Comparison of calculated and measured arc-back current in large power rectifier systems

EVEN with the high degree of reliability which mercury-arc rectifiers possess today, there are still a few occasions when current is carried in the wrong direction in one of the rectifier tanks and hence an arc back is created. Since this arc-back current can rise to an abnormally high value in a few milliseconds, a means must be provided to limit and to cut off this abnormal current when an arc back does occur. In order to be able to specify the characteristics of the rectifier tanks, the rectifier transformer, and the anode and cathode circuit breakers, it is essential that the magnitude of this arc-back current can be calculated with a reasonable degree of accuracy. Methods and procedures for calculating this arc-back current have been described in numerous books and papers,1¿6 and these calculations have been extremely useful in permitting the design of rectifier equipments which have reliably supplied large quantities of d-c power. On occasions when field tests have been performed and measurements of this arc-back current have been made, an abnormally large discrepancy has usually been found to exist between the calculated and the measured values of this current with the calculated values being appreciably higher than the measured values. While the calculated results were on the safe side, they have, on occasions, required the installation of additional impedance in the a-c system and have therefore increased the initial installation cost and may have penalized the operating performance of this equipment.