“Principle of thermoelectric effect: As shown in Figure 1, two different colors are used to represent two different metal materials. Terminals A and B are used for temperature measurement in a normal temperature environment, and are called cold ends. Point C is the measured terminal. Due to the thermoelectric effect, the temperature between the A terminal and the C terminal and the B terminal and the C terminal is different, so a potential difference will be generated. And because of the difference between the two metal materials, the two potential differences are different. Finally, there is a potential difference between A and B. After measuring the potential difference between AB, the reference metal characteristic value and the cold junction temperature are checked and calibrated. Finally, the temperature value of the corresponding C terminal can be obtained by measuring the potential difference output from the AB terminal.
Thermocouple, thermoelectric effect and thermoelectric effect principle
A thermocouple is a sensor that connects one end of two different materials of metal and uses the thermoelectric effect to measure temperature.
In 1821, German scientist Thomas John Seebeck discovered the inverse effect of the thermal effect of electric current: that is, when different temperatures are applied to the two ends of a piece of wire, an electromotive force will be generated at both ends of the wire, and there will be an electric current in the wire after the loop is closed. flow past. This phenomenon is called the “thermoelectric effect”, also called the “Seebeck effect”.
Principle of thermoelectric effect: As shown in Figure 1, two different colors are used to represent two different metal materials. Terminals A and B are used for temperature measurement in a normal temperature environment, and are called cold ends. Point C is the measured terminal. Due to the thermoelectric effect, the temperature between the A terminal and the C terminal and the B terminal and the C terminal is different, so a potential difference will be generated. And because of the difference between the two metal materials, the two potential differences are different. Finally, there is a potential difference between A and B. After measuring the potential difference between AB, the reference metal characteristic value and the cold junction temperature are checked and calibrated. Finally, the temperature value of the corresponding C terminal can be obtained by measuring the potential difference output from the AB terminal.
Chinese standardized thermocouples have been produced in accordance with IEC international standards on January 1, 1988, and eight standardized thermocouples of S, R, B, K, J, T, N, and E are designated as China’s unified design thermocouples (as shown in the figure) 2).
The metals used in S, R, and B thermocouples are relatively expensive, so the price is relatively high; the metals used in K, T, J, N, and E thermocouples are relatively cheap, so the relative price is relatively cheap. The following describes the temperature measurement range, advantages and disadvantages of these types of thermocouples:
S, R, B type thermocouple
The long-term maximum operating temperature of S, R-type and B-type thermocouples is 1300℃ and 1600℃, respectively, and the short-term maximum operating temperature is 1600℃ and 1800℃ respectively. Advantages: It has the highest accuracy, best stability, wide temperature measurement zone, long service life, etc. It has good physical and chemical properties, thermoelectric potential stability and good oxidation resistance at high temperatures, and is suitable for oxidizing and inert atmospheres.
The S-type thermocouple has excellent comprehensive performance. The S-type thermocouple conforms to the international temperature standard. It has been used as an interpolating instrument for the international temperature standard for a long time. Although “ITS-90” is stipulated that it will no longer be used as an interpolating instrument for the international temperature scale in the future, the International Temperature Advisory Committee (CCT) believes that the S-type thermocouple can still be used to approximate the international temperature scale.
The overall performance of the R-type thermocouple is equivalent to that of the S-type thermocouple. Type B thermocouple is similar to S and R, but it is not suitable for reducing atmosphere or atmosphere containing metal or non-metal vapor. But its obvious advantage is that it does not need to use compensation wire for compensation, because the thermoelectric potential is less than 3μV in the range of 0~50℃.
Disadvantages of T, R, and B thermocouples: This type of thermocouple has a low thermoelectric potential rate, low sensitivity, reduced mechanical strength at high temperatures, and is very sensitive to pollution, and precious metal materials are expensive.
K, N, E, J, T type thermocouple
The temperature measurement range and advantages and disadvantages are shown in Table 1:
Supplement: The N-type thermocouple overcomes two important shortcomings of the K-type thermocouple: the K-type thermocouple is between 300 and 500 ℃, and the thermoelectromotive force is unstable due to the short-range order of the nickel-chromium alloy crystal lattice; at 800 ℃ About, the thermoelectromotive force is unstable due to the preferential oxidation of nickel-chromium alloy.
The advantages and disadvantages of thermocouples
a. Wide temperature range: from low temperature to jet engine exhaust, thermocouples are suitable for most practical temperature ranges. The temperature range of the thermocouple is between 200°C and +2500°C, depending on the metal wire used.
b. Rugged and durable: The thermocouple is a durable device with good shock and vibration resistance and is suitable for dangerous and harsh environments.
c. Fast response: Because of their small size and low heat capacity, thermocouples respond quickly to temperature changes, especially when the sensing junction is exposed. They can respond to temperature changes within hundreds of milliseconds.
d. No self-heating: Because the thermocouple does not require an excitation power source, it is not easy to self-heat, and it is safe in itself.
a. Complex signal conditioning: A large amount of signal conditioning must be performed to convert the thermocouple voltage into a usable temperature reading. For a long time, signal conditioning consumes a lot of design time, and improper processing will introduce errors, resulting in reduced accuracy.
b. Low accuracy: In addition to the inherent inaccuracy of the thermocouple due to the metal characteristics, the measurement accuracy of the thermocouple can only reach the measurement accuracy of the reference junction temperature, generally within 1°C to 2°C.
c. Susceptible to corrosion: Because the thermocouple is composed of two different metals, under some working conditions, corrosion over time may reduce accuracy. Therefore, they may need protection; and maintenance is essential.
d. Poor noise immunity: When measuring millivolt-level signal changes, noise generated by stray electric and magnetic fields may cause problems. Twisted thermocouple wire pairs may greatly reduce magnetic field coupling. Using shielded cables or routing and shielding in metal conduits can reduce electric field coupling. The measuring device should provide hardware or software signal filtering to effectively suppress the power frequency (50 Hz/60 Hz) and its harmonics.
Selection elements of thermocouple and thermal resistance
We can choose thermocouple and thermal resistance according to the following elements.
The temperature range that needs to be measured: a thermocouple is generally selected for a temperature above 500°C, and it depends on the application environment for a temperature below 500°C.
Measurement range selection: The temperature measured by the thermocouple generally refers to the “point” temperature, and the thermal resistance is usually used to measure the temperature of the space.
Cold junction compensation
Due to the principle of the thermoelectric effect. Therefore, an additional temperature sensor is needed to measure the temperature of the reference point, which is what we often call the cold junction compensation point.
Several common cold junction compensation sensors are as follows:
1. Thermistor: fast response and small package. However, linearity is required and accuracy is limited, especially in a wide temperature range. Requires excitation current, which will generate self-heating and cause drift. The overall system accuracy after combining the signal conditioning function is poor, which is only suitable for applications with low measurement accuracy and low cost.
2. Resistance temperature measuring device (RTD): Compared with thermistor, RTD is more accurate, stable and linear in characteristics, but the package size and cost are higher than thermistor. Because of the need for a well-matched excitation source and sampling circuit, the design is relatively more complicated and requires better peripheral components. Using RTD as a thermocouple measurement system for cold junction compensation usually requires higher system-level precision.
3. Integrated temperature sensor: The integrated temperature sensor is an integrated temperature measuring element made by semiconductor technology. Through semiconductor process technology, the information obtained by temperature measurement and other analog units is digitally output, with high integration, and system-level accuracy far below 1°C can be obtained. The peripheral circuit design is simple, and it can directly communicate with the MCU. The cold junction compensation scheme for the high-precision thermocouple acquisition system is also the simplest to use and design.
Integrated temperature sensor ADT7320
David Liu, an engineer of the technical authorized agent Excelpoint Shijian, introduced a typical ADI temperature sensor for cold junction compensation-ADT7320. The functional block diagram is shown in Figure 3:
ADT7320 is a high-precision digital temperature sensor that uses a 16-bit ADC to monitor and digitize temperature parameters with a resolution of 0.0078°C. By default, the ADC resolution is set to 13Bit (0.0625°C). The principle is that the internal temperature sensor generates a voltage proportional to the absolute temperature. This voltage is compared with the internal reference voltage and then input to the precision digital modulator.
The internal temperature sensor has high accuracy and linearity in the entire rated temperature range, and no user correction or calibration is required.
In addition, it has an over-temperature alarm function, which guarantees functional safety. The external output port, INT and CT make it possible to directly send an interrupt signal to the back-end MCU through a 10K pull-up resistor in the case of ultra-high temperature or low temperature.
Analog front end for thermocouple measurement that ADI can provide
The AD8494/AD8495/AD8496/AD8497 thermocouple amplifiers provide a simple, low-cost solution for the front end of the thermocouple temperature measurement signal conditioning. For the thermocouple sampling terminal, common-mode interference signals, ESD and overvoltage protection are often considered in field design.
AD849x is specially optimized for measuring and amplifying the signals of J-type and K-type thermocouples. It not only integrates ESD and overvoltage protection functions at the front end through CMOS process, but also has excellent common-mode resistance to 5mV/°C system-level linearity. Response Vout=(TMJ × 5 mV/°C) +VREF, Where TMJ represents the measured junction temperature of the thermocouple. The comparison of AD849x homologous differentiation is shown in Figure 5:
ADI can provide integrated solution of thermocouple measurement
David Liu introduced that ADI can provide a variety of thermocouple measurement integration solutions.
AD7124-4/AD7124-8 is a 24bit ADC as the core, internal high integration MUX, PGA, REF, etc., for the complete solution of thermocouple and thermal resistance measurement with direct access type. It can achieve high resolution, low noise performance and low nonlinearity error capability.
The on-chip low-noise PGA can flexibly adjust the gain programming range (1, 2, 4, 8, 16, 32, 64, 128) through software to adjust the amplitude of the input signal to reach the effective sampling range of the ADC. The gain stage has high input impedance, and the input leakage current does not exceed 3.3 nA in full power mode and 1 nA (typical value) in low power mode.
The circuit shown in Figure 6 is a reference design for a typical thermocouple using RTD for cold junction compensation. Use two analog input pins to connect thermocouples (AIN2, AIN3), and three-wire RTD circuits (AIN1, AIN6, AIN7). AIN2 and AIN3 are configured as fully differential input channels to measure the voltage generated by the thermocouple. For this circuit, as shown in Figure 6, the thermocouple is floating. To bias the thermocouple to a known level, enable the VBIAS voltage generator on AIN2 to bias the thermocouple to the following value:
The thermocouple measurement is an absolute measurement, so a reference voltage source is required, using the built-in 2.5 V reference voltage source of the AD7124-4/AD7124-8.
For cold junction compensation, an excitation current source is used to excite the RTD. This current is generated from AVDD and flows to AIN1. Figure 6 details the analog pins and their configuration.
For this circuit, the cold junction circuit uses the reference input REFIN1(±). The current flowing through the 4-wire RTD (used for cold junction measurement) also flows through the precision reference resistor, generating a reference voltage. The voltage generated on this precision reference resistor is proportional to the voltage on the RTD, so the fluctuation of the excitation current will be eliminated. Since the reference buffer has been enabled, the margin required for normal operation (AVDD − 0.1 V and AVSS + 0.1 V) must be satisfied. The margin of 0.125 V (500 μA × 250 Ω) is provided by the 250 Ω ground resistance, as shown in Figure 6.
LTC298X measures various temperature sensors and digitally outputs the results (in °C or °F), with 0.1°C accuracy and 0.001°C resolution. LTC298X can measure the temperature of almost all standard (B, E, J, K, N, S, R, T type) or custom thermocouples, automatically compensate the cold junction temperature and achieve linearization of the results. The device can also use standard 2, 3 or 4 wire RTDs, thermistors and diodes to measure temperature. It has 20 reconfigurable analog inputs and supports many sensor connection and configuration options. LTC298X includes excitation current source and fault detection circuit suitable for each temperature sensor.
The LTC298X can directly interface with ground-referenced sensors without the need for level shifters, negative supply voltages, or external amplifiers. All signals are buffered and synchronized digitized by three high-precision, 24-bit ΔΣ ADCs driven by an internal 10ppm/°C (maximum) voltage reference.
AD7124 or LTC298X
Accuracy: LTC298x has an accuracy of 0.1°C, and AD7124 has an overall system accuracy of ±1°C in the measurement temperature range of −50°C to +200°C.
Channel: cold junction compensation. Taking 4-wire RTD as an example, LTC298x can measure 15 thermocouples, AD7124-4 can measure 2 thermocouples, and AD7124-8 can measure 6 thermocouples.
Relative cost factor: LTC298X is higher than AD7124, but it provides more sampling channels while reducing the designer’s related requirements for calibration. The AD7124-X, which has a relatively low cost, also has higher system-level sampling accuracy, but has fewer measurable channels, and requires the designer to spend a certain amount of effort on system calibration.
Protection: LTC298x series products have automatic detection functions for burnout, short circuit and failure.
For thermocouple temperature collection, Shijian can provide professional, accurate, and flexible ADI thermocouple measurement products and solutions, as well as system-level sampling solutions, bringing convenience to designers!