Pat is correct. It is a heat transfer mechanism that is important. Matching the bare bulb surface temperature with a known ignition temperature for a specific material is not a scientifically reliable methodology.

Dimensional lumber has an ignition temperature that is much less than the bulk diffusion flame temperature. Now think about trying to ignite a 2 x 4 with a Bic lighter. An analysis of the flame temperature vs. the ignition temperature will predict ignition. A heat transfer analysis would not predict ignition. An experiment would be consistent with the no ignition.

In addition, changing the boundary conditions will change the heat transfer, which can change the ultimate result. Covering the surface of the bulb with a cotton sock will change the heat transfer due to changing the energy balance at the surface.

Although NFPA 1033 does not explicitly call out heat transfer as preresiquate knowledge, an investigator has to understand heat transfer to be able to practically and reliably apply such to the analysis of fires.

Douglas J. Carpenter, MScFPE, CFEI, PE, FSFPE
Vice President & Principal Engineer
Combustion Science & Engineering, Inc.
8940 Old Annapolis Road, Suite L
Columbia, MD 21045
(410) 884-3266
(410) 884-3267 (fax)
www.csefire.com