Return voltage-a reliable diagnosis for water treed cables?

To characterize the state of deterioration of water treed cable insulation different methods are in use. The problems with all diagnostic tools are firstly the sensitivity and secondly the lack of selectivity. Except for very severely deteriorated cables, the commonly used signals like loss tangent or depolarization currents are not able to detect selectively water tree induced degradation. In some cases the methods are not even able to guarantee that the signal is not just the result of the sound insulating material itself. The return voltage and especially its nonlinear dependence on the magnitude of the poling voltage is claimed to be a good measure of the state of degradation of hv-cables. A survey of data collected in the field shows that this assumption is not always true if only one point of each return voltage curve (e.g. the maximum value) is used to calculate a diagnostic parameter. If, in contrast to common practice, the complete time dependence of the return voltage curve is considered, more meaningful and characteristic information about the state of deterioration due to water trees can be gained. This is documented by the results of field-experiments