Calibration by Comparison of Platinum Resistance Thermometers Using Slow Resistance Bridges

Abstract

Platinum resistance thermometers (PRTs) are calibrated at the highest level in fixed points, as specified in the International Temperature Scale of 1990 (ITS-90). However, in order to reduce cost and time, platinum resistance thermometers can also be calibrated by comparison. In the temperature range from −100 °C to 300 °C, it is possible to achieve uncertainties as small as 5 mK. A PRT is calibrated by comparison by comparing its resistance reading with the temperature reading of a reference thermometer, placed at the same temperature inside the temperature-controlled calibration medium. The reference thermometer is commonly also a resistance thermometer, so the intrinsic measurement problem is the simultaneous measurement of two resistances. Four methods that perform this measurement are presented in this article. A special emphasis is given to the measurement with slow bridges. Slow bridges are not able to produce a stable resistance reading within 20 s after the connection of the PRT, so they are often considered to be unsuitable for calibration by comparison of PRTs. To overcome this problem, a special method that directly measures the ratio between the reference PRT and PRT under calibration is presented and analyzed. The analysis and measurement results proved that this method and consequently the slow resistance bridges are capable of performing calibration by comparison of PRTs at least at the same level as the conventional methods.