In this article, an enhanced flame spread model is used to simulate a rail car compartment fire test. The model was found to be able to reproduce the following experiential results: the predicted progressive burning locations are consistent with the experimental record; the predicted temperatures and heat fluxes at various locations essentially follow the measured trends; and the predicted onset of flashover is within 9% of the measured time of 180 s. The sensitivity of the predicted time to flashover is assessed using 18 fire scenarios in which the uncertainties in the measured material properties are systematically examined. The time to flashover is found to be most sensitive to changes in seat material properties. For the investigated rail car compartment, the impact of porosity of the overhead luggage rack structure on time to flashover is also examined and found to be significant for small ignition source fires.

No abstract

No abstract

No abstract

A method to predict time to flashover in ISO 9705 room corner fire tests based on cone calorimeter data is proposed in this paper. The method involves classification of materials and estimation of time to flashover as two steps in sequence. In the first step, the investigated material is classified into discriminating material groups. In the second step, the time to flashover is calculated with a derived formula for the corresponding material group, which contains material density, time to ignition and heat release rate collected from small‐scale experiments. Compared with two existing models, the proposed method has improved the prediction precisions in both the time to flashover and classification of material categories for a wide range of materials. Copyright © 2012 John Wiley & Sons, Ltd.