The behaviour of charge on the surface of polymeric materials

In the high voltage industry one of the factors limiting the performance of a pulse power system is the build-up of charge on the surface of the insulators. In these systems solid insulators are used to support the conductors. When operating pulse power systems in repetitive mode, the accumulation of charge on the insulators after each pulse can promote "surface flashover" and cannot be ignored as a limiting factor in the design of the system.The aim of this project was to enhance knowledge of the effect of accumulated charge on the surface conductivity, the surface charging/discharging behaviour and the surface charge decay behaviour of polymeric insulators. The project was carried out in 2 stages.Firstly surface charging/discharging transient currents were observed using a concentric cylindrical measurement system for uncharged PMMA and PMMA samples charged by exposure to corona. The charging/discharging transients were analysed to characterise possible conduction mechanisms and the DC steady state currents were used to calculate the surface conductivities. Charging the sample surfaces with corona led to an increase in the surface conductivities of the samples. The behaviour of the charging/discharging transients were complex and fitted well against 3 and 4 term exponential functions. Differences in behaviour were also observed due to sample thickness.;In the second stage of the project surface charge decay measurements were performed on PMMA samples that were subjected to corona charging. This was done to determine the relative importance of surface conductivity and charge neutralisation to the decay of surface charge. A novel charge probe based on varactors was developed and calibrated to perform these measurements. This system was novel in its use on measurements of surface charge on polymeric materials. The measured surface potentials from the varactor probe were used to calculate the surface charge density distributions. It was observed that most of the charge accumulates near the electrodes. Analysis of the total surface charge showed that the rate of charge decay was higher when the decay occurs through the surface conduction mechanism as compared to conduction through gas neutralisation, where no significant charge decay was observed. The rates of charge decay for thin and thick samples, which were corona charged at a grid voltage of 1 kV, were observed to be 0.027 {cedil}{9D}{9E}{aelig}C/min and 0.048 {cedil}{9D}{9E}{aelig}C/min, respectively, for neutralisation through the gas; while the corresponding rates of charge decay through surface conduction were observed to be 0.096 {cedil}{9D}{9E}{aelig}C/min and 0.18 {cedil}{9D}{9E}{aelig}C/min, respectively. Evidence of the crossover effect has been observed in certain geometries. Correlations have been made between the charge decay measurements and the transient behaviours observed in the first part of the project. The charge decay behaviours were found to be in good correlation with the surface conductivities observed for corona treated samples where an external field was applied across the sample surface.In the high voltage industry one of the factors limiting the performance of a pulse power system is the build-up of charge on the surface of the insulators. In these systems solid insulators are used to support the conductors. When operating pulse power systems in repetitive mode, the accumulation of charge on the insulators after each pulse can promote "surface flashover" and cannot be ignored as a limiting factor in the design of the system.The aim of this project was to enhance knowledge of the effect of accumulated charge on the surface conductivity, the surface charging/discharging behaviour and the surface charge decay behaviour of polymeric insulators. The project was carried out in 2 stages.Firstly surface charging/discharging transient currents were observed using a concentric cylindrical measurement system for uncharged PMMA and PMMA samples charged by exposure to corona. The charging/discharging transients were analysed to characterise possible conduction mechanisms and the DC steady state currents were used to calculate the surface conductivities. Charging the sample surfaces with corona led to an increase in the surface conductivities of the samples. The behaviour of the charging/discharging transients were complex and fitted well against 3 and 4 term exponential functions. Differences in behaviour were also observed due to sample thickness.;In the second stage of the project surface charge decay measurements were performed on PMMA samples that were subjected to corona charging. This was done to determine the relative importance of surface conductivity and charge neutralisation to the decay of surface charge. A novel charge probe based on varactors was developed and calibrated to perform these measurements. This system was novel in its use on measurements of surface charge on polymeric materials. The measured surface potentials from the varactor probe were used to calculate the surface charge density distributions. It was observed that most of the charge accumulates near the electrodes. Analysis of the total surface charge showed that the rate of charge decay was higher when the decay occurs through the surface conduction mechanism as compared to conduction through gas neutralisation, where no significant charge decay was observed. The rates of charge decay for thin and thick samples, which were corona charged at a grid voltage of 1 kV, were observed to be 0.027 {cedil}{9D}{9E}{aelig}C/min and 0.048 {cedil}{9D}{9E}{aelig}C/min, respectively, for neutralisation through the gas; while the corresponding rates of charge decay through surface conduction were observed to be 0.096 {cedil}{9D}{9E}{aelig}C/min and 0.18 {cedil}{9D}{9E}{aelig}C/min, respectively. Evidence of the crossover effect has been observed in certain geometries. Correlations have been made between the charge decay measurements and the transient behaviours observed in the first part of the project. The charge decay behaviours were found to be in good correlation with the surface conductivities observed for corona treated samples where an external field was applied across the sample surface