A Guide for establishing primary AC-DC transfer standard using ET2001 ADS system

Transcription

1 A Guide for establishing primary AC-DC transfer standard using ET2001 ADS system (Tentative Version 2.01, 12/Apr./2007) Nano-Electronics Research Institute / AIST, Japan

2 About This Manual This manual, "A Guide for establishing primary AC-DC transfer standard using ET2001 ADS system", provides detailed technical information about the operation of type-et2001 AC-DC Standard system (hereafter referred to as ADS system), and its application for the evaluation of thermal converters to establish a primary AC-DC transfer standard. Contents of the other manuals are as follows: Quick-Start Manual for ET2001 ADS system This manual provides technical information for the use of ET2001 ADS system, including comprehensive guidance for installation and initial setting-up of the equipment. "Technical Reference for ET2001 ADS system Hardware" This manual provides detailed technical information about hardware of the ET2001 ADS system, including the interfacing commands and circuit descriptions. "Technical Reference for ET2001 Thermal Converters" This manual provides detailed technical information about the special TVCs, LF-TVC, HF- TVC, and MF-TVC, to be combined with the ET2001 ADS system. The organization of the manual is to first provide a simple overview of AC-DC transfer standard using the ET2001 ADS system. This is followed by examples for setting up the equipment and program in Chapters 2-3. Subsequent chapters provide information on the operation of FRDC- DC, AC-LF, and AC-DC difference measurements (Chapters 4-6), followed by examples on the evaluation of TC modules in Chapter 7. Supplementary information is given in Chapter 8. chapter 1 Concepts of ADS system chapter 2 Configure System chapter 3 Set-up Software Introductory guide for AC-DC standard chapter 4 FRDC-DC Measurement chapter 5 AC-LF Measurement chapter 6 AC-DC Measurement chapter 8 Supplementary Information FAQs, Credits Contact Address chapter 7 Evaluation of Thermal Converter Guide for evaluation of AC-DC difference of TCs

3 Important Notes (PS&USB Module) CAUTION! --- Check correct voltage setting for the AC line voltage before applying the ac power to the PS&USB module. (FRDC Module) CAUTION! --- Output current of more than 30 ma may results in the permanent damage for analog switches in the output circuit. NOTE --- Output voltage lower than 1V may not have sufficient output stability. (DSS Module) CAUTION! --- Output current of more than 50 ma may results in the permanent damage in the output buffer circuit. NOTE --- Output voltage lower than 1V may not have sufficient output stability. NOTE --- Output current of more than 30 ma may cause instability in the output and increase in the distortion of the waveform. (TC/AMP Module) CAUTION! --- Incorrect setting of the [Nominal Input] in [TVC specification] may result in the overloading of the thermal converter element. CAUTION! --- Incorrect setting of the [Input resistance] in [TVC specification] may result in the overloading of the FRDC and/or DSS output circuit. NOTE --- When connecting a TVC to AMP module, the Input-Lo/Guard of the AMP module should be connected to Output-Lo of FRDC/DSS module via Output-Lo/Chassis of the TVC. NOTE --- All the TC/AMP modules must have corresponding specification files stored in the "Etrace\TCspecs" folder.

4 List of Abbreviations AC-AC [AC to AC (difference)]: Relative AC-DC difference with respect to fixed frequency. AC-DC [AC to DC (difference)]: Error in comparing rms AC quantity with DC quantity. AC-LF [AC to Low Frequency AC (difference)]: LF characteristic in AC-AC difference. ADC [Analog to Digital Converter]: Converts output EMF of TC element to digital quantity. AMP [Amplifier (module)]: TC module without dedicated TC element inside the chassis. CPDC [Chopped DC (mode)]: Quasi steady-state DC waveform with periodical off-times. DVM [Digital Voltmeter]: Replaced by AMP module in ET2001 ADS system. ADS [(ET2001) AC-DC Standard (system)]: The equipment described in this manual. DDS [Direct Digital Synthesis (chip)]: A semiconductor chip for generation of sine-wave. DSS [Digital Sine-wave Synthesizer (circuit)]: A circuit (module) for generation of sine-wave. FRDC [Fast-Reversed DC]: Quasi-rectangular waveform with equal rms power in AC and DC. HF-TVC [High-Frequency Thermal Voltage Converter]: A TVC for high-frequency reference. GPIB [General-Purpose Interface Bus]: Interface bus commonly used in measuring instruments. LF-TVC [Low-Frequency Thermal Voltage Converter]: A TVC for low-frequency reference. MDFR [Modified FRDC (mode)]: Modified FRDC waveform with periodical off-times. MDR [Miniature Delta Ribbon (cable)]: A standard cable used in ET2001 ADS system. MES [Measurement]: A thermal AC-DC transfer standard to be calibrated (UUT). PS [Power Supply (circuit)]: Power supply circuit in USB&PS module. REF [Reference]: Reference standard by which UUT standard is calibrated. TC [Thermal Converter]: Thermal Converter to be used in AC-DC comparison. TC [Thermal Converter (module)]: Combination of a TC element and a ADC circuit. TVC [Thermal Voltage Converter]: Thermal Converter for AC voltage standard. USB [Universal Serial Bus]: Serial Interface Bus used in ET2001 ADS system USB&PS [USB and Power Supply (module)]: Power-supply module with USB interface. UUT [Unit Under Test]: A thermal AC-DC transfer standard to be calibrated (MES).

5 Table of Contents 1. Introduction Background AC-DC Transfer Standard AC-DC Difference Thermal Converter Determination of Frequency Characteristics DC Characteristics Low-Frequency Characteristics High-Frequency Characteristics Principles of Measurements FRDC-DC Difference Measurement ACLF-AC Difference Measurement AC-DC Difference Measurement Set-up ET2001 ADS System System Overview FRDC System Configuration Measurement Sequence AC-LF System Configuration Measurement Procedure AC-DC System Configuration Measurement Procedure Set-up ET2001 ADS Software Software Components Manual Operation Initial Set-up Procedure Manual Operation Setting Parameter Register Reference Point Register Measurement Procedure Register Measurement Parameter Specify File Name to Save Results Register Measurement Options Specify Address to Send Data Start/Stop Measurement FRDC-DC Difference Measurement Executing Measurement Measurement Procedure Data Format... 43

6 4.2. Curve-fitting Evaluation of Uncertainty Type-A Uncertainties Type-B Uncertainties Uncertainties in Sensitivity Coefficient Combined Uncertainty AC-LF Measurement Executing Measurement Measurement Procedure Data Format Evaluation of Uncertainty Type-A Uncertainties Type-B Uncertainties Uncertainties in Sensitivity Coefficient Combined Uncertainty AC-DC Difference Measurement Executing Measurement Measurement Procedure Data Format Evaluation of Uncertainty Type-A Uncertainties Type-B Uncertainties Uncertainties in Sensitivity Coefficient Combined Uncertainty Evaluation and Calibration of TC modules General Scheme Evaluation of a Reference Thermal Converter Method of Evaluation DC Characteristics Low Frequency Characteristics High Frequency Characteristics Over-all Characteristic Evaluation of Uncertainty Calibration of a Thermal Converter Comparison of AC-DC transfer difference Evaluation of Uncertainty Remote Calibration Supplementary Information FACs Trouble Shooting Acknowledgements Contact Address... 75

7 Appendix Appendix-A Data Examples...179

8 1. Introduction 1.1. Background DC voltage standards precise to 10-9 are achievable with Josephson junction devices. High precision for AC voltages, even to 10-7 are much more difficult to measure, so that the method to set an AC standard most commonly used is through the "transfer" or comparison with a precision DC standard. This is achieved by comparing the RMS power of the AC voltage with that of the standard DC using a thermal converter. AC voltage standards in the frequency range between 10 Hz and 1 MHz can thus be derived. However, because every electrical system is subject to noises from many sources, with many dependent on the frequency, predetermined corrections must be added to the measurements to compensate for these and to calibrate the equipment under test. Especially troublesome are non-negligible heating or cooling during the DC input mode. These Thomson and Peltier thermoelectric effects give rise to frequency-independent AC-DC differences at a 10-6 level [1]. Because of the difficulties in avoiding or evaluating the thermoelectric effects in a thermal converter, only a limited number of national metrology institutes have been able to establish independent primary standards of AC-DC transfer. Since 2001, the AIST in Japan [3] uses the Fast-Reversed DC (FRDC) method to evaluate the frequency independent AC-DC difference in the transfer standard. FDRC was developed in the 1990's, aiming at the experimental determination of the thermoelectric effects in thermal converters [2]. The FRDC method is based on the assumption that, if the frequency of the polarity reversal in a rectangular FRDC waveform is much faster than the thermoelectric time constant, the thermoelectric effects do not affect the temperature distribution. This document describes a second-generation FRDC instrument, ET2001 AC-DC Standard (ADS) System, developed at AIST in cooperation with Sunjem Co. Ltd, Japan. The new instrument includes not only the FRDC source, but also a complete miniature AC-DC comparator system, consisting of a DSS module, TC/AMP modules, and HF-TVC. The system may be used to establish an independent primary AC-DC transfer standard in calibration laboratories AC-DC Transfer Standard AC-DC Difference The ac voltage is defined by the root-mean square (rms) value of the sinusoidal waveform: V AC (rms) 1 T { V(t) } 2 dt (1.1) T 0-8 -

9 In accordance with this definition, it is possible to compare an ac voltage with a dc voltage by alternately applying them to the same heater in a thermal converter (TC) and measuring the temperature rise with a thermocouple. When dc and ac voltages that result in equal power output are applied to the input of an ideal thermal converter, the resultant EMFs are the same. In the case of an actual TC, however, the output EMFs are influenced by the effect of non-joule heating and the frequency dependent characteristics of the heater circuit. The AC-DC transfer difference is conveniently defined by the following equation. δ AC DC V AC V DC V DC E AC = E DC (1.2) The quantities E DC and E AC represent the output EMFs of the thermocouple when the dc voltage V DC and the ac voltage V AC, respectively, are applied to a thermal converter. Steady-State DC Input Output AC(Sine wave) V DC V AC (t) E DC E AC ( f ) Thermal Converter Fig. 1.1 Thermal converter for AC-DC transfer Standard Thermal Converter The most accurate AC-DC transfer standards are realized by the use of "thermal converters". Single-Junction Thermal Converters (SJTCs) were developed in the 1950s. The structure of a typical SJTC element is shown in Fig A thin filament-heater and a thermocouple are enclosed in an evacuated glass bulb. The thermocouple junction is in thermal contact with the heater at the midpoint of the heater, but is electrically insulated from it by a bead of glass or ceramic. Multijunction Thermal Converters (MJTCs) were developed in the 1970s-1980s. The MJTCs are designed to suppress the Thomson and Peltier effects. These are the main cause of the AC-DC transfer difference around 1 khz. The structure of type-jstc04 thermal converter element, developed at AIST in cooperation with Nikkohm Co., is illustrated in Fig The thermopile (thermocouples connected in series) is formed on a thin polyimide membrane supported by an alumina (Al2O3) frame. The heater is formed on an AlN (aluminum nitride) chip mounted on the polyimide membrane. The thermal converters are capable of comparing the joule heating between ac and dc modes at 0.1 ppm level, and are widely employed as the primary standard in most national standard laboratories

10 Heater Support Lead Bead Glass Bulb Thermocouple Fig. 1.2 Structure of Single-Junction Thermal Converter Heater (NiCr) Input-Hi Thermocouple (Bi/Sb) Output-Hi Polyimide Film Alumina Frame Input-Lo Output-Lo AlN Chip Fig. 1.3 Structure of JSTC04 Multi-Junction Thermal Converter There are three main causes of the AC-DC transfer difference: (1) Thermoelectric effect (dc offset): When the dc current is passed through the heater of a thermal converter, non-joule heating/cooling takes place along the heater due to thermoelectric effects such as the Thomson or Peltier effect. In the case of SJTC with standard construction, an AC-DC difference of a few ppm is observed due to the thermoelectric effects. (2) High-frequency characteristic: In the frequency range above 10 khz, the skin effect of the conductor and the stray inductance and capacitance in the input circuit become significant. When a standard-design SJTC-element is combined with a current-limiting metal-film resistor of 1kΩ, the effect to the AC-DC transfer difference is of the order of 0.1 ppm, 1 ppm, and 100 ppm at the frequency of 10 khz, 100 khz, and 1 MHz, respectively. (3) Low-frequency characteristics: The thermal time constant of a standard-design SJTC-element is about 1 s. At frequencies below 100 Hz, double-frequency thermal ripple is created due to insufficient thermal inertia. In the case of a standard SJTC, the effect to the AC-DC difference is of the order of 0.1 ppm and 10 ppm at 100 Hz and 10 Hz, respectively

11 The AC-DC transfer difference of a thermal converter over the 100 Hz to 1 MHzfrequency range is given in equation 1.3. The three components are as follows: LF: low frequency component HF: high frequency component TE: thermoelectric effects δ AC DC δ LF ( f ) + δ HF ( f ) +δ TE (1.3) The typical frequency characteristic of a thermal converter is illustrated in the figure. Frequency characteristic of a thermal converter can be evaluated using special thermal converters (HF-TVC and LF-TVC), as described in subsections and The thermoelectric effects, which occur at the dc-mode, give the frequency-independent offset in the AC-DC difference. Since both the low-frequency characteristic and the high-frequency characteristic reduce below 0.1 ppm in the frequency range between 100 Hz and 10 khz, the AC-DC difference in this range is predominately a DC offset due to the thermal effects and is not frequency dependent. δ AC-DC Thermal Ripple Stray L,C Hz 10 khz δ TE DC Offset f Fig. 1.4 Frequency characteristic of a thermal converter 1.3. Determination of Frequency Characteristics DC Characteristics The major source of the frequency-independent AC-DC difference is the second-order Thomson effect. The typical temperature distribution along the heater due to Joule-heating is shown in Fig. 1.5(a). When the Thomson effect is present, the electric current influences the heat flow and modifies the temperature gradient along the heater. The change in temperature distribution due to the Thomson effect is shown in Fig. 1.5(b). The Thomson effect can result in a temperature distribution as large as 0.1 K. However, most of the effects are canceled to the first-order by reversing the polarity of the current and taking the mean. The temperature gradient due to the second-order Thomson effect is of the order of a few mk, as shown in Fig. 1.5(c), and contributes

12 to the AC-DC difference at the ppm level. The AC-DC difference due to the second-order Thomson effect can be evaluated using the formula by Widdis: δ ac dc = 1 σ 2 θ 0 12 ρk (1.4) The symbols σ, θ 0, ρ, and κ represent the Thomson coefficients, mid-point temperature-rise, electric resistivity, and the thermal conductivity of the heater. In the case of standard SJTCs, the thermal transfer difference is of the order of a few parts in The thermoelectric effects in thermal converters can be evaluated by the "fast-reversed dc" method. Joule Component ΔT 100K (a) x 1st-order Thomson Effect DC- ΔT 0.1K x (b) 2nd-order Thomson Effect DC+ ΔT 1mK x (c) Fig. 1.5 Temperature distribution along heater Low-Frequency Characteristics When sinusoidal voltage of frequency f is applied to a TVC (thermal voltage converter), joule heating varies with double-frequency, 2f. If the frequency is sufficiently high, i.e., if the thermal time constant τ is much longer than the period of the double-frequency heating (τ>>1/f), the variation of temperature becomes negligible due to the thermal inertia of the heater. At frequencyies below 100 Hz, thermal inertia of the heater becomes insufficient to suppress the double-frequency thermal ripple. The thermal ripple causes the AC-DC difference of a TVC due to the imperfections in the SJTC elements: (a) Non-linearity of input-output characteristic of TVC. (b) Frequency dependence of the heater-resistance. (c) Imperfect averaging of the voltage ripple in EMF output

13 In the case (c), the effect to the AC-DC difference may be reduced by use of a low-pass filter or by setting the integration time of a DVM to the multiple of the input frequency. While in cases (a) and (b), the effects are based on the thermal properties of the SJTC elements, and the contribution to the AC-DC difference has to be evaluated. The low-frequency performance of a thermal converter may be evaluated by two methods using the ET2001 ADS system, i.e., (1) Impedance-matching method and (2) AC-LF measurement (synthesized-waveform method). In the case of the Impedance-matching method, AC-DC difference of a TVC is compared against another TVC of the same type using a special comparison circuit, such that one TVC is operated at a much smaller power level than the other. In the case of the AC-LF measurement, a synthesized waveform source (DSS module) is used as a reference standard, and the change in the output EMF from the TVC is measured by a DVM (AMP module). These methods are described in detail in a separate technical reference "TC manual" High-Frequency Characteristics Operating at above 10 khz, the frequency characteristic of the TVC-input circuit due to the skin effect, dielectric loss, and the stray inductance and capacitance becomes non-negligible compared with thermoelectric effects. Beyond 100 khz, the frequency characteristic contributes more than 1 ppm and becomes the dominant term in the AC-DC transfer difference. Since the impedance of the input circuit determines the frequency characteristic of a TVC, it is quite important do define the reference plane of the input circuit from which the AC-DC difference is defined. Usually, the reference plane is taken at the center of a TEE connector directly connected to the input of a TC or TC module. The primary standard in the high-frequency characteristic of AC- DC difference is realized by a specially designed TVC (HF-TVC), which has a special construction so that its high-frequency characteristic is calculable from its structure and dimensions. The design and the performance of the HF-TVC is described in detail in a separate technical reference "TC manual" Principles of Measurements FRDC-DC Difference Measurement Until the late 1990's, the ac-dc transfer standard in Japan had an uncertainty of 10 ppm. This precision was not good enough to calibrate new instrumentation. Application of the "Fast-Reversed DC" (FRDC) method, developed at PTB (Physikalisch-Technische Bundesanstalt), has allowed ten-fold improvement in the uncertainty of the national standard. The purpose of the method is to evaluate the thermoelectric transfer difference experimentally, as illustrated in Fig For simplicity, only the Thomson effect along the heater is shown in the figure. When dc current passes through a thermal converter, the temperature distribution is modified due to the Thomson effect as shown in Fig. 1.6(a). When the current is reversed, the polarity of the Thomson effect is also reversed, resulting in a different temperature distribution along heater as shown in Fig. 1.6(b). The

14 characteristic time constants of the change in the temperature distribution due to the Thomson and Peltier effects are determined by the structure and material of the heater, hereafter called "thermoelectric time constants". In the case of FRDC mode, if the reversal of the current is slow enough compared with the thermoelectric time constants, the same temperature distribution along the heater is obtained as that for the steady-state dc, as shown in Fig. 1.6(c). Hence the average output EMF of thermal converter in the slow-reversing mode is equal to the mean output EMF for DC+ and DC- modes, and the FRDC-DC difference becomes zero. On the other hand, if the reversal of the current is fast enough, thermoelectric effects do not have enough time to develop during one current direction, and the influence of thermoelectric effects is reduced to zero, as shown in Fig. 1.6(d). In this case, the FRDC-DC difference equals to the thermoelectric effect in dc modes. DC[+] DC[-] Slow Switching Fast Switching Joule component Thomson component (a) (b) (c) (d) x x Fig. 1.6 Thermoelectric effects in thermal converters with the FRDC waveform. In the FRDC-DC difference measurement, rectangular-waveform are synthesized by switching between a positive dc source (DC+) and a negative dc source (DC-) as illustrated in Fig The switching is performed using high-speed analog switches. If the switching is performed in a perfect way, a high-precision rectangular ac waveform is obtained whose rms power is equal to the mean of the two dc sources. The rectangular waveform synthesized in this way is called the Fast- Reversed DC (FRDC) waveform, and the circuit for producing the FRDC waveform is called the FRDC source. Following the definition of the AC-DC difference of a thermal converter given by (1.5), an "FRDC-DC difference" δ FRDC-DC is defined as follows: δ FRDC DC V FRDC V DC V DC E FRDC =E DC (1.5) Here, E FRDC represents the EMF for the FRDC waveform, and E DC represents the mean EMF for the DC+ and DC- waveform. A modified waveform shown in Fig. 1.8 is used in the actual FRDC sources. Since the same number of positive edges and negative edges are included in DC and FRDC waveform, the effect from switching transients and high frequency components are canceled between the DC and FRDC modes

15 DC+ DC- Thermal Converter nv Detector Fig FRDC-DC difference measurement circuit. DC(-) FRDC DC(+) Fig. 1.8 Modified DC and FRDC waveforms. In this method, the Nono-Voltmeter measures the output EMF voltages from the TVC. If the difference between the voltages V FRDC and V DC is small, the input-output characteristic can be approximated to be linear in the low voltage range. In this case, the following approximation is possible: V FRDC V DC + [ E FRDC E DC ] k Where, k = ΔE ΔV. (1.6) Here, ΔE represents the change in the EMF output from the TVC when a small change in the input voltage ΔV is applied. Substituting (1.6) to (1.5), the FRDC-DC difference δ FRDC-DC is determined by the following equation: δ FRDC DC E FRDC E DC ne DC ( ) ( ) Where, n = ΔE / E ΔV /V (1.7) The 'normalized index n is of the order of 2 for the TVCs with square-output characteristics. In the case of modified waveform, the quantities E FRDC and E DC represent the average EMFs for the two MDFR modes and for the two CPDC modes respectively, as defined by, E FRDC E MDFR(1) + E MDFR(2 ) 2, E DC E CPDC+ + E CPDC 2 (1.8)

16 ACLF-AC Difference Measurement As discussed in section 1.3.2, the low-frequency performance of a thermal converter may be evaluated by two methods using the ET2001 ADS system, i.e., (1) Impedance-matching method and (2) the AC-LF measurement (synthesized-waveform method) which is described in this section. In this measurement, a synthesized waveform source (DSS module) is used as a reference standard. The change in the output EMF from the TVC is measured by a DVM (AMP module). The "ACLF-AC difference" δ ACLF is defined using the following definition. δ ACLF V( f ) V( f 0 ) V( f 0 ) E ( f )=E( f 0 ) (1.9) Here, E(f) represents the EMF at the test frequency f, and E(f 0 ) represents the EMF at the reference frequency f 0. The schematic diagram of the measurement circuit is shown in figure 1.9. In this method, the main detector is the DVM (ADC in TC module) which measure the output EMF voltages from the TC element. AC Thermal Converter nv Detector Fig. 1.9 AC-LF Measurement Circuit. As in the case of the FRDC-DC difference measurement, if the difference between the voltages V(f) and V(f 0 ) is small, the input-output characteristic can be approximated to be linear in the small voltage range. In this case, the following approximation is possible: V AC ( f ) V AC ( f 0 )+ [ E( f ) E( f 0 )] k Where k = ΔE ΔV. (1.10) Substituting (1.10) to (1.9), the ACLF difference δ ACLF is determined by the following equation: δ ACLF E( f ) E( f 0 ) ne( f 0 )

17 ( ) ( ) Where, n = ΔE / E ΔV /V (1.11) Normalized index "n" is of the order of 2 for the TVCs with square-output characteristics AC-DC Difference Measurement The purpose of the AC-DC difference (comparison) measurement is to determine the relative difference in the AC-DC difference between two TVCs, usually specified as TC(X) and TC(S). The ET2001 ADS system performs the AC-DC difference measurement based on the dual-channel method. The schematic diagram of the dual channel method is shown in Fig In this method, the two nv-detectors measure the output EMF voltages of the two TVCs separately. DC- DC+ AC TC(X) TEE TC(S) nv Detector nv Detector Fig AC-DC difference measurement circuit. The input-output characteristics of the two TVCs are shown in Fig The EMF output of a TVC is approximately proportional to the square of the input voltage. The output-quantity X DC and S DC represent the EMF outputs from TVC(X) and TVC(S) for the dc input voltage V DC. Similarly, the output-quantity X AC and S AC represent the EMF outputs for the ac input voltage V AC. The inputquantity V X and V S represent the ac input voltages which produce the same EMF voltage (X DC, S DC ) as in the case of applying the dc input voltage V DC. Using the definition of the AC-DC difference of a TVC given in (1.2), the relative AC-DC difference between TVC(X) and TVC(S) is deduced as δ X δ S V X V S V DC. (1.12) X AC = X DC S = S If the difference between the dc input voltage V DC and ac input voltage V AC is small, the inputoutput characteristic of the two TVCs may be approximated to be linear in the small voltage range. In this case, the following approximation is possible:

18 ( ) k X ( ) k S V X V AC + X DC X AC V S V AC + S DC S AC Where k X = ΔX ΔV, k S = ΔS ΔV. (1.13) Here, ΔX and ΔS represent the change in the EMF output from TVC(X) and TVC(S) when a small change in the input voltage ΔV is applied. Substituting (1.13) to (1.12), the relative AC-DC difference δ X δ S is determined by the following equation: δ X δ S S AC S DC n S S DC X AC X DC n X X DC Where n X = ΔX / X DC ΔV / V DC ( ) ( ), n = ( ΔS / S DC) S ( ) ΔV / V DC (1.14) The normalized indices n X and n S are of the order of 2 for the TVCs with square-output characteristics. Some of the semiconductor-based AC-DC transfer standards, like Fluke 792A or Datron 4920, have linear output characteristics, resulting in normalized indexes close to unity. Output (X) E AC (X) E DC X DC X AC S DC (S) E AC (S) E DC S AC V AC V S V X V DC Intput Fig Input-Output Characteristic

19 2. Set-up ET2001 ADS System 2.1. System Overview The new AC-DC transfer standard system was developed at AIST in cooperation with SunJEM Co. The system consists of four main components, i.e., FRDC module, Digital Sine-wave Synthesizer (DSS) module, Thermal Converter (TC) or Amplifier (AMP) module, and Power Supply/Interface (PS/IF) module. The appearance of the FRDC, DSS and TC modules are shown in figure 2.1. The complete set of modules can be packed into a B4-file size attaché case, and the whole system can be transported to the calibration site if required. A. FRDC Module FRDC module is a rectangular-waveform voltage source to be used in FRDC-DC difference measurement. It produces accurate FRDC and DC waveforms with amplitudes from 1 V to 10 V and switching frequencies between 0.1 Hz and 10 khz. The design of the FRDC circuit is based on the "source A/B switching" scheme [2], in order to establish the equality of rms values between the FRDC and DC waveforms. B. DSS Module DSS module generates highly stable sinusoidal ac and dc outputs to be used in AC-DC difference measurements. The module is based on a direct-digital-synthesizer (DDS) device, and generates frequencies between 10 Hz and 1 MHz at rms voltages from 1 V to 10 V. In the evaluation of lowfrequency characteristics of thermal converters, the DSS module is used as a reference in ac voltage standard. C. TC Module TC module is a digital-output thermal converter. The module consists of a thermal converter element, a precision A/D converter as a nv detector, a D/A converter for offset compensation, and an optically isolated digital control circuit. The AMP module is almost identical to the TC module, except that it does not contain a dedicated thermal converter element, and the type-n input connector is replaced with a low-thermal DC input connector to be combined with an external thermal converter such as HF-TVC. D. USB&PS Module USB&PS module provides isolated DC power sources to the main modules. Switching regulator circuits are avoided to minimize the effect of high-frequency interference to the sensitive nano-volt detection circuit. The module also provides an optically isolated USB-to-serial interface circuit between a PC controller and the FRDC, DSS, and TC modules

20 Fig. 2.1 ET2001 ADS system 2.2. FRDC System FRDC-DC difference measurement is performed using a FRDC module and one TC module to be evaluated. The TC module may be replaced by a combination of conventional thermal converter and an AMP module. The standard procedure for the measurement is as follows: (1) Configuration --- Connection of the three modules (FRDC, TC/AMP, USB&PS) and setting-up the control program. (2) Setting Parameter --- Registering various measurement conditions and measurement options step-by-step. (3) Execute Measurement --- Performing a fully automated FRDC-DC difference measurement. (4) Data Analysis --- Determination of a thermoelectric transfer difference of the thermal converter Configuration CASE Measurement with a TC module (1) Connect an AC Power cable and a USB cable to USB&PS module. Then turn on the power switch. (2) Check green light in front panel, and connect a 26-pin MDR cable to FRDC/DSS port. (3) Connect the other end of the 26-pin MDR cable to a FRDC module. (4) Connect a 20-pin MDR cable to TC#1(X) port of USB&PS module. (5) Connect the other end of the 20-pin MDR cable to the TC (or AMP) module. (6) Connect the N-R input connectors of the FRDC module and the TC (AMP) module with an NP-NP adopter or with a NP-NP cable no longer than 50 cm. CASE Measurement with a thermal converter and an AMP module

21 Same as CASE 1 except that the TC module is replaced by a combination of a thermal converter and an AMP unit. The Input-Lo and the Output-Lo of the thermal converter must be connected to the Input-Lo of the AMP module. FRDC module TC module Fig. 2.2 FRDC measurement Measurement Sequence The FRDC-DC difference of a TC module is measured using a measurement sequence [*+-//-+*], where the symbols represent: [*]: Modified FRDC (A+/B-): MDFR[1] mode [+]: Chopped dc (A+/B+): CPDC[+] mode [-]: Chopped dc (A-/B-): CPDC[-] mode [/]: Modified FRDC (A-/B+): MDFR[2] mode At each FRDC output mode, the output EMF from the TC element is measured by the internal AD converter, and the digital data are transferred to the controller (PC) via the PS/IF module. The readings are averaged for specified number of reading, and the FRDC-DC difference is calculated using the formula (2.7) described in section The complete FRDC measurement sequence is described in section AC-LF System The ACLF system is for an AC-AC difference measurement at low-frequency range (<100Hz), to evaluate low-frequency characteristic of a TC module. The ACLF measurement uses fixedsampling-per-period output-mode of the DSS module, and low-frequency characteristic of a TC module is evaluated using the DSS module as a reference standard. The TC module may be replaced by a combination of conventional thermal converter and an AMP module. The standard procedure for the measurement is as follows: (1) Configuration --- Connection of the three modules (DSS, TC/AMP, USB&PS) and setting-up the control program

22 (2) Setting Parameter --- Registering various measurement conditions and measurement options step-by-step. (3) Execute Measurement --- Performing a fully automated ACLF measurement. (4) Data Analysis --- Determination of a thermoelectric transfer difference of the thermal converter Configuration CASE Measurement with a TC module DSS module TC module Fig. 2.3 ACLF measurement (1) Connect an AC Power cable and a USB cable to USB&PS module. Then turn-on the power switch. (2) Check green light in front panel, and connect a 26-pin MDR cable to FRDC/DSS port. (3) Connect the other end of the 26-pin MDR cable to a DSS module. (4) Connect a 20-pin MDR cable to TC#1(X) port of USB&PS module. (5) Connect the other end of the 20-pin MDR cable to the TC (or AMP) module. (6) Connect the N-R input connectors of the DSS module and the TC (AMP) module with an NP- NP adopter or with a NP-NP cable no longer than 50 cm. CASE Measurement with a thermal converter and an AMP module. Same as the configuration shown above except that the TC module is replaced by a combination of a thermal converter and an AMP unit. The Input-Lo and the Output-Lo of the thermal converter must be connected to the Input-Lo of the AMP module Measurement Procedure Three-mode measurement sequences, [AC(ref) / AC(test) / AC(ref)], are used to measure the difference in the EMF output with frequency, while eliminating the influence of linear drift in DSS output and EMF output voltage. AC(test): Sinusoidal AC output at test frequency

23 AC(ref): Sinusoidal AC output at reference frequency To avoid the effects from transient response after mode-switching, controller waits for specified period (normally 10s - 20 s) before integrating the reading from the TC/AMP module. The AC-AC difference is calculated using the formula (2.11) described in section The complete ACLF measurement sequence is described in section AC-DC System The AC-DC system is for an AC-DC or AC-AC difference measurement using a DSS module at fixed-sampling-per-period mode. This measurement determines relative difference between two TC modules, TC(X) with respect to TC(S). The TC module(s) may be replaced by a combination of conventional thermal converter(s) and AMP module(s). The standard procedure for the measurement is as follows: (1) Configuration --- Connection of the four modules (DSS, TC/AMP(X), TC/AMP(S), USB&PS) and setting-up the control program. (2) Setting Parameter --- Registering various measurement conditions and measurement options step-by-step. (3) Execute Measurement --- Performing a fully automated AC-DC difference comparison measurement. (4) Data Analysis --- Determination of a thermoelectric transfer difference of the thermal converter. In the comparison of the AC-DC difference in the higher frequency range (>10 khz), frequency characteristic of a TVC including the connecting circuit between the TVCs becomes significant. Hence, it is widely accepted to define the branch-point of a TEE connector, which connects the TVCs and the AC/DC source, as the 'reference-plane' from which the frequency characteristic of a TVC is evaluated. In the case of standard TVC using type-n receptacle (N-R) as the input connector, the center of N-type TEE connector (N-TA-RRR) is taken as the reference plane Configuration CASE Measurement using TEE connector (1) Connect an AC Power cable and a USB cable to USB&PS module. Then turn-on the power switch. (2) Check green light in front panel, and connect a 26-pin MDR cable to FRDC/DSS port. (3) Connect the other end of the 26-pin MDR cable to a DSS module. (4) Connect a 20-pin MDR cable to TC#1(X) port of USB&PS module. (5) Connect the other end of the 20-pin MDR cable to the TC (or AMP) module. (6) Connect another 20-pin MDR cable to TC#2(S) port of USB&PS module. Also, connect the other end of the 20-pin MDR cable to the TC (or AMP) module. (7) Connect the N-R input connectors of the DSS module and the two TC modules (or TC) with an NP-NP adopter or with a NP-NP cable no longer than 50 cm

24 TC module(x) TEE TC module(s) DSS module Fig. 2.4 ACDC measurement CASE Measurement using a HF-TVC and a TC module. TEE connector is replaced by a built-in TEE inside a HF-TVC. The output of the HF-TVC is connected to AMP module. AMP module HF-TVC TC module DSS module Fig. 2.5 ACDC measurement Measurement Procedure [1] AC-DC Difference measurement

25 A set of two four-mode standard measurement sequences, [AC/DC+/DC-/AC] and [AC/DC- /DC+/AC] are used in turn to eliminate the influence of linear drift and to check the dependence on the sequence in DC+ and DC-: AC: Sinusoidal AC output at test frequency DC+: Steady-state positive DC output. DC-: Steady-state positive DC output. [2] AC-AC Difference measurement] As in the case of ACLF measurement, three-mode measurement sequences, [AC(ref)/AC(test)/ AC(ref)], is used to measure the difference in the EMF output with frequency. In either measurements, controller waits for specified period (normally 10s - 20 s) before integrating the reading from the TC/AMP module(s). The AC-DC or AC-AC difference is calculated using the formula (2.14) described in section The complete measurement sequence for the AC-DC or AC-AC difference measurements are described in section

26 3. Set-up ET2001 ADS Software 3.1. Software Components The package of ET2001 control software is composed from the following items. (1) Main program ET01vXIII ---Executable basic program. (2) "Defsetting" File ---Preset values for et2001 control program (3) "ErrorLog" file ---Log file to report errors. (4) "TCspecs" folder ---Folder in which information on the registered TC modules are recorded. ---IMPORTANT--- All TC/AMP modules must have corresponding data file in this folder. (5) "Procedures" folder ---Folder in which measurement procedures are registered. (6) "MesResults" folder --- A default folder in which measurement data is to be recorded. (7) "Tools" folder ---Folder which contains USB Interface driver (D10606) and utility program (8) "Manual" folder

27 (9) "ET2001" package folder Installer for ET2001 program Manual Operation Initial Set-up Procedure (1) Double click ET01vX icon in the installed directory to start program. (2) Configuration Prompt Message window will appear. (The color of the LED stays green, showing that no power is applied to the MDR ports, and it is safe to connect or disconnect modules.) Check the following condition: (a) USB cable is connected. (b) USB&PS module is Powered ON. (c) All the modules are connected to the USB&PS module. Check proper configuration for each specific measurements (FRDC, AC-LF, AC-DC, etc). Then click Proceed button to apply power to the modules. (The color of the LED will turn to yellowish green, showing that now it is NOT safe to connect or disconnect modules.)

28 (3) When AMP modules are connected to port-1, a message window will appear prompting the input of description for the TVC connected to the AMP module. Check OK to proceed. (Not necessary for TC module.) (4) [TC-X] Specification window will appear. Chose appropriate description (or edit and save as a new list.) Check Select to proceed. (Not necessary for TC module.) To make a new entry, type in the information on the thermal converter to be measured. To protect the TCs and/or FRDC/DSS modules from overloading, following information is required in the specification list: Nominal Input: Maximal input voltage to be applied to TC. Nominal Output: Output EMF at the nominal input voltage. Input Resistance: Current compliance is calculated from this

29 Time Constant: Gives warning if measured value differs. Output Curve: for calculating EMF output voltage. (5) When AMP modules are connected to port-2, repeat the same procedure for [TC-S] Specification window (6) After a few seconds, a message window will appear notifying the completion of initial set-up procedure. Check OK to proceed to manual control or setting measurement parameters. In the case of an AC-DC difference measurement, the initially displayed windows are: (b) (a) (c) (d) (e) (f) (a) Main window: (b) DSS module control/display: (c) TC-X module control/display: (d) TC-X module control/display: (e) Output EMF monitor: (f) Temperature monitor: Manual Operation (a) Main window Use Manual Control menu to control modules manually, or use Measurement menu or SET button to proceed to the next stage

30 (b-1) FRDC module control/display This window is displayed when an FRDC module is connected to Port-1 (FRDC measurement). This window shows (1) Serial Number of the FRDC module, (2) Output voltage, (3) Output frequency, (4) Output modes, (5) EXEC/STBY condition, (6) Waveform parameter, and (7) Adjustment parameters for the four sources A+/A-/B+/B-. (b-2) DSS control/display This window is displayed when a DSS module is connected to Port-1 (ACDC/ACAC measurement). This window shows (1) Serial Number of the DSS module, (2) Output voltage, (3) Output frequency, (4) Output modes (AC/DC/ACLF), (5) EXEC/STBY condition, (6) Buffer operation, and (7) Adjustment parameters for the AC/DC+/DC- sources and offset in AC

31 (c) TC modules control/display These windows show (1) Serial Number of TC/AMP module, (2) DVM reading, (3) Temperature monitor inside the TC/AMP module, (4) Gain/Offset setting, and (5) parameters describing the TC/AMP modules connected to Port-2 and Port-3, respectively. Only one window is displayed for FRDC and ACLF measurements. (d) Output EMF monitor This window displays the output trace from the TC/AMP module(s) during the integration period. Green line represents reading from TCX, and yellow line represents one from TCS

32 (e) Temperature monitor This window displays the change in the temperature inside the TC/AMP module(s). It displays the initial warm-up condition, and usually shows the temperature several degrees higher than the ambient temperature after stabilization. Green line represents reading from TCX, and yellow line represents one from TCS Setting Parameter Register Reference Point [Case 1 / FRDC measurement] FRDC measurement has no reference point and this window is skipped automatically. [Case 2 / ACLF measurement] Reference Frequency for AC-LF Set reference frequency (default 100 Hz) for an AC-AC(ref) difference measurement. Then select resolution (number of sampling per period) to 1/512 (default) or 1/1024. Click OK button to proceed

33 [Case 3 / ACDC difference measurement] Set reference to DC for an AC-DC difference measurement or to some reference frequency (default 1kHz) for an AC-AC(ref) difference measurement. Resolution of waveform (number of sampling) is fixed to 1/32 micro sec Register Measurement Procedure Use Edit/Save button to make a new procedure or modify an existing procedure. Use Select button to proceed

34 Measurement Procedure ---Editing Procedure Use Add/Delete/Register buttons to edit an existing procedure

35 The voltage level must be in the range between 1 V to 10.2 V, and should not exceed more than 120% of the nominal input voltage of TC. The allowable frequency range is from 10 Hz to 1 MHz for AC-DC and AC-AC measurements and 5 Hz to 200 Hz for AC-LF measurement. Use NewFile/Save buttons to either create a new file or overwrite the existing procedure, and use EXIT button to exit this window and continue setting parameters Register Measurement Parameter Start-up Wait: Waiting time for initial heat-up of thermal converter element. The first test point may be used as a dummy measurement for additional start-up wait. Wait for Trigger: Waiting time before accumulation of ADC reading is started, e.g., during the mode change between (AC/DC+/DC-). Roughly set as time constants x10. ADC Integration: Accumulation (integration) time for ADC reading, e.g., repetition number for one Voltage/Frequency setting. Should be roughly equal to the Wait for Trigger

36 Measurement/block: Repetition number for one Voltage/Frequency setting. Should be larger than 5 to calculate standard deviation accurately Specify File Name to Save Results. Specify directly and name of the file to which the measurement data are stored

37 Initial default directory is C:\Program Files\Etrace\MesResults. Any other directory may be selected, and the new directory will be recorded as a new default in to the DefSetting file Register Measurement Options Input Measurement Option Recommended options are, Drift Allowance : 10 ppm/min. (May not be sufficient for most accurate measurements) Ending Option : Go to stand-by mode. (power-off modules, safer option) Index Measurement Option : Always Measure Index. (for quick check use Skip-Measurement option) Specify Address to Send Data. Set Report Option

38 After registering all the parameter or options, click OK to go back to the main menu. (The measurement will NOT start automatically.) 3.4. Start/Stop Measurement After registering all the parameter or options, the SET button will change to GO button, ready for a fully automated FRDC-DC/ACLF-AC/AC-DC/AC-AC difference measurement. Hit the GO button to start measurement. (The GO button will change to STOP button, which enables the abortion of the measurement.)

39 [Automated Measurement] Typical appearance of display (screen) during the automated FRDC-DC difference measurement is shown below. The fourth parameter RT displayed on the bottom of the main widow shows the approximate remaining time before the measurement will be completed. [ End of measurement]

40 After finishing the measurement, the program go to the stand-by mode showing the Configuration Prompt Message window, ready to restart another FRDC measurement or change configuration to ACDC or ACLF measurements

41 4. FRDC-DC Difference Measurement 4.1. Executing Measurement Measurement Procedure The measurement procedure of an FRDC-DC difference measurement is basically the same as that for standard AC-DC difference measurements. Flow-chart of the measurement procedure is shown in Fig 4.1. Start Initialize Input Parameters Measurement Block #1 Measurement Block #n Set Meas. Parameters Measure Index "n" Adjust Four Sources (A±/B±) Measurement Sequence #1 Measurement Sequence #m FRDC [1] Mode DC [+] Mode DC [-] Mode FRDC [2] Mode FRDC [2]Mode DC [+] Mode Change Mode Wait for Stabilization Measurement Block #N Measurement Sequence #M DC [-] Mode FRDC [1] Mode Calculate δ DVM Reading Ending Procedure Store Data to DISK Re-adjust Sources END Fig. 4.1 flow-chart of FRDC-DC difference measurement program

42 After registering all the parameters or options, as described in chapter 3, the program will go to stand-by mode, ready for a fully automated FRDC-DC difference measurement. When "GO" button is pressed, the program will apply voltage to the TC module, and waits for a specified period of time (normally 10 minutes) to avoid the effects from initial warm-up drift. Then the program repeats procedure (1) to (5) at each test points (Measurement Loops) as follows. (1) Measurement of sensitivity index The control program measures the normalized sensitivity index n at each test point. The normalized sensitivity indices are measured by changing the input voltage by dv (normally 0.1%). Influence of drift of the output voltage is removed by a measurement sequence [(V-dV ) / (V+dV ) / (V-dV)]. NOTE After repeating the index-measurement 10 times, the program proceeds to the next stage with warning message that the drift is too large. (2) Adjustment of voltage level The FRDC and CPDC waveforms are generated by combining four independent voltage sources (A+, A-, B+, B-) inside FRDC module. The amplitudes of the outputs from the four sources are automatically adjusted within 0.01% relative to each other. This process reduces AC components in the power and prevents low frequency thermal ripple. NOTE ---- After repeating the adjustment five times, the program proceeds to the next stage with warning message that the adjustment is not sufficient. (3) Measurement sequence In the case of FRDC-DC difference measurement, the following eight-mode measurement sequence is used to eliminate the influence of linear drift in FRDC output and EMF output voltage: [FRDC/CPDC+/CPDC-/FRDC/FRDC/CPDC-/CPDC+/FRDC] To avoid the effects from transient response after mode-switching, program waits for specified period (normally 10 s) before reading the EMF from TVC. Then the readings of the DVM are integrated for 10 s ~ 20 s and average values and standard deviations are calculated. (4) Determination of FRDC-DC difference The FRDC-DC difference is calculated using formula (2.7). The measurement sequences are repeated for 10 times, and average value and standard deviation for the FRDC-DC difference is obtained. (5)Storing measurement data After measurement sequence, measurement conditions and measurement data of each test point are stored to hard disk of measurement controller. The recorded items are listed in the following subsection. After measurements for all test points are executed, summary of measurement data are stored to the hard disk of the system controller. Then instruments are reset to initial condition preparing for the

43 exit from the measurement program. In the case of standard measurement condition, one measurement loop takes about one hour. For a set of 16 standard test points from 0.1 Hz to 5kHz repeated twice, whole measurement (33 points) takes approximately 32 hours Data Format The results from the FRDC-DC difference measurement are stored into the specified data-file using the same format as displayed in the [Data Recorded to File] window. The data-file consists of the following records. (1) Title "Data from FRDC-DC difference measurement." (2) Revision Number of the control program (3) Main header common to all measurement-blocks, including: (3-1) Comment of the measurement, (3-2) ID (serial) number of FRDC module, (3-3) Name, ID (serial) number, and description of TC/AMP module, (3-4) Input resistance of TC and fixed Dummy-resistance (1kohm), (3-5) Off-time, (3-6) Number of repetition for one measurement blocks, (3-7) Waiting time before the ADC integration, and for initial warm-up time. (3-8) Number of ADC sampling. (4) Time constant of TC (measured). (5) Data for one set of measurements, consisting of: (5-1) Measurement block number, (5-2) Date and Time of each measurement block, (5-3) Output Level/Mode for each block, (5-4) Switching Period/Frequency for each block, (5-5) Result of TC Index measurement, (5-6) Result of Source adjustments, (5-7) Results of one measurement-sequence " + // + ", consisting of: (5-7-1) Measurement-sequence number, (5-7-2) Time of each measurement-sequence, (5-7-3) Temperature inside TC/AMP module, (5-7-4) EMF outputs for each mode (" + /"), (5-7-5) Average standard deviation of EMF outputs in ppm, (5-7-6) FRDC-DC difference for each sequence in ppm, (5-8) Average FRDC-DC difference for each measurement-block (5-9) Standard deviation of FRDC-DC difference in ppm. (6) Summary of the measurement. (7) Error/Warning message during the measurement

44 4.2. Curve-fitting Thermoelectric transfer difference δ TC is the most important quantities to be evaluated by the FRDC-DC difference measurement of a TC. Most of the MJTCs, including JSTC04 and JSTC05, are successfully designed to reduce the thermoelectric transfer difference within the detection sensitivity (<0.1ppm). Typical results from FRDC-DC difference measurement for SJTCs and MJTCs are shown in Fig SJTC S SJTC S12-57 MJTC N SJTC S6-8 MJTC PTB MJTC GL32472 SJTC S10-28 SJTC S Switching Frequency (Hz) Fig. 4.2 Typical results of FRDC-DC difference measurement. In the case of SJTCs with measurable thermoelectric transfer difference, the time constants can be evaluated by curve-fitting the data of δ FRDC-DC to a formula that determines the frequencydependence of δ FRDC-DC. δ FRDC DC = δ TE 2τ TE T SW tanh T SW 2τ TE. (4.1) A simple mathematical model of the thermoelectric effect in a thermal converter has been introduced in [7]. In this model, the thermoelectric effect was represented by an excess current i TE =δ TE i 0 which responds exponentially with a time-constant τ TE. Please refer to ref.[7] for more detailed information Evaluation of Uncertainty In this section, the uncertainty is estimated for the FRDC-DC difference measurement. The sources of uncertainty are divided into two categories, namely, Type-A and Type-B. The type-a

45 uncertainties can be evaluated from actual measurement as the standard deviation of the data, while the type-b uncertainties have to be estimated using different methods depending on the nature of the sources of uncertainty Type-A Uncertainties The type-a uncertainty in the FRDC-DC difference measurement for a "test"-tc module (UUT) is contributed from either uncertainty related to FRDC output ΔV or that to the output EMF voltage ΔE from TC. Considering the square input-output characteristic of standard TVC, the uncertainty components in ΔE contributes with factor one-half to the FRDC-DC difference: dv V = 1 de n E, (4.2) The type-a uncertainty is composed of the following components: (1) Stability/noise of FRDC module output In the case of an FRDC module of the ET2001 ADS system, typical thermal drift is specified to be <10 ppm/deg. Though linear drift in the output is compensated by the standard measurement sequence [MDFR(1), CPDC(+), CPDC(-), MDFR(2), MDFR(2), CPDC(-), CPDC(+), MDFR(1)], and most of the non-linear fluctuation averages out for the normal integration period of 10 s, it can still contribute to the type-a uncertainty. Since the change in the output of the FRDC module causes the change in the EMF output of the TCs, and the temperature coefficients of TC-output are much larger than that of the FRDC circuit, the effect to the type-a uncertainty will be included in the effect from the stability of TC module output discussed in the following paragraph. On the other hand, typical short-term stability (0.03Hz to 3Hz) of the output of the FRDC module is specified to be <10 ppm p-p. The equivalent low-frequency noise from the FRDC circuit is calculated as (10 / 5) / 3 = 1.2 ppm/ Hz. For normal measurement sequence (40 s for positive and 40 s for negative), contribution to the type-a uncertainty is estimated to be 1.2 x ( 2 / 40) = 0.26 ppm for each sequence, or 0.26 / 10 = 0.08 ppm for the average of 10 measurement sequence. (2) Stability of TC EMF output In the case of JSTC04 TC elements used in the standard TC modules, the temperature coefficients of the output EMFs are of the order of 100 ppm/k, much higher than that of the output of FRDC module. Hence the thermal guarding of the TC module against the change of the ambient temperature is critically important for a precision measurement. The ET2001 control software suspends measurement until proper drift-condition (<10 ppm/min) is established. The linear component of the drift in the EMF output is compensated by the standard measurement sequence, as in the case of the drift of an FRDC module. In this case, second-order drift during one set of measurement (1 min) should not exceed one-half of the linear drift. Hence, fluctuation in the output of a TC module (including the fluctuation in the output of FRDC module) should not affect the measurement results by more than 5 x (1/2) =2.5 ppm for each sequence. The contribution to FRDC-DC difference for the average of 10 measurement sequence is estimated to be 2.5 / (2* 10) = 0.4 ppm, taking eq. (11.2) into account

46 (3) Thermal noise in TC output In addition to the effect of temperature variation, a fluctuation is contributed by the Johnson noise of the thermocouple of TC. In the case of a JSTC04 TC elements which has 400 Ω EMF output impedance, the Johnson noise of the thermocouple e n is estimated as, e n (rms) (J / K) 300(K) 400(Ω), (4.3) 2.6 nv / Hz In this case, the thermal noise from the TC is 0.43 ppm/ Hz with respect to the smallest total output EMF of 3 mv (for JSTC04C-200 at 1V test voltage), taking square characteristic of TC into account. For normal measurement sequence (40 s for positive and 40 s for negative), contribution to the type-a uncertainty is estimated to be 0.43 x ( 2 / 40) = 0.1 ppm for each sequence, or 0.1 / 10 = 0.03 ppm for the average of 10 measurement sequence. (4) Resolution of Detector The ADC circuit, combined with back-up circuit, has typical resolution of better than 100 nv/ Hz. This resolution amounts to 14.3 ppm/ Hz in the voltage resolution with respect to the total output EMF of 7 mv, or 7.2 ppm/ Hz in the resolution for the FRDC-DC difference measurement considering the square characteristic of the EMF output. Hence, the resolution of the detector usually dominates the over-all resolution of the measurement system. For normal measurement sequence (40 s for positive and 40 s for negative), contribution to the type-a uncertainty is estimated to be 7.2 x ( 2 / 40) = 1.6 ppm for each sequence, or 1.6 / 10 = 0.5 ppm for the average of 10 measurement sequence at 7mV. Assuming that measurement sequence will be repeated 10 times for each test frequency point, the overall type-a uncertainty for the average of the 10 measurements is estimated to be 0.7 ppm. Since the type-a uncertainty are strongly dependent on the measurement conditions, such as the stability of the ambient temperature or possible interference from external noise, actual type-a uncertainty must be calculated from the standard deviation (spread) of the 10 measurement sequence Type-B Uncertainties Sources of type-b uncertainty in the FRDC-DC difference measurement for a "test"-tc module (UUT) is as follows: (1) Memory Effect When an analog switch change from ON state to OFF state, electric charges are trapped in the FET channel. When the switch becomes ON state again, these charges are released and injected to the output current, resulting in the positive FRDC-DC difference proportional to switching frequency

47 In the case of FRDC modules of ET2001 ADS system, the source A/B switching scheme is employed in order to suppress the effect. The results from the FRDC-DC difference measurements do not show such linear dependence with switching frequency, and hence the contribution of this item is estimated to be within the maximum-resolution of the measurement system, i.e., 0.3 ppm. The estimated standard (1s) uncertainty is 0.3/ 3 = 0.17 ppm, assuming uniform distribution. (2) Interference Between the Sources To obtain the equal RMS power for FRDC and DC modes, it is essential that there is no interference between the Source A and Source B. This requirement may be confirmed using the "isolation-test circuit" described in section The contribution to the measurement uncertainty should be within the maximum-resolution of the measurement system, i.e., 0.3 ppm. The estimated standard (1σ) uncertainty is 0.3/ 3 = 0.17 ppm, assuming uniform distribution. (3) Effect of off-time If the period of the OFF-state is not long enough, the switching transients do not converge during the off time, and there is a possibility of correlation between the waveform from the sources A and B. The effect can be checked experimentally by changing the off-time between 5µs to 200 µs. If no change in the FRDC-DC difference is detected, the contribution to the measurement uncertainty is estimated to be within the maximum-resolution of the measurement system, i.e., 0.3 ppm. The estimated standard (1σ) uncertainty is 0.3/ 3 = 0.17 ppm, assuming uniform distribution. (4) Mismatching of rms power As in the case of AC-DC difference comparison measurement, mismatching of rms power between the FRDC mode and the dc mode can cause thermal ripple, and contribute to the uncertainty due to the nonlinear output characteristic of the TC under test. Hence, in the measurement program, all the four sources (A±, B±) should be adjusted for equal outputs to within 100 ppm before each measurement. If this precaution is taken, the contribution to the FRDC-DC difference should be much smaller than 0.1 ppm. (5) Output resistance of the source Since the output impedance of the FRDC source is not zero, the value obtained in this condition deviates from that for pure voltage mode to that for pure current mode. The degree of the deviation is estimated by the ratio of output impedance of the source (0.1 Ω) to the minimum input resistance of the TVC (100 Ω), and should be smaller than 0.1Ω 100Ω = Since the absolute values of FRDC-DC difference for normal TCs are smaller than 10 ppm, the deviation is estimated to be much smaller than 0.1 ppm. (6) High-Frequency Components Some types of RF coaxial cables use Cu coated Fe wires as the inner conductors. If these cables are used to connect the TVC with the FRDC source, it can cause FRDC-DC difference that increases linearly with switching frequency due to a skin-effect. This effect occurs at the voltage mode, and becomes significant for TVCs with low input resistance (<100Ω). Other magnetic materials such as

48 iron-clips should also be avoided. If these precautions are taken, the contribution from the skineffect should be much smaller than 0.1 ppm. (7) Curve fitting In the case of SJTC with measurable thermoelectric transfer difference, frequency-independent part of the AC-DC transfer difference of TVC is determined from a curve fitting of FRDC-DC difference data to the theoretical formula. The uncertainty in the curve fitting is evaluated from relative deviation of the FRDC-DC difference data from the theoretical curve around 1kHz. This uncertainty component may be omitted for MJTCs with non-measurable thermoelectric transfer difference Uncertainties in Sensitivity Coefficient The sources of uncertainty in the sensitivity coefficient measurement and its contribution to the FRDC-DC difference results are evaluated in the following paragraph. (1) Resolution of Detector The relationship between the resolution of the EMF voltages (ΔE) and the error in the "index" measurements n ( ) n is given by. ( ) n n = ΔE E DC ΔE E DC. (4.4) The DAC circuit of the TC module has typical resolution of 14 ppm/ Hz in the voltage resolution with respect to the smallest output EMF of 7 mv (see technical reference "Hardware manual"). For normal measurement sequence (10 s for -0.1%, 10 s for +0.1%, and 10 s for -0.1%), contribution to the type-a uncertainty is estimated to be 14 x ( (1/10 +1/20)) = 5.2 ppm. Combining the values ΔE/E DC for ±0.1% change in the input, the uncertainties in the index measurement ( n) n are estimated to be <1.3x10-3. This estimation can be confirmed by repeating the index measurement more than 10 times and calculating the standard deviation. The relationship of the uncertainty in the index measurement ( n) n and the uncertainty contributed to the FRDC-DC difference is given by: [ ( δ FRDC DC )] 2 E FRDC E DC ne DC 2 n n 2 (4.5) In the FRDC-DC difference measurement, the ac voltage is adjusted to the dc voltage such that E FRDC should equal to E DC within 100 ppm. Hence the uncertainty contributed to the FRDC-DC difference measurement is estimated to be smaller than 0.13 ppm. The estimated standard (1σ) uncertainty is 0.13/ 3 = 0.08 ppm, assuming uniform distribution in the adjustment. (3) Output Linearity of FRDC module

49 Non-linearity in the voltage source can also be a source of error in the index-measurement. Using (4.5), the uncertainty contributed to the FRDC-DC difference is evaluated to be: ( ΔV) ( ) δ FRDC DC δ FRDC DC V DC ΔV. (4.6) V DC In the case of an FRDC module of ET2001 ADS system, the linearity (ΔV)/ΔV of the output of FRDC for ±0.1% change should not exceed more than 1%. Hence, the uncertainty contributed to the AC-DC difference is estimated to be smaller than 1 % of the measured value. Assuming that the FRDC-DC difference of the TC modules are <10 ppm, the uncertainty contributed to the AC- DC difference is estimated to be 0.1 ppm in the worst case. The estimated standard (1σ) uncertainty is 0.1/ 3 = 0.06 ppm, assuming uniform distribution in the frequency characteristic. (3) Input Linearity of TC/AMP module The TC/AMP module of the ET2001ADS system uses CS5532 Σ-D A/D converter, which has integral non-linearity better than 10 ppm. Following the same calculation as in the case of the resolution of detector (1), the contribution to the uncertainty is calculated as 0.13 x (10/5.2) = 0.25 ppm. The estimated standard (1σ) uncertainty is 0.25/ 3 = 0.15 ppm, assuming uniform distribution in the frequency characteristic. Since the measurements of the sensitivity coefficients ("index") are performed only once for each test frequency point (or once for all the test points), the uncertainty in the index-measurement must be evaluated as a type-b component Combined Uncertainty The sources of uncertainty in the FRDC-DC difference measurement, performed on a 500Ω-input JSTC04 element at 2V to 5V, are summarized in Table 4.1. The combined uncertainty is evaluated by taking root-sum-square of all the uncertainty components

50 (Numbers represents one standard deviation in µv/v) Table 4.1 Uncertainty Budget for FRDC measurement

51 5. AC-LF Measurement 5.1. Executing Measurement Measurement Procedure The measurement procedure of an ACLF-AC difference measurement is basically the same as that for standard AC-DC difference measurements, except that the standard sequence (AC/DC[+/-]/AC) is replaced by (AC[test]/AC[ref]/AC[test]). Flow-chart of the automated measurement routine is shown in Fig 5.1. Start Initialize Input Parameters Set Meas. Parameters Measurement Block #1 Measure Index "n" Measurement Block #n Measure at AC [ref] Measurement Sequence #1 Measure at AC [test] Measure at AC [ref] Change Mode Wait for Stabilization Measurement Sequence #m Calculate δ DVM Reading Measurement Block #N Measurement Sequence #M Re-adjust Sources Ending Procedure Store Data to DISK END Fig 5.1 flow-chart of ACLF-AC difference measurement program

52 After registering all the parameter or options, as described in chapter 3, the program will go to stand-by mode, ready for a fully automated ACLF-AC difference measurement. When "GO" button is pressed, the program will apply voltage to the TC module and wait for a specified period of time (normally 10 minutes) to avoid the effects from initial warm-up drift. Then the program repeats the following procedure (1) to (4) at each test points (Measurement Loops). (1) Measurement of sensitivity index The control program measures the normalized sensitivity index n of TC at each test point. The normalized sensitivity indexes are measured by changing the input voltage by dv (normally 0.1%). Influence of drift on the output voltage is removed by a measurement sequence [(V-dV) / (V+dV) / (V-dV)]. NOTE After repeating the index-measurement 10 times, the program proceeds to the next stage with warning message that the drift is too large. (2) Measurement sequence A measurement sequence, [AC[ref] / AC[test] / AC(ref)], is used to eliminate the influence of linear drift in DSS module output and EMF output voltage. The sequence is repeated for a specified number of times (normally ten times) for each measurement point, and the average value and standard deviation are calculated from the set of ten measurements. After each mode-switching, the controller waits for a specified period (normally15 seconds) to avoid the output-transients, and then takes the reading from ADC for a specified period (normally15 seconds). (3) Determination of AC-AC difference The AC-AC differences δ between the reference frequency and the test frequencies are calculated by using formula (2.11). (4) Storing measurement data After measurement sequence, measurement conditions and measurement data of each test point are stored to the hard disk of the measurement controller. The recorded items are listed in the following sub-section. After measurements for all test points are executed, a summery of measurement data are stored to the hard disk of the system controller. Then instruments are reset to the initial condition preparing for the exit from the measurement program. In the case of standard measurement conditions, one measurement loop takes about half an hour. For a set of 12 standard test points from 4 Hz to 100Hz, repeated two times (total 24 points), whole measurement takes approximately 16 hours Data Format The results from the ACLF-AC difference measurement are stored into the specified data-file using the same format as displayed in the [Data Recorded to File] window. The data-file consists of the following records. (1) Title "Data from ACLF-AC difference measurement."

53 (2) Revision Number of the control program (3) Main header common to all measurement-blocks, including: (3-1) Comment of the measurement, (3-2) ID (serial) number of DSS module, (3-3) Name, ID (serial) number, and description of TC/AMP module, (3-4) Number of repetition for one measurement blocks, (3-7) Waiting time before the ADC integration, and for initial warm-up time. (3-8) Number of ADC sampling. (4) Time constant of TC (measured). (5) Data for one set of measurements, consisting of: (5-1) Measurement block number, (5-2) Date and Time of each measurement block, (5-3) Test Voltage for each block, (5-4) Test/Reference Frequency for each block, (5-5) Result of TC Index measurement, (5-6) Result of Source adjustments, (5-7) Results of one measurement-sequence, consisting of: (5-7-1) Measurement-sequence number, (5-7-2) Time of each measurement-sequence, (5-7-3) Temperature inside TC/AMP module, (5-7-4) EMF outputs for each mode ([Ref]/[Test]/[Ref]), (5-7-5) Average standard deviation of EMF outputs in ppm, (5-7-6) ACLF-AC difference for each sequence in ppm, (5-8) Average ACLF-AC difference for each measurement-block (5-9) Standard deviation of ACLF-AC difference in ppm. (6) Summary of the measurement. (7) Error/Warning message during the measurement Evaluation of Uncertainty In this section, the uncertainty is estimated for the Low-Frequency AC-AC difference (ACLF) measurements. The sources of uncertainty are divided into two categories, namely, Type-A and Type-B. The type-a uncertainties can be evaluated from actual measurement as the standard deviation of the data, while the type-b uncertainties have to be estimated using different methods depending on the nature of the sources of uncertainty Type-A Uncertainties The type-a uncertainty in the evaluation of low-frequency characteristics of a "test"-tc module (UUT) is composed of the following four components: (1) Stability of DSS module output

54 In the case of the measurement with the ET2001 ADS system, typical thermal drift and short-term stability (0.1Hz to 10Hz) of the output of the DSS module is specified to be <10 ppm/deg and <2 ppm, respectively. Though linear drift in the output is compensated by the standard measurement sequence (AC[ref], AC[test], AC[ref]), and most of the non-linear fluctuation averages out for the normal integration period of 10 s, it can still contribute to the type-a uncertainty. Since the change in the output of the DSS module causes the change in the EMF output of the TCs, the effect to the type-a uncertainty will be included in the effect from the stability of TC module output discussed in the following paragraph. On the other hand, typical short-term stability (0.03Hz to 3Hz) of the output of the DSS module is specified to be <10 ppm p-p. The equivalent low-frequency noise from the DSS circuit is calculated as (10 / 5) / 3 = 1.2 ppm/ Hz. For normal measurement sequence (20 s for AC[MES] and 20 s for AC[REF]), contribution to the type-a uncertainty is estimated to be 1.2 x ( 2 / 20) = 0.38 ppm for each sequence, or 0.38 / 20 = 0.09 ppm for the average of standard 20 (10x2) measurement sequence. (2) Stability of TC module output As in the case of the FRDC-DC difference, the ET2001 control software suspends measurement until proper drift-condition (<10 ppm/min) is established. In this case, second-order drift during one set of measurement (1 min) should not exceed one-half of the linear drift. Hence, fluctuation in the output of TC module (including the fluctuation in the output of DSS module) should not affect the measurement results by more than 5 x (1/2) =2.5 ppm for each sequence. The contribution to ACLF-AC difference for the average of standard 20 (10x2) measurement sequence is estimated to be 2.5 / (2* 20) = 0.28 ppm. (Please refer to section 4.3.2) (3) Thermal noise in TC output As in the case of the FRDC-DC difference, the effect of thermal noise of the 400 ohm thermocouple e n is estimated to be 0.43 ppm/ Hz. For normal measurement sequence (20 s for AC[MES] and 20 s for AC[REF]), contribution to the type-a uncertainty is estimated to be 0.43 x ( 2 / 20) = 0.14 ppm for each sequence, or 0.14 / 20 = 0.03 ppm for the average of standard 20 (10x2) measurement sequence.(please refer to section 4.3.2) (4) Resolution of Detector As discussed in section 4.3.2, the effect of The DAC circuit, combined with back-up circuit, has typical resolution of better than 100 nv/ Hz. This resolution amounts to 25 ppm/ Hz in the voltage resolution with respect to the total output EMF of 4 mv, or 12.5 ppm/ Hz in the resolution for the ACLF measurement considering the square characteristic of the EMF output. For normal measurement sequence(20 s for AC[MES] and 20 s for AC[REF]), contribution to the type-a uncertainty is estimated to be 12.5 x ( 2 / 20) = 3.9 ppm for each sequence, or 3.9 / 20 = 0.87 ppm for the average of standard 20 (10x2) measurement sequence. (5) Effect of thermal ripple When test frequency is below 100 Hz, and the integration time is not the exact multiple of the inverse of the test frequency, thermal ripple in the output of the thermal converter may not be

55 averaged out and can contribute to the fluctuation in the ACLF measurement. In the case of standard TC elements for ET2001 ADS system, JSTC04 and JSTC05, which have relatively large thermal time constant of 2.8s and 6s respectively, no increase in the standard deviation is observed. The total type-a uncertainty for each measurement sequence is estimated to be 1.28 ppm, taking the RSS of the contribution from each source of uncertainty. Since the type-a uncertainty are strongly dependent on the measurement conditions, such as the stability of the ambient temperature or possible interference from external noise, actual type-a uncertainty must be calculated from the standard deviation (spread) of the 10 measurement sequence. The calculated standard deviation must be confirmed to be within the estimated value for the worst situation Type-B Uncertainties Sources of type-b uncertainty in the evaluation of low-frequency characteristic of a "test"-tc module (UUT) using a DSS module as the reference standard is as follows: (1) DC offset in AC-output mode The dc-offset voltage in the ac-mode causes a first-order thermoelectric effect in the EMF output. The DSS module has the capability of adjusting the DC offset with resolution of 0.02%. In the case of a TC elements that have reversal error of <100 ppm, the dc-offset of 0.02% may cause a systematic error of 0.02 ppm in the AC-DC difference measurement. (2) Effect of Dead-time In the case of ET2001 system, the input voltages are switched by high-speed analog switches, and the off-time is smaller than 1 _s. The effect of the off-time to the AC-DC difference measurement is expected to be smaller than 0.01 ppm, and will not be included in the uncertainty budgets. (3) High-frequency spurious noise The synthesized sinusoidal waveform contains high-frequency spurious noise, mainly from quantization noise of 10-bit D/A converter and clock feed-through at the sampling frequency. The sampling frequency is 512 or 1024 times the basic (test) frequency and <1 MHz in the case of ACLF-LF difference measurement. Since the frequency characteristic at 1Mz is smaller than 100 ppm and total power of the spurious noise is specified to be smaller than 1 ppm (-60dB), contribution to uncertainty in the ACLF-AC difference measurement is estimated to be smaller than 0.01 ppm in the worst case Uncertainties in Sensitivity Coefficient The sources of uncertainty in the sensitivity coefficient measurement and its contribution to the AC- LF measurement results are evaluated in the following paragraph. Since the index measurement is performed at DC output mode, the procedure for the evaluation of uncertainty is almost exactly the same as in the case of the FRDC-DC difference measurement

56 (1) Resolution of Detector Contribution of the resolution of the EMF voltages (ΔE) to the measurement uncertainty via the "index" measurements is the same as in the case of the FRDC measurement and is given by the following two equations. ( ) n n = ΔE E DC ΔE E DC. (5.1) [ ( δ ACLF )] 2 E( f ) E( f ) 0 ne( f 0 ) 2 n n 2 (5.2) The DAC circuit of the TC module has typical resolution of 25 ppm/ Hz in the voltage resolution with respect to the smallest output EMF of 4 mv. For normal measurement sequence (15 s for -0.1%, 15 s for +0.1%, and 15 s for -0.1%), contribution to the type-a uncertainty is estimated to be 25 x ( (1/15 +1/30)) = 7.9 ppm. Combining the values ΔE/E DC for ±0.1% change in the input, the uncertainties in the index measurement ( n) n are estimated to be < This estimation can be confirmed by repeating the index measurement more than 10 times and calculating the standard deviation. In the AC-LF measurement, the voltages at test frequencies are adjusted with respect to the reference frequency within 20 ppm. Hence the uncertainty contributed to the AC-LF measurement is estimated to be smaller than 0.04 ppm. The estimated standard (1σ) uncertainty is 0.04/ 3 = 0.02 ppm, assuming uniform distribution in the adjustment. (2) Linearity in Output of DSS module Non-linearity in the voltage source can also be a source of error in the index-measurement. Using (5.2), the uncertainty contributed to the AC-DC difference is evaluated to be: ( ΔV ) ( ) δ ACLF δ ACLF V DC ΔV. (5.3) V DC In the case of a DSS module of ET2001 ADS system, the linearity (ΔV)/ΔV of the output of DSS for ±0.1% change should not exceed more than 1%. Hence, the uncertainty contributed to the AC- DC difference is estimated to be smaller than 1 % of the measured value. Assuming that the frequency characteristic of the TC module is better than 10 ppm from 10 Hz to 1kHz, the uncertainty contributed to the AC-DC difference is estimated to be <0.1 ppm. The estimated standard (1σ) uncertainty is 0.1/ 3 = 0.06 ppm, assuming uniform distribution in the frequency characteristic. (3) Input Linearity of TC/AMP module The TC/AMP module of the ET2001ADS system uses CS5532 Σ-D A/D converter, which has integral non-linearity better than 10 ppm. Following the same calculation as in the case of the

57 resolution of detector (1), the contribution to the uncertainty is calculated as 0.04 x (10/7.9) = 0.05 ppm. The estimated standard (1σ) uncertainty is 0.05/ 3 = 0.03 ppm, assuming uniform distribution in the frequency characteristic. Since the measurements of the sensitivity coefficients ("index") are performed only once for each test frequency point (or once for all the test points), the uncertainty in the index-measurement must be evaluated as type-b Combined Uncertainty The sources of uncertainty in the ACLF-AC difference measurement are summarized in Table 5.1. The combined uncertainty is evaluated by taking root-sum-square of all the uncertainty components. (Numbers represents one standard deviation in µv/v) Table 5.1 Uncertainty Budget for AC-LF measurement

58 6. AC-DC Difference Measurement 6.1. Executing Measurement Measurement Procedure The procedure of an AC-DC or an AC-AC difference measurement employs the standard sequence (AC/DC+/DC-/AC) or (AC/ AC[REF]/AC). Flow-chart of an automated measurement routine is shown in Fig 6.1. Start Initialize Input Parameters Measurement Block #1 Set Meas. Parameters Measure Index "n" Adjust Sources (DC- by DC+) Measurement Block #n Measurement Block #N Adjust Sources (AC by DC) Measurement Sequence #1 Measurement Sequence #m Measurement Sequence #M AC Mode DC [+/-] Mode DC [-/+] Mode AC Mode Calculate δ Re-adjust Sources Change Mode Wait for Stabilization DVM Reading Ending Procedure Store Data to DISK END Fig. 6.1 flow-chart of AC-DC difference measurement program

59 After registering all the parameter or options, as described in chapter 3, the program will go to stand-by mode, ready for a fully automated AC-DC (or AC-AC) difference measurement. When "GO" button is pressed, the program will apply voltage to the TC module and waits for a specified period of time (normally 10 minutes) to avoid the effects from initial warm-up drift. Then the program repeats the following procedure (1) to (6) at each test points (Measurement Loops). (1) Measurement of sensitivity indices The control program measures the normalized sensitivity indices of two TVCs n X and n S at each test point. The normalized sensitivity indices are obtained by changing the input voltage by dv (normally 0.1%). Influence of drift of the output voltage is removed by a measurement sequence [(V dv) / (V+dV) / (V dv)]. NOTE After repeating the index-measurement 10 times, the program proceeds to the next stage with warning message that the drift is too large. (2) Adjustment of DC voltages The negative DC output voltage (DC ) is adjusted to within 0.01% with respect to the positive DC input modes (DC+). NOTE ---- After repeating the adjustment five times, the program proceeds to the next stage with warning message that the adjustment is not sufficient. (3) Adjustment of AC voltage Before adjusting the level of AC output voltage from DSS module, DC offset voltage in the AC output waveform is adjusted within 0.01% (target 20 ppm). Then the AC output voltages at test frequency, is adjusted to within 0.01% (target 20 ppm) of the average value of two DC input modes (DC+, DC ). NOTE ---- After repeating the adjustment five times, the program proceeds to the next stage with warning message that the adjustment is not sufficient. (4) Measurement sequence A four-mode measurement sequence is used to eliminate the influence of linear drift in DSS module output and EMF output voltage. Two slightly different sequence, [AC/ DC+ / DC- / AC] and [AC/ DC- / DC+ / AC] is executed in turn to check possible hysteresis-effect in DC modes. For an AC-AC difference measurement, the sequence is replaced by (AC/ AC[REF]/AC). The sequence is repeated for specified number (normally ten times) for each measurement point, and the average value and standard deviation are calculated from the set of ten measurements. After each modeswitching, the controller waits for a specified period (normally 15 seconds) to avoid the outputtransients, and then take the reading from ADCs for a specified period (normally 15 seconds). (5) Determination of AC-DC difference

60 The difference in the AC-DC differences (δ x δ s ) between the two TVCs, TVC-X and TVC-S, are calculated by using formula (2.14). (6) Storing measurement data After measurement sequence, measurement conditions and measurement data of each test point are stored to hard disk of measurement controller. The recorded items are listed in the following subsection. After measurements for all test points are executed, summery of measurement data are stored to the hard disk of the system controller. Then instruments are reset to initial condition preparing for the exit from the measurement program. In the case of standard measurement condition, one measurement loop takes about one hour. For a set of 17 standard test points from 10 Hz to 1MHz repeated twice (total 34 points), whole measurement takes approximately 20 hours Data Format The results from the AC-DC (AC-AC) difference measurement are stored into the specified datafile using the same format as displayed in the [Data Recorded to File] window. The data-file consists of the following records. (1) Title "Data from AC-DC (AC-AC) difference measurement." (2) Revision Number of the control program (3) Main header common to all measurement-blocks, including: (3-1) Comment of the measurement, (3-2) ID (serial) number of DSS module, (3-3) Name, ID (serial) number, and description of TC/AMP modules, (3-4) Number of repetition for one measurement blocks, (3-7) Waiting time before the ADC integration, and for initial warm-up time. (3-8) Number of ADC sampling. (4) Time constant of TC (measured). (5) Data for one set of measurements, consisting of: (5-1) Measurement block number, (5-2) Date and Time of each measurement block, (5-3) Test Voltage for each block, (5-4) Test (& Reference) frequency for each block. (5-5) Results of TC Index measurement, (5-6) Results of Source adjustments, (5-7) Results of one measurement-sequence, consisting of: (5-7-1) Measurement-sequence number, (5-7-2) Time of each measurement-sequence, (5-7-3) Temperature inside TC/AMP modules, (5-7-4) EMF outputs for each mode, (5-7-5) Average standard deviation of EMF outputs in ppm,

61 (5-7-6) AC-DC (AC-AC) difference for each sequence in ppm, (5-8) Average AC-DC (AC-AC) difference for each measurement-block (5-9) Standard deviation of AC-DC (AC-AC) difference in ppm. (6) Summary of the measurement. (7) Error/Warning message during the measurement Evaluation of Uncertainty In this section, the uncertainty is estimated for the AC-DC difference measurement. The sources of uncertainty are divided into two categories, namely, Type-A and Type-B. The type-a uncertainties can be evaluated from actual measurement as the standard deviation of the data, while the type-b uncertainties have to be estimated using different methods depending on the nature of the sources of uncertainty Type-A Uncertainties The Type-A uncertainty in the calibration of a "test"-tc module (UUT) against the reference TC module (REF) is composed of the following four components: (1) Stability of the DSS module output In the case of the measurement with the ET2001 ADS system, typical thermal drift and short-term stability (0.1Hz to 10Hz) of the output of the DSS module is specified to be <10 ppm/deg and <2 ppm, respectively. As in the case of an FRDC or an AC-LF measurement, linear drift in the output is compensated by the standard measurement sequence [AC(mes), AC(ref), AC(ref), AC(mes)], and most of the non-linear fluctuation averages out for the normal integration period of 10 s. Furthermore, in the case of an AC-DC difference measurement which includes two TC elements of the same type, fluctuations in output EMF voltages from the two TC elements tend to cancel each other. Since the change in the output of the DSS module causes the change in the EMF output of the TCs, the effect to the type-a uncertainty will be included in the effect from the stability of TC module output discussed in the following paragraph. On the other hand, typical short-term stability (0.03Hz to 3Hz) of the output of the DSS module is specified to be <10 ppm p-p. The equivalent low-frequency noise from the DSS circuit is calculated as (10 / 5) / 3 = 1.2 ppm/ Hz. For normal measurement sequence (20 s for AC and 20 s for DC), contribution to the type-a uncertainty is estimated to be 1.2 x ( 2 / 20) = 0.38 ppm for each sequence, or 0.38 / 20 = 0.09 ppm for the average of standard 20 (10x2) measurement sequence. (2) Stability of TC module output As in the case of the FRDC-DC difference or ACLF-AC difference measurement, the ET2001 control software suspends measurement until proper drift-condition (<10 ppm/min) is established. In this case, second-order drift during one set of measurement (1 min) should not exceed one-half of the linear drift. Hence, fluctuation in the output of TC module (including the fluctuation in the output of DSS module) should not affect the measurement results by more than 5 x (1/2) =2.5 ppm for each sequence. The contribution to AC-DC or AC-AC difference for the average of standard

62 (10x2) measurement sequence is estimated to be 2.5 / (2* 20) = 0.28 ppm. (Please refer to section 4.3.2) (3) Thermal noise in TC output As in the case of the FRDC measurement or ACLF measurement, the effect of thermal noise of the 400 ohm thermocouple e n is estimated to be 0.43 ppm/ Hz. For normal measurement sequence (20 s for AC[MES] and 20 s for AC[REF]), contribution to the type-a uncertainty is estimated to be 0.43 x ( 2 / 20) = 0.14 ppm for each sequence, or 0.14 / 20 = 0.03 ppm for the average of standard 20 (10x2) measurement sequence.(please refer to section 4.3.2) (4) Resolution of Detector As discussed in section 4.3.2, the effect of The DAC circuit, combined with back-up circuit, has a typical resolution of better than 100 nv/ Hz. This resolution amounts to 25 ppm/ Hz in the voltage resolution with respect to the total output EMF of 4 mv, or 12.5 ppm/ Hz in the resolution for the ACLF measurement considering the square characteristic of the EMF output. For a normal measurement sequence (20 s for AC and 20 s for DC), contribution to the type-a uncertainty is estimated to be 12.5 x ( 2 / 20) = 3.9 ppm for each sequence, or 3.9 / 20 = 0.87 ppm for the average of standard 20 (10x2) measurement sequence. (5) Effect of thermal ripple When test frequency is below 100 Hz, and the integration time is not the exact multiple of the inverse of the test frequency, thermal ripple in the output of the thermal converter may not be averaged out and can contribute to the fluctuation in the AC-DC difference measurement. In the case of standard TC elements for ET2001 ADS system, JSTC04 and JSTC05, which have relatively large thermal time constants of 2.8s and 6s respectively, no increase in the standard deviation is observed. The total type-a uncertainty for each measurement sequence is estimated to be 7.x ppm, taking the RSS of the contribution from each source of uncertainty. Since the type-a uncertainty are strongly dependent on the measurement conditions, such as the stability of the ambient temperature or possible interference from external noise, actual type-a uncertainty must be calculated from the standard deviation (spread) of the 10 measurement sequence. The calculated standard deviation must be confirmed to be within the estimated value for the worst situation Type-B Uncertainties Sources of type-b uncertainty in the calibration of a "test"-tc module (UUT) against the reference TC module (REF) is as follows: (1) DC offset in AC-output mode The dc-offset voltage in the ac-mode causes a first-order thermoelectric effect in the EMF output. The DSS module has the capability of adjusting the DC offset with resolution of 0.02%. In the case

63 of TC elements that have reversal error of <100 ppm, the dc-offset of 0.02% may cause a systematic error of 0.02 ppm in the AC-DC difference measurement. (2) Frequency Characteristic of the circuit At higher frequencies (>100 khz), non-negligible effect may be contributed from parasitic impedance of the input cable. The effect is strongly dependent on the system configuration, and it is very important to evaluate the effect experimentally by changing the length of the input cable or by changing the earth-guard configuration. In the case of the ET2001 ADS system, this effect is not observed, or at least within the resolution (0.9 ppm) of the system. The estimated standard (1σ) uncertainty is 0.9/ 3 = 0.5 ppm, assuming uniform distribution. (3) Effect of Dead-time In the case of ET2001 system, the input voltages are switched by high-speed analog switches, and the off-time is smaller than 1 _s. The effect of the off-time to the AC-DC difference measurement is expected to be smaller than 0.01 ppm, and will not be included in the uncertainty budgets. (4) Harmonic distortion in AC output Distortion of the sinusoidal waveform produces in higher order frequency components. Combined with the frequency characteristic of the input circuit, the higher-order components may cause error in rms power in the input voltage. Contribution to the measurement uncertainty is estimated by multiplying the specified total harmonic distortion (up to 10th order) with possible frequency characteristic in the input circuit as 0.01 ppm for 100kHz, and 1.0 ppm for >100kHz to 1MHz. (5) High-frequency spurious noise The synthesized sinusoidal waveform contains high-frequency spurious noise, mainly from quantization noise of 10-bit D/A converter and clock feed-through at the sampling frequency. The sampling frequency is fixed at 32 MHz in the case of normal an AC-DC or an AC-AC difference measurement, and total power is smaller than 1 ppm (-60dB). Hence, the contribution to uncertainty in the AC-DC or an AC-AC difference measurement is estimated to be smaller than 1 ppm in the worst case Uncertainties in Sensitivity Coefficient The sources of uncertainty in the sensitivity coefficient measurement and its contribution to the AC- DC difference results are discussed in the following paragraphs. (1) Resolution of Detector The relationship between the resolution of the EMF voltages (ΔX, ΔS) and the error in the "index" measurements is given by. n X n X = ( ΔX) X DC ΔX X DC,

64 n S n S = ( ΔS) S DC ΔS S DC. (6.1) The DAC circuit of the TC module has typical resolution of 25 ppm/ Hz in the voltage resolution with respect to the smallest output EMF of 4 mv. For a normal measurement sequence (15s for -0.1%, 15 s for +0.1%, and 15 s for -0.1%), contribution to the type-a uncertainty is estimated to be 25 x ( (1/15 +1/30)) = 7.9 ppm. Combining the values ΔX/X DC ΔS/S DC for ±0.1% change in the input, the uncertainties in the index measurement ( n) n are estimated to be < This estimation can be confirmed by repeating the index measurement more than 10 times and calculating the standard deviation. The relationship of the uncertainty in the index measurement ( n) n and the uncertainty contributed to the AC-DC difference comparison ( δ X δ S ) is given by: [ ( δ X δ S )] 2 S S AC DC n S S DC 2 n S n S 2 + X X AC DC n X X DC 2 n X n X 2 (6.2) In the AC-DC difference comparison measurement, the ac voltage is adjusted to the dc voltage such that X AC should equal to X DC within 100 ppm. Hence the uncertainty contributed to the AC- DC difference comparison measurement is estimated to be smaller than 0.28 ppm. The estimated standard (1σ) uncertainty is 0.28/ 3 = 0.16 ppm, assuming uniform distribution in the adjustment. (2) Linearity in Output of DSS module Non-linearity in the voltage source can also be a source of error in the index-measurement. Using (3.3), the uncertainty contributed to the AC-DC difference is evaluated to be: ( ) ( δ X δ S ) ( δ X δ S ) ΔV V DC ΔV. (6.3) V DC In the case of a DSS module of ET2001 ADS system, the linearity (ΔV)/ΔV of the output of DSS for ±0.1% change should not exceed more than 1%. Hence, the uncertainty contributed to the AC- DC difference is estimated to be smaller than 1 % of the measured value. Assuming that the frequency characteristic of the TC modules is better than 100 ppm from 10 Hz to 1kHz, the uncertainty contributed to the AC-DC difference is estimated to be <1.0 ppm. The estimated standard (1σ) uncertainty is 1.0/ 3 = 0.6 ppm, assuming uniform distribution in the frequency characteristic. (3) Input Linearity of TC/AMP module The TC/AMP module of the ET2001ADS system uses CS5532 Σ-D A/D converter, which has integral non-linearity better than 10 ppm. Following the same calculation as in the case of the resolution of detector (1), the contribution to the uncertainty is calculated as 0.28 x (10/7.9) = 0.36 ppm. The estimated standard (1σ) uncertainty is 0.36/ 3 = 0.21 ppm, assuming uniform distribution in the frequency characteristic

65 Since the measurements of the sensitivity-coefficients ("index") are performed only once for each test frequency point (or once for all the test points), the uncertainty in the index-measurement must be evaluated as type-b Combined Uncertainty The sources of uncertainty in the AC-DC or AC-AC difference measurement are summarized in Table 6.1. The combined uncertainty is evaluated by taking the root-sum-square of all the uncertainty components. (Numbers represents one standard deviation in µv/v) Table 6.1 Uncertainty Budget for AC-DC and AC-AC difference measurement

66 7. Evaluation and Calibration of TC modules 7.1. General Scheme The ET2001 ADS may be used in two different stages in AC-DC transfer standard, i.e., (1) evaluation of a Reference TC module (REF) as a primary AC-DC transfer standard, and (2) calibration of other thermal converters ( UUT or Unit Under Test) using the Reference thermal converter. A simplified schematic diagram of traceability chain of AC-DC transfer standard is shown in Fig In the first stage, a reference TC module is calibrated using an FRDC module, a DSS module, a HF-TVC and a LF-TVC. The first stage realizes the treceability of the Reference TC module to the SI unit, as described in section 1.3. In the second stage, thermal converters under test (UUT) or conventional TVCs combined with AMP modules are calibrated by AC-DC difference comparison measurements using the Reference thermal converter. Calibration Fig Schematic diagram for the calibration of thermal converters Evaluation of a Reference Thermal Converter Method of Evaluation As described in section 1.3, the AC-DC transfer difference of the Reference TC module is evaluated by the combination of (1) frequency independent DC offset, (2) low-frequency characteristic below 100 Hz, and (3) high-frequency characteristic above 10 khz. In the following sub-sections, an evaluation of a TC module, SN , which has a 500Ω-input JSTC04 thermal converter element, is explained in detail. After the evaluation, the TC module may be used as a REF TC module to be used as a reference standard in the comparison measurements

67 Fig Evaluation of Reference TC module DC Characteristics The frequency independent part (DC offset) of the AC-DC transfer difference of the TC module was evaluated by an FRDC-DC difference measurement at reversing frequency between 0.1 Hz and 1 khz. A typical result from FRDC-DC measurement, performed on a TC module SN , is given in figure 7.3. As the TC module shows negligibly small thermoelectric effect, it is not possible to determine the thermoelectric time constant. Hence, the thermoelectric transfer difference of the TC module was determined by the average of the FRDC-DC difference between 100 Hz to 1 khz, to be ppm, ppm, and 0.01 ppm at 3V, 5V, and 7V, respectively

68 Fig Result from FRDC-DC difference measurement Low Frequency Characteristics The low-frequency characteristic (10 Hz Hz) of the TC module (SN ) was evaluated by AC-AC difference measurement using an LF-TVC (SN66014) as a reference standard. The LF- TVC was combined with an external 200 Ω resistor, and was used as a reference standard at 5V. Figure 7.4 shows the result of an AC-AC difference measurement performed on the TC module at test frequencies from 4 Hz to 600 Hz, with reference frequency at 1 khz. From the result, the lowfrequency characteristics the TC module is evaluated by a curve-fitting of the data to the second order polynomial in (1/f) to be: δ [ppm] 7.9 f [Hz ] 6.33 f [Hz ]. (7.1) LF Fig Result of AC-AC difference measurement with a LF-TVC High Frequency Characteristics The high-frequency characteristic (10 khz 1 MHz) of the Reference TC module (SN ) was calibrated using a HF-TVC (SN ) as a reference standard. As described in a

69 seaparate technical referece "TC manual", frequency characteristics of the 500-Ω input HF-TVCs are estimated to be: ( 0.0 ± 7.5) γ [ppm] [MHz]. (7.2) 500 f Figure 7.5 shows the result from the AC-AC difference comparison measurement between the TC module and the HF-TVC with reference frequency at 1 khz. From the result of the comparison measurement, the high-frequency characteristics of the Reference TC module (SN ) is evaluated by a curve-fitting of the data to the second order polynomial to be: 2 2 δ [ppm] f [MHz] 5.25 f [MHz ]. (7.3) HF Fig Result of AC-AC difference measurement with a HF-TVC Over-all Characteristic The over-all frequency characteristic of the AC-DC difference of the TC module is determined by the three measurements described in the preceding subsections, i.e., (1) DC offset by FRDC-DC difference measurement, (2) LF characteristic by AC-AC difference measurement with LF-TVC, and (3) HF characteristic by AC-AC difference measurement with HF-TVC. In total, the AC-DC difference of the Reference TC module (SN ) at 5V is characterized by the following formula:

70 δ Total [ppm] f 1 [Hz 1 ] 6.33 f 2 [Hz 2 ] f [MHz] 5.25 f 2 [MHz 2 ]. (7.4) Fig Evaluated AC-DC difference of a Reference TC module (SN ) Evaluation of Uncertainty The sources of uncertainties in the evaluation of a REF TC module are summarized in Table 7.1. The combined uncertainty is evaluated by taking root-sum-square of the uncertainty contributions from the three measurements, i.e., (1) FRDC-DC difference measurement, (2) AC-AC difference measurement with LF-TVC, and (3) AC-AC difference measurement with HF-TVC. The thermoelectric transfer difference determined by the FRDC-DC difference measurement is a DC quantity, and the uncertainty related to the measurement contributes to the AC-DC difference of the reference thermal converter at all the frequencies, i.e., 10 Hz to 1 MHz. The uncertainty related to the low-frequency characteristic of the reference thermal converter is contributed from (a) the frequency characteristic of the LF-TVC, and (b) the AC-AC difference measurement with the LF- TVC. The uncertainty related to the high-frequency characteristic of the reference thermal converter is contributed from (a) the frequency characteristic of the HF-TVC, (b) effect from the Built-in TEE, and (c) the AC-AC difference measurement with the HF-TVC

71 Table 7.1. Uncertainty budget for self-calibration 7.3. Calibration of a Thermal Converter Comparison of AC-DC transfer difference After evaluating the Reference TC module (REF), the AC-DC difference comparison measurement will be performed between the Reference TC module and a TC module under test (UUT), and the TC module under test is calibrated against the Reference TC module. In most cases, intermediate (or Transfer ) standards are used for the calibration of UUT TC module to avoid repeated use of the Reference thermal converter. The TC modules under test (UUT) are usually standards sent from the client (secondary) standard laboratories to be calibrated by the host (primary) laboratories. This is the typical procedure to establish the traceability-chain between the host laboratories and the client laboratories. Calibration Fig Calibration of a TC module under test (UUT)

72 The Transfer standards may also be used for an inter-laboratory comparison between two independent laboratories with primary AC-DC transfer standards. Though the ET2001 ADS system is capable of realizing an independent AC-DC transfer standard, it is quite important to check its conformity with the other independent laboratories. The conformity can most easily be verified by transferring a TC module and compare calibration results between the laboratories. In ether case, the calibration involves two calibration institutes, i.e., the "host/primary" institute (S) and client/secondary institute (X). The client laboratory prepares a calibrated traveling standard (TCX), and sends the TCX to the host laboratory. Then the host laboratory carries out the calibration of the TCX against its primary standard, and then issues the calibration certificate. This is the conventional scheme of the inter-laboratory calibration of AC-DC transfer standard. Fig Inter-laboratory calibration (conventional scheme) The AC-DC difference of the intermediate (Transfer) TC module is determined by the combination of (1) results from the AC-DC difference comparison measurement between the Transfer TC module and the Reference TC module, and (2) AC-DC difference of the Reference TC module evaluated by the procedure described in the previous section. Figure 7.9 shows the result of calibration of a Traveling TC module (SN ) by the reference TC module (SN ). The two TC modules have the same type of the TC elements (type JSTC04E), and hence shows similar frequency dependence in the AC-DC transfer difference

73 Fig. 7.9.Calibration of a Traveling TC module by the reference TC module Evaluation of Uncertainty The sources of uncertainties in the evaluation of a transfer TC module (TRANS) or TC module under test (UUT) is summarized in Table 7.2. The combined uncertainty is evaluated by taking root-sum-square of the uncertainty contributions, i.e., (1) uncertainty in the evaluation of the AC- DC difference of the Reference TC module, and (2) uncertainty in the AC-DC difference comparison measurement(s) between the two (or three) TC modules. Table 7.2. Uncertainty budget for the calibration of a TC module

74 Remote Calibration In the case of the AC-DC transfer standard, the remote calibration is performed using a thermal converter as a traveling standard. The traveling standard is calibrated by a pilot (host) laboratory, and is sent to another (client) calibration institute, where it is compared with a client's reference thermal converter by AC-DC difference comparison measurements. The schematic diagram of the remote calibration is illustrated in figure ET2001 ADS system is especially suited for the remote calibration, on condition that both the host and client institute has the ET2001 ADS systems. In this case, a traveling TC module prepared by the host institute will be sent to the client institute by post. The TC module of the ET2001 system is quite tough for the vibration or temperature change, and the no special care is required for the transportation. After receiving the traveling TC module, client institute can simply connect the traveling TC module with their TC module (UUT) and start the automated measurement program. After finishing the comparison measurement, all the measurement data, including the fluctuation of ambient temperature and identification of TC modules are automatically sent to the host laboratory via . When necessary, the measurement data can be sent to the host laboratory by throughout the measurement, so as to confirm proper measurement conditions. It is also possible to remote-control the measurement program from the host institute using a remote-login operation of the Internet. Fig Remote or "E-trace" calibration