Journal of Fire Sciences

The mechanism of flashover in a compartment fire is investigated using nonlinear theory through a two-layer zone model. To simplify the illustration of the approach, there is only one variable, i.e. the upper smoke temperature, with the smoke layer interface height being fixed. Effects of different factors, such as the geometry of the compartment, the thermal and chemical properties of the combustible content, and the ventilation conditions, affecting flashover are discussed. Results show that the occurrence of flashover depends on many factors, among which, heat release rate is the most important.

A series of experiments was conducted in a small-scale wind tunnel to investigate the effect of forced flow on forward smoldering propagation in flexible polyurethane foam material located horizontally. In the wind tunnel, the air at a flow rate in the range of 0.01-1.5m/s passed over the foam layer surface. Foam samples of different lengths with the same cross-sectional area were studied. From the experimental results, it is seen that the foam length has no clear influence on the smolder propagation velocity. With an increase of air flow rate, firstly, the smolder propagation velocity increases and reaches a maximum at an air flow rate of around 0.8 m/s due to the increased mass transfer of oxygen into the smolder reaction region. It decreases with further increase of air flow rate as heat losses becomes dominant compared to the positive effect of the increase of oxygen transfer into the reaction zone. The variation of the steady smoldering temperature with the air flow rate follows the same trend as that of the smolder propagation velocity. In this paper, a simple analysis is also presented to analyze the co-effect of the mass transfer of oxygen to the smoldering reaction region and the heat loss to the surrounding on the smolder propagation under our current experimental conditions.

A series of fire experiments was performed to study the gas temperature development and charring behaviour of timber construction compartments. The measured gas temperatures were 300-500 C lower than predicted by the parametric temperature-time curves of Eurocode 1 for unprotected and insufficiently protected timber structures. For structures protected with a double layer of gypsum plasterboard, the agreement was better. In all cases, however, the prediction was on the safe side. One protective layer of gypsum plasterboard delayed the onset of charring for 20 min. Using two layers, the delay time was doubled. Reduced charring rates for long exposure times were observed, resulting from the increasingly protective effect of the char layer withtime.