Hajime Urai

Development of partial discharge detection and diagnostic methods of vacuum circuit breaker

There is a possibility of partial discharge (PD) occurrence due to an abnormality of the insulation and the aging of the vacuum circuit breaker (VCB) in cubicle-type gas insulated switchgear (C-GIS). PD may contribute to a risk that leads to breakdown in the worst case, and cause failure of the equipment. Thus, knowing the characteristics of the PD internal VCB is required. In this work, a development of PD detection and diagnostic method of VCB is introducedA model VCB consisting of a vacuum interrupter (VI) was connected in series with a transformer. The internal pressure of VI was set by a turbo molecular pump via flexible hose, insulating tube and leak valves. The internal pressure of VI was verified up to 10-2 Pa by a digital metal ionization gauge. The diagnostic method was studied using a number of sensors with different specification. We have used detection resistance, horn antenna, transient earth voltage (TEV) sensor and three different frequency bands to detect PD signals.

Construction of measurement system of PD phenomena in medium vacuum region of vacuum interrupter

Vacuum interrupter (VI) is an alternative device that contributes to reduce usage of SF6 gas insulated apparatus in the future. For the past research, we have investigated the partial discharge (PD) phenomena and determined the gas pressure in VI based on PD patterns from 1 Pa to kPa order. We revealed that the pressure below and above 260 Pa were attributed to Townsend and Streamer-like discharge, respectively. We realized that it is also important to identify the PD phenomena and determine the gas pressure of VI based on PD at the pressure where the PD starts to occur, which is at the left side of Paschen curve (below 1 Pa order) for detection purposes. Thus, this paper presents the construction of PD measurement system for investigating PD phenomena in medium vacuum region of VI. The measurement system was confirmed free from unwanted corona discharge. We also identified the partial discharge inception voltage (PDIV) and PD pulse waveform from 0.2 to 5 Pa. The PDIV was identified to follow the left side of Paschen curve. In addition, the occurrence PD in medium vacuum of VI could be characterized as Townsend-like discharge.

Development of diagnostic methods for vacuum leakage from vacuum interrupter by partial discharge detection

Investigation is made to establish a monitoring method for vacuum leakage from a vacuum interrupter (VI) using a discharge detection method. AC rated voltage is applied to a practical VI filled with air to cause partial discharge (PD) inside VI. A diagnostic method is studied using a number of sensors with different specification. An attempt is made to compare the detection sensitivity of PD for the various sensors. Experimental results reveal that the rise time and magnitude of PD signal waveform detected with each sensor vary depending on discharge phenomena occurring at different internal pressures of VI. As a result, it is suggested that PD detection with the suitable sensors can determine the internal pressure in VI, along with distinguishing PD signal from noise.

Measurement of hot gas exhaust characteristics in SF 6 circuit breaker with small model interrupter

High-voltage circuit breakers have continuously been developed to reduce their size and operating energy. In the current interruption process, hot gas is generated and exhausted out of the conducting exhaust tube. The hot gas causes electrical breakdown at the exit of the exhaust tube. The hot gas characteristics in the exhaust tube were measured to clarify the dielectric strength. Two measurement methods, the small gap discharge method and the extra-fine thermocouple, were used to estimate the temperature of the hot gas. In the small gap discharge method, the temperature was deduced from the breakdown voltage at the small gap by referencing the critical electric field strength. The critical electric field strength is the electric field strength where ionization coefficient is equal to electron attachment one. An extra-fine thermocouple with a 25 micron tip was also used to measure the temperature on a trial basis. The response speed of the thermocouple was slower than that of transient change of temperature in the interruption process. First-order response lag was compensated by de-convolution with fast Fourier transform (FFT). As a result, the peak value of the temperature in the conducting exhaust tube agreed well with the results estimated by the other method.

Estimation of interruption performance in high-voltage circuit breakers with a modified method for arc parameter evaluation

Interruption performance of high-voltage circuit breakers has been quantitatively evaluated through current zero measurements during interruption tests. The interruption capabilities of the circuit breakers are able to be estimated by means of an arc model calculation. The parameters of the arc model are determined through the current zero measurements. In order to simulate more precisely the success or failure of current interruption, a modified extraction method of arc parameters is proposed in this paper. The arc parameters are calculated by combining an equivalent circuit calculation and a non-linear least square fitting for current zero measurement data. The arc model calculation with the modified extraction method of arc parameters is applied to the development and design of an actual high-voltage circuit breaker. The capacitance required for 80 kA interruption can be estimated based on the current zero data measured during 50 kA interruption tests. The test results of interruption and failure agree well with the predicted performance. The proposed method consisting of the arc parameter extraction from current zero data and characterization of the parameters as the function of arc time constant can evaluate short-line fault interruption performance in high-voltage circuit breakers.