SIM Regional Comparison of AC-DC VOLTAGE TRANSFER DIFFERENCE. SIM.EM-K6a, SIM.EM-K9, SIM.EM-K11 and SIM.EM-Supplementary 120 V / 53 Hz FINAL REPORT

Transcription

1 SIM Regional Comparison of AC-DC VOLTAGE TRANSFER DIFFERENCE SIM.EM-K6a, SIM.EM-K9, SIM.EM-K11 and SIM.EM-Supplementary 120 V / 53 Hz FINAL REPORT January 2004 December /98

2 SIM. AC-DCC Voltage Transfer Difference CONTENTS Page 1. Introduction 2. Definition of the measurand 3. Travelling standard description 4. Participants 5. Circulation of the travelling standard 6. Measurement methods 7. Environmental conditions during transportation 8. Measurement results 8.1 SIM. EM-K6a results 8.2 SIM. EM-K9 results 8.3 SIM. EM-K11 results V / 53 Hz supplementary point results 9. Reference value 10. Degree of equivalence 10.1 Degree of equivalence SIM. EM-K6a 10.2 Degree of equivalence SIM. EM-K Degree of equivalence SIM. EM- K Degree of equivalence 120 V / 53 Hz supplementary point 11. Link with the CCEM key comparison 11.1 Link SIM.EM-K6a - CCEM-K6a 11.2 Link SIM.EM-K9 - CCEM-K Link SIM.EM-K11- CCEM-K Corrective actions 13. Acknowledgements 14. References Appendix A. Reports of the participants Final Report 2/ /98

3 SIM. AC-DCC Voltage Transfer Difference 1. Introduction In the Sistema Interamericano de Metrología (SIM) there are several National Metrology Institutes (NMIs) having calibration and measurement capabilities in the ac-dc voltage transfer difference, but only three NMIs have participated in the CCEM Key Comparisons of ac-dc transfer difference, CCEM-K6a, CCEM-K9 and CCEM-K11. Three comparisons, SIM.EM-K6a, SIM.EM-K9, SIM.EM-K11, were proposed to assess the measurement capabilities of the remaining NMIs in the SIM region, in ac-dc voltage transfer difference. The proposed test points were selected to link the results of such comparisons with the equivalent CCEM Key Comparisons, through the three NMIs participating in both. Additionally, a fourth comparison, SIM.EM-Supplementary, was proposed, power/energy meter calibration capabilities. in support of the SIM NMIs CENAM volunteered to provide the travelling standard (TS) and to pilot and coordinate the comparison. It was agreed that the comparison references values were to be based on the results provided by the laboratories with participation in the key comparisons. The comparisons started in January 2004 and the measurements were concluded in December The Draft A was issued by the pilot laboratory and was reviewed by the participants in 2006, comments were added and the final results are reported in this document. 2. Definition of the measurand The measured quantity is the ac-dc voltage transferr difference of the travelling standard: δ = (Vac - Vdc) / Vdc where: δ is the ac-dc voltage transferr difference of the travelling standard Vac is the rms value of the ac input voltage Vdc is the dc input voltage which when reversed produces the same mean output voltage of the transfer standard as Vac. The measurement points for the different comparisons are shown in Table I. SIM.EM-K6a 3 V / 1 khz 3 V / 20 khz 3 V / 100 khz 3 V / 1 MHz Table I. Test points SIM.EM-K9 SIM..EM-K11 1 kv / 1 khz 100 mv / 1 khz 1 kv / 10 khz 100 mv / 20 khz 1 kv / 20 khz 100 mv / 100 khz 1 kv / 50 khz 100 mv / 1 MHz 1 kv / 100 khz SIM.EM-Supplementary 120 V / 53 Hz Final Report 3/ /98

4 SIM. AC-DCC Voltage Transfer Difference Accessories: Description: Model: Serial Number: Description: Model: Serial Number: Description: Model: Serial Number: Description: Model: Serial Number: 3. Travelling standard description The travelling standard was an ac-dc thermal transfer standard, model Fluke 792A, fitted with accessories for voltage measurements. Description: AC-DC Transferr Standard Fluke (with a ype-nm/type-nf adapter attached) Model: 792A Serial Number: V Range Resistor, Fluke 792 A Power Pack, Fluke 792 A Power Pack Cable, Fluke 792 A N/A Handbook Fluke 792A N/A N/A The travelling standard belongs to the pilot laboratory, CENAM, who has maintained calibration of this device since The stability observed is within the manufacturer s specification and within the uncertainty obtained by the pilot laboratory. The following graphs show the measurements that the pilot laboratory made before the departure of the travelling standard in 2003 and during the comparison in The graphics show that the stability evaluated [2], shown in dashed lines, is good enough to serve as a travelling standard. Final Report 4/ /98

5 SIM. AC-DCC Voltage Transfer Difference δ [µv/v] V / 1 khz δ [µv/v] V / 20 khz 18/02/ δ [µv/v] /02/ /07/03 09/05/03 28/07/03 16/10/03 09/05/03 28/07/03 16/10/03 3 V / 100 khz δ [µv/v] 1 kv / 1 khz 06/10/03 25/12/03 14/03/04 02/06/04 21/08/04 09/11/04 04/01/04 24/03/04 12/06/04 31/08/04 19/11/04 04/01/04 24/03/04 12/06/04 31/08/04 19/11/04 18/02/03 09/05/03 28/07/ V / 1 MHz 40 δ [µv/v] /06/03 17/08/03 06/10/03 25/11/03 14/01/04 04/03/04 23/04/04 12/06/04 01/08/04 20/09/04 09/11/04 δ[µv/v] /07/03 06/10/03 25/12/03 14/03/04 16/10/03 04/01/04 24/03/04 12/06/04 31/08/04 19/11/04 1 kv / 10 khz 02/06/04 21/08/04 09/11/04 Final Report 5/ /98

6 SIM. AC-DCC Voltage Transfer Difference δ [µv/v] kv /20 khz δ [µv/v] kv / 50 khz 18/07/03 35 δ [µv/v] /07/ /11/03 06/10/03 25/12/03 06/10/03 25/12/03 1 kv / 100 khz 100 mv / 1 khz 14/01/04 04/03/04 23/04/04 12/06/04 01/08/04 20/09/04 09/11/04 29/12/04 14/03/04 02/06/04 21/08/04 09/11/04 δ [µv/v] 14/03/04 02/06/04 21/08/04 09/11/04 18/07/03 06/10/03 20 δ[µv/v] /04/01 28/07/01 05/11/01 13/02/02 24/05/02 01/09/02 10/12/02 20/03/03 28/06/03 06/10/03 14/01/04 23/04/04 01/08/04 09/11/ mv / 20 khz δ [µv/v] /11/03 14/01/04 04/03/04 25/12/03 14/03/04 02/06/ V / 53 Hz 23/04/04 12/06/04 01/08/04 20/09/04 09/11/04 29/12/04 21/08/04 09/11/04 Final Report 6/ /98

7 SIM. AC-DCC Voltage Transfer Difference m V / 100 khz mv / 1 MHz δ[µv/v] δ [µv/v] /11/03 14/01/04 04/03/04 23/04/04 12/06/04 01/08/04 20/09/04 09/11/04 29/12/ /11/03 14/01/04 04/03/04 23/04/04 12/06/04 01/08/04 20/09/04 09/11/04 4. Participants NRC NIST INMETRO INTI CENAM National Research Council, Canada Contact: Peter Filipski, Peter.Filipski@nrc-cnrc.gc.ca National Institute of Standard and Technology, USA Contact: Thomas Lipe, Thomas.lipe@nist.gov Joseph Kinard, Joseph.Kinard@nist.gov Instituto Nacional de Metrologia, Normalizaςao e Qualidade Industrial, Brazil. Contact: Giovanna Borghi, gbalmeida@inmetro.gov.br Edson Afonso, eafonso@inmetro..gov.br Instituto Nacional de Tecnología Industrial, Argentina Contact: Lucas Di Lillo, ldili@inti.gov.ar Héctor Laiz, laiz@inti.gov.ar Administración Nacional de Usinas y Transmisiones Eléctricas, Uruguay Contact: Alfredo Spaggiari, aspaggiari@ute.com.uy Daniel Slomovitz, d.slomovitz@ieee.org Centro Nacional de Metrología, México (Pilot Laboratory) Contact: Sara Campos, scampos@cenam. mx Final Report 7/ /98

8 SIM. AC-DCC Voltage Transfer Difference 5. Circulation of the travelling standard A measurement period of six weeks for each participant was agreed upon. This period included time for clearing the customs, receiving, preparation, making measurements and shipping to the next laboratory. Each laboratory, except CENAM, measured the travelling standard once. The travelling standard returned to CENAM after measurement at NIST, NRC and the South American NMIs. It was agreed upon to hand- during the customs clearance. Flight tickets Mexico-Argentina-Mexico, Uruguay-Brazil-Uruguay and carry the travelling standard when sent to South America, to avoid any possible damage and delays Brazil-Mexico-Argentina were financed by the grant from the Organization of the American States. The pilot laboratory sent the travelling standard to the first participating laboratory in January The whole comparison was carried out without an undue delay during the After the measurements in the five laboratories, the travelling standard returned to the pilot laboratory in November The pilot laboratory performed the last measurements in December The schedule followed by the travelling standard is shown on Table II. Table II. Schedule followed by the travelling standard. Participant Laboratory CENAM (Mexico) NIST (United States) CENAM (Mexico) INTI (Argentina) (Uruguay) INMETRO (Brazil) CENAM (Mexico) NRC (Canada) CENAM (México) Transport of the Date of measurements travelling standard December 2003 January and February 2004 Courier Service February 27 th 2004 Courier Service 4 th March To 16 th April 2004 Hand Carried 18 th May to 22 nd June 2004 Hand Carried June 25 th 2004 Hand Carried August 2004 Hand Carried September to October 2004 Courier Service November 2004 Courier Service In order to conduct the comparison in a timely fashion, it was helpful that the participant laboratories knew the customss procedures for their port of entry to get the importation permission prior to the arrival of the travelling standard, as each country had different procedures and different periods of answer. During the circulation of the travelling standard there was a constant contact between the sending laboratory, the receiving laboratory and the pilot laboratory, to communicate the departure and arrival dates of the travelling standard. Final Report 8/ /98

9 SIM. AC-DCC Voltage Transfer Difference 6. Measurement methods The participants were requested to submit a detailed report of their measurements, which are attached as an Appendix A of this document. The following information was extracted from the reports. All participants used a two channel method. Threee participants used a nanovoltmeter to read the output voltage of their standards and an 8 ½ digits multimeter to read the output voltage of the travelling standard. Two participants used nanovoltmeters to read both the output of their standard and the outputt of the travelling standard. Five participants used two sources with an external relay to apply the AC and the DC voltage to the instrument under test. Two participants indicated the AC-DCC sequence used to minimize the thermal drift; NRC reported using the sequencee AC, DC +, DC -, AC, except for the 1 kv measurement points, in which case NRC applied the sequence AC,DC +,AC,DC -,AC. CENAM used the sequence AC,DC +,AC,DC -,AC exclusively. For frequencies below 1 MHz, NIST used as plane of reference the center of a type-gr874 Tee connector, the travelling standard being connected to it by a type-gr874 to type-n adapter. At 1 MHz NIST used as a plane of reference the center of a type-n tee connector. INTI,, INMETRO and CENAM used as plane of reference the center of a type-n Tee connector and NRC used the center of a special asymmetrical tee, having as one arm a type-gr874 connector and as the other arm a type-n connector. To calibrate the travelling standard the participating laboratories used the standards shown in Table IIII a, their sources of traceability are shown in Table III b. Only four laboratories participated in all comparisons. NIST participated in SIM.EM-K6a and SIM.EM-K9. INTI decided to withdraww their resultss from the comparison SIM.EM-K6a, after discovering a systematic error. See section 12 for corrective actions. Laboratory NIST INTI INMETRO CENAM NRC SIM.EM-K6a SJTC 1 V SJTC + RR TTS Fluke 792A 1 V PMJTC Ω RR Table III.a Standards used for measuring the travelling standard Standard used for measuring the travelling standard SIM.EM-K9 SJTC + RRR PMJTC kωω RR PMJTC kωω RR 1 V SJTC + RR TTS Fluke 792A 1,5 V PMJTC kω RRR MJTC Ω RR 1 V PMJTC +200 kω RR SIM.EM-K11 PMJTC - 10:1 RVD Micropotentiometer Ballantine 1 V SJTC - 10:1 RVD TTS Fluke 792A TTS Fluke 792A micropotentiometer H Ω SJTC SIM.EM-Supplementary PMJTC + RR 1 V SJTC + RR TTS Fluke 792A 1 V PMJTC + 30 kω RR MJTC+ 6 kω RR SJTC: Single Junction Thermal Converter PMJTC: Planar Multijunction Thermal Converter TTS: Thermal Transfer Standard RVD: Resistive Voltage Divider RR: Range Resistor Final Report 9/ /98

10 SIM. AC-DCC Voltage Transfer Difference Laboratory NIST INTI INMETRO CENAM NRC Table III. b Sources of traceability SIM.EM-K6a SIM.EM-K9 SIM.EM-K11 Own realization PTB at 3 V PTB at 1 V Step up in voltage Step down in voltage procedure procedure PTB at 100 V PTB at 3 V PTB at 3 V Step up in voltage procedure Step down in voltage procedure PTB at 3 V PTB at 1 V Step up in voltage procedure PTB at 1 kv PTB at 1 V Step up in voltage procedure PTB at 100 mv PTB at 1 V Step down in voltage procedure Own realization SIM. EM-Supplementary PTB at 3 V Step up in voltage procedure PTB at 100 V Step up in voltage procedure PTB at 120 V PTB at 1 V Step up in voltage procedure NRC and NIST have their own calculable standards to provide traceability to the working standards used during the comparison. INTI,, INMETRO and CENAM used standards whose values are traceablee to the Physikalisch Technische Bundesanstalt (PTB) standards. INMETRO standardss were directly calibrated by the PTB at the comparison test points. INTI, and CENAM performed voltage step up and step down procedures to derive the values of their working standards at 3 V, 1 kv and 100 mv. The reports show that in all laboratories the environmental travelling standard were at (23 ± 1) CC and (45 ± 15) % of R.H. conditions during the calibration of the 7. Environmental conditions during transportation A temperature, humidity and pressure data logger traveled with the travelling standardd to monitor the ambient conditions during the transport. There was no evidencee that the environmental condition changes had an effect on the stability of the TS and thus influenced the final measurements. 8. Measurement results This section summarizess the results reported by the participants. They reported the expanded uncertainty with a coverage factor of k =2, except for who reported their results with a coverage factor of k=1. In sections 9 and 10 the expanded uncertainty is evaluated, taking into account the degrees of freedom reported by the participants, at a confidence level of 95,45 %. Final Report 10/ /98

11 SIM. AC-DCC Voltage Transfer Difference 8.1 SIM.EM-K6a. results Table IV. Results at 3 V, ac-dc voltage transfer difference δ i and its uncertainty U δ δi. Laboratory i NIST INMETRO CENAM NRC Calibration date Jan to Feb, th May to 22 nd June, th June to 28th July, rd to 20 th August, 2004 Sept to Oct, 2004 SIM.EM-K6a 3 V δi δ and U δ i (k=2,0) µv/v 1 khz 20 khz δi U δ i δi u δ i δi U δ i 4 4 δi U δ i 6 6 δi U δ i khz MHz SIM.EM-K9 results Table V. Results at 1 kv, ac-dc voltage transfer difference δ I and its uncertainty U δ δi. Laboratory i NIST INTI 4 th Calibration date Jan to Feb 2004 March to 16 th April th May to 22 nd June 2004 INMETRO 28 June to 28 th July 2004 CENAM 7 th to 24 th August 2004 NRC Sept to Oct δi δi U δ δi Uδi δ 16 δi u δ SIM.EM-K9 1 kv δi δ and U δ i (k=2,0) µv/ V 1 khz 10 khz 20 khz 50 khz 4 i 11 5 δi Uδi δ δi Uδi δ δi Uδi δ khz Final Report 11/ /98

12 SIM. AC-DCC Voltage Transfer Difference 8.3 SIM.EM-K11 results Table VI. Results at 100 mv, ac-dc voltage transfer difference δ I and its uncertainty U U δi. Laboratory i INTI INMETRO CENAM NRC SIM.EM-K mv Calibration δi date and U δ i (k=2,0) µv/ V 1 khz 20 khz 1000 khz 1 MHz 4 th March to 16 th April th May to 22 nd June th June to 28 th July 2004 δi U δ i δi u δ i δi U δ i th to 30 August δi U δ i Sept to Oct 2004 δi U δ i SIM.EM-Supplementary point at 120 V / 53 Hz Table VII. Results at 120 V, ac-dc voltage transfer difference δ I and its uncertainty U U δi, Laboratory i INTI INMETRO CENAM NRC Calibration date 4 th March to 16 t April th May to 22 nd June th June to 28th July th to 24 th August 2004 Sept to Oct 2004 th SIM.EM-Supplementary 120 V δ and U δ (k=2,0) µv/ V 53 Hz 2.2 δ i U δi δ i u δi δ i U δi δ i U δi δ i U δi Final Report 12/ /98

13 SIM. AC-DCC Voltage Transfer Difference 9. Reference value The reference value, δsim, was calculated as the weighted mean [3] of the reported values from the laboratories that took part in the corresponding CCEM key comparison; these are NIST, NRC and INTI. where: u δ i δ SIM = n i=1 n i=1 w i w δ i i 1 with w i = 2 u δ is the standard uncertainty associated with the reported δ i values. i (1) The standard uncertainty of the reference values, u δ, was evaluated as [3][4] : SIM uδ SIM = 1 n i= 1 w i (2) The measurement results of NRC, NIST and INTI were considered independent; since NRC and NIST have their own calculable standardss and INTI has traceability to the values of the uncorrelated PTB standards. The expanded uncertainty of the reference value was evaluated as: U δ = ku SIM δ SIM (3) where k was estimated at a confidence level of 95,45 % taking into account the reported effective degrees of freedom as: where ν eff are the effective degrees of freedom of the δ SIM and υ δ i are the effective degrees of freedom associated with the δi. υ eff = n i i= 1 u 4 δ w i n = 1 SIM υ w δ i i u δ i 4 (4) Final Report 13/ /98

14 SIM. AC-DCC Voltage Transfer Difference 10. Degree of equivalence The degree of equivalence (D i ) between the i-th participant with respect to the reference value ( δ SIM) was evaluated as follows: Di = δ δi -δ SIM (5) For the laboratories without contribution to the reference value, the expandedd uncertainty of D i ( U estimated as[4]: D i ) was U D i = k Di u 2 δ i + u 2 δsim (6) For the laboratories with contribution to the reference value, the expanded estimated as[4]: uncertainty of D i ( U D i ) was U D i = k Di u 2 δi - u 2 δsim (7) where kdi, for laboratories without contribution to the reference value, was estimated at a confidence level of 95,45 %, taking into account the effective degrees calculated as, υ eff = u 4 δi υ δ i u 4 Di 4 u δ + υ SIM effδ SIM (8) For laboratories that contributed to the reference value, the effective degrees of freedom were calculated as: υ eff = u δ i * (1 i n i = 1 υ δ i w ) w i 4 u Di 4 + n i= i+ 1 u δ * i wi n i = w 1 i υ δ i 4 (9) Final Report 14/ /98

15 SIM. AC-DCC Voltage Transfer Difference 10.1 Degree of equivalence SIM.EM-K6a At 3 V, the reference values were evaluated as the weighted mean of the values from NRC and NIST. Correlations between the results weree not considered because the laboratories that did not contribute to the reference value, had neither traceability to the values of NRC nor to the values of NIST. Degree of equivalence Di and its uncertainty U at 3 V / 1 khz, in µv/v; for k=2.0 δ SIM.EM-K6a 3 V / 1 khz = 0.3 µv/v ; U δs SIM.EM -K6a =1.2 µv/v Laboratory i δi U δ i Di U D i NIST INMETRO CENAM NRC Di D i µv/v NIST Laboratory i 3 V/ 1 khz INMETRO CENAM NRC Degree of equivalence Di and its uncertainty U at 3 V / 20 khz, in µv/v; for k=2.0 3 V / 20 khz δ SIM.EM-K6a = 2.3 µv/v ; Uδ = 2.1 µv/v Laboratory i NIST INMETRO CENAM NRC δi U δ i SIM.EM-K6a Di U D i Di D i µv/v NIST 3 V/ 20 khz INMETRO Laboratory i CENAM NRC Degree of equivalence Di and its uncertainty U at 3 V / 100 khz, in µv/v. Di 3 V / 100 khz δ SIM.EM-K6a = 16.3 µv/v ; Laboratory i NIST INMETRO CENAM NRC U δs SIM.EM-K6a U δ i Di U D i = 5.8 µv/v δi k D i µv/v NIST 3 V/ 100 khz Laboratory i INMETRO CENAM NRC Final Report 15/ /98

16 SIM. AC-DCC Voltage Transfer Difference Degree of equivalence Di and its uncertainty U at 3 V / 1 MHz, in µv/v. Di 3 V / 1 MHz δ SIM.EM-K6a = 5.6 µv/v ; U δsim M.EM-K6a = 12.4 µv/v Laboratory i δi U δ i Di U D i k NIST INMETRO CENAM NRC D i µv/v NIST Laboratory i 3 V/ 1 MHz INMETRO CENAM NRC The results at 3 V show a degree of equivalence lower than 3 µv/v at 1 khz, 4 µv/v at 20 khz, 4 µv/vv at 100 khz, and 5 µv/v at 1 MHz, except for one laboratory at 1 MHz. All the differences between the reported values and the reference value are within the uncertainty reported by the participants Degree of equivalence SIM.EM-K9 For the points at 1 kv, the reference values were evaluated as the weighted mean of values from NRC, NIST and INTI. As mentioned in the point 6 of this document, the values of, INTI and CENAM have traceability to the PTB standards at 100 V, 3 V and 1 V, whereas INMETRO have traceable values to PTB standards at 1 kv. Correlation between, INMETRO and CENAM may exist with respect to INTI values, at 1 V and 3 V levels, since they are traceable to PTB standards. It should be considered, that the voltage step up procedures that these NMIs conducted in order to scale from 1 V or 3 V up to 1 kv, ensures that the effects of the correlation between the laboratories at 1 V and 3 V can be considered negligible at 1 kv. Degree of equivalence Di and its uncertainty U at 1 kv /1 khz, in µv/v; for k=2.0 Di 1 kv / 1 khz δ SIM.EM-K9= 7.6 µv/v ; Uδ = 6.9 µv/v SIM.EM-K9 Laboratory i NIST INTI INMETRO CENAM NRC δi U δ i Di U Di D i µv/v NIST 1 kv /1 khz INTI Laboratory i INMETRO CENAM NRC Final Report 16/ /98

17 SIM. AC-DCC Voltage Transfer Difference Degree of equivalence Di and its uncertainty U at 1 kv /10 khz, in µv/v. 1 kv / 10 khz δ SIM.EM-K9= 3.3 µv/v ; = 7.2 µv/vv Laboratory i NIST INTI INMETRO CENAM NRC U δsim M.EM-K9 δi U δ i Di Di U Di k D i µv/v NIST 1 kv/ 10 khz INTI Laboratory i INMETRO CENAM NRC Degree of equivalence Di and its uncertainty U at 1 kv /20 khz, in µv/v. Di 1 kv / 20 khz δ SIM.EM-K9= µv/v ; = 8.0 µv/v Laboratory NIST INTI INMETRO CENAM NRC U δsim.e EM-K9 i δi U δ i Di U Di k D i µv/v NIST 1 kv/ 20 khz INTI Laboratory i INMETRO CENAM NRC Degree of equivalence Di and its uncertainty U at 1 kv /50 khz, in µv/v Di 1 kv / 50 khz δ SIM.EM-K9= µv/v ; = 9.7 µv/vv Laboratory i NIST INTI INMETRO CENAM NRC U δsim M.EM-K9 Uδi δ Di U Di δi k D i µv/v NIST INTI 1 kv/ 50 khz Laboratory i INMETRO CENAM NRC Final Report 17/ /98

18 SIM. AC-DCC Voltage Transfer Difference Degree of equivalence Di and its uncertainty U at 1 kv /100 khz, in µv/vv Di 1 kv / 100 khz δ SIM.EM-K9= µv/v ; = 16.8 µv/ /V Laboratory i NIST INTI INMETRO CENAM NRC U δsim M.EM-K9 Uδi δ Di U Di δi k D i µv/v NIST 1 kv/ 100 khz INTI Laboratory i INMETRO CENAM NRC The results at 1 kv show a degree of equivalence smaller than 4 µv/v at 1 khz; 7 µv/v at 10 khz; 12 µv/v at 20 khz; 17 µv/v at 50 khz and 36 µv/ /V at 100 khz. All the differences between the reported values and the referencee value are within the uncertainty reported by the participants Degree of equivalence SIM.EM-K11 At 100 mv, the referencee values weree evaluated as the weighted mean of values from NRC and INTI. As mentioned in section 6, the values of, INTI and CENAM have traceability to the PTB standards at 3 V and 1 V, whereas INMETRO have traceable values to PTB standards at 100 mv. Correlation between, INMETRO and CENAM may exist with respect to the INTI values at 1 V and 3 V, but the measurement process, associated with the step down procedures, ensures that the uncertainty at 1 V and 3 V is very low compared with the uncertainty at 100 mv. The effects of the correlation between the laboratories can be considered negligiblee at 100 mv. At the CCEM K-11 errors were considered due to the power supply voltage and to the temperature and the humidity coefficients of the travelling standard, but they have been considered insignificant [9]. Degree of equivalence Di and its uncertainty δ SIM.EM-K 100 mv / 1 khz 11= 9.2µV/V ; U δ =7.8 µv/v Laboratory i δi INTI INMETRO 8.0 CENAM 7.5 NRC 9.5 SIM.EM-K11 U δ i Di UD i U Di at 100 mv /1 khz, in µv/ /V; for k=2.0 D i µv/v INTI 100 mv/ 1 khz Laboratory i INMETRO CENAM NRC Final Report 18/ /98

19 SIM. AC-DCC Voltage Transfer Difference Degree of equivalence Di and its uncertainty δ SIM.-EM-K 100 mv / 20 khz K11= -0.4 µv/v ; U δ = 7..3 µv/v Laboratory i δi INTI INMETRO -2.0 CENAM -6.0 NRC -1.8 SIM.EM-K11 U δ i Di UD i U Di at 100 mv /20 khz, in µv/v; for k=2.0 D i µv/v INTI Laboratory i 100 mv / 20 khz INMETRO CENAM NRC Degree of equivalence Di and its uncertainty δ SIM.EM-K 100 mv / 100 khz K11= 23.4 µv/v ; U δ = 9.9µV/V Laboratory i δi INTI INMETRO 18.0 CENAM 21.6 NRC 24.1 U δ i SIM.EM-K11 Di UD i U Di at 100 mv /100 khz, in µ V/V; for k=2.0 D i µv/v INTI 100 mv / 100 khz Laboratory i INMETRO CENAM NRC Degree of equivalence Di and its uncertainty δ SIM.-EM-K 100 mv / 1 MHz K11=69.8 µv/v ; U δ = µv/v SIM.EM-K11 Laboratory i INTI INMETRO CENAM NRC δi U δ i Di UD i U Di at 100 mv /1 MHz, in µv/v; for k=2.0 D i µv/v INTI 100 mv / 1 MHz Laboratory i INMETRO CENAM NRC Final Report 19/ /98

20 SIM. AC-DCC Voltage Transfer Difference The results at 100 mv show a degree of equivalence smaller than 5 µv/v at 1 khz, 6 µv/v at 20 khz 6 µv/v at 100 khz, and smaller than 55 µv/v at 1 MHz. All the differences between the reported values and the reference value are within the uncertainty reported by the participants Degree of equivalence at 120 V/ 53 Hz At 120 V and 53 Hz, the reference value was evaluated as the weighted mean of values from NRC and INTI. The values of INTI and CENAM are traceablee to the PTB standards at 1 V, whereas INMETRO and have traceable values to PTB standards at 100 V and 120 V, respectively. Correlation between, INMETRO and CENAM may exist with respect to INTI values at 1 V, but the long measurement process, associated with the voltage step up procedures, makes the uncertainty at 1 V low compared with the uncertainty at 120 V. The effects of the correlation between the laboratories at 1 V level can be considered negligible at 120 V. Degree of equivalence Di and its uncertainty 120 V / 53Hz δ SIM.EM-1.9= 4.2 µv/v ; =2.8 µv/v U δsim.e EM 1.9 UD i Laboratory i INTI INMETRO CENAM NRC δi U δ i Di U Di k at 120 V /53 Hz, in µv/vv D i µv/v INTI Laboratory i 120 V / 53 Hz INMETRO CENAM NRC The resultss at 120 V show for all participants a degree of equivalence smaller than 3 µv/v. All the differences between the reported values and the reference value are within the uncertainty declared by the participants. 11. Link with the CCEM Key Comparison The resultss of CCEM-K6a, CCEM-K9 and CCEM-K11 key comparisons are available. Rather than evaluating the differences between the pairs of laboratories, only the differences between the resultss of participants in the SIM comparison, not participating in the CCEM comparisons, and the CCEM key comparison reference value (KCRV) were calculated. The link between a result of an i-th laboratory participating in the SIM comparison with respect to the reference value of the corresponding CCEM key comparison is estimated using (10) [5]. D link i = δ i SIM - δ = ( δ - SIM ) i SIM δ + ( δ SIM - δ KCRV + KCRV ) (10) Final Report 20/ /98

21 SIM. AC-DCC Voltage Transfer Difference where: δ i SIM δ KCRV δ i - SIM δ SIM δ SIM - δ KCR RV is the reported value of the i-th laboratory participating in the SIM comparison, is the reference value of the corresponding CCEM comparison, is the degree of equivalence of the i-th laboratory participating in SIM comparison with respect to the reference value of SIM comparison, evaluated in the section 10 of this document (D i ), is the difference between the references values of both comparisons. This term will be evaluated using the weighted mean of the differences between reference values of the laboratories participating in both comparisons, [5]. The last term of (10) can be evaluated using the values determination of the reference values of SIM and CCEM: of the laboratories participating in the (δ - SIM δkc CRV ) i = (δi - CCEM δ KCRV ) - (δi SIM - δsim ) (11) Equation (11) is equivalent to: ( δ SIM - δ KCRV ) i = D iccem - D isim (12) The weighted mean of the differences between the reference values equals δ - δ SIM K KCRV n i= 1 (D i CCEM Each difference D i has its own uncertainty, = n i= 1 - D 1/u u i SIM 2 (D 2 )/u (D i CCEM D i SIM between reference values was estimated as the root of the sum square of the u as follows: -, u i - D CCEM i SIM. ) D i ) SIM D i CCEM, then the expandedd uncertainty of the difference D i (13) u D ( i CCEM - D i SIM ) i = u 2 D i SIM + u 2 D iccem (14) The uncertainty of the weighted mean of the differences between reference values is then: 1 (15) u( δ δkcrv ) = SIM 1 2 u Finally the uncertainty of D link i is equal to: n i= 1 (D i D i SIM CCEM ) U D = link i k u 2 D i SIM 2 + u δ -δ ( SIM KCRV ) (16) Final Report 21/ /98

22 SIM. AC-DCC Voltage Transfer Difference with k=2, because the participants reported results at k =2.0 and the references values at SIM and CCEM comparisons were reported at k = Link SIM.EM-K6a CCEM-K6a Link with the CCEM-K6a, D i link and its uncertainty U Di li ink, at 3 V / 1 khz, in µv/v 3 V / 1 khz Laboratory i Di link 2.6 INMETRO 0.6 CENAM 0.5 UDi link D i link µv/v 14 3 V / 1 khz Laboratory i INMETRO CENAM 3 V / 1 khz SIM.EM-K6a CCEM-K6a NIST NRC δ KCRV KCRV U δ D i SIM U D isim D i CCEM U D i CCEM Link with the CCEM-K6a, D i link and its uncertainty U Di li ink, at 3 V / 20 khz, in µv/v 3 V / 20 khz Laboratory i Di link 4.2 INMETRO 2.2 CENAM 0.7 UDi link D i link µv/v V / 20 khz Laboratory i INMETRO CENAM 3 V / 20 khz SIM.EM-K6a CCEM-K6a NIST NRC δ KCRV KCRV U δ D i SIM U D isim D i CCEM U D i CCEM Final Report 22/ /98

23 SIM. AC-DCC Voltage Transfer Difference Link with the CCEM-K6a, D i link and its uncertainty U Di li ink, at 3 V / 100 khz, in µv/v 3 V / 100 khz 3 V / 100 khz 20 Laboratory i INMETRO CENAM Di link UDi link 20.2 D i link µv/v Laboratory i INMETRO CENAM 3 V / 100 khz SIM.EM-K6a CCEM-K6a NIST NRC δ KCRV KCRV U δ D i SIM U D isim D i CCEM U D i CCEM Link with the CCEM-K6a, D i link and its uncertainty U Di li ink, at 3 V / 1 MHz, expressed in µv/v 3 V / 1 MHz Laboratory i INMETRO CENAM Di link UDi link D i link µv/v V / 1 MHz Laboratory i INMETRO CENAM 3 V / 1 MHz SIM.EM-K6a CCEM-K6a NIST NRC δ KCRV 5,6 121 U δkcrv 12,4 6,7 D i SIM -4,6 4,4 U D isim 12,6 12,3 D i CCEM 8,0-10,0 U D i CCEM Final Report 23/ /98

24 SIM. AC-DCC Voltage Transfer Difference 11.2 Link SIM.EM-K9 CCEM-K9 Link with the CCEM-K9, D i lin nk and its uncertainty U Di link k, at 1 kv / 1 khz, in µv/v 1 k V / 1 khz Laboratory i INMETRO CENAM Di link UDi link D i link µv/v kv / 1 khz Laboratory i INMETRO CENAM 1 kv / 1 khz SIM.EM-K9 CCEM-K9 NIST NRC INTI δ KCRV KCRV U δ D i SIM U D isim D i CCEM U D i CCEM Link with the CCEM-K9, D i lin nk and its uncertainty U Di link k, at 1 kv / 10 1 k V / 10 khz Laboratory i INMETRO CENAM Di link UDi link 26.7 D i link µv/v khz, in µv/v 1 kv / 10 khz Laboratory i INMETRO CENAM 1 kv / 10 khz SIM.EM-K9 CCEM-K9 NIST NRC INTI δ KCRV KCRV U δ D i SIM U D isim D i CCEM U D i CCEM Final Report 24/ /98

25 SIM. AC-DCC Voltage Transfer Difference Link with the CCEM-K9, D i lin nk and its uncertainty U Di link k, at 1 kv / 20 khz, in µv/v 1 k V / 20 khz Laboratory i INMETRO CENAM Di link UDi link D i link µv/v kv / 20 khz Laboratory i INMETRO CENAM 1 kv / 20 khz δ KCRV U δkcrv D i SIM U D isim D i CCEM U D i CCEM SIM.EM-K9 CCEM-K9 NIST NRC INTI Link with the CCEM-K9, D i lin nk and its uncertainty U Di link k, at 1 kv / 50 khz, in µv/v 1 k V / 50 khz Laboratory i INMETRO CENAM Di link UDi link 23.5 D i link µv/v kv / 50 khz INMETRO Laboratory i CENAM 1 kv / 50 khz δ KCRV U δkcrv D i SIM U D isim D i CCEM U D i CCEM SIM.EM-K9 CCEM-K9 NIST NRC INTI Final Report 25/ /98

26 SIM. AC-DCC Voltage Transfer Difference Link with the CCEM-K9, D i lin nk and its uncertainty U Di link k, at 1 kv / 1000 khz, in µv/vv 1 kv / 100 khz 1 k V / 1000 khz Laboratory i INMETRO CENAM Di link UDi link D i link µv/v INMETRO Laboratory i CENAM 1 kv / 100 khz δ KCRV U δkcrv D i SIM U D isim D i CCEM U D i CCEM SIM.EM-K CCEM-K NIST NRC INTI Link SIM.EM-K11- CCEM-K11 Link with the CCEM-K11, D i link and its uncertainty U Di li ink, at 100 mv / 1 khz, in µv/v 100 mv / 1 khz Laboratory i INMETRO CENAM Di link UDi link 18.1 D i link µv/v mv / 1 khz Laboratory i INMETRO CENAM 100 mv / 1 khz δ KCRV U δkcrv D i SIM U D isim D i CCEM U D i CCEM SIM.EM-K11 CCEM- K11 INTI NRC Final Report 26/ /98

27 SIM. AC-DCC Voltage Transfer Difference Link with the CCEM-K11, D i link and its uncertainty U Di li ink, at 100 mv / 20 khz, in µv/v 100 mv / 20 khz Laboratory i INMETRO CENAM Di link UDi link D i link µv/v m V / 20 khz Laboratory i INMETRO CENAM 100 mv / 20 khz δ KCRV U δkcrv V D i SIM U D isim D i CCEM U D i CCEM SIM.EM-K11 CCEM- K11 INTI NRC Link with the CCEM-K11, D i link and its uncertainty U Di li ink, at 100 mv / 100 khz, in µv/v 100 mv / 100 khz Laboratory i INMETRO CENAM Di link UDi link D i link µv/v mv / 100 khz Laboratory i INMETRO CENAM 100 mv / 100 khz δ K KCRV U δkc CRV D i SIM U D i SIM D iccem M U D i CCEM SIM.EM-K11 CCEM- K11 INTI NRC Final Report 27/ /98

28 Link with the CCEM-K11, D i link and its uncertainty U Di li ink, at 100 mv / 1 MHz, in µv/v 100 mv / 1 MHz Laboratory i INMETRO CENAM Di link UDi link D i link µv/v SIM. AC-DCC Voltage Transfer Difference 100 mv / 1 MHz Laboratory i INMETRO CENAM 100 mv / 1 MHz δ KCRV U δkc CRV D i SIM U D i SIM D iccem M U D i CCEM SIM.EM-K11 CCEM- K11 INTI NRC Corrective actions: Concerning the INTI results at 3 V, last July 2007, the ac-dc transfer laboratory of INTI was audited by Dr. Klonz from PTB, at that time INTI compared their measurement at 3 V against the PTB values. The results are shown in the following table, they are in agreement with the PTB results, within the uncertainties declared. Laboratory ac-dc voltage transfer difference at 3 V δi and U δ i (k=2, 0) µv/v 1 khz 20 khz 100 khz 1 MHz INTI δ INTI U δ INTI PTB δ PTB U δ PTB Acknowledgements CENAM acknowledges the participants for their time invested in doing measurements, preparing reports and dealing with administrative procedures for clearing customs and sending the travelling standard to the next laboratory. Thanks are given to the participants for the comments made on the protocol and the Final Report 28/ /98

29 SIM. AC-DCC Voltage Transfer Difference drafts. Special acknowledgement is given to the participants in the CCEM Key comparisons; NRC, NIST and INTI, because thanks to their participation the link could be extended to the other SIM-region for his collaboration. It NMIs. CENAM thanks Dr. Harold Sanchez, at the Instituto Costarricence de Electricidad, is also gratefully acknowledged the financial support of the Organization of American States, which made it possible to hand carry the travelling standard in South America. The collaboration of Dr. David Avilés and Dr. René Carranza protocol and on the Drafts is gratefully acknowledged. from CENAM, for their comments on the 14. References [1] Guidelines for CIPM key comparison (Appendix F to the Mutual recognition of national measurements standards and of measurements certificates issued by national metrology institutes (MRA). March [2] Mary Gibbons Natrella. NBS. Experimental Statistics. August, 1963 [3] Dietrich, C.F., Uncertainty, Calibration and Probability, [4] BIPM, IEC, IFCC, ISO, IUPAC, IUPAP, Measurement, (1993) OIML Guide to the Expression of Uncertainty in [5] Liefrink F., Dierikx E.F., Heimeriks J.W., Eklund G., Flouda I., Funck T., Helistö P., Jakab A., Janssen J.-T., Jeanneret B., Jensen H.D., Hetland P.O., Lindic M., Lo-Hive J.-P., Nicolas J., Nunes M., Power O., Raso Alonso F.I., Reymann D., Selçik S., Streit J., Vrabcek P., Waldmann W., EUROMET project no 429: Comparison of 10 V electronic voltage standards, Metrologia, 2003, 40, Tech. Suppl., [6] Reymann D., Link between the comparison EUROMET.EM.BIPM-K11.b and the ongoing comparison BIPM.EM-K11.b, Metrologia, 2003, 40, Tech. Suppl., [7] Klonz M., CCEM-K6.a: key comparison of ac-dc voltage transfer standards at the lowest attainable level of uncertainty, Metrologia, 2002, 39, Tech. Suppl., [8] C.J. van Mullem, E.F. Dierikx and J.P.M. de Vreede. Key comparison CCEM.K6.c of ac-dc voltage transfer standards at selected frequencies between 1 MHz to 100 MHz. Final Report February [9] K.E. Rydler and V. Tarasso. CIPM key comparison CCEM K11 and CCEM-K11.1 of ac-dc Voltage transfer difference at low voltages. Final report. March 2007 [10] A. Poletaeff. CCEM K9 comparison of ac-dc high voltage standards. Final report. Final Report 29/ /98

30 SIM. AC-DCC Voltage Transfer Difference APPENDIX A REPO ORTS OF THE PARTICIPANTS Final Report 30/ /98

31 31/98

32 32/98

33 33/98

34 34/98

35 35/98

36 36/98

37 37/98

38 38/98

39 39/98

40 40/98

41 41/98

42 42/98

43 43/98

44 44/98

45 45/98

46 46/98

47 47/98

48 48/98

49 49/98

50 50/98

51 51/98

52 52/98

53 53/98

54 54/98

55 55/98

56 56/98

57 57/98

58 58/98

59 59/98

60 60/98

61 61/98

62 62/98

63 63/98

64 64/98

65 65/98

66 66/98

67 67/98

68 68/98

69 69/98

70 70/98

71 71/98

72 72/98

73 73/98

74 74/98

75 75/98

76 76/98

77 77/98

78 78/98

79 79/98

80 80/98

81 81/98

82 82/98

83 83/98

84 84/98

85 85/98

86 86/98

87 87/98

88 88/98

89 89/98

90 90/98

91 91/98

92 92/98

93 93/98

94 94/98

95 95/98

96 96/98

97 97/98

98 END OF THE REPORT 98/98