Intercomparison of Pt/Pd thermocouples calibrated at the freezing points of Cu (or Au) and Ag
Project Description
The project is intended to assess the uncertainties associated with the calibration of thermocouples and to establish the equivalence of calibrations at high temperature fixed points.
Thermocouples are no longer an interpolating instrument of the international temperature scale, therefore for many years no intercomparison of thermocouples between National Metrology Institutes have been carried out. Nevertheless thermocouples are used and calibrated in large numbers, at the NMIs mostly at temperature fixed points. The fixed points used for this calibration are usually not suitable for measurements with radiation thermometers, and it is therefore difficult to estimate the uncertainty of these calibrations, which are considered to be traceable to ITS-90.
New developed Pt-Pd thermocouples allow in principle uncertainties of 100 mK (k=2) and are therefore the ideal instruments for this intercomparison.
It is proposed that a pilot intercomparison is organised with only a small number of participants Two stable Pt/Pd thermocouples will be provided by PTB. One of them will be used for the intercomparison of the Cu fixed point cells of the participants. Well defined annealing procedures must be used for the measurements. The other thermocouple should be calibrated at the freezing point of Cu and Ag using more or less standard calibration procedures.
Final Report 2005-03-22
The subject of the project was the intercomparison of Pt/Pd thermocouples calibrated at the freezing points of copper (t90 = 1084.62 °C) and silver (t90 = 961.78 °C). Furthermore the freezing temperatures of different copper cells should be compared.
PTB circulated two stable Pt/Pd thermocouples. One of them was annealed and only used for the intercomparison of the copper fixed point cells of the participants. The second thermocouple was annealed and calibrated at the freezing points of Cu and Ag. Furthermore it was used to measure immersion profiles at both fixed points.
The results of the comparison are as follows: The agreement of the emfs measured by the participants at the freezing points of copper by using both Pt/Pd thermocouples is within a temperature equivalent of about 0.15 K and also within the combined uncertainties for k = 1 of a single calibration performed at the participating laboratories. The measured emfs of the thermocouple used at the freezing point of silver show an agreement within a temperature equivalent of about 0.07 K which is within the combined uncertainty of each measurement at this fixed point. From the results it can be concluded that Pt/Pd thermocouples are suitable as transfer standards for the dissemination of temperatures and the approximation of the ITS-90 because of their stability within about 0.1 K for temperatures up to the freezing point of copper. Furthermore, it was confirmed that the freezing temperatures of the different copper fixed point cells used at most of the participating laboratories agree within a temperature equivalent of about 50 mK.
The results of the intercomparison have been presented and will be published in the proceedings of the 9th International Symposium on Temperature and Thermal Measurements in Industry and Science, Cavtat 2004.