Summary
Comparison tests between Japanese national-standard-class measuring systems for switching-impulse high voltage and reference measuring systems of 5 highvoltage laboratories in Japan are carried out. In the comparison tests, switching-impulse high-voltage waveforms of 200/2000 ?s and of 300/2400 ?s are employed. The peak values of the applied voltage are 180 kV, 260 kV, 340 kV, 420 kV or 500 kV in the positive or in the negative polarity. The switching-impulse voltage is applied 10 times at each voltage level. The tested reference measuring systems are calibrated by comparing measured parameters, which are peak value, front time and time-to-half value. The calibration factors are derived as quotients of the measured parameters by the standard measuring system and those by a tested reference measuring system. The uncertainties of the measurements by the reference measuring systems are evaluated combining standard uncertainty of the measurement by the standard measuring system, statistical type-A uncertainty in the comparison test, uncertainty due to non-linearity, and uncertainty due to influence of the nominal epoch. The evaluated uncertainties of the measurement by all the reference measuring systems are less than the requirements of IEC60060-2: 2010. Using measured data in the comparison tests, influences of software packages employed for the calculation of the parameters of the measured switching impulse voltage waveforms on the calibrated values are discussed. In the discussion, the parameters of the measured switching impulse voltage waveforms are recalculated by independently developed 2 software packages. From this discussion, it is found that the calculated front time and the time to half value have biases originated from individually employed software packages. When the software package employed in the standard measuring system is also employed in the reference measuring systems, the influence of the biases can be cancelled and the evaluated uncertainties in the measurement of fronttime and time-to-half-value can be lowered.
Additional informations
| Publication type | ISH Collection |
|---|---|
| Reference | ISH2017_253 |
| Publication year | |
| Publisher | ISH |
| File size | 2 MB |
| Pages number | 6 |
| Price for non member | Free |
| Price for member | Free |
