| Abstract | A detailed description of the existing National Research Council of Canada's (NRC) one-zone fire growth model is presented. Several changes to the model are recommended based on recent experimental observations. The model was compared to the results of three full-scale fire tests performed withdifferent loadings of flexible polyurethane foam in an ASTM/ISO standard room. An accurate prediction of the experimental mass loss rate was found to have the greatest importance in predicting overall fire behavior. The experimental mass loss rate was observed to increase more quickly than the predicted rate and to decay more slowly as a function of fuel depletion. These differences were attributed, respectively, to an underprediction of the flame spread rate on the fuel bv the model. and to an (assumed) tendency of polyurethane f oams to resist oxygen penetration and flame spread as burning progresses. For flashover fires, the inability of the one-zone model to distinguish between the hot ceiling layer and the rest of the compartment resulted in an underprediction of the external heat flux to the fuel, which further impeded the growth of the model fire. Increasing the flame spread rate by a factor of 2.3. introducing an appropriatedecay function for the mass loss rate, and-incorporating a hot layer temperature in the estimation of the heat flux to the fuel resulted in excellent predictions of the mass loss rate as well as all the other experimental results, demonstrating the suitability of the proposed modelling. |
|---|
