Computational compensation of systematic errors accompanying non-equilibrium thermocouple measurements

Application of thermocouple sets of various complexity as temperature sensors in porous materials can be accompanied with systematic errors related to the very high differences between the thermal conductivities of thermocouple forming metals and measured samples. Since the higher measurement uncertainties can then make some presumed utilizations of measured temperature fields questionable, this paper introduces an approach which can overcome this drawback. Using a computational model, it is able to separate parameters of particular constituents of the sample-thermocouple system that helps to filter out the systematic errors by means of extracting the pure material parameters. The applicability of the approach is demonstrated on a one-sided high-temperature heating experiment, in which the temperature field in a sample is monitored and subsequently used for the determination of thermal conductivity. The effect of the computational compensation procedure is found to be very significant. The up to 64 degrees C temperature differences between the raw data and corrected values can cause systematic thermal conductivity errors of up to 46%. Depending on temperature, the corrected thermal conductivity is then by up to 0.89 W m(-1) K-1 lower.