Thermocouple measurement errors and precautions

Abstract: Thermocouples are one of the simplest and most common temperature sensors. However, if not taken seriously during use, it can lead to larger measurement errors. Considering the existing issues, the main factors affecting measurement errors include thermocouple insertion depth, response time, thermal radiation, and thermal impedance, as well as uneven thermocouple wires, flow errors in armored thermocouples, selective oxidation of K-type thermocouples, K-state, operating atmosphere, insulation resistance, and degradation of thermocouples. Precautions should be taken during use. To improve measurement accuracy and extend the service life of thermocouples, it is helpful to some extent.
introduce
In existing temperature measurement systems, the most commonly used temperature sensor - thermocouple, due to its simple structure, is often mistaken for "thermocouple with two wires connected to a cabinet", but in reality, it is not. Although the structure of thermocouples is simple, there are still various problems that may arise during use. For example, incorrect installation or use can cause greater measurement errors, and even qualified thermocouple testing may be due to improper operation and failure, such as carburization in a reducing atmosphere. If not taken seriously, K-type thermocouples may also exceed tolerances due to selective oxidation.
In order to improve measurement accuracy, reduce measurement errors, and extend the service life of thermocouples, users are required to not only have instrument operation skills, but also knowledge in physics, chemistry, and materials. Based on years of practice and more detailed reference to relevant information, the measurement errors of thermocouples and their preventive measures.
2. Heterogeneity effect of thermocouple wires
(1) The thermocouple material itself is uneven
Thermocouples are tested in the measurement room, and according to the rules, the depth of insertion into the testing furnace is only 300mm. Therefore, the test results of each thermocouple can only reflect or reflect the behavior starting from the measurement of the 300mm long thermocouple wire. However, when the length of the thermocouple is long, most of the double wires are in the high-temperature zone. If the thermocouple wire is uniform, according to the uniform loop rule, the measurement results are independent of the length. However, uneven thermocouple wires, especially low-grade metal thermocouple wires, can generate local thermoelectric electromotive forces under temperature gradients, known as parasitic electromotive forces. The error caused by parasitic potential is called heterogeneity error.
In the existing validation programs for precious metals and low alloy thermocouples, the heterogeneity of thermocouples has not yet been provided, and only in the thermocouple wire standards do thermocouple wires have some uneven requirements. Calculate the thermal electromotive force using the head to tail testing method. Conventional thermocouple wire production plants generate uneven thermoelectric electromotive force according to national standards to meet product requirements.
(2) Thermocouple wires are made of different materials
For the thermocouples of the new system, even if the uneven thermal EMF can meet the requirements, repeated processing can lead to distortion of the processed bent thermocouples, which will lose uniformity. Moreover, due to the long-term use of even numbered wires at high temperatures, such as thermocouples inserted into industrial furnaces, they will degrade or even wire along the direction of the long wire. As the temperature increases, the degradation will increase. When the component is in a place where the temperature gradient decreases, It can also generate electromotive force superimposed on the total thermal EMF due to parasitic measurement errors.

3. Shunt error of armored thermocouples
(1) Diversion error
Huaju Group's carburizing furnace equipped with armored thermocouples is only allowed for one week. In order to explore the reasons, the author went to the site for inspection and found no abnormalities, so they had to take the test results from the measurement room and found them to be qualified. So what's the problem? Finally, based on the on-site characteristics of the installed thermocouples, it was found that the problem was caused by incorrect diversion of armored thermocouples.
The so-called shunt error is measured using armored thermocouples. When the temperature of the middle part of the thermocouple exceeds 800 ° C, due to the decrease in insulation resistance, the thermocouple indicates an abnormal phenomenon, which is called shunt error. According to the principle of uniform circuit, thermocouple temperature measurement is only related to the temperature at both ends of the measurement and reference ends, and is independent of the temperature distribution in the middle. However, due to the fact that the insulation of armored thermocouples is MgO powder, the temperature rises by 100 ° C, and the insulation resistance decreases by an order of magnitude. When the middle part is at high temperature, there must be a leakage current, resulting in a branching error in the output potential of the thermocouple.
(2) Diversion error conditions
Armored thermocouple inserted into the furnace, specifications and experimental conditions: diameter ф 4.8mm, 25m long, with a central heating zone length of 20m and a temperature of 1000 ℃. In this experiment, the temperature difference between the thermocouple measurement and the middle section is 200 ℃. If the temperature at the measuring end is higher than the middle part, a negative error will occur; On the contrary, a positive error is generated. If the temperature difference between the two is 200 ° C, the diversion error is about 100 ° C. This cannot be ignored, as factors such as shunt error conditions and the type and diameter of armored thermocouples [2]
The influencing factors and countermeasures of diversion error
It is necessary to understand the influencing factors of diversion errors and take appropriate measures to reduce or eliminate the impact of diversion errors. The results indicate that the shunt error of armored thermocouples is caused by high temperature.
(1) Armored thermocouple diameter
For K-type armored thermocouples with a length of 9m (MgO insulation), only the middle part of the thermocouple is heated. The experimental results show that the magnitude of the diversion error is inversely proportional to the square root of its diameter (the diameter is too small to comply with this rule), which means that the smaller the diameter, the greater the diversion error.
When the central temperature exceeds 800 ℃, ф 3.2mm armored thermocouples may produce shunt errors. However, ф 6.4mm and ф When the middle temperature of the 8mm armored thermocouple is 900 ℃, the shunt error has not been detected yet. about ф 6.4mm (hotline diameter of Φ 1.4mm) and ф 8mm (heating wire diameter is ф A 2.0mm armored thermocouple generates a diameter when the temperature in the middle part is 1100 ℃ ф 8mm armored thermocouple. The diversion error is only ф Half of 6.4mm. The value (50%) is near the square of the wire diameter of two armored thermocouples (1.42/2.02), as well as the square ratio of the electrode wire diameter, which is the resistance ratio of the wire electrode. Therefore, in order to reduce diversion errors, the diameter of the armored thermocouple should be selected.

(2) Temperature of the middle section
If the temperature in the middle part exceeds 800 ℃, there may be a diversion error, and its size will increase exponentially with the increase of temperature. Therefore, except for the measuring end, other parts should be avoided above 800 ℃ as much as possible.
1) The length and position of the heating zone in the middle section
When the temperature in the middle of the heating zone is higher than 800 ℃, the longer the heating zone is and the farther it is from the measurement end, the greater the diversion error. Therefore, the length of the heating strip should be as short as possible and should not be heated from the measuring end to reduce diversion errors.
(3) Resistance of thermocouple wires
When the diameter of the sheathed thermocouple is the same, the shunt error will increase as the resistance of the thermocouple wire increases. Therefore, it is better to use small resistance thermocouple wires. For example, compared to K-type thermocouples, S-type armored thermocouples with the same diameter will reduce the shunt error by 40%. Therefore, S-type thermocouples can be used to measure the distribution of furnace temperature field, which is costly but more accurate.
(4) Insulation resistance
At high temperatures, the resistivity of oxides decreases exponentially with increasing temperature. The shunt error mainly depends on the insulation characteristics of high-temperature components. The lower the insulation resistance, the easier the shunt error is. When the insulation resistance increases by 10 times or decreases to 1/10, the shunt error will decrease to 1/10 or 10 times. To reduce diversion errors, armored thermocouples should be used as much as possible to increase insulation thickness. If the above measures are ineffective, assembled thermocouples must be used.
5. Short range ordered structural changes (K state)
K-type thermocouples, in the temperature range of 250 ℃ -600 ℃, form short-range ordered structures due to changes in their microstructure, which can affect thermoelectric potential and error, known as the K-state [3]. This is a unique lattice change of Ni Cr alloy. When the Cr content is within the range of 5-30%, there is an ordered disorder transition in the atomic lattice. Errors caused by changes in Cr content and temperature. Heat the K-type thermocouple from 300 ℃ to 800 ℃ and remove it at a point of 50 ℃ to measure the potential. At 450 ℃, the maximum deviation reaches 4 ℃, and within the range of 350-600 ℃, it is a positive deviation. Due to the presence of K-state, the temperature rise or cooling test results of K-type thermocouples are inconsistent. Therefore, it is clearly specified in the low-cost thermocouple verification program that the test sequence is: temperature test point by point from low temperature to high temperature. In addition, at the 400 ℃ test point, not only is the heat transfer effect poor and it is difficult to achieve thermal equilibrium, but it also occurs within the maximum range of K state error. Therefore, carefully judge whether the point is qualified.
The phenomenon of short-range ordered structural changes in Ni Cr alloys not only exists in K-type but also in the positive electrode of E-type thermocouples. However, the E-type thermocouple is only 2/3 of the K-type as a variation. In short, the K state is related to temperature and time, and the deviation will greatly change when the temperature distribution or thermocouple position changes. It is difficult to accurately evaluate the magnitude of the deviation.