Assessment of minicrucible fixed points for thermocouple calibration, through an international comparison
The aim of this study is to estimate the achievable accuracy in thermocouple calibration by means of minicells fixed points. The budget of errors associated with fixed point calibrations generally involves two main uncertainties: the influence of chemical impurities and spurious heat flux effects. To reduce these uncertainties INM designed minicells (about 15 g of silver) where the pure metal is enclosed in a pure graphite crucible. The length and thickness of the thermometer well were calculated to keep the difference in temperature between the hot junction and the solid-liquid interface, due to the thermal losses along thermocouples' legs, within 50 mK. Extensive experimental work was performed on silver cells to study their long term stability and the homogeneity of a batch. To measure possible differences in calibrations of type S thermocouples between minicells and large classical cells, silver minicells from the same batch were sent to four European national laboratories. The discrepancies between results spread over a 0.17 C range. These values are consistent with the budget of uncertainties presented, dealing with a 0.15 K calibration and overall 3 sigma accuracy. In the near future, it will be possible to have a complete set of minicells fixed points, providing an inexpensive facility for accurate and reliable home calibration.
Bibliographic Reference: Paper presented: 7th International Symposium on Temperature, its Measurements and Control, Toronto (CA), April 28 - May 1, 1992
Availability: Available from (1) as Paper EN 36912 ORA
Record Number: 199210838 / Last updated on: 1994-12-02
Original language: en
Available languages: en