NUMERICAL PREDICTION OF GLOWING COMBUSTION OF GLUE LAMINATED LARCH WOOD HEATED BY CONSTANT RADIANT HEAT FLUX

Abstract

　To construct large-scale buildings, principal construction elements such as beams and columns shall be fire resistant. In case of timber structures, fire resistance is needed so that construction element shall withstand the standard fire for a specified time and so that the self-burning shall be stopped until the elements are cooled down to normal ambient temperature. Self-extinguishment is a complex phenomenon that involves various external factors and the properties of wood. Self-extinguishment is confirmed by experiments in practice. However, it is expected to predict the occurrence of self-extinguishment by analytical methods for use in design and development of wide varieties of wooden members.

　Based on the two-dimensional unsteady heat conduction equation, Harada et al. have created a model that incorporates the effects of reactions peculiar to wood combustion: moisture evaporation, pyrolysis of volatile components, glowing combustion (char oxidation), shrinkage and cracking.

　In this paper, the rate of decrease in the residual ratio due to glowing combustion was derived from the results of experiments on glulam heated by constant radiant heat flux using a cone calorimeter. The derived rate equation was applied to the heat conduction model to reproduce the cone calorimeter experiments. The calculated values showed that the samples burned more violently than the experimental value. The reason is that the analysis did not consider the oxygen concentration distribution within the char layer.

　The glowing reaction rate equation was modified in consideration of the influence of the oxygen concentration distribution within the char layer, based on the following assumptions. (1) The oxygen concentration distribution decays exponentially with depth from surface. (2) The effect of shape changes due to shrinkage and cracking of the char layer is ignored. (3) The change in the surface position of the sample is ignored.

　As a result of calculation using the modified glowing reaction rate equation, the maximum value of the surface temperature and the maximum value of the mass loss rate were in good agreement between the calculated value and the experimental value. On the other hand, the values of charred depth and burnout depth were smaller in the calculations than in the experiments in case of low heating intensity. When the heating intensity was 30kW/m², the calculation value and the experimental value of charred depth and burnout depth agreed very well. Further verification is required for the cause of the difference between the calculated value and the experimental value at low heating intensity.