Characterizing Drift Behavior in Type S Thermocouples to Predict In-use Temperature Errors

• Published in: International journal of thermophysics Vol. 41; no. 1
• Main Authors: Webster, E., Saunders, P.
• Format: Journal Article
• Language: English
• Published: New York Springer US 01.01.2020; Springer Nature B.V
• Subjects: Annealing; Classical Mechanics; Condensed Matter Physics; Drift; Geophysics; High temperature; Industrial applications; Industrial Chemistry/Chemical Engineering; Metallurgy; Physical Chemistry; Physics; Physics and Astronomy; Rhodium; Rhodium base alloys; Tempmeko 2019; TEMPMEKO 2019: Selected Papers of the 14th International Symposium on Temperature and Thermal Measurements in Industry and Science; Thermocouples; Thermodynamics; Uncertainty; Noble-metal; Rare-metal; Thermocouples; Type S; Drift; Inhomogeneity
• ISSN: 0195-928X; 1572-9567
• DOI: 10.1007/s10765-019-2579-0

The Type S thermocouple continues to be used as a primary reference for industrial applications and as a secondary reference in many national metrology institutes (NMI). The Type S thermocouple, in a well annealed state, can yield temperature measurements accurate to better than ± 0.2 °C up to approximately 1100 °C. Metallurgical changes, even in the best cared for thermocouples, quickly lead to drift in use. The main causes of drift are reversible processes occurring in the Pt/Rh alloy thermoelement. Most NMIs will periodically apply a high-temperature (1100 °C) anneal to restore the thermocouple to a state close to that of the original. But, for many second-tier laboratories and industrial users, this annealing procedure is not available or is impracticable. Therefore, these users are faced with measurements that are subject to accumulated or continuous drift errors. Fortunately, recent research has shown that the Pt-10%Rh thermoelement changes with temperature in a predictable way, enabling forecasting of drift as a function of exposure time and temperature. This study aggregates drift results for several Type S thermocouples from different manufacturers after exposure to MSL’s gradient furnace in the range 100 °C to approximately 900 °C. Numerical techniques are then used on these data to anticipate the likely in-use temperature errors for other Type S thermocouples. The method can provide improved confidence for end-users of Type S thermocouples where regular annealing is either difficult or impossible.