Final Report Key Comparison COOMET.EM.BIPM-K10.b. Comparison of the 10 V Josephson Voltage Standards COOMET 542/RU/11. A.S. Katkov 1, P.A.

Size: px

Start display at page:

Download "Final Report Key Comparison COOMET.EM.BIPM-K10.b. Comparison of the 10 V Josephson Voltage Standards COOMET 542/RU/11. A.S. Katkov 1, P.A."

Stephanie Robinson
5 years ago
Views:

1 Final Report Key Comparison COOMET.EM.BIPM-K10.b Comparison of the 10 V Josephson Voltage Standards COOMET 54/RU/11 A.S. Katkov 1, P.A.Chernyaev 1 D.I.Mendeleyev Institute for Metrology (VNIIM), Moskovsky pr., 19, St. Petersburg, Russia a.s.katkov@vniim.ru RUP Belarussian State Institute of Metrology (BelGIM), Starovilensky tract 93, 0053 Minsk, Belarus electron@belgim.belpak.minsk.by December 014 Corrected for misprint on 7 December Abstract The 10 V Josephson voltage standard of BelGIM was compared with the similar standard of VNIIM in August-November 014 in accordance with the option В of the BIPM.EM-K10.b comparison protocol. The standard of VNIIM based upon the Josephson effect was used as a transfer standard to measure the output voltage by BelGIM. The final results are in good agreement with the reference value supplied by the BIPM voltage standard: the relative voltage difference was 0, at the combined relative standard uncertainty of for the nominal voltage of 10 V. Table of Contents Introduction Participants and schedule.... Transfer standard and requirements for measurements Measurement methods Results... 6 Conclusion... 1 References... 1

2 Introduction In the Mutual Recognition Arrangement (MRA) it is stated that the metrological equivalence of national measurement standards will be determined by a set of key comparisons chosen and organized by the Consultative Committees of the CIPM working in close cooperation with the Regional Metrology Organizations (RMO s). The RMO key comparison (COOMET Project No. 54/RU/11) supplementing the COOMET.EM.BIPM-K10.b comparison was carried out to link COOMET national laboratories. This comparison allow a link between the values of the national voltage standard of BelGIM (Belarusian State Institute of Metrology) and the BIPM (International Bureau of Weights and Measures) due to the result obtained by VNIM in the framework of the on going on site BIPM Josephson comparison (BIPM.EM-K10.b). In the KCDB, the deviation of the result obtained by BelGIM from the BIPM value will be indicated with value of the combined uncertainty obtained in measurements BelGIM-VNIIM and VNIIM- BIPM. In the period from to , BelGIM compared its voltage standard with the VNIIM primary voltage standard by means of the VNIIM Josephson voltage transfer standard. 1. Participants and schedule Table 1 shows the measurements schedule of the BelGIM-VNIIM comparison. Table 1 Institute Country Transfer Standard measurements BelGIM VNIIM Belarusian State Institute of Metrology D.I. Mendeleyev Institute for Metrology Belarus Russia In the course of comparison, the transfer standard was transported to BelGIM by train as a hand luggage. After completion of measurements at BelGIM, the transfer standard was returned to VNIIM in the same way.

3 3. Transfer standard and requirements for measurements The voltage standard of VNIIM based on the Josephson effect with an output voltage of 10 V, which had participated in the framework of the ongoing on site BIPM Josephson comparison (BIPM.EM-K10.b), was used as a transfer standard (TS) [1-3]. In these measurements the output voltages of the TS and the used standard were connected through null detector. The TS with its Josephson SINIS microcircuit isolated from the chassis, could be isolated from the body of the reference frequency source of 10 MHz and connected to a microwave generator battery supply. The microwave generator circuit was isolated from the TS frame. Built-in filters allowed TS to operate with or without TS chassis grounding. After dipping in a Dewar with liquid helium the TS probe was kept there for at least 4 hours prior to the measurements. During the measurements, the voltage polarity of the TS and the standard were reversed using the biasing current source of the JVS. Measurement conditions including the laboratory room temperature, relative humidity and air pressure were monitored. The TS frequency was synchronized with the frequency of the compared standard. 3. Measurement methods The TS was measured at BelGIM from to using BelGIM voltage standard. After the measurements the TS was transported to VNIIM, where it was measured against the VNIIM standard on For the realization of 10 V voltage both BelGIM and VNIIM standards use type SIS Josephson microcircuits. During the comparisons the measurements of voltage difference of the TS and the used standard are performed. The measurements are carried out for two polarities and the results are averaged. 3.1 The BelGIM standard The BelGIM voltage standard is based on the SupraVOLTcontrol system [4]. It s serial number is 004. The system features a 3-channel (channels A, B and C) Josephson Voltage Standard (JVS) with microprocessor-based control and allows calibration of secondary voltage standards and external voltmeters. During the comparison the channel A was mainly used. To maintain a high accuracy of the nanovoltmeter (used as a null detector), at the beginning of each working day the system option «Calibration nanovoltmeter» was used, after which the amplification factor was saved in the system s memory. Example of nanovoltmeter gains measured during four days before and after measurements present on Figure 1. The selection of an optimal exposure frequency (close to 75GHz) was made using the «Searching optimal frequency» system option.

4 4 Figure 1. Gain of nanovoltmeter The BelGIM standard has different modes preset in the used software. For measurements by the transfer standard, the secondary standards calibration mode (M1) and comparison mode (M) were used. In the M1 mode the polarity reversal of Josephson array connection and polarity of the secondary standard connection was used for each sampling point of the secondary standard output voltage. To determine each of these values, up to 40 measurements were made (0 for each polarity) lasting for about 3 seconds. During the measurement the Josephson microcircuit was disconnected completely from the ground and from the BelGIM standard s electronic unit. In the M1 mode the EMFs along the TS probe was excluded by using the mean of two measurements of plus and mines voltage on TS. In the M mode the BelGIM standard measured the voltage realized by the transfer standard as follows: 1. The BelGIM standard changed the preset frequency which allowed obtaining a voltage difference of 1 μv, the maximum allowed difference measured by the nanovoltmeter having been set to be 500 μv;. The result of a single measurement includes readings for positive (+) and negative (-) Josephson voltage offsets in both systems. For each polarity 0 readings of voltage difference were made, NPLC = 1;

5 5 3. The number of single measurements was 8; 4. The result that was used for the further measurement analysis was the average of the single measurements. The time of a single measurement reading was about 10 minutes. Time to obtain the result from 33 readings was 5.5 hours. The resistance of measuring terminals of the BelGIM standard was 3 Ω, the resistance of the measuring terminals leakage (based on the research) was 90 GΩ. During the measurements TS Dewars and BelGIM standard s chassis were grounded at the same point connected to the ground connector of the room where the BelGIM standard was located. Measurements were carried out for different modes of operation of the measurement circuit, including: Enabling and disabling of the galvanic coupling between the TS and the reference frequency source of 10 MHz; Using of a TS microwave generator at different frequencies; Using of a TS microwave generator with the power line supply or battery supply; Using of a TS microwave generator with a power amplifier or without it; Measurements with switched off cooling fan in the TS microwave generator; Helium vapors pressure change in the BelGIM standard s Dewar; Using of the different measurements channels of the BelGIM standard; Using of the modes M1 and M of the BelGIM standard. 3. Transfer standard VNIIM TS [1] includes a 10 V Josephson junctions array made at the PTB (Germany) by the SINIS technology. TS has an cryoprobe with Josephson array, shielded from the Earth magnetic field. Cryoprobe design allows it to change its size, which provides compactness of the TS during transportation. Circuit outputs contain low frequency LC filters at 4 K temperature. TS microwave generator frequency is changed within 68 7 GHz and has an output of about 40 mw, which is increased up to 60 mw when using the power amplifier. Microwave waveguide is galvanically isolated from the measuring circuit, which allows to reduce the influence of interference sources on the reproduced voltage and the instruments participating in the comparisons. Output voltage control and monitoring of TS parameters is performed by compact bias unit based on batteries. The resistance of the TS measuring outputs is 10 Ω, resistance of the measuring outputs leakage is more than 100 GΩ.

6 6 3.3 VNIIM standard The TS voltage is measured with the VNIIM standard [3] using nanovoltmeter Keithley 18A. The TS voltage is defined as a mean voltage value measured at the two polarities of the VNIIM standard output voltage. In addition to changing the TS and VNIIM standard polarity, the nanovoltmeter connection polarity was changed. One voltage measuring reading includes U+ (10 s), U- (10 s), pause of 10 s and repeat of the measurements with a change of the nanovoltmeter polarity. The time to obtain one reading can be up to 80 s, the time to obtain a result from 60 readings could attain 1 hour 0 minutes. The resistance of the VNIIM standard measuring outputs was 10 Ω, the resistance of the measuring outputs leakage was more than 100 GΩ. Measurements were carried out for different operation modes of the measurement circuit, including: Enabling and disabling of the galvanic coupling between the TS and the reference frequency source of 10 MHz; TS microwave generator power line supply or battery supply; Measurements with switched off cooling fan in the TS microwave generator. No impact of changes in the measuring circuit on the measuring results was detected. 4. Results 4.1 BelGIM measuring results In order to obtain the best accuracy of the results, the measuring circuit was investigated during the measurements. Preliminary measurements of the TS voltage (10 V) were carried out on 5 Aug. 014 to identify and eliminate possible defects (e.g. due to transportation) and to assess the compatibility of the systems. The optimal grounding points were chosen, the ability of systems to operate under galvanic isolation of the reference signal of 10 MHz for TS and the effect of switching the TS generator s power line supply to the battery were tested. The modes of the BelGIM standard were also tested. The dependence of the measured voltage on the pressure change in the Dewar with liquid helium of the BelGIM standard was determined. The dependence at kpa pressure jump (open output of the Dewar) is shown in Figure. The comparison was performed with the dewar open to the atmospheric pressure.

7 7 Figure. Dependence of voltage difference between two quantum standards of kpa pressure jump in BelGIM dewar The TS voltage measurements in M mode were carried out on 6 Aug It was also found that the cooling fan installed in the TS generator increased in standard deviation of single measurements. The measurements of TS voltage under different modes of operation of the TS and BelGIM standard were continued on and (3.1). The TS operated with and without the power amplifier; the frequency of TS generator was changed (from 69.5 GHz to GHz, and the TS voltage from V to V respectively); the TS generator cooling fan was disabled, the connection of the reference TS source was changed; TS generator s supply was changed; the pressure in the BelGIM standard s Dewar was changed by installing a throttle on the way of the liquid helium vapor outflow; the adjustment of the BelGIM standard characteristics was carried out, including the selection of the optimal microwave frequency and nanovoltmeter amplification factor. The measurement results are shown in Figure 3.

8 8 5 разность U_БелГИМ и U_ЭС, нв UBelGIM - UTS, / nv номер измерения Number of measurements Figure 3. The results of measurements of the voltage difference between the BelGIM standard and the TS. The measurements were conducted at the TS microwave generator frequency of GHz include points 1 14, the rest of them were carried out at the frequency of 69.4 GHz; The TS microwave generator power amplifier was used in points 4 to 31; Isolation of the galvanic coupling of the reference frequency for the TS was used in points 1 to 18; Disabling of the TS microwave generator s cooling fan was made in points 19 33; The pressure excess of kpa in the Dewar of the BelGIM standard over the atmospheric pressure corresponds to points 1 to 0, 7, 8, in other points there were no excess; Battery supply for the TS microwave generator was used in points 3, 33; Channel A of the BelGIM standard was used in points 1 to 31, channel B in point 3, channel C in point 33; BelGIM standard mode M was used in points 1 to 3, mode M1 in point 33. Experimental data obtained in different measurement modes suggest that the measurements were substantially independent. Results of the TS and BelGIM standard comparisons:

9 9 Voltage difference U BelGIM U TS i1 U i D = 1.08 nv; Uncertainty Type A u A i1 U i D 3 = 0.3 nv; Uncertainty Type B u B u B _ BelGIM u B _ TS =1.3 nv; Combined uncertainty u C = 1.3 nv. Uncertainty Type B estimation for the results of the TS and BelGIM standard comparisons is calculated in accordance with the uncertainty budget, given in Table. Table. Uncertainty Type B budget for the TS and BelGIM standard comparisons Contributing factors Type Uncertainty, nv BelGIM TS [] Frequency bias B Isolation leakages B Nanovoltmeter B 0.5 Flatness of step B - -* Combined B * - influence of current do not detected 4. VNIIM measuring results Upon return from BelGIM on 9 Apr. 014, the TS and the VNIIM standard were compared at VNIIM. The voltage was measured under different TS operation modes, including the TS working without power amplifier with the TS microwave generator cooling fan disabled; the connection of the reference TS source was changed; TS generator supply was changed (power line supply or battery supply). The microwave frequency of the TS and the standard of VNIIM was chosen to

10 10 be GHz. The total voltage of the measured difference included thermo voltage of both standards and the nanovotmeter voltage bias (less than 0.5 μv). The measurement results are shown in Figure 4. Measurement mode with the positive polarity of the nanovoltmeter connection included points 1 to 10, 1 to 30, 41 to 50, other points had negative polarity; Isolation of the galvanic coupling of the reference frequency for the TS was used in points 1 to 50; Disabling of the TS microwave generator cooling fan was used in points 11 to 60; Battery supply for the TS microwave generator was used in points 1 to 30. The 60 measurements used in the calculation of the final result are not strongly correlated so the standard deviation of the mean of the 60 measurements is used as the type A uncertainty. 15 разность, нв UVNIIM - UTS / nv Number номер of измерения measurements Figure 4. Results of the voltage difference measurements between the VNIIM standard and the TS.

11 11 Results of the TS and the VNIIM standard comparison: Voltage difference U TS U VNIIM i1 U i D = -0.1 nv; Uncertainty Type A u A i1 U i D 59 = 0.7 nv; Uncertainty Type B u B u B _ VNIIM u B _ TS =1.1 nv; Combined uncertainty u C = 1.3 nv. Uncertainty Type B estimation for the results of the TS and VNIIM standard comparisons was calculated in accordance with the uncertainty budget given in Table 3. Table 3. Uncertainty Type B budget for the TS and VNIIM standard comparisons [] Contributing factors Type VNIIM Uncertainty, nv TS Frequency bias B Isolation leakages B Nanovoltmeter B 0.01 Combined B Reference value The reference value is the BIPM voltage standard value, which is assumed to be timeindependent. The reference voltage value is disseminated from the BIPM by means of the results obtained in the comparisons of the TS and BIPM, VNIIM and BelGIM standards. 4.4 Degree of equivalence with respect to reference value The degree of equivalence D between the BelGIM and BIPM voltage standards is calculated on the basis on the BIPM-VNIIM and VNIIM-BelGIM comparisons using the TS:

12 1 DBelGIM-BIPM = DVNIIM-BIPM + (DBelGIM-TS + DTS-VNIIM) with the uncertainty uc BelGIM-BIPM = uc VNIIM-BIPM + uc BelGIM-TS + uc TS-VNIIM The results of the voltage standards comparisons are shown in Table 4. Table 4. The results of the voltage standards comparisons using the TS Standards compared Degree of equivalence D, nv Combined uncertainty, uc, nv Source VNIIM - BIPM [] BelGIM TS * TS VNIIM Total: BelGIM - BIPM * The value does not consist type B uncertainty of TS as it is included in TS-VNIIM measurements The degree of equivalence between BelGIM and BIPM voltage standards at a 10 V voltage was obtained at the level of nv with the combined uncertainty of.6 nv. Conclusion The conducted comparisons demonstrated high metrological characteristics of the BelGIM voltage standard and its reliable operation. References 1. A. S. Katkov, V. E. Lovtsyus. A compact comparison standard based on the Josephson effect with an output voltage of 10 V. Measurement Techniques vol. 54, issue 7, 011. p S Solve et al 011 Metrologia , Comparison of the Josephson voltage standards of the VNIIM and the BIPM (part of the ongoing BIPM key comparison BIPM.EM-K10.b) 3. A. S. Katkov, S. Solve. Key comparisons of the VNIIM and BIPM voltage standards. Measurement Techniques vol. 54, issue 11, 01. p Supracon. Josephson Voltage Standard supravoltcontrol. MANUAL. 007.

Comparison of the Josephson Voltage Standards of the NIMT and the BIPM

Comparison of the Josephson Voltage Standards of the NIMT and the BIPM (part of the ongoing BIPM key comparison BIPM.EM-K10.b) S. Solve, R. Chayramy, M. Stock, S. Pimsut* and N. Rujirat* Bureau International