included quantitative backdraft experiments in a half residential-scale apparatus. The dimensions of the full scale compartment was constructed to give dimensions comparable to those of the half scale apparatus. For this comparison 21 experiments were conducted. Results include: energy flow rates, species concentrations (CH4, 0 2, C02), temperature and pressure histories, vent flows and gravity current velocities. Experiments retained consistent conditions, with only two controlled variables; gas flow rate and time. Varying these the total unbumt hydrocarbon (methane) concentration in the compartment prior to a backdraft was varied. The most significant uncontrolled variable was the ambient wind condition. This had a substantial effect on the results of each experiment. Conclusions that may be gained from this research are that scale has no obvious effect on backdraft. For a backdraft to occur the required total unbumt hydrocarbon concentration must be above 15%