Elisabeth Lindell

Influence of rise time on partial discharge extinction voltage at semi-square voltage waveforms

This work presents measurements of the partial discharge (PD) extinction voltage in three different types of test objects, using semi-square voltages with 2 ¿s and 100 ¿s rise time. A needle creating corona discharges, a twisted pair specimen commonly used for testing motor insulation and a paper/oil test object modelling the turn-to-turn insulation of a transformer winding were investigated, presenting extinction voltages between approximately 1 kV and 8 kV. For the twisted pair specimen the PD extinction voltage was significantly lower at the shorter rise time, whereas for the other test objects, just a small or no difference could be detected. The conclusion is that depending on what range of rise times and what insulation system that are studied, there may be an influence of the rise time of the applied voltage on the PD extinction voltage.

Measurement of partial discharges at rapidly changing voltages

This work presents a measuring system for PDs at repetitive voltages with short rise times, based on a coupling device adapted for this kind of voltages and a method for removing contributions from the applied voltage. The potential of the system has been demonstrated through measurements of PDs around a needle in air at voltages of 5.5 kV - 9.5 kV with rise times down to 10 mus. The observations obtained with this PD source prove that the measuring technique described enables detailed studies of PD properties under repetitive rapidly changing voltage waveforms in the investigated range of voltages and rise times.

Evaluation of a PD measuring system for repetitive steep voltage waveforms

Measurements of partial discharges (PDs) at high frequency voltages is important but not a straightforward task. In this paper a PD measuring system developed for use at repetitive, steep voltage waveforms is presented, which is based on electrical detection using capacitive decoupling of the PD signals. The system utilizes only a low order filter and allows the presence of significant remnants of the applied voltage in the signal used for PD detection. Analytical calculations that estimate the reachable sensitivity of this method are presented, and for an example with a test object of 100 pF capacitance, a sensitivity in the range of 1 - 10 pC/kV at 1 μs rise time was obtained. Further, the transfer function of the measuring system has been measured and used for reconstruction of the voltage across the test object, the latter consisting of both the applied square-like voltage and voltage drops due to PDs. This is an attractive possibility since the applied high voltage signal at the test object can be recalculated from the same signal as is used for PD detection. In addition, the reconstruction of the PD voltage drop could provide information about the PD processes, provided that the sampling frequency is high enough.

Stochastic detection of partial discharges

The traditional method to detect partial discharges relies heavily on the use of high-pass filters. Whereas this is a very effective approach for normal power voltage shapes, severe problems arise when partial discharges are to be detected under waveforms generated by modern power electronics. Therefore, a technique that utilizes the random occurrence of partial discharges for detection has been developed and used for semi-square voltages. This technique has evolved since the first publications and new applications have been found. The present paper introduces the technique and discusses the necessary analysis elements, requirements imposed as well as some examples intended to show possible application areas.

Study of Partial Discharges at the Positive and Negative Voltage Flanks of Semi-Square Voltages

Experiences from the use of a measuring system for detection of partial discharges (PDs) at a repetitive, unipolar, square-like voltage with a rise time of 2.0 mus, are here presented. PDs were detected during the major part of the voltage flanks, even when the duty cycle of the applied voltage was not completely stable. The voltage was applied for relatively long time and one PD appeared at each voltage flank. As the measurement progressed however, the time intervals within which the PDs appeared shifted. This indicates that comparisons of results from measurements performed at e.g. different rise times or voltage levels can be difficult, due to the inherent stochastic variations in the PD appearance. One example of a time interval where the PDs at the negative voltage flank all appeared during the falling part of the voltage is also shown. Under such circumstances, it is important to be able to measure the PDs also during the voltage flanks to avoid that the test object is considered to be PD free.