T.R. McComb

Comparative evaluation of computer methods for calculating the best-fit sinusoid to the digital record of a high-purity sine wave

A comparative evaluation of programs used in the “equivalent bit” test is described. Four methods of determining the best-fit sine wave were tested using computer generated data contaminated to varying degrees with white noise. These data were supplemented with actual records of sine waves obtained using fast digitizers with 6- to 12-bit resolution and record lengths of between 512 to 16 000 words. The programs were evaluated for relative accuracy, absolute accuracy, and efficiency, both in terms of how often each method converged and also of how much computer time each method took to converge.

Measuring Characteristics of the Fastest Commercially-Available Digitizers

Measurements of the dynamic performance of eight different digitizers are presented and analysed to show the types and magnitudes of error to which each digitizer is prone. The impact of these errors on the measurements of fast single-shot transients, such as front chopped impulses with times-to-chopping in the range of 500 ns (the fastest covered by the standards) to a few ns, is examined. The suitable range of application of each digitizer is deduced.

International comparison of HV impulse measuring systems

Present standards for qualifying HV impulse measuring systems by unit-step-reponse parameters are complex and difficult to apply and some systems, which have response parameters within the limits of the standards, have unacceptable errors. This paper takes the first step in providing a simplified method based on simultaneous measurements of an HV impulse by a reference system and the system under test. Comparative measurements have been made in four National Laboratories and the relative differences are reported. The results are discussed and the further work which is required is outlined.

Comparative measurements of HV impulses to evaluate different sets of response parameters

Results are reported of comparative measurements made in 14 HV (high-voltage) laboratories in ten different countries. The theory of the proposed methods of characterizing the dynamic behavior is given, and the parameters to be used are discussed. Comparative measurements made using 95 systems based on 53 dividers are analyzed. This analysis shows that many of the system now in use, even though they fulfil the basic response requirements of the standards, do not meet the accuracy requirements. Because no transfer measurements were made between laboratories, there is no way to detect similar errors in both the system under test and the reference system. Hence, the situation may be worse than reported. This has led to the recommendation that comparative measurements should be the main route for quantifying industrial impulse measuring systems. >

Standards for digital recorders for measurements in high-voltage impulse tests

Abstract The background to the first standards for digital recorders for high-voltage impulse measurements is reviewed briefly. The emphasis in the first issues is given and the import of the decisions made in these early documents presented. The changes in the latest revision of Institute of Electrical and Electronics Engineers (IEEE) Std. 1122-1998 are reviewed briefly and contrasted with the approach in the latest Committee Draft for Voting on the revision of International Electrotechnical Commission (IEC) Document 1083-1. The importance of the changes in these documents is discussed. New work in the field is summarized and recent developments in digital recorders are discussed. The present status of digital measurements is summarized and proposals for future improvements to the standards are put forward.

Evaluation of practical definitions of parameters for smooth impulses and impulses with small amounts of distortion (overshoot and/or oscillations)

The existing definitions of the parameters of high-voltage impulses were derived to be used on measurements made on an oscillogram and hence they were based on a small number of readings which would usually have uncertainties of reading of about /spl plusmn/1% (or more) of the read parameter. Records taken on digital recorders may have much greater precision (and the possibility of lower uncertainties) and in some cases their precision may be increased by appropriate processing of the recorded data. This paper discusses definitions of the parameters used to define impulses and how they can be modified to facilitate automated data processing while maintaining a reasonable correlation with the very-large established database gathered using the old definitions. The parameters are evaluated using these new definitions on records of standard test impulses and the values obtained are compared with those obtained using the old definitions evaluated with either non-linear curve fitting or smoothing.