Bolarin S. Oyegoke

New Techniques for Determining Condition of XLPE Cable Insulation from Polarization and Depolarization Current Measurements

This paper describes how a commercially available test set has been used to make preliminary determinations of condition of XLPE cable insulation at remote sites in NE Queensland. Polarization and depolarization currents are measured with an injection voltage of 5 kV with observed depolarisation period of 60 seconds. It is shown that misleading results may be obtained by this technique due to rapid drying out of cable insulation after removal of the cable 50 Hz normal supply and laboratory experiments to systematically examine this effect are reported. Procedures needed to obtain consistent field test results are recommended as are diagnostic criteria to provided an indication of cable insulation condition.. It is concluded that whilst the technique can provide evidence of poor insulation quality, it is not possible to distinguish between water trees in insulation and poor joint insulation; additional research work to solve this problem is suggested.

Water migration in degraded XLPE cables

During dielectric response measurements on samples of XLPE 22 kV cable that were known to have been severely-degraded in service by water trees, it was observed that the insulation of the cable appeared to degrade considerably after water-immersed cable sample were connected to the rated voltage AC supply. This paper describes results of systematic experiments made to investigate this factor further using polarisation and depolarisation current (PDC) equipment to measure the condition of cable samples. It is found that the cable condition appears to improve from poor to good condition within a few days after short circuiting of the previously-energised cable commenced. Conversely it is found that the condition of the cable appears to degenerate over time following re-application of the rated voltage to the cable samples. It is hypothesised that electrostatic forces maintain moisture in the trees during energisation and that the moisture retreats from the trees when the supply voltage is removed. The implication of this phenomenon on dielectric response measurements is discussed in the paper.

Asset Management of Medium Voltage Cable Networks

This paper describes the background to research currently underway associated with asset management (AM) of degraded cable networks in North Queensland. In the paper the main failure mode of cables installed twenty years ago is identified as the growth of water trees through the cable insulation. The costs of managing water tree degradation in a large distribution network are considered and the economics of cable replacement in a regulated monopoly are examined. Condition monitoring techniques for assessing the condition of cables are reviewed and the possibility of silicone liquid injection as an alternative to cable replacement is discussed: a research methodology to support improved cable AM is also presented.

The effect of cable terminations on dielectric response measurements

This paper presents findings on dielectric response measurements on field aged and new medium voltage XLPE cables using a non-destructive frequency domain technique. Emphasis is laid on the laboratory and field acquired data interpretation. Results presented in this paper shows possible error being introduced into the judgement of water tree deterioration when the influence of the cable terminations is not properly considered.

Improved condition assessment of XLPE insulated cables using the isothermal relaxation current technique

Well-known procedures using the isothermal relaxation current method (IRC) use an empirically-derived ageing factor (the A-factor) to estimate the condition of cables. The A-factor is calculated from plots of the product of instantaneous value of IRC and the time to that value from the start of the current, against the log of time. A-factors are computed from a standard formula using constants determined from curve fits of measured depolarization currents. From measurements of the breakdown strength of samples of real and degraded cables it has been found that A-factors do not provide reliable estimates of XLPE cable condition. However, if semi-conducting material of the cable insulation is known and cables are classified according to semi-conducting material type it is possible to get better correlation between conditions of cables as indicated by A-factors and AC breakdown voltage. This paper contains a novel analysis of result from previous researcher and also includes result from tests on other cables. The refined procedure is found to give reasonable value of A-factor for all cable tested.