An Experimental Investigation on the Influence of Thermal Feedback on Pre-Flashover Fires
The aim of this project is to perform an experimental study on the influence of the thermal feedback on the burning behavior of well ventilated pre-flashover fires. For the purpose an experimental method has been developed. Here the same identical objects are tested under free burn conditions and in two different rooms, which only are varied by linings of significantly different thermal inertia. As all linings were non-combustible the heat release rate could be found without the influence of thermal feedback and for two different levels of thermal feedback. The ISO 9705 Room Corner Test facility was chosen as the same measuring equipment could be used for all the tests. Using this method, 10 experiments were performed with three different sizes of heptane pools and three experiments were carried out with a block of flexible polyurethane foam. In addition to these 13 experiments, 16 experiments carried out by Carleton University and NRC-IRC performed on seven different types of fire loads representing commercial premises, comprise the tests used for the study. The results show that for some of the room test the heat release rate increased due to thermal feedback compared to free burn for a pre-flashover fire. Two phenomena were observed, that relate well to theory was found. In an incipient phase the heat release rate rose with the temperature of the smoke layer/enclosure boundaries. This increase was also found to depend on the flammability properties of the burning object. The results also documented a simple relation that can be used for estimating the impact of thermal feedback for pre-flashover design fires. A rapid increase of the heat release rate commenced after the incipient phase. This is seen as thermal runaway caused by the energy gain in the smoke layer exceeding the energy that can be lost through the boundaries. The increase of the heat release rate after the onset of thermal feedback did not seem to be dominated by either temperature of the smoke layer/enclosure boundaries or the type of materials of the burning object. The onset point of thermal runaway was found to depend on the thermal inertia of the linings as well as the flammability parameters of the burning object. This correlates well with theory. At the onset point of thermal runaway the smoke layer temperature was found to be as low as 300°C for linings with very low thermal inertia, which makes the onset point significantly below the traditional flashover criterion for the smoke layer of 5-600°C. This indicates that caution should be used when using this criterion for rooms with very low thermal inertia. Given the profound difference between room burn conditions and free burn, the results show that free burn results should also be used with cation for prediction of pre-flasover design fires in rooms.