| Summary: | This investigation generated new experimental data on premixed gas/air vented explosions. A small (0.01 m3) and medium scale (0.2 m3) cylindrical vessels were used with L/D of 2.8 and 2 respectively, with range of vent area coefficients Kv of 2.7-21.7. The initial set of experiments considered free venting, so that the flame propagation during the venting process was laminar and also the short distance of the vessels would reduce the effects of flame self-accelleration. Covered vents were later used with vent static burst pressure Pstat from 35 to 450mb in the 10L vessel. Different gas mixtures were used throughout this work including methane-air (10%), propane-air (4 and 4.5%), ethylene-air (6.5 and 7.5%), and hydrogen-air (30 and 40%) gas mixtures. The ignition position at the far end opposite the vent and central location mid-way the length of the vessels were compared. Current venting guidance is based on experimental vented explosions with central ignition, but this work shows that end ignition opposite the vent is the worst case. The current design procedures for the protection of explosions using venting is shown to be inadequate for hydrogen-air explosions. New data has been presented which indicates that for hydrogen explosions, the vent flow behaves differently as compared to other gas mixtures investigated. Hence, the need for more research in hydrogen-air mixtures in order to have better understanding of hydrogen venting process. Experimental data from the current work also shows that multiple vents and vent shapes have significant effects on explosion overpressure and flame speeds. This is contrary to the assumption of the current venting standards. The effect of static burst pressure on explosion venting was shown to be quite different to that in the design standards, which is supported by other work in larger vessels. Other aspects of vent design that the standards say are not important were shown to be significant: the number of vents, the position of the vent, the shape of the vent, the ignition position. Laminar flame venting theory was shown to be a good predictor of the results and those from the literature where larger vessels were used.
|
|---|
