Fire safety

摘要

Stage 1 involved a comprehensive study of known similar work carried out in various countries. Stage 2 involved the instrumentation of the former British Coal large-scale fire gallery, followed by a programme of testing. In the first instance, measurements were taken without belting samples in order to 'characterise' the gallery and derive relationships that would form the basis of requirements for mid-scale equipment. This was followed by measurements taken with three belts of differing fire performance: a 'good' belt by current acceptance standards, a 'borderline' belt by the same standards and a belt with less fire resistance that would have been expected to fail. These measurements provided information on the following: Relationships between air velocity measurements taken at the front of the gallery with a portable anemometer, the array of anemometers behind the sample and the differential pressure transducer in the ducting. Values for the power of the calibration fires (heat release rates) calculated from the quantity of gas consumed, oxygen consumption measurements and temperature rise values. Equations describing the temperature rise curves. The behaviour of the three different belt types, including the extent of fire damage, temperature profiles along the length of the sample and flame velocities. Stage 3 saw the work transferred to Fenner Dunlop's technical facility at Marfleet, Hull, where the mid-scale MSHA test gallery had been identified as a potentially suitable starting point for the development of a laboratory test. This work followed the same route as Stage 2, with instrumentation of the gallery, characterisation of the gallery with calibration fires (no belting) and subsequent testing of the three belt types. Throughout this work changes were made to several variables including: fuel type, fuel flow rate, burner design, burner position, trestle design/height and air flow. Stage 4 provided for a wider range of belts, for which there was a known history of results on the large-scale high-energy test, to be tested on the new mid-scale facility. These showed that a mid-scale test was feasible. It also showed that the test was able to distinguish between different qualities of cover on the same carcass. It showed that the repeatability of the test was good and that belts that were not fire-resistant caught fire readily and had to be extinguished. The belt that had met the 10 minute test requirement gave a similar performance in the new test and all of the belts that were tested, which represented the complete range of those currently accepted for underground use, using the high-energy test performed well in the new test. From these results it was concluded that the new test could be used to simulate with confidence the performance of belts in the large-scale gallery.