Calibration of reference thermocouple of a high temperature X-ray diffraction camera is made by nearly the same method as used in the author's former report. Amount of temperature deviation is calculated quantitatively as a function of temperature under various conditions of thermocouple wires and the specimen holder rod, which are their thickness, length, emissivity, thermal conductivity, relative situation in the furnace etc. A new coefficient δ is introduced in the calculation in order to treat the size effect of the thermocouple head. The so-called gap effect is also considered. At low temperatures, the temperature deviation is mainly due to thermal flow through thermocouple wires, while, at high temperatures, the gap effect becomes predominant. Maximum temperature deviation occurs, at about 450°K in vacuum and at about 650°K in air if the gap effect is neglected. Very large deviations increasing with temperature are observed, and the calibration curves are strongly affected by the relative situation of the thermocouple head and the specimen when the gap effect is large. Temperature distribution on the specimen holder rod is also analysed.
The results are compared with experimental data and discussion is made for obtaining favorable conditions to minimize the temperature deviation.