Citation

Khan MM, Alpert RL. J. Fire Sci. 2003; 21(1): 41-54.

Copyright

(Copyright © 2003, SAGE Publishing)

DOI

10.1177/0734904103021001004

PMID

unavailable

Abstract

An empirical flame spread parameter (dimensional), which is a function of peak chemical heat release rate and ignition behavior, obtained from bench-scale ASTM E-2058 tests, is used to correlate the flame spread behavior of eighteen thin polymeric rooflight materials in large-scale ASTM E-108 tests. Although this empirical parameter appears to correlate the extent of flame spread in large-scale tests of most rooflight materials, including melting types, the parameter fails to predict the flame spread behavior of several nonmelting materials. Data have also been analyzed using a dimensionless parameter, based on an existing concurrent flame spread model that includes the consideration of burnout for thin materials. This parameter, a flame acceleration factor (FAF), is a function of ignition time, burning duration and chemical heat release rate obtained from E-2058 tests. A correlation between FAF and the extent of flame spread measured in large-scale E-108 tests is presented. Consistent with the model prediction, flame spread remains within the flame exposure region in E-108 tests when FAF <0. Flame spread is distinctly beyond the exposure source when FAF >0. An exception is the case when there is extensive melting, a phenomenon not taken into account by the FAF model.

Language: en