This study was designed to be an initial step to investigate the potential of low flow nozzles as part of a retrofit flashover prevention system in residential homes with limited water supplies. Not all homes have water supplies that can meet the needs of a residential sprinkler system. Current alter- natives, such as including a supplemental tank and pump, increase the cost of the system. These homes could benefit from an effective fire safety system with lower water supply requirements. The experiments in this study were conducted in a steel test structure which consisted of a fire room attached to a hallway in an L-shaped configuration. Three types of experiments were conducted to evaluate nozzles at different flow rates and under different fire conditions. The performance of the nozzles was compared to the performance of a commercially available residential sprinkler. The first set of experiments measured the distribution of the water spray from each of the nozzles and the sprinkler. The water spray measurements were made without the presence of a fire. The other two sets of experiments were fire experiments. The first set of fire experiments were designed to measure the ability of a water spray to cool a hot gas layer generated by a gas burner fire. The fire source was a propane burner which provided a steady and repeatable flow of heat into the test structure. Two water spray locations were examined, in the fire room and in the middle of the hallway. In each position, the burner was shielded from the water spray. The results showed that for equivalent conditions, the nozzle provided greater gas cooling than the sprinkler. The tests were conducted with a fire size of approximately 110 kW, and water flow rates in the range of 11 lpm (3 gpm) and 19 lpm (5 gpm). The second set of fire experiments used an upholstered sofa as the initial source of the fire with the water spray located in the same room. As a result of the compartment size and water spray distribution, the nozzle flowing water at 23 lpm (6 gpm) provided more effective suppression of the fire than the sprinkler flowing 34 lpm (9 gpm) did. The nozzle was similarly effective with the ignition location moved 1.0 m (3.2 ft) further away. However, the nozzle failed to suppress the fire with a reduced water flow rate of 11 lpm (3 gpm). The results of this limited study demonstrate the potential of low flow nozzles, directly flowing water on to the fuel surface, with the goal of preventing flashover. Additional research is needed to examine larger room sizes, fully furnished rooms, and shielded fires to determine the feasibility of a reduced water flow flashover prevention system.
Calculation of Critical Water Flow Rates for Wildfire Suppression