Vol.3 No.1 2010

44/110

Research paper : Improving the reliability of temperature measurements up to 1550 ℃ (M. Arai et al.)−41−Synthesiology - English edition Vol.3 No.1 (2010) Hideki OguraJoined the National Research Laboratory of Metrology, Agency of Industrial Science and Technology in 2000. Assigned to the Thermometry Section, Temperature and Humidity Division, National Metrology Institute of Japan, AIST in 2001. Worked at LNE-Cnam, the national metrology institute of France, for one year and four months from 2007, and engaged in research for high-temperature thermocouple and eutectic point. Has worked on the R&D of thermocouples and fixed points. Currently, as the vice-chairman of the working group of the Temperature Measurement Subcommittee, under the 36th Committee on Industrial Instrumentation of the Japan Society for the Promotion of Science, investigates the thermocouple calibration technique and the verification of reliability, in cooperation with the calibration laboratories. In this paper, was in charge of the research of thermocouple stability and the development of eutectic point. Masaya IzuchiJoined the National Research Laboratory of Metrology, Agency of Industrial Science and Technology in 1980. Assigned to the Thermometry Section, Temperature and Humidity Division, National Metrology Institute of Japan, AIST in 2001. Worked on the R&D of temperature standard for thermocouples and calibration service. Helps with the assessment and management of JCSS as the technical advisor and the subcommittee member of the technical committee to ensure the reliability of traceability. In this paper, was in charge of the uncertainty evaluation and the establishment of quality management system for thermocouple calibration.Discussions with Reviewers1 Motivation of the researchComment and question (Akira Ono, AIST)This paper describes excellent Type 2 Basic Research and Product Realization Research where the traceability system for the thermocouples in Japan was designed from a bird’s-eye perspective, various elemental techniques were integrated while developing a new elemental techniques, and the traceability system acceptable by society was constructed, all while working with the international movement. Since the many thermocouples are commonly used the social and industrial effect of the increased reliability of the temperature measurement is immense.One issue, the engineers mainly of the iron and steel manufacturing industry have often indicated that there are some serious problems in the reliability of thermocouples at high temperatures. On the other hand, thermocouples have been generally used as thermometers for a very long time, and some have said that there might not be no more room for advancement, and the research in Japan has not been active. With this background, I am quite surprised the authors sought a new research subject in the thermocouples, and were able to construct a traceability system for thermocouples with exceedingly high reliability. What was your motivation for starting this research, and what do you think was the factor of success? Can you please reply based on your experiences as researchers who were directly involved.Answer (Masaru Arai, Hideki Ogura, and Masaya Izuchi)The greatest motivation was to set our minds on tackling the inhomogeneity of thermocouples. As mentioned in the paper, the thermocouple is affected by temperature gradient due to inhomogeneity. Therefore, in other national metrology institutions, there is a limitation tacked on to the thermocouple calibration that the value is applicable only if it is used under the same condition that the thermocouple was tested. But that means the user of the calibrated thermocouple could not use the values as they are. When we started the research, we faced the issue of how to handle the uncertainty evaluation considering that the inhomogeneity caused bias rather than variation, and that this increased with longer exposure to high temperature. We attempted to solve the problem by taking the following procedures: (1) reduce the inhomogeneity of the thermocouple itself, (2) evaluate the inhomogeneity appropriately, and (3) check the adequacy of the evaluation method by having the users submit the temperature distribution data of the calibration device they possess.The reason we succeeded is, using perhaps an overused phrase, we simply “never gave up and went all the way.” Following the inhomogeneity increase over time is a work that requires endless patience, but one of the factors of success was we developed a highly accurate calibration device and established an efficient and precise thermocouple evaluation method using this device. The precise evaluation to the fine level as we did this time was never done before because it took so much time and effort. Yet now, the evaluations can be done much more efficiently, and we can conduct experiments in many different conditions. As a result, we were able to build a traceability system for thermocouples with exceeding reliability.2 Factors for the stabilization of Pt/Pd thermocouplesComment and question (Jun Hama, Evaluation Division, AIST)Authors clarified heat treatment conditions to reduce the drift and inhomogeneity, which cause the uncertainty sources of calibration of the Pt/Pd thermocouples, and developed uncertainty evaluation method to the establishment of the calibration method. These results are helpful to the development of the transfer standards with high precision and to supply them to industry. To understand these results more clearly and to consider the possibility of further increasing their reliability, please teach why the drift and inhomogeneity stabilize in the case of the Pt/Pd thermocouples, although the authors described the guidelines for heat treatment conditions to reduce the uncertainties due to these factors. In the type R thermocouples, why don’t the drift and inhomogeneity decrease at similar heat treatment temperature?Answer (Masaru Arai, Hideki Ogura, and Masaya Izuchi)In Fig. 9(c), only the tendency for increased Seebeck coefficient is presented schematically, but in case of the Pt/Pd thermocouple, there are temperature ranges where the Seebeck coefficient may become large or small by exposure to the temperature range from room temperature to 1300 °C. Therefore, by conducting preliminary heat treatment for sufficiently long time at an appropriate temperature, the change in the integral value of the electric field generated in the wire along the thermocouple can be kept very low. Moreover, in the Pt/Pd thermocouple, the change in the Seebeck coefficient due to exposure tends to become saturated over time, and the emf ultimately stabilizes. On the other hand, in the type R thermocouple, the composition of the Pt-Rh alloy continues to change at around 1000 °C. Therefore, the Seebeck coefficient continues to decrease without saturation, and as a result, the drift does not become saturated and the emf continues to decrease.3 Microscopic factors for the instability of the Pt/Pd thermocouple

※このページを正しく表示するにはFlashPlayer9以上が必要です