Корекція похибок дрейфу та набутої неоднорідності у термоелектричному перетворювачі з керованим профілем температурного поля

Abstract

Набута термоелектрична неоднорідність термопар та її негативний вплив на похибку вимірювання температури термоелектричними перетворювачами відомі понад століття. Сучасні дослідники вважають її основним джерелом похибки вимірювання температури, а саме явище однозначно негативним. Однак останнім часом запропоновано ефективні методи боротьби з цією похибкою. Розроблено підхід, що дає змогу використати набуту термоелектричну неоднорідність термопар для оцінювання метрологічних характеристик термоелектричних перетворювачів. Приобретенная термоэлектрическая неоднородность термопар и ее негативное влияние на погрешность измерения температуры термоэлектрическими преобразователями известны больше столетия. Современные исследователи считают ее главным источником погрешности измерения температуры, а само явление однозначно негативным. Однако в последнее время предложены эффективные методы борьбы с этой погрешностью. В статье предложен подход, позволяющий использовать приобретённую термоэлектрическую неоднородность термопар для оценки метрологических характеристик термоэлектрических преобразователей. Thermocouples are the most popular sensors of temperature used in measuring praxis for temperatures in the range 600–2500 ºC, in spite of their drawbacks. The most important among them is their error, which is often too big for many cases in industry and science. The total error of thermocouples is much greater than that of their measurement channels. The main errors proper for thermocouples are as follows: 1. Considerable initial deviation of their conversion characteristic (CC) from the nominal one. The likely deviation in CC for the most popular type of thermocouples (type K) may reach 5,5 °C at 600 °C and 8 °C at 1100 °C. 2. Considerable drift of CC during operation at high temperatures, that is a change of CC in time. It may vary in the range of 0,5 °C to 10 °C for the mentioned above type K of thermocouples during 1000 hours at 600 °C or at 1100 °C respectively. 3. Thermoelectric inhomogeneity of thermocouple legs acquired during operation at high temperatures. The error due to inhomogeneity may reach 10 °C when measuring 1100 °C during 1000 hours using type K thermocouples, or even more in some cases. The error due to acquired inhomogeneity stems from changes in thermocouple legs at high temperatures in time. These changes are caused by the effect of chemical and physical processes (such as oxidation, diffusion, recrystallization etc.) in legs. If a thermocouple is split into imaginative sections, each section operates at its own temperature. That is the reason why, during prolonged operation, CC of each section changes in time in accordance with its particular operating temperature. If the temperature field along the thermocouple legs changes, the temperatures of each section changes correspondingly. Therefore, the error of each section is not constant, so the total error of a thermocouple varies even when the temperatures of the measuring and the reference junctions remain constant. This is an appearance of the error due to thermoelectric inhomogeneity when the developed emf depends on the distribution of temperature along thermocouple legs. Many researchers consider that it is the main reason of thermocouple error. Sometimes the high thermocouple error after extended use is thought to be inevitable and impossible to correct. However, recent studies have discovered new methods for decreasing the influence of the error due to acquired inhomogeneity on measurements of temperature using thermocouples. There are three stages of understanding of the error due to inhomogeneity during the history of thermocouples. The first stage began in 1906, when the problem of the inhomogeneity in thermocouples was posted. Since then the problem has been studied by various researchers. In 1976 professor Kirenkov drawn conclusion that inhomogeneous thermocouples cannot be used in measuring praxis because of a big error which is impossible either to correct orcompensate. No methods how to deal with the error due to acquired during operation inhomogeneity were suggested.This is the second stage of the problem – direct prohibition of use of inhomogeneous thermocouples. The third stage began in 1984, when some methods of correction for the error due to inhomogeneity were suggested. The first of them was thermocouple calibration in situ, or in a temperature field similar to that of operation. Also the method for computation of the thermocouple error in an operating conditions based on the thermocouple calibration in a laboratory was suggested in the same year. However, the most effective method for compensating the error due to inhomogeneity was suggested in 2006. The idea of the method is to stabilize a temperature field along a thermocouple using additional subsystems for temperature control. Implementation of the method requires a new sensor which was suggested in 2006 and was called the thermocouple with controlled profile of temperature field. This is the third stage of the problem. At this stage some effective methods to compensate the error due to inhomogeneity were suggested. This paper suggest the next stage, that is the fourth one, which implies development of the method suggested in 2006 and the thermocouple with controlled profile of temperature field. The idea of the stage is to use thermoelectric inhomogeneity of thermocouples to assess the metrological properties of thermoelectric thermometers.