Quantification of the thermal hazard from metallic and organic dust flash fires

Abstract

An in-house constructed nominally 20-L flash fire apparatus was used to evaluate and compare the flash fires fueled by an organic dust (non-dairy coffee creamer) a metal dust (aluminum) and a flammable gas (methane). Dispersion was achieved using an injection system similar to the injection systems found in standard Siwek 20-L combustion chambers and a 10-J spark igniter was used to ignite the fuels. A heat flux gauge, thermocouples, a high-definition video recorder and an infrared wavelength camera were used to evaluate the severity of the flash fires. Multiple concentrations of dusts and a stoichiometric methane mixture were tested. All measurement methods showed reasonable agreement when ranking the severity of different deflagrations, but thermocouple and heat flux gauge measurements were sensitive to the position of the flame relative to the sensors, leading to some inconsistency. Analysis of IR measurements taken at 30 Hz provided fireball dimensions, growth rates, and relative temperatures for dust-fueled deflagration, but were unable to accurately assess the high-turbulence premixed methane flash fires due to the high flame speeds and 30 frames per second limitation. The IR camera was also limited to a maximum temperature of 1200 °C, which is inadequate for the temperature of some metal dust deflagrations, including aluminum and requires adjustment of the material emissivity, which would require additional analysis and testing. Measurement strategies for a next-generation flash fire testing apparatus are proposed based on the results of this study.