A high-precision and high-stability K-type thermoelectric pile type heat flow sensor

Abstract

The application provides a high-precision and high-stability K-type thermoelectric pile type heat flow sensor, relates to a heat flow measuring technical field, and comprises a sensitive head, a signal processing circuit and a packaging shell, wherein core devices are all device. The sensitive head is a circular structure with a diameter of 10 mm, is internally provided with an annular gradient heat conduction groove and a double-layer stress buffer layer, is centrally provided with a K-type thermoelectric pile array formed by a plurality of pairs of K-type thermocouples connected in series, electrodes of the K-type thermoelectric pile array are combined with a base through a Pt-Rh alloy transition layer and are high-temperature oxidation resistant electrodes, a surface of the high-temperature oxidation resistant electrodes is covered with a nanocrystalline alumina dense layer, and 8 mV output is corresponding to 1 MW / m2 heat flow. The signal processing circuit comprises signal amplification, temperature measurement, cold end compensation, filtering, voltage stabilization, double-channel differential temperature drift real-time offset and dynamic gain calibration modules, and temperature drift is inhibited. The packaging shell is made of 310S stainless steel, has a temperature resistance of greater than or equal to 800 DEG C and is internally provided with a temperature field homogenization isolation layer. The sensor has compact structure, high precision and good stability, and is suitable for a 600 DEG C high-temperature environment.

Description

Technical Field

[0001] This invention relates to the field of heat flow measurement technology, and specifically to a high-precision, high-stability K-type thermopile heat flow sensor. Background Technology

[0002] Heat flux sensors are the core components for measuring the heat flux density on the surface of an object. Their measurement accuracy and stability are directly related to the reliability of high-temperature equipment operation monitoring and thermal process control. In high-temperature scenarios such as metallurgy and aerospace, it is often necessary to measure ambient temperatures of around 600℃ and heat flux densities at the level of 1MW / m². At the same time, there are high requirements for the size, accuracy, stability and device integration of the sensor.

[0003] Existing K-type thermopile heat flux sensors have significant shortcomings: the sensing head is too large, making it unsuitable for confined installation spaces, and the mismatch between the sensing structure and the sensing head size affects heat flux conduction efficiency and measurement response speed; the signal processing circuit design is flawed, and the weak millivolt-level signal output by the thermopile is easily interfered with during amplification, making it difficult to meet practical requirements for measurement accuracy; the temperature measurement circuit and the heat flux signal acquisition circuit have poor coordination, resulting in poor temperature compensation and insufficient stability under high-temperature environments; and the core components rely on imports, which not only increases production costs but also leads to supply chain instability and inconvenient maintenance.

[0004] Furthermore, existing thermopile sensors generally suffer from problems such as high thermal stress, severe edge heat dissipation, easy electrode oxidation and detachment, and difficulty in suppressing temperature drift at 600℃, resulting in poor long-term operational stability and failing to meet the requirements of high-precision high-temperature monitoring. Therefore, there is an urgent need to develop a K-type thermopile heat flux sensor with a compact sensing head, reasonable circuit design, and the use of all necessary components, which can meet the requirements of high precision and high stability, in order to overcome the shortcomings of existing technologies. Summary of the Invention

[0005] To address the aforementioned technical issues, a high-precision and high-stability K-type thermopile heat flow sensor is provided. This sensor boasts advantages such as compact structure, high measurement accuracy, good stability, modular design, and strong adaptability. Through innovative structural design and circuit algorithm optimization, it solves problems such as thermal stress, edge heat dissipation, electrode oxidation, and temperature drift at high temperatures, enabling simultaneous and accurate measurement of heat flow and temperature under high-temperature conditions.

[0006] The technical means employed in this invention are as follows: A high-precision, high-stability K-type thermopile heat flow sensor includes a sensing head, a signal processing circuit, and a packaging shell. The signal processing circuit is electrically connected to the sensing head and integrated inside the packaging shell. The sensitive head is circular, with concentric ring-shaped gradient heat conduction grooves on the periphery to constrain the heat flow to be concentrated and conducted to the central region; it is equipped with a double-layer stress buffer layer to buffer high-temperature thermal stress. The central region of the sensitive head is provided with a K-type thermopile array composed of multiple pairs of K-type thermocouples connected in series. A platinum-rhodium alloy transition layer to enhance the bonding force is provided between the electrodes of the K-type thermocouples and the substrate. The electrodes are made of high-temperature oxidation resistant material and the surface is covered with a dense layer of nanocrystalline alumina. The signal processing circuit includes a signal amplification circuit, a temperature measurement circuit, a cold junction compensation module, a filtering module, a voltage regulation power supply module, a dual-channel differential temperature drift real-time cancellation module, a dynamic gain calibration unit, and a microcontroller. The microcontroller, as the digital computing core, performs dual-channel differential temperature drift real-time cancellation and dynamic gain calibration, performs differential operations on the temperature drift signals of the K-type thermopile array and the temperature measurement circuit, and adaptively adjusts the amplification gain according to the deviation of the amplified signal. The encapsulation shell is provided with a temperature field homogenization isolation layer.

[0007] Furthermore, the double-layer stress buffer layer includes an inner buffer layer and an outer buffer layer. The inner buffer layer is attached to the substrate of the K-type thermopile array, and the outer buffer layer is connected to the annular gradient heat conduction groove, forming a composite buffer structure.

[0008] Furthermore, the electrode includes a nickel-chromium alloy layer and a gold alloy layer bonded to the surface of the nickel-chromium alloy layer, and the platinum-rhodium alloy transition layer is deposited between the substrate and the electrode using a magnetron sputtering process.

[0009] Furthermore, multiple pairs of the K-type thermocouples are evenly arranged in a ring array within the central region, and the total output voltage of the K-type thermopile array is linearly related to the measured heat flux density.

[0010] Furthermore, the microcontroller connects to the first analog-to-digital converter (ADC) chip and the second ADC chip via two independent analog SPI communication interfaces. The first ADC chip is located at the rear end of the signal amplification circuit, and the second ADC chip is located at the rear end of the temperature measurement circuit, enabling synchronous data acquisition between the heat flow channel and the temperature channel. The microcontroller's on-chip non-volatile memory stores the temperature drift coupling coefficient, gain correction coefficient, and segmented temperature compensation curve, which are automatically read and initialized upon power-up.

[0011] Furthermore, the dynamic gain calibration unit monitors the deviation of the amplified signal from the theoretical standard value in real time. When the deviation of multiple consecutive samples exceeds the calibration trigger criterion, the reference gain is corrected with a gain correction coefficient. The corrected gain is limited to the protection range. After the deviation converges, the optimal gain parameter is fixed to the on-chip memory of the microcontroller.

[0012] Furthermore, the dual-channel differential temperature drift real-time compensation module compensates for the temperature drift of the output voltage of the K-type thermopile array based on the real-time temperature difference between the cold junction of the thermopile channel and the temperature measurement channel and the temperature drift coupling coefficient; when the error of a single compensation calculation exceeds the resampling criterion and continuous recalculation still exceeds the limit, the segmented temperature drift coupling coefficient is automatically updated.

[0013] Furthermore, the temperature measuring circuit includes a K-type thermocouple, a cold junction compensation module, a signal preamplifier module, an anti-interference filter module, an ADC analog-to-digital converter module, and a digital signal output module connected in sequence; the hot junction of the K-type thermocouple is attached to the surface of the sensing head.

[0014] Furthermore, the K-type temperature measuring thermocouple is an armored thermocouple. The main body of the armored thermocouple is arranged and fixed along the wiring groove inside the encapsulation shell, and the cold junction lead is housed in the low-temperature zone inside the encapsulation shell, which is far away from the sensing head.

[0015] Furthermore, the filtering module includes an analog filtering submodule and a digital filtering submodule. The analog filtering submodule includes a low-pass filter circuit located in the signal amplification circuit. The digital filtering submodule uses a moving average filtering algorithm run by the microcontroller and has a built-in jump interference rejection criterion. The voltage regulation power supply module includes a first power supply branch and a second power supply branch that are independent of each other. The first power supply branch supplies power to the signal amplification circuit, the dual-channel differential temperature drift real-time cancellation module, and the dynamic gain calibration unit. The second power supply branch supplies power to the temperature measurement circuit and the cold junction compensation module.

[0016] Compared with the prior art, the present invention has the following beneficial technical effects: This invention provides a high-precision, high-stability K-type thermopile heat flux sensor. It optimizes the sensing head structure (annular gradient heat conduction groove, double-layer stress buffer layer, Pt-Rh alloy transition layer, and nanocrystalline alumina dense layer), and innovates the signal processing circuit design (dual-channel differential temperature drift real-time compensation module and dynamic gain calibration unit). It employs all core components and is equipped with a complete standardized manufacturing process and multi-condition graded calibration method. The manufacturing stage sequentially completes the following steps: sensing substrate grooving, ceramic fiber buffer layer bonding, Pt-Rh transition layer evaporation, K-type thermopile array annular arrangement welding, nickel-chromium-gold alloy electrode sealing, and nanocrystalline alumina dense layer plasma spraying. Then, the signal processing circuit is surface-mounted and soldered, the dual-channel regulated power supply is partitioned and wired, and the overall sealing and temperature field isolation layer is filled. The calibration stage is divided into three steps: low temperature reference calibration, 600℃ high temperature heat flux gradient calibration, and long-term temperature drift aging calibration. Relying on the standard radiant heat flux source and constant temperature furnace, the dynamic gain parameter is entered, the dual-channel temperature drift compensation coefficient is solidified, and the cold end compensation curve is stored. This effectively solves the problems of high thermal stress, severe edge heat dissipation, easy oxidation and detachment of electrodes, and difficulty in suppressing temperature drift of existing sensors at high temperatures. It realizes synchronous and accurate measurement of heat flux and temperature, with a compact structure, good stability, and is suitable for high-temperature industrial scenarios, solving many defects of existing technologies. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a flowchart of the signal processing circuit of the present invention; Figure 2 This is a pin diagram of the TPA1286U-SO1R of the present invention; Figure 3 This is a pin diagram of the SGM8622 of this invention; Figure 4 This is a pin diagram of the SGM8541 of this invention; Figure 5 This is a pin diagram of the CS1237 of the present invention; Figure 6 This is a structural diagram of the sensor of the present invention; Figure 7 This is a structural diagram of the sensor protective layer and thermal resistance layer of the present invention; Figure 8 This is a structural diagram of the PT1000 sensor of the present invention; Figure 9 This is a diagram of the nickel-chromium-nickel-silicon functional thin film of the sensor of the present invention; Figure 10 This is a top view of the sensor structure of the present invention; Figure 11 This is a cross-sectional view of the sensor of the present invention. Detailed Implementation

[0019] This invention provides a high-precision, high-stability K-type thermopile heat flux sensor, comprising: a sensing head, a signal processing circuit, and a packaging shell. The signal processing circuit is electrically connected to the sensing head and is integrated inside the packaging shell. The core components of the sensing head and the signal processing circuit are all based on [specific technologies / mechanisms]. Specific implementation details are as follows: The sensing head adopts a circular structure with a diameter of 10mm, which is suitable for narrow installation spaces. It has an annular gradient heat conduction groove and a double-layer stress buffer layer inside. The annular gradient heat conduction groove is distributed in a concentric ring, with a groove width of 0.5-1mm and a groove depth of 0.3-0.5mm. It can effectively reduce edge heat dissipation, ensure the concentration and uniformity of heat flow conduction, and reduce the impact of edge heat dissipation on measurement accuracy. The double-layer stress buffer layer is made of high-temperature resistant ceramic fiber material with a thickness of 0.2-0.3mm. It is divided into an inner buffer layer and an outer buffer layer. The inner layer is attached to the thermopile array substrate, and the outer layer is connected to the annular gradient heat conduction groove. It can effectively alleviate thermal stress under high temperature environment, prevent deformation and damage of the sensing head structure, and improve long-term working stability.

[0020] The central region of the sensitive head features an 8mm diameter heat flow sensing structure, which is a K-type thermopile array composed of 80 pairs of K-type thermocouples (nickel-chromium / nickel-silicon material) connected in series. The K-type thermopile array uses high-temperature oxidation-resistant electrodes (nickel-chromium / gold alloy material). A Pt-Rh alloy transition layer with a thickness of 0.05-0.1mm is provided between the electrodes and the substrate to enhance the bonding force between the electrodes and the substrate and prevent high-temperature oxidation and detachment of the electrodes, making it suitable for high-temperature operating environments up to 600℃. The selected K-type thermocouples are industrial-grade devices with a temperature measurement range of 0-1000℃, stably adaptable to high-temperature operating environments up to 600℃. The 80 pairs of thermocouples are evenly arranged in a ring array within the 8mm diameter sensitive area, ensuring the concentration and uniformity of heat flow conduction. The total output voltage of the thermopile exhibits a good linear relationship with the measured heat flow density; when the heat flow density is 1MW / m², the output voltage is 8mV, ensuring the linearity and accuracy of the heat flow measurement. The surface of the sensitive head is covered with a dense layer of nanocrystalline alumina with a thickness controlled between 0.1 and 0.2 mm. Compared with traditional alumina coatings, it has better density, which can not only play a good role in insulation and protection, reducing external electromagnetic interference and environmental corrosion, but also does not affect the heat flow conduction efficiency, thus ensuring the accuracy of the measurement signal.

[0021] like Figure 1As shown, the signal processing circuit is based on the thermocouple voltage difference measurement principle and mainly includes a signal amplification circuit, a temperature measurement circuit, a cold junction compensation module, a filtering module, a voltage stabilization power supply module, a dual-channel differential temperature drift real-time cancellation module, and a dynamic gain calibration unit. Each module uses domestically produced components to collaboratively achieve accurate acquisition and processing of heat flow and temperature signals. This circuit adds a domestically produced STC8A8K64S4A12 8-bit microcontroller as the core of the entire digital operation chain. The on-chip dual-channel differential temperature drift real-time cancellation algorithm of this invention is not the built-in hardware logic of the ADC chip, but is completed by the microcontroller's on-chip program. The microcontroller reads the original sampling data of the CSE7761 heat flow channel and the CS1237 temperature measurement channel ADCs through the analog SPI bus, thus completing the missing programmable computing hardware carrier in the original text. The TPA1286U, SGM8622, TL431, LDC31855, 74HC595, and two voltage regulator chips used throughout this paper only perform basic analog and simple digital conversion functions such as signal amplification, reference power supply, analog-to-digital conversion, level driving, and power supply regulation. They do not have complex data processing capabilities such as iterative calculation, differential fitting, and mean filtering. The collaborative working logic of each module is clear and closed-loop linked. Based on the voltage regulation power supply module, two sets of voltage regulator chips output stable voltages of 5.0V and 3.3V respectively, providing fluctuating power supply to all other modules and ensuring stable operation of each unit. The weak 8mV heat flow signal output from the K-type thermopile array of the sensitive head is first connected to the signal amplification circuit, where it is linearly amplified 625 times by a high-precision instrumentation amplifier. Simultaneously, the dynamic gain calibration unit relies on the STC8A8K64S4A12 eight-bit microcontroller to monitor the stability of the amplified signal in real time. The equivalent amplification gain is automatically adjusted through the microcontroller's built-in dynamic gain adaptive algorithm to ensure amplification accuracy. (Adjusted amplification gain is shown in the image.) for: ; In the formula, the reference gain Gain correction factor , The deviation between the measured voltage and the theoretical standard value is used as the calibration trigger criterion. A deviation > ±0.05%FS after five consecutive sampling signals is set. The adjusted gain is limited to a protection range of 600-650 times. Convergence is determined when the deviation is ≤ ±0.05%FS after a single iteration, and the optimal gain parameter is stored in the microcontroller's on-chip Flash. Synchronously with the heat flow signal acquisition, the K-type thermocouple in the temperature measurement circuit collects the ambient temperature in real time and transmits the temperature signal to the cold junction compensation module. This module, built using an LDC31855 chip, an NTC thermistor, and an SGM8541 operational amplifier, generates a compensation voltage in real time to offset the error caused by cold junction temperature drift. The amplified and cold junction compensated signal is then processed by the analog and digital filtering modules. The analog filtering relies on two sets of SGM8622 operational amplifiers to build a 10Hz low-pass filter circuit, while the digital filtering uses an 8-point moving average algorithm run by the microcontroller. The nth filtered output value... for: ; in, For the i-th corrected (e.g., temperature drift compensation) sampled voltage value, the system ADC sampling rate is set to 10 SPS, and the filter output refresh period is 0.8 s. Simultaneously, the algorithm incorporates a jump interference rejection criterion. If the difference between a single-point sampled value and the window mean exceeds ±3 times the sampling standard deviation, it is determined to be high-frequency electromagnetic spike interference and the abnormal data is discarded. Valid historical sampled values ​​are reused for mean calculation, thoroughly filtering out high-frequency interference and power supply noise. The dual-channel differential temperature drift real-time cancellation module, as the core stabilization unit, separately acquires the original temperature drift signals from the ADC outputs of the thermopile array and the temperature measurement circuit. The STC8A8K64S4A12 eight-bit microcontroller executes the dual-channel differential temperature drift real-time cancellation algorithm to complete the real-time differential calculation of the two temperature drift sampled values, and the compensated heat flow voltage output... for:

[0022] in, The original output voltage of the thermopile channel, and the temperature drift coupling coefficient. The calibration value range is 0.12~0.18μV / ℃. This module measures the real-time cold junction temperature difference between the thermopile channel and the temperature measurement channel. A resampling trigger criterion is set if the single-time compensation calculation error exceeds 0.02℃. If the error still exceeds the limit after three consecutive recalculations, the segmented coupling coefficient is automatically updated. Accurate compensation is achieved by relying on the temperature drift error of the synchronous coupling between the reference channel and the measurement channel. Ultimately, the temperature drift compensation error is ≤ ±0.02℃, effectively suppressing interference caused by the device's own temperature drift and ambient temperature changes under high-temperature environments. The complete execution flow of this module is as follows: the microcontroller synchronously reads the dual-channel ADC values, converts the two cold junction temperatures and calculates the temperature difference, substitutes them into the correction formula to output the compensated heat flow voltage, and sends the compensated data into the moving average filter ring buffer. Each device module has a clear division of labor and works collaboratively to ultimately achieve accurate acquisition and processing of heat flow and temperature signals.

[0023] The STC8A8K64S4A12 microcontroller added in this invention serves as the sole digital processing core of the device. Existing basic function chips only handle signal front-end conditioning and analog-to-digital conversion, unable to independently perform complex digital algorithm calculations. The microcontroller connects to the CSE7761 and CS1237 analog-to-digital converter chips via two independent analog SPI communication interfaces, enabling synchronous acquisition of dual-channel heat flow and temperature data. Its on-chip non-volatile Flash memory can independently store factory calibration parameters such as temperature drift coupling coefficient, gain correction coefficient, and segmented temperature compensation curves. Upon power-up, the device automatically reads these calibration parameters to complete compensation initialization. The mathematical models, decision thresholds, runtime sequences, and parameter boundaries of all core algorithms in this invention are embedded within the microcontroller program, autonomously achieving real-time end-to-end computation of gain adaptive calibration, dual-channel temperature drift differential cancellation, and digital smoothing filtering under high-temperature conditions, thus addressing the deficiencies of the original technical solution, such as the lack of algorithm carriers and incomplete technical parameters. Specific implementation details are as follows: 1. Signal amplification circuit: such as Figure 2As shown, the core component is the TPA1286U-SO1R high-precision instrumentation amplifier. Its pin functions are as follows: Pin 1 (IN-) is the inverting input of the differential signal, forming a differential input pair with pin 4 (IN+) (non-inverting input), used to receive the positive and negative terminals of the measured signal. This effectively suppresses common-mode noise, making it suitable for measurement scenarios with small signals and high noise levels. Pins 2 and 3 are both RG gain setting terminals. Connecting a resistor of a specific value between them sets the amplifier gain. The gain is 1x when open-circuited, and gradually increases as the external resistor value decreases. Pin 5 (V-) is the negative power supply terminal. It can be connected to a negative voltage for dual power supply and to the common ground for single power supply. Pin 6 (REF) is the reference voltage terminal, used to set the input voltage. The DC bias level of the output signal is usually connected to the common ground when powered by dual power supplies, allowing the output to swing around zero level. When powered by a single power supply, it is usually connected to the midpoint voltage of the power supply to avoid signal distortion. Pin 7 (OUT) is the signal output terminal, which can output the amplified signal. Its rail-to-rail output characteristic allows the signal swing to be close to the power supply voltage. Pin 8 (V+) is the positive power supply terminal, which can be connected to the positive voltage or the positive terminal of the power supply when powered by a single power supply. When using it, a decoupling capacitor should be added near the power supply pin to suppress noise and ensure stable operation of the amplifier. In addition, the zero-drift architecture of this chip can achieve extremely low input offset voltage and temperature drift, which is suitable for high-precision measurement scenarios. The bottom heat sink of the MSOP8 package can be connected to the V- terminal to improve heat dissipation performance and operating stability.

[0024] like Figure 6 As shown, the sensor head of this invention adopts a circular, integrated, dense structure with a diameter of 10mm. The overall structure features a layered layout with a central functional area and an outer thermal buffer zone. It consists of a substrate, a double-layer stress buffer layer, a K-type thermopile array, a Pt-Rh alloy transition layer, an anti-oxidation electrode, and a nanocrystalline alumina dense layer, arranged sequentially from the inside out. This clear hierarchical structure and high assembly precision effectively adapt to long-term high-temperature operating environments up to 600℃, while also accommodating confined installation conditions. A concentric annular gradient thermal groove is formed around the periphery of the sensor head. The groove parameters are uniformly 0.5–1mm in width and 0.3–0.5mm in depth, evenly surrounding the central heat flow sensitive area. This overcomes the shortcomings of traditional sensors, such as uneven planar heat conduction and excessively rapid heat dissipation at the edges, confining heat flow to the central array area and significantly improving the uniformity and consistency of heat flow acquisition and response.

[0025] like Figure 7As shown, the double-layer stress buffer layer is made of high-temperature resistant alumina ceramic fiber material, with a single layer thickness controlled at 0.2–0.3 mm. It consists of an inner buffer layer and an outer buffer layer to form a composite buffer structure, with the layers tightly bonded without gaps. The inner buffer layer is completely bonded to the surface of the thermopile array substrate, which isolates the substrate stress and buffers thermal deformation. The outer buffer layer is close to the inner wall of the annular gradient heat conduction groove, which can offset the local thermal stress generated by the groove structure under high temperature conditions. The double-layer gradient buffer structure absorbs the deformation caused by high temperature thermal expansion and contraction, avoiding overall warping, cracking, and structural detachment of the sensitive head, and ensuring long-term high-temperature working stability. Figure 10 This is a top view of the sensor structure of the present invention. Figure 11 This is a cross-sectional view of the sensor of the present invention.

[0026] The 8mm circular area at the center of the sensing head is the core heat flow sensing area. Inside, an array of 80 pairs of nickel-chromium / nickel-silicon K-type thermocouples connected in series is evenly arranged in a ring. The array is well-organized and uniformly spaced, ensuring the linearity and consistency of the heat flow signal acquisition. Figure 9 As shown, the K-type thermocouple of this invention uses a nickel-chromium-nickel-silicon functional thin film. A 0.05–0.1 mm thick Pt–Rh alloy transition layer is added between the array electrodes and the ceramic substrate, and uniformly deposited using a magnetron sputtering process. This significantly improves the adhesion between the electrodes and the insulating substrate, solving the problems of electrode delamination, peeling, and poor contact at high temperatures. The electrode body is made of a nickel-chromium-gold alloy material resistant to high-temperature oxidation, possessing excellent high-temperature oxidation and corrosion resistance, and can withstand long-term high-temperature conditions up to 600°C, eliminating electrode oxidation failure and signal drift problems. The outermost layer of the sensing head is covered with a 0.1–0.2 mm thick dense nanocrystalline alumina layer, prepared using a plasma spraying process. The coating has dense and uniform grains, is free of pores and cracks, and has insulation protection, corrosion resistance, and electromagnetic interference resistance properties, without hindering heat flow penetration and conduction, thus balancing protection performance and measurement accuracy.

[0027] The overall fabrication process of the sensing head of this invention is simple and suitable for mass production of devices. The specific fabrication steps are as follows: First, a high-precision CNC micro-carving machine is used to groove the ceramic substrate, precisely machining a concentric annular gradient heat-conducting groove. The substrate is then ground, degreased, and pre-treated with high-temperature annealing to eliminate residual stress from machining. Subsequently, an inner and outer high-temperature resistant ceramic fiber buffer layer is sequentially laid through a bonding and curing process, and then high-temperature compacted and shaped to ensure that the buffer layer is tightly bonded to the substrate and the groove wall. Next, a Pt-Rh alloy transition layer is uniformly deposited on the functional area of ​​the substrate using magnetron sputtering equipment, with precise control of the coating thickness. Then, multiple pairs of K-type thermocouples are uniformly arranged in an annular pattern and welded in series to form a thermopile array through a precision spot welding process. Antioxidant alloy electrodes are then assembled and compacted and cured. Finally, a nanocrystalline alumina coating is sprayed onto the overall outer surface of the sensing head using plasma spraying technology, cured at a constant temperature, and polished to complete the overall structural encapsulation. The final sensing head has a compact structure, stable layers, and excellent high-temperature resistance, which can meet the requirements of long-term high-precision high-temperature heat flow monitoring.

[0028] The TPA1286U-SO1R device has a common-mode rejection ratio better than 100dB at room temperature, an adjustable gain range of 1 to 1000 times, a gain error better than 0.002%, an input offset voltage temperature drift of less than 0.8μV / ℃, excellent noise performance, and good high-temperature stability, which can meet the weak signal amplification requirements in a high-temperature environment of 600℃. The system is equipped with an SGM2203-5.0YK3LG / TR low-dropout linear regulator, with an input voltage range of 2.5-12V and a stable output voltage of 5.0V with an output voltage accuracy error of less than 1%. It features built-in thermal shutdown and short-circuit protection, providing stable power to the instrumentation amplifier and peripheral devices. A TL431 precision reference voltage source is used, providing a stable output reference voltage (typically 2.5V) for the instrumentation amplifier. A CSE7761-SSOP16 AD converter chip is selected, with a conversion rate ≥15SPS and a linearity error ≤±0.002%FS, converting the amplified analog signal into a digital signal. High-precision metal film resistors (accuracy ≤0.1%, temperature coefficient ≤5ppm / ℃) and high-frequency ceramic capacitors (capacitance stability ≤1%) are used.

[0029] like Figure 3As shown, two SGM8622XMS / TR high-precision operational amplifiers are selected, packaged in SOP8. Pin 1 is the output terminal OUTA of channel A, which outputs the amplified signal from channel A. Pins 2 and 3 are the inverting input terminal -INA and the non-inverting input terminal +INA of channel A, respectively, used to input the signal to be amplified and, in conjunction with an external feedback network, to achieve different amplification functions. Pin 4 is the negative power supply terminal -Vs, which can be connected to a negative voltage when powered by dual power supplies, and grounded when powered by a single power supply. Pins 5 and 6 are the non-inverting input terminal +INB and the inverting input terminal -INB of channel B, respectively, forming the signal input pair for channel B. Pin 7 is the output terminal OUTB of channel B, outputting the amplified signal from channel B. Pin 8 is the positive power supply terminal +Vs, connected to the system positive power supply. A decoupling capacitor should be placed near the power supply pin to suppress power supply noise and ensure stable chip operation. One amplifier is used to build a low-pass filter circuit (cutoff frequency 10Hz) to filter out high-frequency interference, while the other is used as a voltage follower to isolate the load.

[0030] The signal amplification circuit is connected as follows: the output of the K-type thermopile is connected to the non-inverting and inverting inputs of the TPA1286U-SO1R instrumentation amplifier, respectively. An external high-precision metal film resistor sets the amplification gain to 625 times, amplifying the 8mV weak signal to a 0-5V standard signal with an amplification gain error ≤0.1%. The SGM2203-5.0YK3LG / TR voltage regulator chip is connected to an external power supply, outputting a stable 5.0V voltage. A parallel filter capacitor in the power supply circuit filters out power ripple. The input of the CSE7761-SSOP16AD chip is connected to the output of the instrumentation amplifier, converting the analog signal into a digital signal before transmitting it to the subsequent signal processing unit. The dynamic gain calibration unit works in conjunction with the signal amplification circuit, monitoring the stability of the amplified signal in real time and automatically adjusting the amplification gain with a calibration error ≤±0.05%, avoiding gain drift from affecting measurement accuracy.

[0031] 2. Temperature measuring circuit: such as Figure 8As shown, this device achieves accurate measurement of ambient temperatures from 0 to 600°C based on the thermoelectric potential difference principle of a K-type thermocouple. It comprises a K-type temperature-sensing thermocouple, a cold junction compensation module, a signal preamplifier module, an anti-interference filter module, an ADC analog-to-digital converter module, and a digital signal output module, all connected in sequence. The K-type temperature-sensing thermocouple uses an industrial-grade WRNK-191 armored thermocouple probe. It features a metal-armored encapsulation structure, with a standard outer diameter of Φ3~6mm and an overall length of 200~300mm, making it a slender, integrated armored component. This structure offers advantages such as resistance to bending, vibration, high temperatures, and oxidation, making it suitable for the confined installation space and vibration conditions of high-temperature industrial equipment. The thermocouple body uses nickel-chromium-nickel-silicon temperature sensing core material, with an overall temperature measurement range of -200~1000℃, fully covering the sensor's rated operating range of 0~600℃; the internal insulation layer uses high-purity high-temperature ceramic material, with an overall temperature resistance of ≥800℃, far exceeding the equipment's normal operating temperature, and can maintain stable insulation performance for a long time, avoiding high-temperature leakage and signal crosstalk problems.

[0032] This invention features a dedicated installation and spatial arrangement structure for the armored thermocouple: the hot junction of the thermocouple is installed using a close-fitting positioning method, with the tip polished and tightly fitted to the outer temperature measurement reference surface of the sensor's sensitive head, ensuring a gapless fit between the hot junction and the temperature measurement area. This enables synchronous and lag-free acquisition of ambient temperature and the temperature of the sensitive head itself. The armored probe body is neatly arranged along the pre-reserved wiring groove inside the encapsulation shell, secured with high-temperature resistant fasteners to prevent probe suspension or movement that could cause measurement errors and structural wear. The cold junction of the thermocouple is led out through high-temperature resistant insulated leads and neatly stored in the low-temperature zone inside the encapsulation shell, away from the high-temperature sensitive head and heating devices. This ensures stable cold junction temperature and prevents direct radiation from the measured high-temperature heat source. The cold junction leads are systematically connected to the downstream cold junction compensation module, providing a stable original temperature measurement signal for subsequent accurate cold junction compensation and temperature signal conditioning. This arrangement is compact and features neat wiring, ensuring both rapid temperature response and temperature measurement consistency, while effectively improving the overall structural reliability and long-term operational stability of the sensor.

[0033] like Figure 4As shown, the cold junction compensation module uses the Changxin Micro LDC31855 chip, the NTC thermistor MF58-103F3950, and the SGM8541 operational amplifier from Sanbang Microelectronics. The SGM8541 is a single-channel operational amplifier in a small 5-pin package. Pin 1 is the signal output terminal OUT, outputting the amplified signal; pin 2 is the negative power supply terminal -Vs, connected to a negative voltage for dual power supply and grounded for single power supply; pin 3 is the non-inverting input terminal +IN; pin 4 is the inverting input terminal -IN, forming a differential input pair that can input the signal to be amplified and set the circuit gain and operating mode via external feedback components; pin 5 is the positive power supply terminal +Vs, connected to the system positive power supply. This chip typically features low power consumption and low offset, making it suitable for signal conditioning scenarios with high space and power consumption requirements. The LDC31855 has a built-in cold junction temperature sensor that collects the ambient temperature of the cold junction in real time and automatically compensates for the cold junction thermoelectric potential drift. The NTC thermistor collects minute temperature changes at the cold junction and converts them into voltage compensation signals through the SGM8541 op-amp. After superposition, the cold junction compensation error is ≤ ±0.1℃.

[0034] The signal preamplifier module and the heat flux amplifier circuit share the TPA1286U-SO1R instrumentation amplifier and TL431 precision voltage reference, with a 50x amplification gain, amplifying the thermoelectric potential of 24.90mV at 600℃ to a standard analog signal of 1.245V. The anti-interference filtering module uses an RC low-pass filter circuit with a cutoff frequency of 10Hz to filter out high-frequency electromagnetic interference and power supply noise, with a response time ≤50ms. Figure 5 As shown, the ADC analog-to-digital conversion module uses the Chipsea CS1237 16-bit high-precision chip, packaged in SOP8. Pin 1 is the reference voltage input REFIN, which can be connected to an external reference voltage to provide a stable reference for analog-to-digital conversion; pin 2 is the ground terminal GND, providing a common ground potential for the chip; pins 3 and 4 are the analog signal differential input terminals AINN and AINP, respectively, used to input the analog signal to be converted; pin 5 is the serial clock input SCLK, receiving the clock signal sent by the master control device to synchronize the data transmission process; pin 6 is the data ready and serial data output terminal DRDY / DOUT, which can output the converted digital signal and provide feedback on the data ready status; pin 7 is the power supply terminal VDD, connected to the system positive power supply; pin 8 is the reference voltage output terminal REFOUT, which can output the internal reference voltage for use by external circuits. When using it, filtering components need to be configured for the power supply and reference pins to ensure conversion accuracy; its conversion rate is ≥10SPS, linearity error is ≤±0.001%FS, and resolution reaches 0.01℃, converting analog temperature signals into digital signals.

[0035] The digital signal output module uses the CR Microelectronics 74HC595 chip, a standard 8-bit serial-in parallel-out shift register. This chip lacks a hardware SPI transceiver core and does not support native SPI bus bidirectional communication. It is primarily used for microcontroller I / O port expansion and serial-to-parallel data conversion, adapting to the device's requirements for regularized digital signal output. Addressing the inherent limitations of traditional bare-board serial signals—such as inability to achieve stable transmission over 5m and susceptibility to signal distortion, packet loss, and interference—this invention adds a long-line drive optimization circuit. A high-speed bus buffer driver chip and a differential signal conversion circuit are matched at the 74HC595 output end to convert the single-ended parallel digital signal into an anti-interference differential transmission signal. Simultaneously, a 100Ω impedance matching current-limiting resistor at the output end eliminates long-line transmission reflection interference, solving the problem of poor integrity in long-distance transmission of single-ended serial signals. This achieves attenuation-free, highly reliable digital signal transmission over distances of ≥5m, balancing port expansion functionality with long-distance signal output requirements.

[0036] The dual-channel differential temperature drift real-time cancellation module collects temperature drift signals from the thermopile array and the temperature measurement circuit, respectively. Relying on the built-in STC8A8K64S4A12 eight-bit microcontroller, it runs the on-chip dual-channel differential temperature drift real-time cancellation algorithm to perform real-time differential calculation and dynamic compensation on the two temperature drift signals. The cancellation error is ≤±0.02℃. It can dynamically correct the measurement deviation caused by device temperature drift and ambient temperature fluctuations under high-temperature conditions, effectively solving the technical problems of difficult high-temperature drift suppression and poor measurement stability of traditional heat flow sensors.

[0037] 3. Filtering Module: This module consists of analog and digital filters, working in conjunction with the signal amplification and temperature measurement circuits. The analog filter comprises a low-pass filter circuit built with the SGM8622XMS / TR operational amplifier in the signal amplification circuit and an RC low-pass filter circuit in the temperature measurement circuit. Both have a cutoff frequency of 10Hz, working together to filter out high-frequency interference, power supply noise, and environmental noise from the entire circuit. A 10nF ceramic capacitor is connected in parallel at the signal input of the temperature measurement circuit to further filter out high-frequency noise in the input signal. The digital filter uses a moving average filtering software algorithm at the output of the LDC31855 chip and after the heat flux signal ADC conversion. The filtering window is set to 8 sampling points, balancing filtering effect and response speed to ensure signal stability.

[0038] 4. Voltage Regulated Power Supply Module: Two sets of voltage regulators are used to power the entire circuit. One set is an SGM2203-5.0YK3LG / TR low dropout linear regulator, equipped with an electrolytic capacitor (100μF) and a ceramic capacitor (0.1μF) to form a filter circuit, outputting a stable 5.0V voltage to power the signal amplification circuit, the dual-channel differential temperature drift real-time cancellation module, and the dynamic gain calibration unit; the other set is a Lixin Microelectronics ET631XX series low dropout linear regulator, equipped with the same specification filter capacitor, outputting a stable 3.3V voltage to power the temperature measurement circuit and the cold junction compensation module; the external 5V power supply is connected to the two sets of voltage regulator chips respectively to ensure that the output voltage fluctuation of the SGM2203 is ≤±0.01V and the output voltage fluctuation of the ET631XX is ≤±0.01V. The two power supply circuits do not interfere with each other, ensuring the measurement accuracy of the circuit.

[0039] The enclosure is made of 310S stainless steel high-temperature alloy, with a high temperature resistance of ≥800℃, and can be adapted to a 600℃ working environment. The overall size of the enclosure matches the sensing head and signal processing circuit, and the surface is treated with anti-corrosion and insulation to avoid environmental corrosion and electromagnetic interference. The enclosure has a built-in temperature field homogenization isolation layer, made of high-temperature resistant insulation cotton with a thickness of 0.5-1mm, which is set to fit the inner wall of the enclosure to equalize the internal temperature field and reduce the impact of external temperature fluctuations on internal components. The enclosure has reserved aviation plugs as signal output interface and power supply interface, which are used to transmit the processed heat flow and temperature signals and to connect to external power supply, respectively.

[0040] The core performance parameters of this sensor are as follows: heat flux measurement range 0-1.2MW / m², measurement accuracy ±0.5%FS; temperature measurement range 0-600℃, measurement accuracy ±0.5℃; 1MW / m² heat flux density corresponds to 8mV output voltage, linearity ±0.1%FS; signal response time ≤50ms; operating current ≤50mA; signal drift ≤0.01mV / h after 72 hours of continuous operation. All core components are proprietary, resulting in low cost and a stable supply chain. It can be widely used in multiple industrial fields such as metallurgy, aerospace, and high-temperature equipment monitoring.

[0041] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A high-precision, high-stability K-type thermopile heat flux sensor, comprising a sensing head, a signal processing circuit, and a packaging shell, wherein the signal processing circuit is electrically connected to the sensing head and integrated inside the packaging shell; characterized in that: The sensitive head is circular, with concentric ring-shaped gradient heat conduction grooves on the periphery to constrain the heat flow to be concentrated and conducted to the central region; it is equipped with a double-layer stress buffer layer to buffer high-temperature thermal stress. The central region of the sensitive head is provided with a K-type thermopile array composed of multiple pairs of K-type thermocouples connected in series. A platinum-rhodium alloy transition layer to enhance the bonding force is provided between the electrodes of the K-type thermocouples and the substrate. The electrodes are made of high-temperature oxidation resistant material and the surface is covered with a dense layer of nanocrystalline alumina. The signal processing circuit includes a signal amplification circuit, a temperature measurement circuit, a cold junction compensation module, a filtering module, a voltage regulation power supply module, a dual-channel differential temperature drift real-time cancellation module, a dynamic gain calibration unit, and a microcontroller. The microcontroller, as the digital computing core, performs dual-channel differential temperature drift real-time cancellation and dynamic gain calibration, performs differential operations on the temperature drift signals of the K-type thermopile array and the temperature measurement circuit, and adaptively adjusts the amplification gain according to the deviation of the amplified signal. The encapsulation shell is provided with a temperature field homogenization isolation layer.

2. The high-precision, high-stability K-type thermopile heat flux sensor according to claim 1, characterized in that, The double-layer stress buffer layer includes an inner buffer layer and an outer buffer layer. The inner buffer layer is attached to the substrate of the K-type thermopile array, and the outer buffer layer is connected to the annular gradient heat conduction groove, forming a composite buffer structure.

3. The high-precision, high-stability K-type thermopile heat flux sensor according to claim 1, characterized in that, The electrode includes a nickel-chromium alloy layer and a gold alloy layer bonded to the surface of the nickel-chromium alloy layer. The platinum-rhodium alloy transition layer is deposited between the substrate and the electrode using a magnetron sputtering process.

4. The high-precision, high-stability K-type thermopile heat flux sensor according to claim 1, characterized in that, Multiple pairs of the K-type thermocouples are evenly arranged in a ring array within the central region, and the total output voltage of the K-type thermopile array is linearly related to the measured heat flux density.

5. The high-precision, high-stability K-type thermopile heat flux sensor according to claim 1, characterized in that, The microcontroller connects to the first analog-to-digital converter (ADC) chip and the second ADC chip via two independent analog SPI communication interfaces. The first ADC chip is located at the rear end of the signal amplification circuit, and the second ADC chip is located at the rear end of the temperature measurement circuit, enabling synchronous data acquisition between the heat flow channel and the temperature channel. The microcontroller's on-chip non-volatile memory stores the temperature drift coupling coefficient, gain correction coefficient, and segmented temperature compensation curve, which are automatically read and initialized upon power-up.

6. The high-precision, high-stability K-type thermopile heat flux sensor according to claim 1, characterized in that, The dynamic gain calibration unit monitors the deviation of the amplified signal from the theoretical standard value in real time. When the deviation of multiple consecutive samples exceeds the calibration trigger criterion, the reference gain is corrected by the gain correction coefficient. The corrected gain is limited to the protection range. After the deviation converges, the optimal gain parameter is fixed to the on-chip memory of the microcontroller.

7. The high-precision, high-stability K-type thermopile heat flux sensor according to claim 1, characterized in that, The dual-channel differential temperature drift real-time compensation module compensates for the temperature drift of the output voltage of the K-type thermopile array based on the real-time temperature difference between the cold junction of the thermopile channel and the temperature measurement channel and the temperature drift coupling coefficient. When the error of a single compensation calculation exceeds the resampling criterion and continuous recalculation still exceeds the limit, the segmented temperature drift coupling coefficient is automatically updated.

8. The high-precision, high-stability K-type thermopile heat flux sensor according to claim 1, characterized in that, The temperature measuring circuit includes a K-type thermocouple, a cold junction compensation module, a signal preamplifier module, an anti-interference filter module, an ADC analog-to-digital converter module, and a digital signal output module, which are connected in sequence; the hot junction of the K-type thermocouple is attached to the surface of the sensing head.

9. The high-precision, high-stability K-type thermopile heat flux sensor according to claim 8, characterized in that, The K-type temperature measuring thermocouple is an armored thermocouple. The main body of the armored thermocouple is arranged and fixed along the wiring groove inside the encapsulation shell. The cold junction lead is stored in the low-temperature zone inside the encapsulation shell, and the low-temperature zone is far away from the sensing head.

10. The high-precision, high-stability K-type thermopile heat flux sensor according to claim 1, characterized in that, The filtering module includes an analog filtering submodule and a digital filtering submodule. The analog filtering submodule includes a low-pass filter circuit located in the signal amplification circuit. The digital filtering submodule uses a moving average filtering algorithm run by the microcontroller and has a built-in jump interference rejection criterion. The voltage regulation power supply module includes a first power supply branch and a second power supply branch that are independent of each other. The first power supply branch supplies power to the signal amplification circuit, the dual-channel differential temperature drift real-time cancellation module, and the dynamic gain calibration unit. The second power supply branch supplies power to the temperature measurement circuit and the cold junction compensation module.

Images