Polyetheretherketone is used in the packaging of high voltage power electronics as a frame material, and the surface discharge behaviors of polyetheretherketone are the main concerns for the packaging design. Pulse width modulation voltage, which shows a variable duty cycle, is applied to polyetheretherketone at working conditions. The surface discharge behaviors of polyetheretherketone under a positive repetitive pulse voltage of 50 Hz with different duty cycles were investigated in this paper. It is generally considered that the surface discharge inception is mainly related to electric field distribution; however, the pulse width of voltage could significantly affect the surface discharge inception characteristics. In this study, the surface discharge inception voltage, time-integrated streamer images, and cumulative surface discharge patterns under a positive repetitive pulse voltage of 50 Hz were obtained. It was found that the surface discharge inception voltage decreased with the increase in the duty cycle. This relationship was explained by the role of time-lag. Time-lag distribution analysis was performed, and the time-lag distribution was mainly determined by stochastic time-lag. The mean stochastic time-lag was obtained by the Laue plot, and the relationship between the mean stochastic time-lag and applied voltage could be well fitted by the inverse power law. Furthermore, the relationship between the surface discharge inception voltage and duty cycle could be exactly fitted by the inverse power law, which was derived from the relationship between the stochastic mean time-lag and applied voltage. The closely related relationships demonstrate that time-lag plays a critical role in the surface discharge inception at different duty cycles.

The surface discharge behaviors of polyetheretherketone (PEEK) in nitrogen under both positive and negative semi-square voltage were investigated using the current pulse detection method by non-inductive resistor and optical signal detection method by photomultiplier and digital camera. A 50Hz positive or negative semi-square voltage was applied to a rod-plane electrode system, and two types of surface discharge in a voltage cycle were found: one forward discharge and one back discharge. The characteristics and initiation mechanism of the two types of surface discharge were explored. The experimental results showed that surface discharges between cycles were independent. Based on the independence, the time-lag characteristics of surface discharge were obtained. An improvement on Laue-Zuber time-lag theory was made to fit the time-lag characteristics of forward discharge and back discharge. The mean time-lag decreased with the increase of voltage for both forward discharge and back discharge, and the relation between mean time-lag and applied voltage could be modeled by inverse power law. Comparative analysis on the characteristics of back discharge and forward discharge under opposite voltage was performed, and the characteristics of back discharge and forward discharge under opposite voltage were homogenous. In addition, the initiation mechanism of the two types of discharge was discussed. The initiation of forward discharge is believed to be mainly determined by the time-lag and the electric field caused by the applied voltage, while the initiation of back discharge is believed to be mainly determined by the time-lag and the electric field induced by the residual surface charge left by the forward discharge in the same cycle. Verification tests under positive semi-square voltage with different fall time were conducted to validate the initiation mechanism.

Electrical insulation capability is one of the most critical challenges for press-pack insulated gate bipolar transistors (IGBTs), but the knowledge of its insulation failure mechanism is much less mature. To understand the insulation failure mechanism, partial discharge (PD) measurement and analysis in press-pack IGBT (PPI) submodule were performed in this study. A dedicated PD measurement system for PPIs was designed and fabricated, and fast oscilloscope, PD detector and ultraviolet (UV) image camera were employed to observe the PD. PD-induced insulation failure phenomenon in the submodule was observed under direct current voltage, and UV images indicated that the discharge occurred at the periphery of the die. Phase-resolved PD (PRPD) analysis method for PPIs was developed and three elementary structures were used to simulate the PD in the electric field reinforcement areas identified by electric field analysis. According to PRPD patterns, the observed PD in the submodule was considered to be mainly caused by a mismatch between the electric field and critical electric field on the surface of passivation layers of the die. The inference was verified at different gas pressures because a decrease of PD inception voltage in the submodule and the bare die was observed at lower gas pressure.

In this study, the surface discharge current pulses of polyetheretherketone (PEEK) material under positive repetitive square voltage in a nitrogen atmosphere are measured. The influences of different voltage amplitudes and frequencies on the detail parameters of forward discharge and backward discharge current pulses are statistically analysed. The results show that as the square voltage amplitude increases, the current pulse amplitude, fall time and pulse width of both forward and backward discharge current increase, and the rise time does not change significantly. As the voltage frequency increases, current pulse amplitude, fall time and pulse width of both forward and backward discharge current decrease, and the rise time does not change significantly. Due to the independence of the discharge at different repetitive cycles, the specific discharge process in one cycle is analysed in detail to explain the influence mechanism of the voltage amplitude and frequency on the discharge current. By mean of the Richardson–Schottky and the Cavallini relaxation model, the relationship between discharge voltage ratio and surface charge, and the decay process of surface discharge are analysed. Furthermore, the influences of amplitude and frequency of the positive repetitive square voltage on the PEEK surface current pulses are explained qualitatively.

Medium-voltage (e.g. 10 kV rated) silicon carbide (SiC) devices have great potentials in medium-voltage variable speed drives. But their high switching dv/dt can increase the voltage stress on motor windings and cause partial discharges. This paper presents a partial discharge study of a medium-voltage form-wound winding under two-level square-wave voltage pulses. A 10 kV SiC device-based test platform is built to generate voltage pulses with high dv/dt. A three-step test approach is proposed and employed to systematically investigate the effects of various voltage parameters on partial discharges. These include voltage rise/fall time, voltage pulse width, pulse repetitive rate, duty ratio, voltage polarity, fundamental frequency, and modulation index. Partial discharge inception voltages (PDIVs) and repetitive partial discharge inception voltages (RPDIVs) of the sample are measured with varied voltage parameters. Test results show that voltage rise/fall time is a major affecting factor which reduces PDIVs of the winding sample by 6.5% when it decreases from 800 ns to 100 ns. Based on test results, a hypothetical partial discharge mechanism is presented to explain the effects of fast voltage rise/fall edges. An empirical equation is also derived to estimate PDIVs of a winding sample under various voltage rise/fall time and pulse width conditions.