STP1082: Predicting the Toxic Hazard of Cable Fires

    Clarke, FB
    President, Benjamin/Clarke Associates, Inc., Kensington, MD

    van Kuijk, H
    President, Benjamin/Clarke Associates, Inc., Kensington, MD

    Steele, S
    President, Benjamin/Clarke Associates, Inc., Kensington, MD

    Pages: 11    Published: Jan 1990


    Abstract

    Despite a growing interest in cable fires, no systematic evaluation of the toxic threat from cables under actual fire conditions has been attempted. The present paper demonstrates the methodology used by Benjamin/Clarke Associates to estimate hazard of cable smoke. The method makes use of small-scale fire property tests and computer modeling. Certain of the predictions have been tested at full scale.

    The small-scale tests provide data on the response of the cables to a radiative heat environment, predicted by the mathematical model. In this work, the NIST cone calorimeter provided mass loss data and the LIFT apparatus was used to measure flame spread parameters. All cable fire data were obtained over a range of imposed flux. Bench-scale measurement of toxic potency of cable smoke was obtained; in some cases, full-scale smoke potency data were obtained as well. The modeling was done by modifying a room fire code (Harvard V) to allow for cable fires. This involved specific calculation of the incident flux to substrates in the hot upper layer of a room fire. Once the flux to a cable bed could be predicted, it was possible to calculate the rate of smoke evolution from the cables, making use of the small-scale test data described above. In a full-scale test of the method, good agreement between predicted and observed mass loss was obtained.

    The methodology was then extended to a series of four cables in two fire scenarios similar to burn studies carried out by the U.S. Navy. The predicted rankings of the cables in these two scenarios were compared with the rankings of the cables by: (1) a toxic potency test; (2) results of a toxic potency test adjusted by the amount of cable present in each case; and (3) a new smoke toxicity test, which makes some allowance for the fire properties of the material. It was found that the bench-scale toxic potency measurements were not good indicators of full-scale performance.

    Keywords:

    fire hazard assessment, fire modeling, cone calorimeter, flame spread, smoke toxicity


    Paper ID: STP23518S

    Committee/Subcommittee: E05.35

    DOI: 10.1520/STP23518S


    CrossRef ASTM International is a member of CrossRef.