Martti Aro

Application of dielectric response measurement on power cable systems

Water treeing is one of the factors leading to failure of medium voltage XLPE cables in long-term service. Increased moisture content inside oil-paper insulated cable is not desirable. To identify water tree degraded XLPE cables or oil-paper cables with high moisture content, diagnostic tests based on dielectric response (DR) measurement in time and frequency domain are used. Review of individual DR measurement techniques in the time and frequency domains indicates that measurement of one parameter in either domain may not be sufficient to reveal the status of the cable insulation. But a combination of several DR parameters can improve diagnostic results with respect to water trees present in XLPE cables or increased moisture content in oil-paper cables. DR measurement is a very useful tool that reveals average condition of cable systems. However, it is unlikely that DR measurement will detect few, but long water trees. In addition, DR cannot locate the defect or water tree site within the cable system. Combination of DR and partial discharge (PD) measurements can improve diagnostic results with respect to global and local defects. However, it is doubtful whether PD test can identify the presence of water trees inside a cable in a nondestructive manner. Further research is needed for more detailed conclusions regarding the status of a particular insulation and for predicting the remaining life of the insulation system.

Selectivity of damped AC (DAC) and VLF voltages in after-laying tests of extruded MV cable systems

The purpose of HV after-laying tests on cable systems on-site is to check the quality of installation. The test on extruded MV cable systems is usually a voltage test. However, in order to enhance the quality of after installation many researchers have proposed performance of diagnosis tests such as detection, location and identification of partial discharges (PD) and tan /spl delta/ measurements. Damped AC voltage (DAC) also called oscillating voltage waves (OVW) is used for PD measurement in after-laying tests of new cables and in diagnostic test of old cables. Continuous AC voltage of very low frequency (VLF) is used for withstand voltage tests as well as for diagnostic tests with PD and tan /spl delta/ measurements. Review on the DAC and VLF tests to detect defects during on-site after-laying tests of extruded MV cable systems is presented. Selectivity of DAC and VLF voltages in after-laying testing depends on different test parameters. PD process depends on type and frequency of the test voltage and hence, the breakdown voltage is different. The withstand voltage of XLPE cable insulation decreases linearly with increasing frequency in log scale. Experimental studies with artificial XLPE cable model indicate that detection of defects with DAC or VLF voltage can be done at a lower voltage than with DC. DAC voltage is sensitive in detecting defects that cause a breakdown due to void discharge, while VLF is sensitive in detecting defects that cause breakdown directly led by inception of electrical trees.

HV impulse measuring systems analysis and qualification by estimation of measurement errors via FFT, convolution, and IFFT

Quality of high-voltage (HV) impulse measurements requires adequate dynamic performance of measuring systems (MS) with respect to the waveforms to be measured. Qualification of MSs for various impulses is a prerequisite for repeatability and reproducibility of HV testing results. Alternative qualification procedures are considered with respect to the standard full lightning impulse including evaluation of significance of its frequency components. Duality of step responses and amplitude-frequency responses is described in detailed examples. Special attention is paid to creeping, encountered with some converting devices. Variations of MS response parameters can lead to improper conclusions when only conventional examination of the dynamic behavior is used. Evaluation of the response parameters is an unnecessary step, which brings additional errors to the HV measurements. The paper is pointing out advantages and practical achievements of the frequency-domain qualification of dynamic properties based on the step response data obtained both experimentally and analytically. Measurement errors of the impulse peak value and time parameters have been evaluated via fast Fourier transform (FFT), convolution, and inverse discrete Fourier transform (IFFT) for a commercial divider. Direct determination of the HV impulse measurement range, no need for any additional parameters, and illustrative graphical presentation of results are the main advantages of this approach. Limitations of different qualification approaches are discussed.

Worldwide Comparison of Lightning Impulse Voltage Measuring Systems at the 400-kV Level

An international comparison of lightning and switching impulse voltage measuring systems was arranged and coordinated by the Helsinki University of Technology (MIKES-TKK), Espoo, Finland, between 1999 and 2002. The number of participants was 26, including the coordinator. This paper summarizes the results obtained by those eight National Metrology Institutes that participated in the comparison. A 400-kV transfer reference measuring system for measuring lightning impulse voltages was prepared by the coordinator. In addition, a 1-kV impulse voltage calibrator system, including calibrators for both standard lightning and switching impulses, was circulated

A calculable impulse voltage calibrator for calibration of impulse digitizers

Design and construction of a calibrator for lightning and switching impulse voltage measuring instruments is presented. The operating voltage range is from 50 mV to 300 V. The estimated uncertainty for the peak value is 0.03% and for front time and time to half value less than 1%.

Calculable impulse voltage calibrator for calibration of impulse digitizers

Design and construction of a calibrator for lightning and switching impulse voltage measuring instruments is presented. The operating voltage range is from 50 mV to 300 V. The estimated uncertainty for the peak value is 0.03% and for front time and time to half value less than 1%.