Behavior of HV cable at short circuit

Abstract

In our previous work, we analyzed an unusual behavior of an HV cable at short circuit when the shield at the power supply side was grounded, and at the load side it was floating. It was shown that the transmission line model is no longer applicable. It was also shown that the cable insulation is overstressed at the load side. Staying within a convenient simplicity of a two wire line approximation it was still possible analyzing salient phenomena of the load breakdown in unterminated shield connection. However, the period of oscillations T for this case was almost four times greater than could be expected from the cable electrical length! We did not address this discrepancy previously; to the best of our knowledge, it was not described in literature. Here, we analyze the reasons for the dramatic increase of T. Again, main tools are lumped circuit simulation and experiments on low-voltage increase model lines. The first suggests that since the currents in the central conductor and the shield flow in the same direction, the resulting circuit inductance is much greater than that in a conventional line discharge. Such inductances have been estimated. Experiments showed that the cable (coiled or spread) and ground layouts have also large influence on T. Experiments with an HV cable at a voltage up to 40 kV increased confidence in the simulation and low-voltage physical modeling results. Overall, simulation and experimental results are in fair agreement.