RESUMEN
The obscuration thresholds for various smoke detectors and combustibles, required as an input parameter in fire simulation, were measured to predict the accurate activation time of detectors. One ionization detector and nine photoelectric detectors were selected. A fire detector evaluator, which can uniformly control the velocity and smoke concentration, was utilized. Filter paper, liquid fuels, and polymer pellets were employed as smoke-generation combustibles. The nominal obscuration thresholds of the considered detectors were 15 %/m, but the ionization detectors activated at approximately 40 %/m and 16 %/m, respectively, on applying filter paper and kerosene. In contrast, the reverse obscuration thresholds were found quantitatively according to the combustibles in the photoelectric detector. This phenomenon was caused by differences in the color of the smoke particles according to the combustibles, which is explained by single-scattering albedo (ratio of light scattering to light extinction). The obscuration thresholds for liquid fuels (kerosene, heptane and toluene) as well as fire types of polymer plastic pellets were also measured for several photoelectric detectors. A database of obscuration thresholds was thereby established according to the detector and combustible types, and it is expected to provide useful information for predicting more accurate detector activation time and required safe egress time (REST).
RESUMEN
Flame retardant cables were investigated using thermo-gravimetric analysis to measure the reference temperature and reference rate required for a fire spread simulation using a Fire Dynamics Simulator (FDS). Sensitivity analysis was also performed to understand the effects of the reference temperature and rate on the pyrolysis reactions. A two-step pyrolysis reaction was typically observed regardless of the cable type, and each pyrolysis reaction could be attributed to single or multiple components depending on the cable type and reaction order. Although the structures, compositions, and insulation performances of the cables differed considerably, the reference temperatures of the two-step pyrolysis reaction were extremely similar regardless of the cable type. Conversely, the reference rates of the different types of cables varied significantly. The sensitivity analysis results indicate that the mean values of the reference temperature and rate are sufficient to simulate the pyrolysis reactions of flame retardant cables. The results obtained herein also suggest that the heat transfer and pyrolysis reaction path associated with the multi-layered cable structure may be more important for accurately determining the ignition and fire spread characteristics, which are attributable to differences in cable structure, composition, and insulation performance.