Preamplifier limitations on the accuracy of Johnson noise thermometers

White, D. R.; Zimmermann, Ernesto

doi:10.1088/0026-1394/37/1/3

Cited by 38 publications

(64 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The junction field effect transistor (JFET) pre-amplifier used for this work is capacitively coupled at the input, which necessitates the use of large value "gate resistors" R g = 20 MΩ to allow a path to ground for the ≈1 pA of gate leakage current of each JFET [13]. These gate resistors create a frequency-independent attenuation in the actual voltage being amplified by each JFET depending on the value of the source resistance.…”

Section: Gate Resistorsmentioning

confidence: 99%

Johnson Noise Thermometry near the Zinc Freezing Point Using Resistance-Based Scaling

et al. 2007

View full text Add to dashboard Cite

Johnson noise thermometry (JNT) is a primary method of measuring temperature which can be applied over wide ranges. The National Institute of Standards and Technology (NIST) is currently using JNT to determine the deviations of the International Temperature Scale of 1990 (ITS-90) from the thermodynamic temperature in the range of 505-933 K, overlapping the ranges of both acoustic gas-based and radiation-based thermometry. Advances in digital electronics have now made viable the computationally intensive and data-volume-intensive processing required for JNT using noise-voltage correlation in the frequency domain. The spectral noise power, and consequently the thermodynamic temperature T , of a high-temperature JNT probe is determined relative to a known reference spectrum using a switched-input digital noise-voltage correlator and simple resistance-scaling relationships. Comparison of the JNT results with standard platinum resistance thermometers calibrated on the ITS-90 gives the deviation of the thermodynamic temperature from the temperature on the ITS-90, T − T 90 .Statistical uncertainties under 50 µK·K −1 are achievable in less than 1 day of integration by fitting the effects of transmission-line time constants over bandwidths of 450 kHz. The methods and results in a 3 K interval near the zinc freezing point (T 90−ZnFP ≡ 692.677 K) are described. Preliminary results show agreement between the JNT-derived temperatures and the ITS-90.

show abstract

Section: Gate Resistorsmentioning

confidence: 99%

Johnson Noise Thermometry near the Zinc Freezing Point Using Resistance-Based Scaling

et al. 2007

View full text Add to dashboard Cite

show abstract

“…Ratio spectra are predominately quadratic below 1 MHz, and the τ Q2 and τ R2 constants are dominated by RC and LC couplings. There are also small 4th-degree frequency terms, where the τ Q4 and τ R4 constants are dominated by LC couplings [10]. The corresponding source impedances, however, do not couple in identical ways in the two networks.…”

Section: Introductionmentioning

confidence: 99%

Progress in Noise Thermometry at 505 K and 693 K Using Quantized Voltage Noise Ratio Spectra

et al. 2010

View full text Add to dashboard Cite

Technical advances and new results in noise thermometry at temperatures near the tin freezing point and the zinc freezing point using a quantized voltage noise source (QVNS) are reported. The temperatures are derived by comparing the power spectral density of QVNS synthesized noise with that of Johnson noise from a known resistance at both 505 K and 693 K. Reference noise is digitally synthesized so that the average power spectra of the QVNS match those of the thermal noise, resulting in ratios of power spectra close to unity in the low-frequency limit. Three-parameter models are used to account for differences in impedance-related time constants in the spectra. Direct comparison of noise temperatures to the International Temperature Scale of 1990 (ITS-90) is achieved in a comparison furnace with standard platinum resis-W. L. Tew (B) Process Measurements Division (836), National tance thermometers. The observed noise temperatures determined by operating the noise thermometer in both absolute and relative modes, and related statistics together with estimated uncertainties are reported. The relative noise thermometry results are combined with results from other thermodynamic determinations at temperatures near the tin freezing point to calculate a value of T − T 90 = +4(18) mK for temperatures near the zinc freezing point. These latest results achieve a lower uncertainty than that of our earlier efforts. The present value of T − T 90 is compared to other published determinations from noise thermometry and other methods.

show abstract

“…For these amplifiers, with typical device parameters, a sensing resistor of 1 k and a cable capacitance of 80 pF, the estimated (by electrical modeling, see [6,7]) relative temperature measurement error is 8 × 10 −7 over the frequency range from 3 kHz to 20 kHz. In contrast to other designs typically employed in JNT, feedback has been used to achieve a gain stability better than 10 −6 for the whole measurement time, thus reducing the need of frequent gain calibrations.…”

Section: Description Of the New Jntmentioning

confidence: 99%

Short Communication: Improvements to INRIM Johnson Noise Thermometer

et al. 2010

View full text Add to dashboard Cite

This paper presents a progress report for the Johnson noise thermometry experiment which is under development at the Istituto Nazionale di Ricerca Metrologica. In order to aim at an uncertainty level better than 10 −5 and to reduce the measurement time, a new setup has been developed. In particular, several modifications have been applied to the experiment described by Callegaro et al. (Metrologia 46: 409, 2009) to improve the traceability to voltage, resistance and frequency standards, and new amplifiers have been designed in order to expand the working frequency range up to 20 kHz, with a sensing resistor of 1 k , while maintaining amplifier induced systematic errors to an acceptable level.

show abstract

Preamplifier limitations on the accuracy of Johnson noise thermometers

Cited by 38 publications

References 17 publications

Johnson Noise Thermometry near the Zinc Freezing Point Using Resistance-Based Scaling

Johnson Noise Thermometry near the Zinc Freezing Point Using Resistance-Based Scaling

Progress in Noise Thermometry at 505 K and 693 K Using Quantized Voltage Noise Ratio Spectra

Short Communication: Improvements to INRIM Johnson Noise Thermometer

Contact Info

Product

Resources

About