Fire Safety Science Digital Archive

The present methods for predicting the occurrence of flashover are reviewed. These methods are based on a closed-form estimate of the upper layer gas temperature which corresponds to the occurrence of flashover using correlations derived from a wide range of test data. It was found that these methods gave reasonable predictions of flashover for enclosures within the range of test data from which the correlations were derived. However, the predictions were found to be not suitable for predicting rapid burning or conditions of flameover due to fire plumes and not the hot gas layer. It was also found to be not suitable for use in high ceilings because cooling of the upper gas layer from plume entrainment was not considered. None of the methods appropriately addressed the time for occurrence of flashover which is an important determinant for assessing life safety. In this paper, a more general methodology for determining both the conditions for flashover to occur and an estimate of the time at which it could occur is outlined. The proposed method is more fundamentally based and can therefore be extended to predict conditions of flashover (and flameover) beyond the limits of the existing methods. Application of the proposed method is presented for predicting the occurrence and time of flameover from a burning storage rack to an adjacent exposed rack based on a developing fire plume. The use of the method in determining the likelihood of occurrence of flashover in high ceiling enclosures is also presented.