Solid-gas interfaces stressed with HV impulses: Surface flashover behaviour and control in the 100s kV regime

The project will entail characterisation of the flashover behaviour of electrode systems bridged with different solid spacer materials, in air environments. The impulsive voltages applied will have peak magnitudes of up to 300 kV, with a time to half-value of the order of hundreds of nanoseconds. The flashover behaviour of cylindrical samples of three different spacer materials (HDPE, Delrin and Ultem), in air with different levels of relative humidity (<10%, ~50% and >90% RH), and in the pressure range 0.5 to 1.5 bar absolute, will be characterised. Methods to modify the surface finish of samples of the chosen materials, with a view to controlling flashover behaviour in humid environments, will be investigated. Open gaps, of the same inter-electrode gap distance but without a solid spacer, will be tested to provide baseline breakdown data for comparison. The electrode profiles used will give rise to a quasi-uniform field distribution, and the polarity effect under the different sets of test conditions will be quantified.
Tests will then be performed in forced airflows, in order to determine the effect on the breakdown characteristics. Contaminants will be introduced into the insulation systems, positioned on the surface of the solid spacers and focussing on triple junctions, and any synergistic effects will be determined. Two different test procedures will be used: a step-up procedure, to determine the peak impulse voltages that initiate flashover under the different sets of test conditions; and overvoltage tests, where the time-to-breakdown of the various insulation systems will be characterised for comparison.
Weibull statistical analysis will be performed on the breakdown voltage results generated using the step-up procedure, to separate the performance of the insulation systems by material and surface finish, at different pressures and levels of relative humidity. This will reveal information such as the sensitivity of the different insulation systems to an increase in applied voltage, and threshold levels of applied voltage, below which breakdown of the insulation system is not expected to occur. The time-to-breakdown data generated in overvoltage tests will be subjected to von Laue statistical analysis, allowing estimation of the relative contributions of the mean statistical time and the formative time to the overall time to breakdown under the various sets of test conditions.
The experimental data generated throughout the project, and the associated discussion of the mechanisms of discharge initiation and development, will be of interest to designers of high-voltage insulation for pulsed power systems, required to operate reliably in humid conditions, and potentially in the presence of surface contaminants. Recommendations will be made on material selection and the surface finish of solid spacers, tailored for different environmental conditions as appropriate.