Premium
Heat release testing of stacked chairs: Analysis of repeatability in a single laboratory
Author(s) -
Hirschler Marcelo M.,
Treviño Javier O.
Publication year - 1997
Publication title -
fire and materials
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.482
H-Index - 58
eISSN - 1099-1018
pISSN - 0308-0501
DOI - 10.1002/(sici)1099-1018(199703)21:2<85::aid-fam598>3.0.co;2-l
Subject(s) - repeatability , environmental science , smoke , combustor , calorimeter (particle physics) , gas burner , stack (abstract data type) , forensic engineering , nuclear engineering , combustion , waste management , engineering , statistics , mathematics , chemistry , computer science , programming language , electrical engineering , organic chemistry , detector
Fire testing of furniture has been a focus of much attention in the 1990s, particularly in terms of the heat it releases in real‐scale fires. One aspect that has received insufficient attention is the case of stacked chairs, often found in places of assembly. Early data have shown that the fire hazard resulting from burning a stack of chairs is often much higher than would have been expected simply from the combined hazard of the individual chairs. A test has been designed, ASTM E1822, in which the flames from a propane gas burner (at 12 l min −1 ) are applied to a vertical stack of 5 chairs for 80 s, and the important fire properties measured. This test was developed with the collaboration of two laboratories; the variables studied were: number of chairs in the stack and duration of exposure. In order to evaluate the repeatability of the method, one laboratory tested six different types of chairs, in triplicate, in a furniture calorimeter. The chairs were donated by different manufacturers for this purpose. A statistical analysis was conducted on seven properties (peak rate of heat release, total heat released, peak rate of smoke released, total smoke released, mass loss, time to peak rate of heat released and initial mass). The overall relative standard deviations for the properties studied ranged between 1% and 20%, which is adequate for a fire test. Thus, the repeatability study was successful. It is of interest that very poor correlation was found between mass loss and heat release, so that the highest fractional mass loss corresponded to the chairs with the lowest heat release rate. Additionally, there is also relatively poor correlation between heat and smoke release rates. The peak heat release rate of the stacks of chairs ranged from almost 300 kW to >1 MW, and the chairs lost between 4% and 40% of their initial mass; only one set of chairs did not release enough heat in any test for a room containing it to go to flashover. Thus, the results suggest that the potential exists for severe fires to develop when such chairs burn. © 1997 John Wiley & Sons, Ltd.