Abstract
A dual-mode auto-calibrating resistance thermometer (DART) is presented. The novel DART concept combines in one instrument the fast and accurate resistance thermometry with the primary method of Johnson noise thermometry. Unlike previous approaches, the new thermometer measures the spectral density of the thermal noise in the sensing resistor directly in a sequential measurement procedure without using correlation techniques. A sophisticated data analysis corrects the thermometer output for both the parasitic effects of the sensor wiring and the amplifier current noise. The instrument features a highly linear low-noise DC coupled amplifier with negative feedback as well as an accurate voltage reference and reference resistor to improve the gain stability over time and ambient temperature. Therefore, the system needs only infrequent calibrations with electrical quantum standards and can be operated over long intervals and a wide temperature range without recalibration. A first prototype is designed for the industrially relevant temperature range of the IEC 60751 (-200 °C to +850 °C); a later extension of the measurement range is being considered. A proof-of-principle measurement with a calibrated Pt100 sensor at room temperature yielded an uncertainty of about 100 µK/K. The final device is expected to reach uncertainties of below 10 µK/K, suitable for accurate measurements of the difference between thermodynamic temperatures and temperatures traceable to the International Temperature Scale of 1990 (ITS-90).
Highlights
Resistance thermometers are widely used in industry and research
The dual-mode auto-calibrating resistance thermometer (DART) represents a novel approach to thermodynamic temperature measurement with low uncertainty over a wide temperature range
Unlike previous highaccuracy thermometers based on noise thermometry, the DART does not require a reference noise source during operation
Summary
Resistance thermometers are widely used in industry and research. They are easy to operate and allow low measurement uncertainties at short measurement times. The temperature is deduced from the temperature dependence of the electrical resistance of the sensor. The noise voltage to be measured is very small, and with a few exceptions (e.g., Ref. 6), the rms value integrated over the signal bandwidth amounts to typically 1 μV at room temperature. This is comparable to the input noise of the lowest noise amplifiers based on junction field-effect transistors (JFETs). The noise thermometer described in Ref. 6 involves the permanent measurement of the sensor resistance with low-frequency tones, which would enable the operation of this device as a resistance thermometer as well.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
