City Research Online

Abstract

This thesis describes a study of the effect of independent variation of electric spark energy and duration on measured minimum ignition energies, for a quiescent gas mixture, and four dust clouds.

To enable such a study, a spark generator was designed and constructed, based on a pulse-modulator principle, that could produce arc-type discharges, of controlled energy and duration, and on a reproducible basis. Particular emphasis was placed on the so-called ‘low-energy’ range, from 0.1 mJ to 20 mJ, where the control of discharge parameters has often been difficult with other discharge circuits.

The fuels chosen for this study were 2.7 % propane-air, a commercial stabiliser powder, Benzanthron, lycopodium and polyacrylonitrile. Of these, the first three were found to ignite below 1 mJ. The ability to ignite dust clouds with spark energies as low as those for hydrocarbon gas-air mixtures was therefore confirmed.

Ignition tests using the gas mixture, where spark-to-flame kernel expansion was studied with a schlieren technique, showed that ignition could be achieved using energies that were an order of magnitude lower than those using traditional capacitive circuits. This result was explained in terms of the increased efficiency in the thermalisation of spark energy using a pulse-circuit, where the rate of energy release in the gap was higher than for a capacitive circuit, for which spark duration is increased using resistive elements.

Ignition tests using the four dust clouds, dispersed in a specially constructed explosion vessel, were made together with high-speed schlieren photographs of the spark-to-flame kernel expansion. Results showed that for a given dust, an optimum spark duration existed at which ignition energy was at a minimum. This was found to coincide with an optimum rate of spark-kernel expansion following discharge. The differences in optimum values of minimum spark ignition energies and durations, between the four powders, were interpreted in terms of the ’induction times’ of the constituent particles. These times depended upon the chemical composition and physical size and shape of the particles.

The dust-ignition work concludes with a simplified model of the spark-ignition process, assuming that a critical energy density in the spark/flame kernel is required for ignition. Both the present results and those from the literature are discussed on the basis of this model. This is shown to account for the general observation that 'pure' capacitive discharges must be of higher energies than those produced when using series resistance or inductance in a capacitive circuit.

The existence of an optimum rate of kernel expansion, at which spark ignition energy was at a minimum, indicated an optimum rate of energy release in the spark gap. As a unique relationship between rate of energy release and spark duration does not exist when comparing different discharge circuit configurations, it was proposed that ’rate of energy release’, rather than 'spark duration', be used as a criterion for measuring minimum ignition energies.

Finally, it is suggested that a fuel be designated two values of minimum ignition energy, one representing a 'hazard assessment' value based on capacitive discharges, and one representing the 'absolute sensitivity' to ignition by a spark discharge.

Publication Type:	Thesis (Doctoral)
Subjects:	Q Science Q Science > QD Chemistry
Departments:	School of Science & Technology > School of Science & Technology Doctoral Theses Doctoral Theses

Export

Downloads