| dc.description.abstract | The purpose of the experiments conducted in this Master's thesis is to predict and explain the change in breakdown voltage when insulating barriers are introduced in a rod-plane gap arrangement. The experiments have been conducted with positive lightning impulse voltage, using the up and down method to determine the 50 % breakdown voltage. A cylindrical rod with rounded tip and radius 3.5 mm was used as the high voltage electrode above a plane grounded electrode. The polycarbonate barriers used were 1 mm thick and of different sizes (4x4 cm, 6x6 cm, 8x8 cm, 16x16 cm, 30x30 cm and 40x40 cm). They were placed at various positions in the 80 mm rod-plane gap to find the optimal combination.The results show that the breakdown voltage of the gap could be increased by the use of barriers, strongly dependent upon their size and position. The largest barrier offered the highest breakdown voltage, an increase of 98.0 % versus the barrier-less rod-plane gap. With the two largest barriers, the optimal position was found to be in the upper part of the gap, 0-10 mm from the high voltage rod tip. The four smaller barriers perform their best around 20 mm from the tip. Literature has suggested that the optimal position is in the range 12-24 mm for this gap [Lebedev et al. 2005], where the breakdown voltage can be over tripled.It has been discovered that placing the smallest barriers close to the high voltage rod tip drops the breakdown voltage to levels below that of the barrier-less gap. A suggested explanation is the strong tangential field present on the barrier surface under these conditions, quickly building up charge on the barrier and leading to breakdown. Streamer inception on the underside of the barrier has not been observed despite the high field strength directly under the rod tip. This is possibly caused by the slightly higher field on the upper side of the barrier, leading to streamer inception which weakens the field under the rod tip. As the barrier size is increased, the voltage drop in the longer streamer path is the dominating factor behind the rise in breakdown voltage. It is recommended to employ barriers of considerable cross-sectional length, preferably twice the gap distance or longer, to ensure satisfactory breakdown performance improvement. An empirical equation for predicting breakdown voltage in barrier insulated rod-plane gaps has been constructed on the basis of the conducted experiments. | nb_NO |
