FAST AND ACCURATE COMPENSATION METHOD IN A THERMOCOUPLE MEASUREMENT, AND A RESPECTIVE DEVICE

Abstract

The present invention relates to measuring temperature (t) using a thermocouple (30), with a first measurement point (33) along a positive conductor (31), and a second measurement point (34) along a negative conductor (32). The arrangement is configured to obtain a Seebeck coefficient(S) for the material pair; and to measure a first temperature (t.sub.1) in the cold end of the thermocouple (30) at a first measurement point (33); and to measure a second temperature (t.sub.2) at a second measurement point (34). Thermal voltages (U.sub.1, U.sub.2) are measured. The arrangement further calculates a weighted average of the temperature (T.sub.ave) by dividing a temperature difference of the first (t.sub.1) and second (t.sub.2) temperatures with a ratio N=U.sub.1/U.sub.2; and calculates temperature (t) in the hot end of the thermocouple (30) based on the obtained Seebeck coefficient(S) for the material pair and the calculated weighted average of the temperature (T.sub.ave).

Claims

1. An arrangement for measuring temperature (t) using a thermocouple (30), where the thermocouple (30) comprises a cold end and a hot end, wherein the arrangement comprises a processor; and a positive conductor (31) of the thermocouple (30) and a negative conductor (32) of the thermocouple (30), where the positive (31) and negative (32) conductors are attached together in the hot end of the thermocouple (30), and where materials of the positive (31) and negative (32) conductors form a material pair; and a first measurement point (33) along the positive conductor (31), and a second measurement point (34) along the negative conductor (32); wherein the arrangement is configured to: obtain a Seebeck coefficient(S) for the material pair, either from previously obtained results in an external data source or by measuring temperature-voltage characteristics of the thermocouple (30); measure a first temperature (t.sub.1) in the cold end of the thermocouple (30), using a first temperature sensor connected to the first measurement point (33); measure a second temperature (t.sub.2) in the cold end of the thermocouple (30), using the first temperature sensor or a second temperature sensor connected to the second measurement point (34); characterized in that the arrangement is further configured to: calculate a weighted average of the temperature (T.sub.ave) by dividing a temperature difference of the first (t.sub.1) and second (t.sub.2) temperatures with a ratio (N) of generated thermal voltages (U.sub.1, U.sub.2) in the positive (31) and negative (32) conductors of the thermocouple (30), respectively, wherein $N=\frac{U_1}{U_2};$ and calculate temperature (t) in the hot end of the thermocouple (30) based on the obtained Seebeck coefficient(S) for the material pair and the calculated weighted average of the temperature (T.sub.ave).

2. The arrangement according to claim 1, characterized in that the weighted average of the temperature (T.sub.ave) is calculated by: $T_{\mathrm{ave}}=t_1-\frac{t_1-t_2}{\frac{U_1}{U_2}}.$

3. The arrangement according to claim 1, characterized in that the first and/or the second temperature sensor is/are connected directly to a connecting pin of a respective conductor (31, 32).

4. The arrangement according to claim 1, characterized in that the arrangement is further configured to: perform temperature measurements of the first (t.sub.1) and second (t.sub.2) temperatures separately, at mutually different times.

5. The arrangement according to claim 1, characterized in that the thermocouple is a K-type thermocouple, consisting of chromel, NiCr, in the positive conductor (31), and alumel, NiAl, in the negative conductor (32).

6. The arrangement according to claim 1, characterized in that the arrangement is further configured to: measure voltage (U.sub.3) as a potential difference between copper wires (35, 36) where voltage (U.sub.3) is defined as: $U_3=U_1-U_2.$

7. The arrangement according to claim 1, characterized in that the processor is further configured to: calculate the temperature of the hot end of the thermocouple (30) with equations based on the Seebeck coefficients of the used materials of the thermocouple (30).

8. The arrangement according to claim 7, characterized in that: obtaining the temperature of the hot end of the thermocouple (30) either from a data table or via a calculation formula, which comprises polynomial correction coefficients of the thermocouple (30), or by another means for transforming the thermal voltage into temperature.

9. The arrangement according to claim 1, characterized in that the Seebeck coefficient(S) for the material pair and the ratio (N) of generated thermal voltages (U.sub.1, U.sub.2) relate to each other as follows: $S=U_1-U_2;N=\frac{S_{+}}{S_{-}},$ where S is a Seebeck coefficient of the material pair, S.sub.+ is a Seebeck coefficient of the positive conductor (31) of the material pair, S.sub. is a Seebeck coefficient of the negative conductor (32) of the material pair, N is the ratio of the thermal voltages (U.sub.1/U.sub.2), U.sub.1 is the thermal voltage of the positive conductor (31) of the material pair, and U.sub.2 is the thermal voltage of the negative conductor (32) of the material pair.

10. The arrangement according to claim 1, characterized in that the processor is further configured to calculate the temperature (t) in the hot end of the thermocouple (30) according to the following steps: measuring the thermal voltage (U.sub.3) in the cold end, and also measuring the temperatures (t.sub.1) and (t.sub.2) in the measurement points (33, 34) of the cold end; calculating the weighted average of the temperature (T.sub.ave) from the cold end temperature results (t.sub.1) and (t.sub.2) which calculation is made according to the type and/or materials of the used thermocouple (30); transforming the weighted average of the temperature (T.sub.ave) to a voltage either using a conversion data table of the used thermocouple type or using a calculation formula involving correction coefficients of the applied thermocouple type; adding the measured thermal voltage to the calculated voltage value, thus obtaining a correct thermal voltage value for the hot end temperature calculation; and converting the correct thermal voltage value to a temperature value using the conversion data table of the used thermocouple type or the calculation formula supplied with the correction coefficients, thus obtaining an accurate hot end temperature value.

11. The arrangement according to claim 1, characterized in that the ratio (N) of generated thermal voltages is configured to be applied as a temperature-dependent parameter either via data table-based information or via polynomial equations.

12. A method for measuring temperature (t) using a thermocouple (30) in an arrangement, where the thermocouple (30) comprises a cold end and a hot end, the arrangement further comprising a processor, and a positive conductor (31) of the thermocouple (30) and a negative conductor (32) of the thermocouple (30), where the positive (31) and negative (32) conductors are attached together in the hot end of the thermocouple (30), and where materials of the positive (31) and negative (32) conductors form a material pair; and a first measurement point (33) along the positive conductor (31), and a second measurement point (34) along the negative conductor (32); wherein the method comprises the steps of: obtaining a Seebeck coefficient(S) for the material pair, either from previously obtained results in an external data source or by measuring temperature-voltage characteristics of the thermocouple (30); measuring a first temperature (t.sub.1) in the cold end of the thermocouple (30), using a first temperature sensor connected to the first measurement point (33); measuring a second temperature (t.sub.2) in the cold end of the thermocouple (30), using the first temperature sensor or a second temperature sensor connected to the second measurement point (34); characterized in that the method further comprises the steps of: calculating, by the processor, a weighted average of the temperature (T.sub.ave) by dividing a temperature difference of the first (t.sub.1) and second (t.sub.2) temperatures with a ratio (N) of generated thermal voltages (U.sub.1, U.sub.2) in the positive (31) and negative (32) conductors of the thermocouple (30), respectively, wherein $N=\frac{U_1}{U_2};$ and calculating, by the processor, temperature (t) in the hot end of the thermocouple (30) based on the obtained Seebeck coefficient(S) for the material pair and the calculated weighted average of the temperature (T.sub.ave).

13. The method according to claim 12, characterized in that the weighted average of the temperature (T.sub.ave) is calculated by the processor by: $T_{\mathrm{ave}}=t_1-\frac{t_1-t_2}{\frac{U_1}{U_2}}.$

14. The method according to claim 12, characterized in that the method further comprises the step of: connecting the first and/or the second temperature sensor directly to a connecting pin of a respective conductor (31, 32).

15. The method according to claim 12, characterized in that the method further comprises the step of: performing temperature measurements of the first (t.sub.1) and second (t.sub.2) temperatures separately, at mutually different times.

16. The method according to claim 12, characterized in that the thermocouple is a K-type thermocouple, consisting of chromel, NiCr, in the positive conductor (31), and alumel, NiAl, in the negative conductor (32).

17. The method according to claim 12, characterized in that the method further comprises the step of: measuring voltage (Us) as a potential difference between copper wires (35, 36) where voltage (Us) is defined as: $U_3=U_1-U_2.$

18. The method according to claim 12, characterized in that the method further comprises the step of: calculating, by the processor, the temperature of the hot end of the thermocouple (30) with equations based on the Seebeck coefficients of the used materials of the thermocouple (30).

19. The method according to claim 18, characterized in that: obtaining the temperature of the hot end of the thermocouple (30) either from a data table or via a calculation formula, which comprises polynomial correction coefficients of the thermocouple (30), or by another means for transforming the thermal voltage into temperature.

20. The method according to claim 12, characterized in that the Seebeck coefficient(S) for the material pair and the ratio (N) of generated thermal voltages (U.sub.1, U.sub.2) relate to each other as follows: $S=U_1-U_2;N=\frac{S_{+}}{S_{-}},$ where S is a Seebeck coefficient of the material pair, S.sub.+ is a Seebeck coefficient of the positive conductor (31) of the material pair, S.sub. is a Seebeck coefficient of the negative conductor (32) of the material pair, N is the ratio of the thermal voltages (U.sub.1/U.sub.2), U.sub.1 is the thermal voltage of the positive conductor (31) of the material pair, and U.sub.2 is the thermal voltage of the negative conductor (32) of the material pair.

21. The method according to claim 12, characterized in that the method further comprises calculating, by the processor, the temperature (t) in the hot end of the thermocouple (30) according to the following steps: measuring the thermal voltage (Us) in the cold end, and also measuring the temperatures (t.sub.1) and (t.sub.2) in the measurement points (33, 34) of the cold end; calculating the weighted average of the temperature (T.sub.ave) from the cold end temperature results (t.sub.1) and (t.sub.2) which calculation is made according to the type and/or materials of the used thermocouple (30); transforming the weighted average of the temperature (T.sub.ave) to a voltage either using a conversion data table of the used thermocouple type or using a calculation formula involving correction coefficients of the applied thermocouple type; adding the measured thermal voltage to the calculated voltage value, thus obtaining a correct thermal voltage value for the hot end temperature calculation; and converting the correct thermal voltage value to a temperature value using the conversion data table of the used thermocouple type or the calculation formula supplied with the correction coefficients, thus obtaining an accurate hot end temperature value.

22. The method according to claim 12, characterized in that the method further comprises the step of: applying the ratio (N) of generated thermal voltages as a temperature-dependent parameter either via data table-based information or via polynomial equations.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0091] FIG. 1 illustrates a traditional single point measurement according to the prior art,

[0092] FIG. 2 illustrates a traditional single point measurement according to the prior art with some exemplary material selections,

[0093] FIG. 3 illustrates an embodiment of the present invention, which offers shorter stabilization times and fast and accurate measurement results, and

[0094] FIG. 4 illustrates how the present invention gives notably different results compared to mere averaging of the positive and negative terminal temperature values of a thermocouple.

DETAILED DESCRIPTION

[0095] The present invention introduces a faster and more accurate arrangement for temperature measurement using a thermocouple.

[0096] A main feature of the present invention is that the temperatures of both measurement wires (i.e. connector terminals, or conductors) of a cold junction of a thermocouple measurement are measured using two digital temperature sensors. The temperature sensors are selected so that they obtain accurate temperature results. The accuracy of the temperature sensors may be ensured by performing calibration of the sensors at first, e.g. using a dry block calibrator, in an embodiment of the invention.

[0097] In an embodiment of the invention, a first temperature sensor is connected directly to a connecting pin of the first connector terminal in the cold end of the thermocouple. Respectively, a second temperature sensor is connected directly to a connecting pin of the second connector terminal in the cold end of the thermocouple in the same embodiment of the invention. In this sense, a temperature measurement of the cold end by a single temperature sensor according to the prior art is now replaced by a measurement performed with two temperature sensors, using a following connection and calculation principle.

[0098] The temperature of the first and the second connector terminals are measured separately in an embodiment of the invention. The separate nature of the temperature measurement means that the first measurement time instant (of the first measurement wire) may be different from the time instant of the second measurement (of the second measurement wire).

[0099] In an embodiment of the invention, a so-called compensation of the cold end temperature of the thermocouple is performed by calculating a weighted average value of the temperatures measured by the two temperature sensors. In an embodiment, the weighted average value calculation means that it is obtained by dividing the temperature difference of t.sub.1 and t.sub.2 in the same ratio as the ratio (meaning:factor) of the voltages generated in the positive and negative conductors of the thermocouple in its cold end.

[0100] For example, if the thermal voltage generated on the positive conductor (=first measurement wire) of the thermocouple is six times greater than the thermal voltage generated on the negative conductor (=second measurement wire), the temperature difference between t.sub.1 and t.sub.2 is divided in the same ratio in an embodiment of the invention. If, for example, t.sub.1=26 C. and t.sub.2=25.4 C., the weighted average temperature is 25.9 C. With this weighted average temperature value, a very accurate measurement result is obtained, even when the temperature of the cold end has not yet stabilized. This has a further advantage of allowing very rapid measurements which significantly shorten the required temperature measurement time.

[0101] In an equation form, if the coefficient with which the negative conductor thermovoltage U.sub.2 needs to be multiplied in order to obtain the positive conductor thermovoltage U.sub.1, is N, then the desired weighted average temperature value is:

[00011]$\begin{array}{cc}{T}_{w,\mathrm{ave}}=t_1-\frac{t_1-t_2}{N},& \left(5\right)\end{array}$

where N is:

[00012]$\begin{array}{cc}N=\frac{U_1}{U_2}& \left(6\right)\end{array}$

[0102] In other words, the equation (5) may also be written as:

[00013]$\begin{array}{cc}{T}_{w,\mathrm{ave}}=t_1-\frac{t_1-t_2}{\frac{U_1}{U_2}}=t_1+\frac{U_2}{U_1}\left(t_2-t_1\right)& \left(7\right)\end{array}$

[0103] It is notable that N of a material pair relates to the relative Seebeck coefficient S.sub.material, relative as follows:

[00014]$\begin{array}{cc}N=\frac{S_{+}}{S_{-}},& \left(8\right)\end{array}$

where S.sub.+ is a Seebeck coefficient of a positive conductor of a material pair, and S.sub. is a Seebeck coefficient of a negative conductor of a material pair.

[0104] With this weighted average calculation principle, the temperature in the hot end of the thermocouple can be calculated accurately even if the temperature between the two connector terminals of the cold end is slightly different between them. The respective cold end temperatures are typically different when environmental conditions have changed recently, and the device is stabilizing (meaning: changing and asymptotically reaching, concerning its temperature) to the new ambient temperature value.

[0105] In the following, illustration of FIG. 3 according to an embodiment of the invention is discussed, and its comparisons to prior art situations of FIGS. 1 and 2 are discussed as well.

[0106] FIG. 3 presents a connection of a thermocouple 30 and used measurement points 33, 34 for the cold end of the thermocouple 30 which are used in the shown embodiment according to the present invention. This may also be called as a thermocouple circuit. The shown thermocouple 30 consists of two wires 31, 32 welded together in one end (the hot end in the left-hand side), where the two wires 31, 32 are made from different metals, thus having mutually different thermoelectric properties. In this embodiment, the positive conductor 31 is made of chromel (NiCr) and the negative conductor 32 is made of alumel (NiAl). In other words, this is a K-type thermocouple. However, the present invention is applicable with any kinds and types of thermocouples, comprising any material pairs. Even semiconductor materials may be used in such thermocouples.

[0107] Thermal voltage and electrical energy in the thermocouple circuit is generated in the connection from the temperature difference between the hot end temperature (marked as t) and the cold end temperature, which actually consists of two separate points 33, 34 (t.sub.1 and t.sub.2, respectively). Measured voltage Us can be defined as a potential difference between copper wires 35, 36 which is the difference of thermal voltages U.sub.1 and U.sub.2. In other words,

[00015]$\begin{array}{cc}{U}_3=U_1-U_2& \left(9\right)\end{array}$

[0108] Naturally, the material pair may be different than the one shown in this illustration. Also copper as the material selected for wires 35, 36 can be changed to another material, in various other embodiments of the invention.

[0109] When the temperature is measured from both terminals 33, 34 of the cold end (as t.sub.1 and t.sub.2) of the thermocouple and the temperature is balanced with a ratio of the expected thermal voltages U.sub.1 and U.sub.2, it is possible to calculate the cold end temperature in real time and also more accurately than with conventional single point measurement presented and discussed earlier in connection of FIGS. 1 and 2.

[0110] In an embodiment of the invention, the temperature of the hot end of the thermocouple can be calculated with equations based on temperature voltage coefficients (Seebeck coefficients) of the used materials (i.e. thermocouple material data tables) or with more detailed data tables that take into account also the nonlinearities of the thermal voltages. Cold end temperatures and the difference between the temperatures can then be taken into account in the calculations of the hot end temperature.

[0111] For different metals and alloys, there exists ready-to-use data tables comprising Seebeck coefficients, which data tables can be used to obtain mutual relationships of the thermal voltage differences between the positive and negative conductor for various material pairs used in the thermocouples. Regarding the present invention, this data needs to be known for each different type of thermocouple which are used with the inventive concept; if the method is applied to several different thermocouple types used in the arrangement. If only a single type of a thermocouple is taken into use in the arrangement (like the used thermocouple having a chromel-alumel material pair in the conductors, i.e. using a K-type of thermocouple as shown in the example of FIG. 3), then the presented arrangement merely requires Seebeck coefficient data table for that single material pair. In the example of FIG. 3, such a data table then consists the material pair of chromel (NiCr) and alumel (NiAl). This data table may be created as a result of earlier conducted measurements by the thermocouple manufacturer. In an embodiment of the invention, the data table information is used in obtaining the hot end temperature t of the thermocouple as described earlier, using also the accurate measurement result of the cold end obtained via the above-described process according to the present invention.

[0112] In more detail, the hot end temperature determination proceeds as follows, in an embodiment of the invention: [0113] 1) Measuring the thermal voltage Us in the cold end, and also measuring the temperatures t.sub.1 and t.sub.2 in the measurement points 33, 34 of the cold end. [0114] 2) Calculating the weighted average of the temperature (T.sub.w,ave) from the cold end temperature results t.sub.1 and t.sub.2 which calculation is made according to the type and/or materials of the used thermocouple 30. [0115] 3) Transforming the weighted average of the temperature (T.sub.w,ave) to a voltage either using a conversion data table of the used thermocouple type or using a calculation formula involving correction coefficients of the applied thermocouple type. [0116] 4) Adding the measured thermal voltage to the calculated voltage value, and thus, a correct thermal voltage value for the hot end temperature calculation is obtained. This correct thermal voltage value corresponds to the 0 C. reference point used in thermocouple data tables and calculation formulas. [0117] 5) Converting the correct thermal voltage value to a temperature value using the conversion data table of the used thermocouple type or the calculation formula supplied with the correction coefficients. Hence, an accurate hot end temperature value is obtained.

[0118] In case proper Seebeck coefficients are not available for the used material pair of the thermocouple from existing sources, it is also possible to test the used materials of the thermocouple by measurements. In an embodiment of the invention, this is performed by measuring the magnitude of an induced thermoelectric voltage in response to a temperature difference across the material; this is the same as the Seebeck coefficient of that particular material. For a material pair, the thermal voltage created in the positive conductor minus the thermal voltage created in the negative conductor (U.sub.1-U.sub.2), is the Seebeck coefficient of that material pair (=S).

[0119] Concluding from the previously discussed matters, the parameter N, as in equations (5) and (6), is obtained from the respective Seebeck coefficient data table for the respective material pair of the used thermocouple. In this way, T.sub.w,ave (as in equation (7)) and finally, t can be obtained with great accuracy.

[0120] In the following, an embodiment of the invention is introduced where the temperature dependency of the thermal voltage ratio in specific thermocouples (i.e. material pairs) is taken into account. In an earlier described simpler embodiment, the thermal voltage ratio N is a calculated and constant value, which is based on general material characteristics of the material pair of the thermocouple. As mentioned above, this earlier described embodiment involves a thermocouple (material pair) specific Seebeck value fixed to a temperature value of 0 C. Now in the present embodiment taking the temperature dependency as a real-life characteristic into account, the value N (i.e. the ratio of the thermal voltages) is determined as a function of temperature, and thus, it is a non-constant value. The value N may be presented in a table form or by a polynomial equation. In this way, the system is able to take into account and correct various temperature dependent or certain temperature value specific deviations, which are present in the currently selected thermocouple applying two specific material wires. Each different type of thermocouple may have its own specific temperature dependent graph in this regard, and even some other anomalies in the system may be taken into account.

[0121] For example, if a wire made of constantan (i.e. CuNi alloy) is considered and a Seebeck coefficient is considered between a temperature range of 100 K . . . 650 K, the Seebeck coefficient decreases from around 15 V/K to around 55 V/K in a slightly non-linear fashion when the temperature increases.

[0122] The presented embodiment with temperature dependent thermal voltage ratio N has a notable advantage when temperatures t.sub.1 and t.sub.2 deviate clearly from one another; this might happen when there is a quick temperature change in the cold end of the thermocouple.

[0123] When the temperature of the cold end has already become completely steady (meaning that t.sub.1=t.sub.2), the Seebeck coefficients in an available database or in polynomial equations concerning the thermocouple type at hand already takes into account the deviations relating to each of the used wires of the thermocouple.

[0124] Therefore, this embodiment is most beneficial in a situation where t.sub.1 and t.sub.2 notable deviate with one another; in rapidly occurring temperature changes.

[0125] Thus, in the described embodiment applying a temperature dependent value of N, in a situation of rapidly varying cold end temperature, the accuracy of the temperature measurement with the described arrangement can be classified as excellent.

[0126] In the earlier described embodiment, where the value of N is considered and applied as a constant value over an applied temperature range, the accuracy of the temperature measurement with the described arrangement can be classified as very good, in situations where the cold end temperatures do not change very quickly.

[0127] Concerning the technical effect obtained by various embodiments of the present invention, the following conclusions are brought up.

[0128] FIG. 4 illustrates how the present invention gives notably different results compared to mere averaging of the positive and negative terminal temperature values (described earlier) of a thermocouple.

[0129] The illustration shows temperature distribution between certain elements of the thermocouple, and in the calculated results in a prior art solution and in accordance with the present invention.

[0130] In this example, the thermal voltage ratio between the positive and the negative conductor is 6.1. The left-hand most vertical bar in the illustration shows the measured temperature of the positive terminal of the thermocouple (in C.). In this example, this temperature (t.sub.1) is 22.6 C. The second vertical bar from the left is the measured temperature of the negative terminal of the thermocouple (in C.). In this example, this temperature (t.sub.2) is 23.1 C.

[0131] In the prior art method, in case of notably different temperatures of the terminals in the cold end due to recent rapid change in the cold end of the thermocouple, it applies mere average value calculation of the measured temperatures of the positive and negative terminals. In this way, the T.sub.ave is in this example=22.85 C. It can be concluded that in this situation, the precision of measurement is mediocre.

[0132] In the present invention, a weighted average temperature value is obtained. This is represented by the fourth bar from the left (=the right-most bar). The weighted average temperature value T.sub.w,ave is calculated as guided in equation (7) in the above description.

[0133] The weighted average temperature value is calculated as:

[00016]$\left(1/\left(6,1\right)\right)*\left(23,1C.-22,6C.\right)+22,6C.~=22,682C.$

and it is shown in the fourth bar of the illustration.

[0134] This temperature differs notably from the above calculated direct average temperature T.sub.ave.

[0135] The weighted average temperature value corresponds accurately to the completely steadied temperature in the cold end of the thermocouple. The present invention can be said to predict the steadied temperature value well before the actual stabilized situation has occurred; i.e. during the dynamic situation where the terminal temperatures (or either of them) are still changing. This is a notable advantage of the present invention.

[0136] When the value of N increases, then the difference between T.sub.ave and T.sub.w,ave gets larger. Hence, with especially larger N's, the moderate accuracy obtained from T.sub.ave can be changed to very precise result with T.sub.w,ave. Furthermore, the very precise result is obtained already during the temperature stabilization period; hence, the measurement time is shortened significantly compared to the prior art. This is a great practical advantage, allowing for instance more temperature measurements to be conducted in the same time period.

[0137] When considering the embodiment where N is a temperature dependent parameter (non-constant), the presented weighted average temperature calculation principle allows correcting or compensating various phenomena or physical realities in the measuring arrangement. For instance, characteristic errors present in the conductors (wirings) of the thermocouple may be corrected when applying the calculation principle according to the present invention. Even some non-desired temperature-dependent characteristics which are caused by the mechanical structure of the measurement assembly may be corrected when a proper polynomial equation or data table is applied. The method according to the present invention is capable to compensate in practice all temperature related or temperature dependent (present with certain temperature values or ranges, for example) errors of the whole thermocouple measurement chain. Hence, a very good accuracy of the temperature measurement results is obtained. Also, this result can be obtained in a very speedy fashion with no defects in the result quality. This characteristic of the present invention may be utilized e.g. in processes, where similar type of temperature cycling process is repeated over and over again.

[0138] The advantages of the present invention are discussed also in the following sections.

[0139] The accuracies of the used temperature sensors and the accuracy of the values within the thermocouple material data table or within the calculation formula affect the measurement accuracy of the presented method according to the invention. With the present invention and its weighted value calculation principle, it is possible to obtain a ten-fold increase (i.e. enhancement) of accuracy already at the starting moments of the measurement, if the present invention is compared to a reference prior art measurement comprising two temperature sensors when using their direct average temperature value as such. Using two temperature sensors and a weighted average temperature calculation principle as described in connection with the present invention, the accuracy of the temperature measurement using a thermocouple is thus improved significantly.

[0140] Besides the accuracy, also the required measurement time needed for obtaining an accurate result decreases significantly. When the environmental conditions, comprising the temperature, change, accurate measurement results can be obtained much faster than in prior art solutions. With current solutions according to the prior art, the operator of the temperature measurement arrangement (i.e. the user) must wait for a certain time period for the conditions to stabilize. This is particularly true when two conductors of a thermocouple have notably different temperatures between one another. In that situation in prior art, the user needed to wait pretty long before the temperatures of the two conductors ended up being quite close, and then proceeding with plain averaging of these temperature values. This stabilization process will take a time, which could be tens of minutes, or even over an hour.

[0141] With the presented arrangement according to the invention, the starting point of the principle is that the temperatures of the two conductors may be substantially different. There could have been a notable change of temperature in the measurement location which still affects the wires of the thermocouple, and possible not in a uniform fashion between the wires. Still, the present invention and its weighted average calculation principle makes it possible to obtain a very accurate temperature result in a very quick (almost instant) manner. It can be concluded that the required measurement time is in the range of seconds compared to the range of tens of minutes of waiting time associated with conventional solutions. With notably shorter waiting times, the operator is able to perform a larger number of temperature measurements in the same time period, without compromising accuracy. Time can be saved to other tasks, for example. Frustration is minimized with the presence of the present invention.

[0142] In the context of the application areas, which are industrial calibrations in industrial environments, such as automation lines, production lines, and factories in general, the advantages are tangible resulting from the very short stabilization times. In industrial processes, the environmental conditions of a product to be processed or manufactured may still vary to a notable extent and/or change rapidly, especially concerning the temperature. Therefore, the available time for doing proper measurements with good measurement accuracy may be very short. In industrial environment, time available for calibrations is usually limited due to scheduled maintenance times and lots of calibrations need to be performed during said maintenance times, whose start and end times are usually well determined e.g. in a production line. The present invention indeed allows the temperature measurements to be performed in a limited available time, which may be a relatively short or even very short time period. Hence, the present invention suits exceptionally well in industrial processes involving industrial calibrations, and for instance processes which require lots of temperature measurements to be performed in a short time span.

[0143] Furthermore, with a single temperature sensor, there are various inaccuracies such as uneven thermal distribution, thermal energy flows (which might act in a complex way), and thermal losses. With the present invention, these inaccuracy sources are handled well, and the measured temperature result based on the inventive principle will be very accurate and it can also be obtained very quickly.

[0144] Even after a long waiting period, the inventive concept of using two temperature sensors and the weighted average calculation principle results in greater measurement accuracy than by using merely a single temperature sensor. Hence, even in that situation, the present invention is advantageous.

[0145] In practice, there is no need for waiting any substantial stabilization time in order for the method according to the invention to work well.

[0146] Hence, technical effects and advantages are notable with the presented embodiments of the present invention, when compared to the various prior art solutions.

[0147] The present invention is not restricted merely to embodiments presented in the above, but the scope of the present invention is determined by the claims.