NPL REPORT TQE4. EUROMET Projects 473 and 612: Comparison of the measurement of current transformers (CTs) EUROMET.EM-S11 Final report

Transcription

1 NPL REPORT TQE EUROMET Projects and : Comparison of the measurement of current transformers (CTs) EUROMET.EMS Final report Stuart Harmon Lesley Henderson NOT RESTRICTED MARCH

2 EUROMET Projects and : Comparison of the measurement of current transformers (CTs) Stuart Harmon and Lesley Henderson Industry and Innovation Division ABSTRACT The Euromet comparison titled Comparison of the measurement of current transformers was carried out over two projects with NPL as pilot laboratory and thirteen other participating European National Measurement Institutes (NMI). Current transformer measurements made by the participating NMIs support a large number of measurements made in the electrical generation, supply and distribution industries in their own countries. They also support many transformer manufacturers who rely on national standards as a source of traceability.

3 Crown copyright Reproduced with the permission of the Controller of HMSO and Queen's Printer for Scotland ISSN National Physical Laboratory Hampton Road, Teddington, Middlesex, TW LW Extracts from this report may be reproduced provided the source is acknowledged and the extract is not taken out of context. Approved on behalf of the Managing Director, NPL by Dr J T Janssen, Knowledge Leader, Time, Quantum and Electromagnetics Team Authorised by Director, Industry and Innovation Division

4 TABLE OF CONTENTS Introduction... Participants... Transfer standard... Measurements... Symbols and definitions... Measurement conditions and methods... Traceability... Transfer standard behaviour... Sample uncertainty budget... Comparison reference value... Results... Conclusion... References... APPENDIX : Types of standard used in comparison... APPENDIX : Measurement methods... APPENDIX : BNMLCIE Results... APPENDIX : Amended uncertainty values from VMT/VMC... APPENDIX : Sample uncertainty budgets... Introduction The Euromet comparison titled Comparison of the measurement of current transformers was carried out over two projects with NPL as pilot laboratory. Project started in April, with ten other participating European National Measurement Institutes (NMI) and project started in January with a further three European NMIs. Current transformer measurements made by the participating NMIs support a large number of measurements made in the electrical generation, supply and distribution industries in their own countries. They also support many transformer manufacturers who rely on national standards as a source of traceability. The current (ratio) errors and phase displacement of each ratio of the uncompensated current transformer transfer standard were determined at a defined frequency, burden and power factor, using each participant s standard measuring method and equipment.

5 Participants Pilot laboratory: L Henderson, S Harmon, A Wheaton, NPL, National Physical Laboratory, Teddington, United Kingdom Project participants and affiliation in order of transfer standard circulation: A Bergmann, E Joons, SP, Swedish National Testing and Research Institute, Borås, Sweden A Deffrennes, W Vancoetsem, G Vandermeersch, LBE, Laborelec, Linkebeek, Belgium H Latzel, G Roeissle, PTB, PhysikalischTechnische Bundesanstalt, Braunschweig, Germany B Pączek, GUM, Central Office of Measures, Warsaw, Poland A Boros, J Gellén, I Szunyogh, OMH, National Office of Measures, Budapest, Hungary G Crotti, IEN, Istituto Elettrotecnico Nazionale Galileo Farraris, Turin, Italy Affiliation at time of publishing: I.N.RI.M., Istituto Nazionale di Ricerca Metrologica R Kämpfer, B Jeckelmann, METAS, Swiss Federal Office of Metrology and Accreditation, BernWabern, Switzerland W Waldmann, BEV, Bundesamt für Eichund Vermessungswesen, Vienna, Austria E Suomalainen, HUT, Helsinki University of Technology, Helsinki, Finland A Rautiainen, P Helistö, MIKES, Centre for Metrology and Accreditation, Helsinki, Finland Project Participants and affiliation in order of transfer standard circulation: R Styblíková, CMI, Czech Metrological Institute, Prague, Czech Republic V Gegevicius, Z Balachoviciené, VMT/VMC, Vilnius Metrology Centre, Vilnius, Lithuania I Blanc, BNMLCIE, Laboratoire Central des Industries Électriques, Fontenayaux Roses, France Affiliation at time of publishing: LNE, Laboratoire national de métrologie et d essais H Çayci, UME, Ulusal Metroloji Enstitüsü, GebzeKocaeli, Turkey

6 Transfer standard Manufacturer: Serial number: Transformation ratios: Class and rating: Smith Hobson Ltd, United Kingdom J,,,,,,,, / A Class. at VA, Hz Measurements Each participating laboratory was asked to make measurements under their normal laboratory conditions using the following guidelines: Transformation ratios: All Burden: VA, cos β = Test frequency; or Hz Temperature: or ± C Rated current:,,,,, & % (% optional) Symbols and definitions The symbols and definitions stated are in accordance with IEC : []. ε X Current (ratio) error: The error which a transformer introduces into the measurement of a current and which arises from the fact that the actual transformation ratio is not equal to the rated transformation ratio. The current error expressed in ppm is given by the formula: ε X (ppm) = ( K n I s I p ) / I p K n Rated transformation ratio. I p Actual primary current. I s Actual secondary current when I p is flowing under the conditions of measurement. Z B Burden: The impedance of the secondary circuit in ohms and power factor. The burden is expressed as the apparent power in voltamperes absorbed at a specified power factor and at the rated secondary current. cos β Power factor for sinusoidal waveforms. I/I n Excitation current, expressed in percent of rated current. δ X Phase displacement: The difference in phase, expressed in µrad, between the primary and secondary current vectors, the direction of the vectors being so chosen that the angle is zero for a perfect transformer. The phase displacement is said to be positive when the secondary current vector leads the primary current vector. ε ref ε X comparison reference values. δ ref δ X comparison reference values. χ ε ε X deviation from comparison reference values ε ref. χ δ δ X deviation from comparison reference values δ ref. U The expanded uncertainty of measurement supplied by each participant, stated as the standard uncertainty multiplied by a coverage factor k.

7 Measurement conditions and methods From each participant s report, each laboratory s measuring conditions and method of calibration are summarised in the following table. Uncertainty values given in the table are those quoted in the participant s report. Laboratory Date measured Note Table : Measurement conditions Method Temperature Burden Frequency Note ºC VA Cos β Hz NPL Note ± ± % & SP / ± ± % ±. LBE / ±. PTB / ±. & GUM / ± %. ±. OMH / ± ± % IEN /. ±.. ±. METAS / ± ± %. BEV / ± ± % HUT / ±. ± % MIKES /. ±... CMI / ±. ±.% VMT/VMC /. ±. ±. BNMLCIE /. ± UME / ± Note : The Date measured is the month in which the last measurements were carried out. Note : Measurements were made by NPL six times during both comparisons. Measurements were made at the start of project in April, two more sets were made during circulation in August and February, and a final set at the end of the project in August. Measurements were made at the start of project in January and at the end of the project in May. Note : Two distinct Methods were used by the participants to calculate the current error and phase displacement of each transfer standard ratio. Descriptions of each method are given in Appendix. Method : Comparison against compensated current comparators and/or standard current transformers with errors measured on homemade or commercial test sets.

8 Method : Calibrated using Rogowski coil and/or current shunts with outputs measured using digital multimeters. Traceability Each participant supplied a statement of traceability to the SI. The following table shows if traceability is to their own national standards or if their traceability is to another national laboratory. Laboratory NPL SP LBE PTB GUM Table : Traceability Traceable to own National Standards Traceable to other National Standards PTB PTB OMH PTB IEN METAS BEV PTB HUT MIKES CMI VMT/VMC PTB BNMLCIE UME Transfer standard behaviour NPL measured the transfer standard six times during both projects, at the start and end of both projects and twice during circulation of project, so that any change of the transfer standard would be detected. The standard deviation of these measurements shows that there was no significant change throughout the period of the comparison. Figure shows the current error deviation from the reference value for the measurements made by NPL for ratio / at I/I n = %, at a burden of VA, unity power factor, at a frequency of Hz and at an ambient temperature of C. Typically the standard deviation of the mean of all NPL s measurements can be shown to be less than ppm.

9 Figure : Ratio /, I/I n = %, Transfer standard behaviour Current error /ppm Apr Jun Aug Oct Dec Feb Apr Jun Aug Oct Dec Feb Apr Jun Aug Oct Dec Feb Apr Date of measurement Figure shows the phase displacement as a deviation from the reference value for the measurements made by NPL for ratio / at I/I n = %, at a burden of VA, unity power factor, at a frequency of Hz and at an ambient temperature of C. Typically the standard deviation of the mean for NPL s phase displacement measurements can be shown to be less than µrad. Figure : Ratio /, I/I n = %, Transfer standard behaviour Phase error /µrad Apr Jun Aug Oct Dec Feb Apr Jun Aug Oct Dec Feb Apr Jun Aug Oct Dec Feb Apr Date of measurement Sample uncertainty budget Table shows a typical uncertainty budget used by NPL in the calculation of its uncertainty values. The uncertainty budget given shows the contributions associated with the measurements made on ratio / at I/I n = %, at a burden of VA, unity power factor, at a frequency of Hz and at an ambient temperature of C. The uncertainty budgets supplied by each of the participants contained all or most of the principle contributions listed below and are given in Appendix.

10 Table : Sample uncertainty budget Current error uncertainty, ratio /, I/I n = % Source of uncertainty Value Type Prob.Dist. Divisor C i u i (y) υ i or (±) ppm (±) ppm υ eff Calibration of bridge and comparator B normal. Error in the bridge B rectangular.. Error due to frequency setting B rectangular.. Resolution of test set. B rectangular.. Error due to burden setting. B rectangular.. Circuit configuration B normal. Error due to current setting B rectangular.. Repeatability. A normal. Combined uncertainty. Expanded uncertainty k=. The contributions for the Calibration of the bridge and comparator and Error in the bridge take into account any error of the test set and standard comparator used in calibration of the transfer standard. The contribution for the error due to the burden setting takes into account the fact that the actual burden was either too high or too low. The change in ε X and δ X for a change in VA was calculated and then a contribution included in the budget. The contribution for the error due to the setting of the test frequency takes into account any deviation from the stated measurement frequency. The contribution for the error due to the current setting covers any inaccuracy in the setting of the applied current, I/I n. The value for repeatability is the standard deviation of the mean for each individual set of measurements. The transfer standard was measured at Hz and Hz by NPL and PTB. The average change between the two frequencies measured by both laboratories is less than ppm for current error and µrad for phase displacement. The influence of temperature on the measured values is not included in the uncertainty budget. The transfer device was measured by NPL at both ºC and ºC. The average difference between ºC and ºC for all ratios at all current levels is less than ppm for current error and µrad for phase displacement. The degrees of freedom (υ i ) for all type B contributions are assumed to be infinite []. The effective degrees of freedom (υ eff ) for all type A contributions are estimated using the WelchSatterwaite formula [] based on the degrees of freedom of the individual uncertainty contributions u i (y) to obtain a coverage factor k. The WelchSatterwaite formula is:

11 v eff = N i u ( y) i u ( y) v i For all measurements made by NPL, the reported expanded uncertainties are based on a standard uncertainty multiplied by a coverage factor of k =, which for a normal distribution provides a level of confidence of approximately %. All participants included a similar statement, quoting a coverage factor of k =, except for IEN s phase displacement results which were quoted at k =. and METAS whose quoted coverage factor was k. Comparison reference value The comparison reference values have been calculated using the weighted mean of each set of measurement results. The draft A report for project used the median, described in NPL report CISE / [], as the estimator of the reference value. This approach was used because it is less influenced by the presence of extreme values. After a participant from project resubmitted results (with the consent of all participants) and with the inclusion of the results from project there was no significant difference in the reference values obtained using either a weighted mean or the median. The difference in the comparison reference values calculated with and without the inclusion of results supplied from laboratories with traceability to other NMIs was generally within the reference value uncertainty. There were several exceptions however, with the most severe being the phase displacement reference value for ratio / at I/I n = %, where the difference was µrads. Therefore the comparison reference values have been calculated using only those laboratories whose results are not correlated to other National Measurement Institutes. From [], the comparison reference value ε ref, calculated as the weighted mean of each set of measurements is given by: N ε ε ref j = u u ref j= where ε j is each participants individual result u j is each participants combined uncertainty for ε j. The standard deviation (standard uncertainty) u ref of ε ref is given by: j u ref = N j= u j A % confidence level is given by: ε ref ± ku ref where k is the coverage factor determined from the t distribution table in []. For calculation of the phase displacement reference values substitute δ for ε.

12 Table gives the comparison reference values, in ppm for current error and µrad for phase displacement, and associated uncertainties (at % confidence level) calculated from the above formulas. The results supplied from LBE, GUM, OMH, BEV and VMT/VMC were not used in the calculation of the comparison reference values as their results are correlated to another national institute. The results supplied by BNMLCIE were not used in the calculation of the comparison reference values as they were received after the release of Draft A, see Appendix. Table : Comparison reference values I/I n Ratio / Ratio / Ratio / % ε ref Uε ref δ ref Uδ ref ε ref Uε ref δ ref Uδ ref ε ref Uε ref δ ref Uδ ref I/I n Ratio / Ratio / Ratio / % ε ref Uε ref δ ref Uδ ref ε ref Uε ref δ ref Uδ ref ε ref Uε ref δ ref Uδ ref I/I n Ratio / Ratio / Ratio / % ε ref Uε ref δ ref Uδ ref ε ref Uε ref δ ref Uδ ref ε ref Uε ref δ ref Uδ ref

13 Results The results obtained from the comparison, are displayed in two ways. The first is a set of graphs showing the current error comparison results for all ratios at I/I n = %. Selected graphs for lower current excitations and phase displacement are also shown. The graphs, Figures to, show each participant s results as a deviation from the reference value and the uncertainty associated with each deviation. The uncertainty associated with each deviation from the reference value is calculated as the combination of each participant s individual expanded uncertainty and the expanded uncertainty of the reference value for that point. Also shown on each graph are the % confidence intervals for that set of results. Following the graphs is a set of tables, Tables to, displaying the results supplied by all participants, as deviations from the comparison reference values for both current error and phase displacement. The uncertainty associated with each deviation from the reference value is also given.

14 Figure : Ratio /, I/I n = %, current error comparison results Laboratory /comparison order NPL SP() SP() LBE PTB GUM OMH IEN METAS BEV HUT MIKES CMI VMT/VMC UME Current error deviation from reference value /ppm Figure shows the current error comparison results for ratio /, at I/I n = %, given as the deviation from the reference value. The error bars for each point show the uncertainty associated with each deviation. The ε ref (central vertical line) and the % confidence intervals (outer vertical lines) are also shown.

15 Figure : Ratio /, I/I n = %, current error comparison results Laboratory /comparison order NPL SP() SP() LBE PTB GUM OMH IEN METAS BEV HUT MIKES CMI VMT/VCM UME Current error deviation from reference value /ppm Figure shows the current error comparison results for ratio /, at I/I n = %, given as the deviation from the reference value. The error bars for each point show the uncertainty associated with each deviation. The ε ref (central vertical line) and the % confidence intervals (outer vertical lines) are also shown.

16 Figure : Ratio /, I/I n = %, current error comparison results Laboratory /comparison order NPL SP() SP() LBE PTB GUM OMH IEN METAS BEV HUT MIKES CMI VMT/VMC UME Current error deviation from reference value /ppm Figure shows the current error comparison results for ratio /, at I/I n = %, given as the deviation from the reference value. The error bars for each point show the uncertainty associated with each deviation. The ε ref (central vertical line) and the % confidence intervals (outer vertical lines) are also shown.

17 Figure : Ratio /, I/I n = %, phase displacement comparison results Laboratory /comparison order NPL SP() SP() LBE PTB GUM OMH IEN METAS BEV HUT MIKES CMI VMT/VMC UME Phase displacement deviation from reference value /µrad Figure shows the phase displacement comparison results for ratio /, at I/I n = %, given as the deviation from the reference value. The error bars for each point show the uncertainty associated with each deviation. The δ ref (central vertical line) and the % confidence intervals (outer vertical lines) are also shown.

18 Figure : Ratio /, I/I n = %, phase displacement comparison results Laboratory /comparison order NPL SP() SP() LBE PTB GUM OMH IEN METAS BEV HUT MIKES CMI VMT/VMC UME Phase displacement deviation from reference value /µrad Figure shows the phase displacement comparison results for ratio /, at I/I n = %, given as the deviation from the reference value. The error bars for each point show the uncertainty associated with each deviation. The δ ref (central vertical line) and the % confidence intervals (outer vertical lines) are also shown.

19 Figure : Ratio /, I/I n = %, current error comparison results Laboratory /comparison order NPL SP() SP() LBE PTB GUM OMH IEN METAS BEV HUT MIKES CMI VMT/VMC UME Current error deviation from reference value /ppm Figure shows the current error comparison results for ratio /, at I/I n = %, given as the deviation from the reference value. The error bars for each point show the uncertainty associated with each deviation. The ε ref (central vertical line) and the % confidence intervals (outer vertical lines) are also shown.

20 Figure : Ratio /, I/I n = %, current error comparison results Laboratory /comparison order NPL SP() SP() LBE PTB GUM OMH IEN METAS BEV HUT MIKES CMI VMT/VMC UME Current error deviation from reference value /ppm Figure shows the current error comparison results for ratio /, at I/I n = %, given as the deviation from the reference value. The error bars for each point show the uncertainty associated with each deviation. The ε ref (central vertical line) and the % confidence intervals (outer vertical lines) are also shown.

21 Figure : Ratio /, I/I n = %, current error comparison results Laboratory /comparison order NPL SP() SP() LBE PTB GUM OMH IEN METAS BEV HUT MIKES CMI VMT/VMC UME Current error deviation from reference value /ppm Figure shows the current error comparison results for ratio /, at I/I n = %, given as the deviation from the reference value. The error bars for each point show the uncertainty associated with each deviation. The ε ref (central vertical line) and the % confidence intervals (outer vertical lines) are also shown.

22 Figure : Ratio /, I/I n = %, current error comparison results Laboratory /comparison order NPL SP() SP() LBE PTB GUM OMH IEN METAS BEV HUT MIKES CMI VMT/VMC UME Current error deviation from reference value /ppm Figure shows the current error comparison results for ratio /, at I/I n = %, given as the deviation from the reference value. The error bars for each point show the uncertainty associated with each deviation. The ε ref (central vertical line) and the % confidence intervals (outer vertical lines) are also shown.

23 Figure : Ratio /, I/I n = %, current error comparison results Laboratory /comparison order NPL SP() SP() LBE PTB GUM OMH IEN METAS BEV HUT MIKES CMI VMT/VMC UME Current error deviation from reference value /ppm Figure shows the current error comparison results for ratio /, at I/I n = %, given as the deviation from the reference value. The error bars for each point show the uncertainty associated with each deviation. The ε ref (central vertical line) and the % confidence intervals (outer vertical lines) are also shown.

24 Figure : Ratio /, I/I n = %, current error comparison results Laboratory /comparison order NPL SP() SP() LBE PTB GUM OMH IEN METAS BEV HUT MIKES CMI VMT/VMC UME Current error deviation from reference value /ppm Figure shows the current error comparison results for ratio /, at I/I n = %, given as the deviation from the reference value. The error bars for each point show the uncertainty associated with each deviation. The ε ref (central vertical line) and the % confidence intervals (outer vertical lines) are also shown.

25 Figure : Ratio /, I/I n = %, current error comparison results Laboratory /comparison order NPL SP() SP() LBE PTB GUM OMH IEN METAS BEV HUT MIKES CMI VMT/VMC UME Current error deviation from reference value /ppm Figure shows the current error comparison results for ratio /, at I/I n = %, given as the deviation from the reference value. The error bars for each point show the uncertainty associated with each deviation. The ε ref (central vertical line) and the % confidence intervals (outer vertical lines) are also shown.

26 Figure : Ratio /, I/I n = %, phase displacement comparison results Laboratory /comparison order NPL SP() SP() LBE PTB GUM OMH IEN METAS BEV HUT MIKES CMI VMT/VMC UME Phase displacement deviation from reference value /µrad Figure shows the phase displacement comparison results for ratio /, at I/I n = %, given as the deviation from the reference value. The error bars for each point show the uncertainty associated with each deviation. The δ ref (central vertical line) and the % confidence intervals (outer vertical lines) are also shown.

27 Figure : Ratio /, I/I n = %, phase displacement comparison results Laboratory /comparison order NPL SP() SP() LBE PTB GUM OMH IEN METAS BEV HUT MIKES CMI VMT/VMC UME Phase displacement deviation from reference value /µrad Figure shows the phase displacement comparison results for ratio /, at I/I n = %, given as the deviation from the reference value. The error bars for each point show the uncertainty associated with each deviation. The δ ref (central vertical line) and the % confidence intervals (outer vertical lines) are also shown.

28 Figure : Ratio /, I/I n = %, current error comparison results Laboratory /comparison order NPL SP() SP() LBE PTB GUM OMH IEN METAS BEV HUT MIKES CMI VMT/VMC UME Current error deviation from reference value /ppm Figure shows the current error comparison results for ratio /, at I/I n = %, given as the deviation from the reference value. The error bars for each point show the uncertainty associated with each deviation. The ε ref (central vertical line) and the % confidence intervals (outer vertical lines) are also shown.

29 Table : Current error deviations from reference values for ratio / GB SE BE DE PL HU IT I/I n NPL SP () SP () LBE PTB GUM OMH IEN % χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε CH AT FI CZ LT TR I/I n METAS BEV HUT MIKES CMI VMT/VMC UME % χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε, is the current error deviation from the reference value for each point supplied by each participant in ppm. Uχ ε, is the expanded uncertainty for each deviation in ppm., indicates that the point was not reported. SP () refers to results given on pages to of their report and SP () refers to results given on pages to of their report.

30 Table : Phase displacement deviations from reference values for ratio / GB SE BE DE PL HU IT I/I n NPL SP () SP () LBE PTB GUM OMH IEN % χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ CH AT FI CZ LT TR I/I n METAS BEV HUT MIKES CMI VMT/VMC UME % χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ, is the phase displacement deviation from the reference value for each point supplied by each participant in µrad. Uχ δ, is the expanded uncertainty for each deviation in µrad., indicates that the point was not reported. SP () refers to results given on pages to of their report and SP () refers to results given on pages to of their report.

31 Table : Current error deviations from reference values for ratio / GB SE BE DE PL HU IT I/I n NPL SP () SP () LBE PTB GUM OMH IEN % χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε CH AT FI CZ LT TR I/I n METAS BEV HUT MIKES CMI VMT/VMC UME % χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε, is the current error deviation from the reference value for each point supplied by each participant in ppm. Uχ ε, is the expanded uncertainty for each deviation in ppm., indicates that the point was not reported. SP () refers to results given on pages to of their report and SP () refers to results given on pages to of their report.

32 Table : Phase displacement deviations from reference values for ratio / GB SE BE DE PL HU IT I/I n NPL SP () SP () LBE PTB GUM OMH IEN % χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ CH AT FI CZ LT TR I/I n METAS BEV HUT MIKES CMI VMT/VMC UME % χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ, is the phase displacement deviation from the reference value for each point supplied by each participant in µrad. Uχ δ, is the expanded uncertainty for each deviation in µrad., indicates that the point was not reported. SP () refers to results given on pages to of their report and SP () refers to results given on pages to of their report.

33 Table : Current error deviations from reference values for ratio / GB SE BE DE PL HU IT I/I n NPL SP () SP () LBE PTB GUM OMH IEN % χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε CH AT FI CZ LT TR I/I n METAS BEV HUT MIKES CMI VMT/VMC UME % χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε, is the current error deviation from the reference value for each point supplied by each participant in ppm. Uχ ε, is the expanded uncertainty for each deviation in ppm., indicates that the point was not reported. SP () refers to results given on pages to of their report and SP () refers to results given on pages to of their report.

34 Table : Phase displacement deviations from reference values for ratio / GB SE BE DE PL HU IT I/I n NPL SP () SP () LBE PTB GUM OMH IEN % χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ CH AT FI CZ LT TR I/I n METAS BEV HUT MIKES CMI VMT/VMC UME % χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ, is the phase displacement deviation from the reference value for each point supplied by each participant in µrad. Uχ δ, is the expanded uncertainty for each deviation in µrad., indicates that the point was not reported. SP () refers to results given on pages to of their report and SP () refers to results given on pages to of their report.

35 Table : Current error deviations from reference values for ratio / GB SE BE DE PL HU IT I/I n NPL SP () SP () LBE PTB GUM OMH IEN % χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε CH AT FI CZ LT TR I/I n METAS BEV HUT MIKES CMI VMT/VMC UME % χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε, is the current error deviation from the reference value for each point supplied by each participant in ppm. Uχ ε, is the expanded uncertainty for each deviation in ppm., indicates that the point was not reported. SP () refers to results given on pages to of their report and SP () refers to results given on pages to of their report.

36 Table : Phase displacement deviations from reference values for ratio / GB SE BE DE PL HU IT I/I n NPL SP () SP () LBE PTB GUM OMH IEN % χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ CH AT FI CZ LT TR I/I n METAS BEV HUT MIKES CMI VMT/VMC UME % χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ, is the phase displacement deviation from the reference value for each point supplied by each participant in µrad. Uχ δ, is the expanded uncertainty for each deviation in µrad., indicates that the point was not reported. SP () refers to results given on pages to of their report and SP () refers to results given on pages to of their report.

37 Table : Current error deviations from reference values for ratio / GB SE BE DE PL HU IT I/I n NPL SP () SP () LBE PTB GUM OMH IEN % χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε CH AT FI CZ LT TR I/I n METAS BEV HUT MIKES CMI VMT/VMC UME % χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε, is the current error deviation from the reference value for each point supplied by each participant in ppm. Uχ ε, is the expanded uncertainty for each deviation in ppm., indicates that the point was not reported. SP () refers to results given on pages to of their report and SP () refers to results given on pages to of their report.

38 Table : Phase displacement deviations from reference values for ratio / GB SE BE DE PL HU IT I/I n NPL SP () SP () LBE PTB GUM OMH IEN % χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ CH AT FI CZ LT TR I/I n METAS BEV HUT MIKES CMI VMT/VMC UME % χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ, is the phase displacement deviation from the reference value for each point supplied by each participant in µrad. Uχ δ, is the expanded uncertainty for each deviation in µrad., indicates that the point was not reported. SP () refers to results given on pages to of their report and SP () refers to results given on pages to of their report.

39 Table : Current error deviations from reference values for ratio / GB SE BE DE PL HU IT I/I n NPL SP () SP () LBE PTB GUM OMH IEN % χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε CH AT FI CZ LT TR I/I n METAS BEV HUT MIKES CMI VMT/VMC UME % χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε, is the current error deviation from the reference value for each point supplied by each participant in ppm. Uχ ε, is the expanded uncertainty for each deviation in ppm., indicates that the point was not reported. SP () refers to results given on pages to of their report and SP () refers to results given on pages to of their report.

40 Table : Phase displacement deviations from reference values for ratio / GB SE BE DE PL HU IT I/I n NPL SP () SP () LBE PTB GUM OMH IEN % χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ CH AT FI CZ LT TR I/I n METAS BEV HUT MIKES CMI VMT/VMC UME % χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ, is the phase displacement deviation from the reference value for each point supplied by each participant in µrad. Uχ δ, is the expanded uncertainty for each deviation in µrad., indicates that the point was not reported. SP () refers to results given on pages to of their report and SP () refers to results given on pages to of their report.

41 Table : Current error deviations from reference values for ratio / GB SE BE DE PL HU IT I/I n NPL SP () SP () LBE PTB GUM OMH IEN % χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε CH AT FI CZ LT TR I/I n METAS BEV HUT MIKES CMI VMT/VMC UME % χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε, is the current error deviation from the reference value for each point supplied by each participant in ppm. Uχ ε, is the expanded uncertainty for each deviation in ppm., indicates that the point was not reported. SP () refers to results given on pages to of their report and SP () refers to results given on pages to of their report.

42 Table : Phase displacement deviations from reference values for ratio / GB SE BE DE PL HU IT I/I n NPL SP () SP () LBE PTB GUM OMH IEN % χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ CH AT FI CZ LT TR I/I n METAS BEV HUT MIKES CMI VMT/VMC UME % χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ, is the phase displacement deviation from the reference value for each point supplied by each participant in µrad. Uχ δ, is the expanded uncertainty for each deviation in µrad., indicates that the point was not reported. SP () refers to results given on pages to of their report and SP () refers to results given on pages to of their report.

43 Table : Current error deviations from reference values for ratio / GB SE BE DE PL HU IT I/I n NPL SP () SP () LBE PTB GUM OMH IEN % χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε CH AT FI CZ LT TR I/I n METAS BEV HUT MIKES CMI VMT/VMC UME % χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε, is the current error deviation from the reference value for each point supplied by each participant in ppm. Uχ ε, is the expanded uncertainty for each deviation in ppm., indicates that the point was not reported. SP () refers to results given on pages to of their report and SP () refers to results given on pages to of their report.

44 Table : Phase displacement deviations from reference values for ratio / GB SE BE DE PL HU IT I/I n NPL SP () SP () LBE PTB GUM OMH IEN % χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ CH AT FI CZ LT TR I/I n METAS BEV HUT MIKES CMI VMT/VMC UME % χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ, is the phase displacement deviation from the reference value for each point supplied by each participant in µrad. Uχ δ, is the expanded uncertainty for each deviation in µrad., indicates that the point was not reported. SP () refers to results given on pages to of their report and SP () refers to results given on pages to of their report.

45 Table : Current error deviations from reference values for ratio / GB SE BE DE PL HU IT I/I n NPL SP () SP () LBE PTB GUM OMH IEN % χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε CH AT FI CZ LT TR I/I n METAS BEV HUT MIKES CMI VMT/VMC UME % χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε Uχ ε χ ε, is the current error deviation from the reference value for each point supplied by each participant in ppm. Uχ ε, is the expanded uncertainty for each deviation in ppm., indicates that the point was not reported. SP () refers to results given on pages to of their report and SP () refers to results given on pages to of their report.

46 Table : Phase displacement deviations from reference values for ratio / GB SE BE DE PL HU IT I/I n NPL SP () SP () LBE PTB GUM OMH IEN % χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ CH AT FI CZ LT TR I/I n METAS BEV HUT MIKES CMI VMT/VMC UME % χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ Uχ δ χ δ, is the phase displacement deviation from the reference value for each point supplied by each participant in µrad. Uχ δ, is the expanded uncertainty for each deviation in µrad., indicates that the point was not reported. SP () refers to results given on pages to of their report and SP () refers to results given on pages to of their report.

47 Conclusion There are several points to bear in mind whilst analysing the results of the comparison. The ambient temperature at which the measurements were made varies from to ºC. There is some variation in the burden at which the measurements were made, although most participants have included contributions in their uncertainty budgets for any deviation from the nominal value. The quoted uncertainties given by each participant were quite varied over all the measured ratios. An extreme example is ratio / at I/I n = % where the quoted uncertainties range from ± ppm to ± ppm for current error, and from ± µrad to ± µrad for phase displacement. In several cases participants have been making current transformer measurements with new measurement systems and techniques and, in one case for the first time, therefore a large amount of experience in the measurement and interpretation of results has been obtained from this comparison. The results supplied by each participant generally show good agreement but with a few exceptions over the whole range of measured values. Deviations from the comparison reference value were mostly within the quoted uncertainties, but again with a few exceptions. It is helpful to analyse these outlying results in more detail and a summary of these is shown in Tables and. Although the measurements were performed in this comparison early in the transition period of the Mutual Recognition Agreement (MRA), it is still useful to consider the results with reference to laboratories declared Calibration and Measurement Capabilities (CMCs) and to propose actions where systematic effects are present. The results in the comparison and in the CMC database are all quoted with an uncertainty confidence of %, so a in occurrence of an outlying result may be expected. There are possible results (current error and phase displacement for ratios), so participants having or lying outside the combined uncertainty of the result and the reference value may be expected on a statistical basis. It is likely that NPL, UME and OMH have results that may be considered on this basis, although one of OMH s outlying results is significantly larger than the relevant declared CMC. There are clearly some laboratories, which have significant systematic effects in their results. These are GUM and BNMLCIE, and their CMC declarations are clearly not supported. The situation for VMT/VMC was similar initially but was improved as they reassessed their analysis of uncertainty, although after they had seen the reference values. This is commented on elsewhere in this report. Two remaining laboratories, SP and IEN, have some ratios with outlying results but possibly too many for a statistical reason. Where there appears to be some systematic effect the laboratory is encouraged to investigate. For IEN, the occurrence of discrepancies looks nonsystematic but possibly the CMC declarations are too low. For

48 SP, there may be a systematic effect for the higher current ratios at low percentages of I/In. And again, it is possible that their CMC declarations are too low. The outlying differences given in Tables and are the difference between the deviations from the reference values, χ ε and χ δ, and the expanded uncertainty, Uχ ε and Uχ δ, for each deviation. Lab. Table : Current error outlying results Extent of outlying results Largest outlying difference Expanded uncertainty Declared CMC UME /, PR ppm ppm none OMH /, PR /, PR ppm ppm ppm ppm ppm ppm SP, and /, PR ppm ppm ppm GUM,,,, and /, PR ppm ppm ppm BNMLCIE,, and /, FR,,, and /, PR ppm ppm ppm ppm ppm ppm VMT/VMC All results, prior to increase of uncertainties none Lab. Table : Phase displacement outlying results Extent of outlying results Largest outlying difference Expanded uncertainty Declared CMC NPL /, PR µrad µrad µrad UME /, PR µrad µrad none OMH IEN SP /, FR /, PR /, FR, and /, PR /, PR, and /, PR µrad µrad µrad µrad µrad µrad µrad µrad µrad µrad µrad µrad µrad µrad µrad µrad µrad µrad GUM,,,,,, and /, PR µrad µrad µrad BNMLCIE /, FR,,,,,, and /, PR µrad µrad µrad µrad µrad µrad VMT/VMC All results, prior to increase of uncertainties none Legend Lab. PR FR Laboratory Partial ratio Full ratio

49 References [] IEC :, Edition., Instrument Transformers Part : Current Transformers [] BIPM, IEC, IFCC, ISO, IUPAC, IUPAP, OIML, Guide to the Expression of Uncertainty in Measurement, International Organisation for Standardization, Geneva, First Edition. [] Cox MG, A discussion of approaches for determining a reference value in the analysis of keycomparison data, NPL Report CISE /,, p. [] Moore WJM and Miljanic PN, The Current Comparator, IEE Electrical Measurement series, Peter Peregrins Ltd, London,, p.

50 APPENDIX : Types of standard used in comparison Table details the type of standard and/or instrument used by each participant in the measurement of the transfer standard. Lab. Rog/Shunt Table : Types of standard used by participants Passive CCC Standard Electronic CCC Standard transformer NPL SP LBE Bridge Passive Electronic Homemade PTB GUM OMH IEN METAS BEV HUT MIKES CMI VMT/VMC BNMLCIE UME Legend Lab. CCC Rog/Shunt Laboratory Compensated current comparator Refers to measurements made using Rogowski coils and/or current shunts

51 APPENDIX : Measurement methods Passive compensated current comparator method Figure shows the basic calibration circuit for the passive compensated current comparator designed by Kusters and Moore [], with a test current transformer connected in series. The compensation winding, which has the same number of turns as the secondary winding, will have a current that is the difference between the comparator secondary winding current and the test transformer secondary current. If i is the nominal secondary current and α and β are respectively the errors of the comparator and the test transformer secondary currents, the compensation winding current is (i + α) (i + β) which is equal to (α β). In linking the detector core, the compensation winding ampereturns due to α will subtract from those of (i + α) due to the secondary winding, giving a resultant due to i, which exactly cancels the primary ampereturns. The detector core is therefore magnetized only by a current β, the error of the test current transformer. An equal and opposite current from the balance control circuit is injected into the compensation winding to give null deflection of the detector. The error β is then given by ir(g + jωc). Figure : Basic calibration circuit Supply Primary circuit Comparator Test C.T. Compensating winding D G C Balance control circuit i r Burden Rogowski and current shunt method Figure shows the Rogowski Coil/current shunt method, (diagram taken from MIKES report) where output voltages from calibrated current shunts/rogowski coils and the DUT (device under test) are measured using calibrated digital multimeters and the DUTs, or in this case, the transfer standard s errors calculated from the ratio of those voltages.

52 Figure : Rogowski/current shunt method Electronically compensated comparator and test set Figure shows the calibration circuit for the electronically compensated comparator and test set. The transformer test set measures the transfer standard (marked DUT in circuit) errors by dividing the differential current into two components. These values are evaluated automatically as current error and phase displacement by the test set electronics. Figure : Electronic measurement system

53 APPENDIX : BNMLCIE Results The following results were received from BNMLCIE on May after Draft A had been agreed by other participants. The results supplied by BNMLCIE have not been used in the calculation of the comparison reference values given in Table. The results in Table are the deviations from the comparison reference values, in ppm for current error and µrad for phase displacement. The expanded uncertainty (k = ) for each deviation is also shown in ppm and µrad. Table : BNMLCIE results as deviations from the comparison reference values I/I n Ratio / Ratio / Ratio / % χ ε U χ δ U χ ε U χ δ U χ ε U χ δ U I/I n Ratio / Ratio / Ratio / % χ ε U χ δ U χ ε U χ δ U χ ε U χ δ U I/I n Ratio / Ratio / Ratio / % χ ε U χ δ U χ ε U χ δ U χ ε U χ δ U

54 APPENDIX : Amended uncertainty values from VMT/VMC Following VMT/VMC investigating the measurement uncertainties regarding their standard transformer, they have supplied recalculated uncertainty values. As these values were supplied after issuing of the comparison reference values they are included in this appendix for completeness. For VMT/VMC values given in Figures to and Tables to, for all current error measurements the expanded uncertainty should be ± ppm (±.%), not ± ppm as stated in the figures/tables. For all phase displacement measurements the expanded uncertainty should be ± µrad (. minutes), not ± µrad as stated in the figures/tables.

55 APPENDIX : Sample uncertainty budgets SP sample uncertainty budget Quantity Testpoints % Guildline % Exp. Uncert. Standard deviation in % *, /ret() Dependence of burden +/ % Exact point of measuring +/ % Resolution in PPM Combined standard uncertainty Expanded uncertainty Estimate [PPM],, Analysis,,, Sensitivity coefficient Standard Uncertainty [PPM] Stated uncertainty Within +/,,,,,,, BEV sample uncertainty budget Current error uncertainty budget quantity X i estimate x i standard uncertainty u ( x i ) probability distribution sensitivity coefficient c i uncertainty contribution u i ( y ) F i,m, normal,, F i,n, rectangular,, F i,b, rectangular,, F i,o, rectangular,, F i, Phase displacement uncertainty budget quantity X i estimate x i standard uncertainty u ( x i ) probability distribution sensitivity coefficient c i uncertainty contribution u i ( y ) δ i,m rad, rad normal,, rad δ i,n rad, rad rectangular,, rad δ i,b, rad rectangular,, rad δ i,o, rad rectangular,, rad δ i, rad where: F i and δ i are the current ratio error and phase displacement respectively, F i,m and δ i,m are the mean value of measurements, F i,n and δ i,n are the error of the standard transformers, F i,b and δ i,b are the standard uncertainty for the calibration of the measuring bridge, F i,o and δ i,o take into account other influences (e.g. influences of the current source).

56 PTB sample uncertainty budget uncertainty budget PTB Braunschweig, section., instrument transformers and highvoltage technique Hz; example: A / A; I / I n = %; EUROMET project ; comparison of CT : ( A to A) / A; April March current error ε X CT N: IW ;..; lfd. Nr. & ; last update:.. quantity estimated value limits distribution standard sensitivity uncertainty type uncertainty coefficient component X i x i u(x i) c i u i(y ) ε DA (SEKAM; n = ). x. x normal. x. x A ε DB (SEKAM). x normal. x. x B ε R (reproducibility). x rectangular. x. x B ε N (standard CT). x. x normal. x. x B ε N (influence of ε DA and δ DA). x. x rectangular. x. x B B (actual burden). VA. % normal. % x /VA. x B ε B (influence of burden). x. x rectangular. x. x B ε MP (influence of I / I n). x rectangular. x. x B ε F (influence of frequency). x rectangular. x. x B ε X = ε DA + ε N + ε N + ε B. x. x measured value (k =) x ± x phase displacement δ X quantity estimated value limits distribution standard sensitivity uncertainty type uncertainty coefficient component X i x i u(x i) c i u i(y ) in crad δ DA (SEKAM; n = ). urad. urad normal. urad. urad A δ DB (SEKAM). urad normal. urad. urad B δ R (reproducibility). urad rectangular. urad. urad B δ N (standard CT). urad. urad normal. urad. urad B δ N (influence of ε DA and δ DA). urad. urad rectangular. urad. urad B B (actual burden). VA. % normal. % urad/va. urad B δ B (influence of burden). urad. urad rectangular. urad. urad B δ MP (influence of I / I n). urad rectangular. urad. urad B δ F (influence of frequency). urad rectangular. urad. urad B δ X = δ DA + δ N + δ N + δ B. urad. urad measured value (k =) : urad urad ; measured value (k =) :. ±. METAS sample uncertainty budget Components Type Distribution Standard uncertainties I/In = % % I/In = % % (AorB) r:rectang. I error Phase I error Phase n:normal (ppm) (min) (ppm) (min) Reference Current Transformer B r.... Measurements Random noise (one measurement) A n.... Nonlinearity of the bridge B r.... Frequency effects (± Hz) B r.... Burden (± % of Pn) B r.... Temperature ( C ± C) B r.... Reproducibility of the setup B r.... Combined standard uncertainty.... (effective degrees of freedom:,, and resp.) Expanded uncertainty U, % Confidence level....

57 GUM sample uncertainty budget Ratio / A/A. Uncertainty budget for ratio error x Uncertainty sources Standard uncertainty for % % x I n Standard uncertainty for % and % x I n Probaility distribution Sensitivity coefficient Calibration of reference current normal. transformer Error of current transformer / / rectangular. bridge Calibration of current normal. transformer bridge Error of burden box / / rectangular. Error of indication of results / / rectangular. Uncertainty type A.. normal. Contribution to the standard uncertainty. Ratio / A/A. Uncertainty budget of phase displacement [min] Uncertainty sources Standard uncertainty for % % x I n Standard uncertainty for % and % x I n Probaility distribution Sensitivity coefficient Calibration of reference current.. normal. transformer Error of current transformer. /. / rectangular. bridge Calibration of current.. normal. transformer bridge Error of burden box. /. / rectangular. Error of indication of results. /. / rectangular. Uncertainty type A.. normal. Contribution to the standard uncertainty.. LBE sample uncertainty budget. Ratio uncertainties (%) σ : value (σ) on the results of our Tettex CT comparator calibration at PTB *, σ : calibration operation own, / σ : precision of the burden *, / σ = σ σ σ + + =, % σ =, % round up to, %. Phase uncertainties (minutes) σ : value (σ) on the results of the PTB * calibration, σ : calibration operation own, / σ : precision of the burden *, / σ = σ σ σ + + =, σ =, round up to,