|dc.description.abstract||Very little research has been conducted in the past on the effects of the insulator surface roughness on the surface flashover voltage of insulators. The surface flashover voltages for several common electrical insulators have been investigated for several different surface roughness groups. Lucite, Lexan, and Celcon samples were prepared with ten different surface roughness groups ranging from treatment with #80 grit sandpaper to 0.05 ]am alumina polish. For each material, ten identically prepared samples were tested for each roughness group. Each sample received twenty 1.2/50 ps double exponential pulses with a fifteen-minute wait between pulses. Although the average first shot voltages were approximately the same for the various materials and roughness groups, different overall average flashover voltages were obtained for each material as a function of roughness.
Further investigation of Lexan was performed by placing two roughnesses on a sample at the same time in order to investigate the importance of the regions at and in between the electrodes. These experiments led to a hypothesis with competing processes. With a rougher surface, the cathode at the triple junction contributes to lower flashover voltages. In the second process, a smoother surface is more susceptible to damage resulting in lower flashover voltages. Explanations of these competing processes are presented.
Other factors that, in theory, take part in the surface flashover phenomena of insulators, which could be influenced by changing the surface profile, are surface and volume resistivities, surface profile path length, surface profile area, electron trajectories, and secondary electron emission yields. A discussion of these factors, along with a discussion of experiments investigating the dependence of the surface and volume resistivities on surface roughness, is presented.||