Elektrot pürüzlülüğünün boşalma olaylarına etkisi

Abstract

Experiments are carried out with 50 Hz AC and DC voltages up to peak value of 140 kV in both polarities, using plane-plane, rod-plane and sphere-plane electrode systems. To obtain various degrees of non-uniformity, radius of 1 mm and 2 mm the sphere and rod electrodes are selected. To obtain desired degree of roughness hemi sphere with heights equal to 2.0 mm, 3.25 mm 5.5 mm are mounted on the plane electrode for each configurations. The electrodes are mounted in a pressure vessel made of polymethylmetacrylate (plexiglass). All electrodes are made of brass. Rod and sphere electrodes are covered with chromium. The lower electrode in both arrangements is a Rogowsky n/2 profile plane electrode with an overall diameter of 80 mm and its edges are rounded with a radius of curvature of 3 mm. For the 50 Hz AC test, with voltages up to 100 kV r.m.s. a 0.22/100 kV, 5 kVA high voltage test transformer is employed. To limit the discharge current at breakdown and reduce the erosion at electrode surfaces, a 50 kohm protecting resistor is connecting between the high voltage test transformer and the electrode system used. Furthermore, the high voltage transformer is disconnected from the mains immediately after breakdown. The peak value of 50 Hz AC voltage is measured by using the method of Chubb and Fortesque. The waveform of the 50 Hz AC voltage applied to the electrode system is observed by means of a capacitive potential divider and Textronix 434, dual beam ossiloscope. The corona inception voltage is determined by observing the voltage drop due to the corona current across a 100 ohm resistor inserted in the earth lead. The same test transformer is used to obtain 140 kV DC voltages. DC corona inception voltage is determined in the same way as for AC measurements, but the across the electrode system is not observed. In all experiments the applied voltage has been raised at a rate of 5 kV/s until breakdown occured. Before starting any experi ments, electrodes are cleaned after mounting the electrodes in the test vessel. The test vessel pressure is increased to 5 bar. After waiting 10 minutes for occuring steady-state gas pressure in vessel, the high voltage is applied to the electrode system, for example, rod-plane, sphere-plane or plane-plane electrode systems. During the tegts, the temparature of the test room is measured 17+2 C. At least, three different values both corona inception and breakdown voltage are taken to the mean value of each measuring point.

When compressed gas was first introduced as an electrical insulant in high voltage power equipment, it was discovered that the insulation strength of the systems was less than that predicted by theory. An explanation for this apparent reduction in the dielectric strength of compressed gas can be given by considering the perturbations of the macroscopic electric field produced either by microscopic roughness of electrode surface ör by conducting particles. That such perturbations of electric field sould play a majör role in the dielectric strenght of gas is due to the effective coefficient of ionization « being a strongly varying function of the electric field strength E, particularly for E=E.., where E,. is the value for which o=0. lum l «. m it will be shown that the effects of the small regions of enhanced field strength caused by the surface roughness can account for the observed limitations on breakdown voltages in gas within the pressure range of engineering importance. Experimental investigations into the effect of electrode surface roughness upon the dielectric strength of strongly electronegative gases have adopted two different approaches; a) the use of geometriçally well-defined artificial protrusions mounted on plane electrode, b) the use of an electrode having nature surface roughness. The former approach provides a well-defined perturbations of the macroscopic electric field, and thus lends itself to an analysis of, for example, onset conditions. vi This thesis is an attempt to discuss this phenoraenon in terms of streamer theory of breakdovm using the second approach. Such an analysis is used to determine the relevant diraensions of artificial protrusions. These dimensions are found to be comparable with those of the roughness associated with production process. However, the dimensions of the artificial protrusions employed have ali been much greater. This suggests an inability to obtain protrusions of the relevant dimensions. Although available artificial protrusions are unrepresentative of practical conditions, the result obtained from the investigations can be used för understanding the effect of the surface roughness. Several forms of artificial protrusion have been used in roughness investigations, e.g. a sphere, a hemisphere, a cylinder, an ellipsoid ör a cone. in this thesis, the hemispheres which have different radius are used. A knowledge of spark breakdown for various gap geometries is essential when designing gas-insulated high voltage apparatus. Many emprical formulae are available from which breakdown ör corona onset field strengths may be calculated. Such formulae are, however, valid only within certain ranges, and extrapolation can lead to large errors. it may, therefore, be of interest to consider the possibilities for a direct application of the physics of gaseous breakdown in the design of gas-insulated high voltage equipment. The classical Townsend theory of the growth of ionization is fundamental to any discussion of breakdown of gaseous dielectrics. it is generally accepted that the Townsend theory of breakdown can account for the onset of breakdown in uniform fields under quasi-static conditions. However, in order to give a detailed description of many observed phenomena of engineering importance, the Townsend theory must be supplemented by the streamer concept of breakdown. Based on these physical concepts, quantitative criteria for the onset of breakdown have been proposed. These are, however, of limited value to high voltage engineer. The main problem is that such criteria are difficult to apply to pratical situations. Another complication is related to fact that the growth of ionization is exponential. This can,owing to the mathematical nature of exponential functions, lead to procedures which appear to give acceptable results. vii A closer analysis may reveal, however, that quite raeaningless properties are hidden in criteria. in this thesis, quantitative criteria is derived för the onset of breakdown in atmosferic air and similar weakly-electronegative gases. Thereafter, a comparable criterion is formulated for strongly-electronegative gases. These criteria contain no quantities other than those which can be obtained directly frora breakdown voltage raeasurements in uniforra field, i.e. from Paschen curve data. No specific data are required for the ionization coefficients of the gas. The advantage, from an engineering point of view, is that it is much easier to perform reliable Paschen curve measurements than it is to measure ionization growth parameters. The criteria can be applied to any nonuniform field configuration provided the electrostatic field distribution is known. Moreover, in many cases, a simple analytical approximation to the field distribution can render an extensive calculation of the electrositatic field unnecessary. Growth of ionization can be approached in two dif- ferent ways. From a molecular point of view, it is the individual ionizing collision processes described in terms of probabilities and collision cross section which are of primarly interest. in high voltage engineering, it is the collapse of the voltage across a gap as a result of gaseous breakdown which is of paramount importance. This phenomenon involves so many collision processes that it is natural to consider the event as a macroscopic phenomenon described in terms of macroscopic swarm parameters such as primary and secondary coefficients of ionization. The number of electrons participating in the ionizing processes is then so large, it makes sence to represent the actual number of electrons, which of course can take only integer values, by a continous function. This makes it possible to apply mathematiçal concepts such as differential and integration to the analysis of the growth of ionization in electron avalanches. in this thesis, the effect of grounded electrode surface roughness to the breakdown voltage of air is investigated for uniform and non-uniform electric field in high pressure. The results obtained from experiments for the roughness electrode surface situations were compared with the smooth electrode surface situations. viii Experiments are carried out with 50 Hz AC and DC voltages up to peak value of 140 kV in both polarities, using plane-plane, rod-plane and sphere-plane electrode systems. To obtain various degrees of non-uniformity, radius of 1 mm and 2 mm the sphere and rod electrodes are selected. To obtain desired degree of roughness hemi sphere with heights equal to 2.0 mm, 3.25 mm 5.5 mm are mounted on the plane electrode for each configurations. The electrodes are mounted in a pressure vessel made of polymethylmetacrylate (plexiglass). All electrodes are made of brass. Rod and sphere electrodes are covered with chromium. The lower electrode in both arrangements is a Rogowsky n/2 profile plane electrode with an overall diameter of 80 mm and its edges are rounded with a radius of curvature of 3 mm. For the 50 Hz AC test, with voltages up to 100 kV r.m.s. a 0.22/100 kV, 5 kVA high voltage test transformer is employed. To limit the discharge current at breakdown and reduce the erosion at electrode surfaces, a 50 kohm protecting resistor is connecting between the high voltage test transformer and the electrode system used. Furthermore, the high voltage transformer is disconnected from the mains immediately after breakdown. The peak value of 50 Hz AC voltage is measured by using the method of Chubb and Fortesque. The waveform of the 50 Hz AC voltage applied to the electrode system is observed by means of a capacitive potential divider and Textronix 434, dual beam ossiloscope. The corona inception voltage is determined by observing the voltage drop due to the corona current across a 100 ohm resistor inserted in the earth lead. The same test transformer is used to obtain 140 kV DC voltages. DC corona inception voltage is determined in the same way as for AC measurements, but the across the electrode system is not observed. In all experiments the applied voltage has been raised at a rate of 5 kV/s until breakdown occured. Before starting any experi ments, electrodes are cleaned after mounting the electrodes in the test vessel. The test vessel pressure is increased to 5 bar. After waiting 10 minutes for occuring steady-state gas pressure in vessel, the high voltage is applied to the electrode system, for example, rod-plane, sphere-plane or plane-plane electrode systems. During the tegts, the temparature of the test room is measured 17+2 C. At least, three different values both corona inception and breakdown voltage are taken to the mean value of each measuring point. IX In the uniform and non-uniform field, by applying 50 Hz AC voltage and positive and negative DC voltage, which is investigated effect of grounded electrode surface roughness on corona inception and breakdown voltages at high air pressure, the following results are obtained; 1) The grounded electrode surface roughness decreases to the breakdown strength of the system in all situations. 2) The effect of surface roughness on the breakdown strength of air increases while the electrode distance and the air pressure is being increased. 3) The breakdown voltage is not much dependence to the radius of the artificial protrusion. Althought in this thesis, the effect of the geometry of protrusion is not investigated, it can be thought that has affected a role. 4) In the negative DC voltages, the breakdown voltages are measured higher than the positive DC and 50 Hz alternative voltages situations. This is the effect of the polarity. 5) The breakdown voltage varies linearly between 0-5 bar pressure range and 2-20 mm electrode distances. The higher pressure and the longer electrode distance are, the higher breakdown voltage is. 6) In the positive DC voltages, the discharges are brighter and noisier than the negative and the 50 Hz alternative voltages. In negative DC voltage, the discharges become suddenly and quite. These are the characteristics to understand what kind of voltage is applied. Consequently, although artificial protrusions having well-defined geometrical shapes can be used to check the validity of onset calculations, the difficulty in obtaining artificial protrusions of dimensions comparable to the roughness assosiated with production processes restricts their usefulness with respect to investigations of discharge development under practical conditions.