ZHANG Jian, YU Chaoyun, ZHAN Cao, et al. Study on Evolution and Failure Mode of Gate Drive Signals of High Voltage Press-pack IGBT[J]. High voltageapparatus, 2023, 59(1): 43-48.
DOI:
ZHANG Jian, YU Chaoyun, ZHAN Cao, et al. Study on Evolution and Failure Mode of Gate Drive Signals of High Voltage Press-pack IGBT[J]. High voltageapparatus, 2023, 59(1): 43-48. DOI: 10.13296/j.1001-1609.hva.2023.01.007.
Study on Evolution and Failure Mode of Gate Drive Signals of High Voltage Press-pack IGBT
Press-pack high voltage IGBT(PPIs)is the core element of modular multilevel converter(MMC)for voltage sourced converter-high voltage direct current(VSC-HVDC)transmission systems. Its gradual thermal aging and resulting failure during the long term operation affects operation performance of MMC. Utilizing the evolution rule of gate drive signals of PPIs to monitor condition of the elements is the important means to achieve condition monitoring of MMC sub-model. Firstly
a power cycling test platform for PPIs is set up to measure periodically the gate drive signal waveform during power cycling test period. The results show that with the increase of aging cycling number
the length of the Miller Plateau of the PPIs gate voltage increases and the gate drive current drops and lags along the groove position. Then
it is found after the microscopic observation of the thermal aged and failed chip surface of PPIs that the gate-oxidization layer accessories cracks occur on surface of chip emitter side. This thermal aging failure mode is the main reason of variation of the gate drive signal. Finally
based on the fully coupled multi-physical fields relationship inside PPIs
a finite element model of PPI is set up to calculate the temperature and pressure distribution of chip emitter surface. The calculation result shows that the thermal stress on the boundary area between additional metal layer and the underlying molybdenum layer is lager which
after a long period of alternating stress fatigue
causes crack germination of the gate-oxide layer and gradually leads to gate failure. The research result in this paper shows it is of certain significance to use the evolution rule of gate drive signals of PPI for the on-line monitoring of gate deterioration condition of PPIs.