Abstract
Thermocouple measurements suffer from errors due to several heat transfer effects. A dominant error source is conduction error due to a difference in the junction and support temperature. Correction for conduction is critical in both steady and transient temperature measurements. The effect of conduction is reduced by using a high length-to-diameter ratio of the wire, but this becomes unfeasible in high-speed flows due to structural limitations. This paper presents a new procedure to perform thermocouple measurement with unprecedented accuracy using a two-wire thermocouple probe.
Our novel numerical correction method necessitates two temperature readings from two different thermocouple diameters. The temperature values are different because the conduction error is a function of their length-diameter ratio and material properties. The two thermocouple readings combined with Conjugate Heat Transfer Simulations (CHT) are used to correct the temperature readings and deliver the actual temperature of the flow free of conduction error. The Conjugate Heat Transfer simulations are run at different temperature boundary conditions and flow Reynolds numbers. This method is generalizable and can correct conduction errors within 0.1K. The paper includes a comprehensive uncertainty quantification approach based on Monte Carlo simulations. This correction was demonstrated experimentally to measure the inlet temperature of a turbine test rig compared against a standard one-wire thermocouple probe, and the uncertainty of the method was quantified.
