| Summary: | 博士 === 國立高雄科技大學 === 工學院工程科技博士班 === 107 === Flashover is a significant phenomenon during the growth of fires within confined spaces and is indicative of conditions that cause severe injury and death. Due to the fire hazard involved by flashover, many studies have investigated this phenomenon. However, the mechanism of flashover is not adequately clarified. After flashover, the indoor combustion situation is affected, and the flame emerging is affected from opening on external wall. Therefore, this study will investigate the effect of pre-/post-flashover on indoor environment and external wall in compartment fires. Four experiments were accordingly conducted: Firstly, according to the fire test methods, specimens are traditionally mounted vertically as a wall or horizontally as a floor. The only exception is the ISO 9705 room corner test in which ceiling material is installed beneath a ceiling. This study was accordingly designed to discuss the test results of ceiling materials in the ISO 9705 room corner test with the testing capacity of the traditional tests to evaluate the feasibility of the traditional tests to rank materials mounted beneath a ceiling. Secondly, the practical relevance of the flashover theory and the practical methods to identify flashover have not been elucidated adequately. This study utilizes an experimental compartment in which ignited fuel exists without or with a secondary fuel, to clarify the necessity of sufficient fuel volatiles and secondary fuels for flashover. The length of the compartment was one-third of the length of an ISO9705 test chamber. In the beginning of the experiments, only the first fuel was ignited. The secondary fuel, if presented, received heat and released volatiles. Thirdly, a previous model for predicting the temperature, both before and after flashover, using adiabatic gas temperatures has been developed. This model has only been verified by experimental data before flashover. This study modified the expression of HRR at flashover, and verified by experiment in an enclosure whose dimension is one third of an ISO 9705 room. Finally, the heat exposure to the external facade from window-like opening locations has not been incorporated into current engineering fire design methods. This study performed experiments in which small-scale enclosures with various opening shapes and locations were underpinned for flames on facades emerging from ventilation-controlled fires at the floor of fire origin.
The results showed: (1) A penetration occurred in the gypsum test and led to a severe fire although flashover was not observed. The results from the traditional tests are obtained from tests that are primarily concerned of the potential of a material leading to flashover. The penetration of flames through ceiling materials cannot be assessed in the other tests. A modification of the traditional test is recommended when ceiling materials are tested. (2) The times to flashover without a secondary fuel approximately equaled with a secondary fuel. A premixed flame does not necessarily indicate flashover. Therefore, only thermal instability is responsible for flashover. (3) This model using the alternative expression can better predict the temperature after flashover. Additionally, the type of flashover was discussed with the prediction of post-flashover temperature. (4) The shape of opening strongly affected the combustion conditions. Opening geometry was divided into two categories: those that combust both inside and outside the enclosure, and combusting primarily outside the enclosure. Two sets of heat flux correlations were derived for the two categories. Additionally, external flame was higher for the category of combustion primarily outside the enclosure, because pyrolyzate primarily burned there.
|
|---|
