Fire Safety Science Digital Archive

Recent testing by the Building and Fire Research Laboratory of the National Institute of Standards and Technology collected a large dataset of species, temperature, velocities, and heat fluxes for a wide range of fuels burning at varying degrees of ventilation inside a reduced scale enclosure. This dataset is extremely well documented with uncertainties well characterized, making it an excellent choice for fire model validation. Selected data from a subset of tests was used as part of a validation exercise using FDS v5. FDS v5 contains a new method of decomposing the mixture fraction coupled with a new combustion model that allows it to predict (rather than assume) CO formation in under-ventilated fires. Eleven tests were simulated and predicted vs. measured data comparisons were made for seven physical quantities distributed over twelve measurement locations. Results show the new combustion model is able to reproduce test conditions over a range of fuels and fire sizes.