PT100 temperature measurement circuit and method for multi-gear current fitting resistance measurement and ADC online calibration

Abstract

The invention discloses a PT100 temperature measurement circuit and method for multi-gear current fitting resistance measurement and ADC online calibration. In order to solve the problem that a PT100 precise temperature measurement system does not have the self-checking capability, an ADC calibration reference resistance network formed by a plurality of precise resistors is utilized, a plurality of reference voltage points are provided, gain drift, offset change and linear deviation can be detected in real time and compensated, and online self-calibration and state monitoring of an ADC in the whole experiment process are achieved. Through error compensation, the overall error is reduced by about 70%. For a PT100 precision temperature measurement system, a fixed current value is usually adopted to measure the resistance, which is easy to interfere; and due to the influence of self-heating factors, the finally measured resistance value is not accurate, and the like. According to the method, multi-gear constant current excitation and a least square resistance fitting method are combined, noise and nonlinear errors are effectively restrained, the resistance measurement precision and the temperature measurement precision of PT100 are greatly improved, and the temperature deviation is controlled within the range of + / -0.03 DEG C to 0.04 DEG C.

Description

Technical Field

[0001] This invention belongs to the field of high-precision temperature measurement technology, and proposes a PT100 temperature measurement circuit and method for multi-level current fitting resistance measurement and online ADC calibration. Background Technology

[0002] Due to its excellent linearity, long-term stability, and high repeatability, the PT100 platinum resistance thermometer is widely used in various precision temperature measurement scenarios and is one of the most commonly used temperature sensors in industrial temperature measurement systems. In testing applications requiring quantitative analysis of temperature changes, the measurement accuracy of the PT100 often directly determines the reliability of subsequent calculations or judgments.

[0003] For example, in safety testing of fine chemical reactions, temperature is not only a fundamental quantity characterizing the reaction process, but also a core parameter used to calculate the reaction exothermic rate, predict adiabatic temperature rise, and assess the risk of thermal runaway. Even a small temperature measurement error can lead to significant deviations in heat flow, kinetic parameters, or hazard assessment results. Therefore, such equipment demands higher accuracy in temperature measurement, and the high-precision PT100 temperature sensing circuit is a key component ensuring the reliability of temperature testing.

[0004] However, the existing PT100 precision temperature measurement system still faces several technical bottlenecks in practical use:

[0005] (1) Lack of effective online self-testing capability: The offset, gain and linearity of the analog-to-digital converter (ADC) may drift under long-term temperature cycling. Traditional systems mostly rely on offline standard resistance boxes for calibration, which cannot detect the ADC status in real time before the test begins. When the ADC drifts slightly, it is difficult for the operator to identify it from the temperature data.

[0006] (2) Insufficient anti-interference capability: In environments containing high-power motors, heating drivers and other strong electromagnetic interference sources, analog signal links are easily affected by common-mode noise, causing temperature signal fluctuations.

[0007] (3) The accuracy of fixed current temperature measurement is limited: a single constant current excitation will aggravate the self-heating effect, amplify the noise effect and reduce the linearity in different temperature ranges, thereby reducing the accuracy of PT100 resistance measurement. Summary of the Invention

[0008] To address the problems in existing PT100 temperature measurement, such as the inability to monitor ADC drift online, insufficient anti-interference capability, and limited accuracy of single constant current measurement, this invention aims to propose a PT100 precision temperature measurement circuit and method that uses multi-level current fitting to determine PT100 resistance and combines it with online ADC calibration and compensation. This method enables online self-checking and compensation of the ADC status, and improves the accuracy and stability of PT100 resistance measurement through multi-level constant current measurement and fitting strategies, thereby achieving high-precision and high-reliability temperature measurement.

[0009] To achieve the above objectives, a first aspect of the present invention provides a PT100 precision temperature measurement circuit with multi-range current fitting resistance measurement and online ADC calibration, comprising:

[0010] An ADC calibration reference resistor network is used to calibrate the ADC's offset error and actual gain online, and to monitor the ADC's linear deviation online.

[0011] The ADC unit, with its input optionally connected to either the PT100 measurement channel or the resistor network, is used to ensure that the PT100 measurement and ADC calibration processes use a unified sampling path to eliminate channel bias.

[0012] The multi-level constant current excitation unit is used to sequentially output multiple current levels to the PT100 to obtain multi-point current-voltage characteristic datasets.

[0013] The channel switcher is used to switch the input path of the ADC between ADC calibration mode and PT100 measurement mode.

[0014] The control unit (MCU) is used to acquire the reference node voltage and perform online ADC calibration; in measurement mode, it acquires the voltage values ​​corresponding to multiple constant currents and calculates the resistance value of the PT100 based on the dataset.

[0015] A second aspect of the present invention also provides a PT100 temperature measurement method for multi-level current fitting resistance measurement and online ADC calibration, applied to the above-mentioned circuit, comprising the following steps:

[0016] S1: Perform an online self-test on the ADC unit to ensure that the ADC's offset, gain, and linearity are all in normal condition before entering the PT100 measurement.

[0017] S2: Switch to the PT100 measurement channel, output a single constant current and collect the corresponding voltage value. Determine whether to enter the PT100 measurement mode based on the temperature change rate: Apply multiple constant current levels to the PT100 in sequence and collect the corresponding voltage values ​​to form a multi-point current-voltage feature dataset.

[0018] S3: Construct a linear model based on the obtained dataset, and calculate the optimal resistance estimate of PT100 using the least squares method; substitute the fitted resistance value into the resistance-temperature characteristic relationship of PT100 to obtain the final temperature value.

[0019] Based on the above technical content, the present invention has the following beneficial effects:

[0020] 1. To address the issue that the PT100 precision temperature measurement system lacks self-testing capability, this invention utilizes an ADC calibration reference resistor network composed of multiple high-precision resistors to provide multiple reference voltage points. This network can detect and compensate for gain drift, offset changes, and linear deviations in real time, achieving online self-calibration and status monitoring of the ADC throughout the entire experimental process. Through error compensation, the overall error is reduced by approximately 70%.

[0021] 2. The PT100 precision temperature measurement system often uses a fixed current value to measure the resistance, which is susceptible to interference and inaccurate due to self-heating. This invention combines multi-stage constant current excitation with a least-squares resistance fitting method to effectively suppress noise and nonlinear errors, significantly improving the resistance and temperature measurement accuracy of the PT100, with temperature deviation controlled within ±0.03~0.04℃.

[0022] In summary, this invention has a compact structure and features online self-diagnosis, error compensation, and high precision. It can be applied to scenarios such as thermal risk assessment of reaction vessels, industrial process control, and precision temperature recording. Attached Figure Description

[0023] Figure 1 This is a layout diagram of a PT100 temperature measurement circuit that combines multi-stage current fitting resistance measurement with online ADC calibration.

[0024] Figure 2 This is a step logic diagram of a PT100 temperature measurement method that combines multi-range current fitting resistance measurement with online ADC calibration.

[0025] Figure 3 This is a comparison chart of the temperature error distribution between single-stage constant current measurement and multi-stage constant current fitting.

[0026] Figure 4 This is a comparison chart of temperature curves between the uncalibrated single-stage constant current output and the fitted multi-stage output. Detailed Implementation

[0027] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0028] This application provides a PT100 temperature measurement circuit with multi-range current fitting resistance measurement and online ADC calibration, including:

[0029] (1) ADC calibration reference resistor network:

[0030] As the core reference for system self-testing, this network consists of several high-precision, low-temperature drift precision resistors connected in series between the reference voltage source and ground, configured as a voltage divider circuit. Its function is to provide the ADC with a series of gradient reference voltage signals covering its measurement range. Unlike traditional offline calibration, this network provides standard signals with known theoretical values ​​during system operation intervals, allowing the system to compare in real time to calculate the ADC's current zero-point offset, gain error, and integral nonlinearity (INL), thereby achieving online quantitative evaluation of the signal chain state.

[0031] (2) High-precision ADC unit (unified acquisition link):

[0032] A high-resolution 24-bit Δ-Σ architecture analog-to-digital converter was employed to construct a unified acquisition chain, meaning that measurement and calibration share the same physical channel. This unit acquires the temperature sensing signal from the PT100 and the calibration signal from the reference resistor network separately through front-end channel switching. This design eliminates inter-channel difference errors introduced by independently setting calibration channels, ensuring that calibration parameters accurately reflect the true drift state of the measurement channel.

[0033] (3) Multi-stage constant current excitation unit:

[0034] It replaces the traditional single constant current source and is configured as a programmable stepped current output mode (the multi-level constant current excitation unit is a built-in function of the analog-to-digital converter chip of the aforementioned high-precision ADC unit). Within a single temperature measurement cycle, it can continuously and rapidly output multiple stable excitation currents of different magnitudes to the PT100 load (e.g., increasing in a preset sequence). The aim is to excite the voltage response of the PT100 at different power levels, thereby obtaining a set of current-voltage (IV) characteristic datasets containing multi-point coordinates, providing a physical data foundation for subsequent mathematical algorithms to eliminate interference and random noise.

[0035] (4) Channel switcher:

[0036] This is used to perform physical isolation switching between PT100 measurement mode and ADC calibration mode. Specifically, in self-test mode, the PT100 input is completely disconnected, and the ADC input is locked to the reference resistor network to avoid external sensor noise interfering with the calibration reference. In measurement mode, the amplified PT100 differential signal is losslessly imported into the ADC to ensure that the signal paths in different operating modes do not interfere with each other.

[0037] (5) Control Unit MCU:

[0038] As the core of the system's computation and control, it incorporates a dual error correction algorithm. On one hand, based on the sampling data of the reference resistor network, it dynamically updates the ADC's error compensation coefficients and subtracts the drift amount in real time from each frame of raw sampling data. On the other hand, in measurement mode, it synchronously controls the scanning timing of multiple constant current sources, acquires multiple sets of corresponding voltage values, and performs least squares fitting calculations. By constructing a resistance-current linear model, the MCU can extract the statistically optimal resistance estimate from multi-point data, thereby effectively suppressing white noise and nonlinear errors in single constant current measurements.

[0039] This application also provides a PT100 temperature measurement method with multi-level current fitting resistance measurement and online ADC calibration, applied to the circuit provided in the above embodiments, including the following steps:

[0040] S1: Perform an online self-test on the ADC unit to ensure that the ADC's offset, gain, and linearity are all in normal condition before entering the PT100 measurement.

[0041] S2: Switch to the PT100 measurement channel, output a single constant current and collect the corresponding voltage value. Determine whether to enter the PT100 measurement mode based on the temperature change rate: Apply multiple constant current levels to the PT100 in sequence and collect the corresponding voltage values ​​to form a multi-point current-voltage feature dataset.

[0042] S3: Construct a linear model based on the obtained dataset, and calculate the optimal resistance estimate of PT100 using the least squares method; substitute the fitted resistance value into the resistance-temperature characteristic relationship of PT100 to obtain the final temperature value.

[0043] Furthermore, S1 includes:

[0044] The control unit controls the multi-channel analog switch to connect the differential input terminal of the ADC to each reference voltage node of the ADC calibration reference resistor network in sequence, while disconnecting the PT100 measurement channel.

[0045] Set the voltage node ratio of the ADC calibration reference resistor network; control the ADC to sequentially acquire the actual voltage values ​​of the voltage nodes, and calculate the corresponding theoretical voltage values ​​based on the resistance ratio;

[0046] Based on the calculation results, the multi-point least squares method is used to fit and compensate the actual gain and offset error of the ADC, while monitoring the linear deviation and issuing an early warning when it exceeds the limit.

[0047] Furthermore, S2 includes:

[0048] S2.1: Switch to the PT100 measurement channel, output a single-level constant current and acquire the corresponding voltage value, and calculate the temperature change rate.

[0049] S2.2: Determine whether the temperature change is stable based on the temperature change rate. If it is stable, proceed to step S2.3.

[0050] S2.3: Apply multiple constant current levels to the PT100 sequentially and collect the corresponding voltage values ​​to form a multi-point current-voltage feature dataset.

[0051] Furthermore, S2.1 specifically includes:

[0052] The control unit controls the channel switcher to select the PT100 measurement channel, and the ADC acquires the PT100 voltage signal at a fixed sampling frequency.

[0053] The control unit calculates the resistance value of PT100 based on the proportional measurement relationship; according to the IEC standard PT100 equation, the resistance value is converted into temperature data to obtain the real-time temperature and calculate the temperature change rate.

[0054] Furthermore, S2.3 specifically includes:

[0055] The multi-level constant current excitation unit is controlled to switch the output level sequentially according to the preset current sequence, and the corresponding voltage signal is collected after the current output of each level stabilizes.

[0056] For each output current level, the voltage drop across the PT100 resistor is collected through the PT100 temperature measurement channel;

[0057] By employing the proportional resistance measurement method and based on the correspondence between the digital and analog quantities converted by the ADC, the gain brought by the pre-amplifier is introduced to derive the digital voltage quantity after ADC sampling and conversion, thus forming a multi-point current-voltage characteristic dataset.

[0058] Furthermore, S3 includes:

[0059] S3.1: Establish a linear model, construct a least squares objective function based on Ohm's law and the multi-point current-voltage characteristic dataset, and calculate the best fitting resistance value for PT100.

[0060] S3.2: Substitute the compensated resistance value into the standard PT100 resistance-temperature relationship formula to calculate the final temperature value.

[0061] S3.3: The control unit outputs the final temperature data and writes it to the data buffer or sends it to the host computer; after completing this temperature measurement cycle, it returns to step S2 and performs temperature change judgment again to form a dynamic and adaptive temperature measurement process.

[0062] like Figure 1As shown, based on the same circuit setup as the above embodiments, this verification example provides a specific implementation of the circuit, which consists of: an analog power supply 1, a four-wire PT100 resistor pluggable terminal block 2, a precision reference resistor 3, a precision resistor network 4, a precision instrumentation amplifier 5, a reference source 6, a multiplexer 7, a 24-bit Δ-Σ analog-to-digital converter 8, an optocoupler isolation chip 9, a digital power supply 10, and an MCU 11.

[0063] The following section details each circuit component and its connection method:

[0064] 1. Analog power supply

[0065] The ADP7118 chip is selected to provide power to the analog circuit section (including the precision instrumentation amplifier 5, the reference source 6, and the 24-bit Δ-Σ analog-to-digital converter 8, etc.). It adopts a single-point grounding design to isolate it from the digital power supply 10, and is used to provide a stable 5V analog power supply.

[0066] 2. Four-wire PT100 resistor pluggable terminal block

[0067] A suitable corrosion-resistant platinum resistance thermometer is selected. The PT100 is a four-wire Kelvin connection, and the voltage across the resistor is output to the input side of the precision instrumentation amplifier 5 in the form of a differential signal.

[0068] 3. Precision reference resistor

[0069] A 1.62KΩ precision resistor (accuracy ±0.01%, low drift) is selected and connected in series with the PT100 to the built-in IDAC constant current source of the 24-bit Δ-Σ analog-to-digital converter 8. It is connected through the Kelvin interface REF0 and REF1 of the ADC chip. It serves as the reference element for proportional measurement, and the stable voltage across its two ends provides a reference voltage for ADC conversion. The accurate calculation of the PT100 resistance value is achieved through formula derivation.

[0070] 4. Precision resistor network

[0071] The network is constructed using high-precision, low-temperature drift resistors, specifically composed of resistor units with resistance values ​​of 1kΩ, 4kΩ, 3kΩ, and 2kΩ connected in series (all with an accuracy of ±0.01%). This network is connected to the input of a 24-bit Δ-Σ analog-to-digital converter 8 via a multiplexer 7, thereby achieving a multiplexed architecture that shares the same ADC chip with the temperature measurement channel.

[0072] In self-test mode, the precision resistor network outputs based on the voltage divider principle. , , and Four sets of fixed-ratio reference voltage signals (of which) (The voltage provided by the system reference source), the above signal is acquired by the ADC and then transmitted to the MCU for processing.

[0073] To overcome the bottleneck of traditional temperature measurement systems lacking online self-testing and achieve autonomous diagnosis of ADC linearity and accuracy, this application's verification example acquires a reference voltage and compares it with theoretical values ​​to calculate offset error, gain error, and linearity error (INL), promptly alerting to system anomalies and preventing invalid experimental data. Simultaneously, this application's verification example utilizes reference voltage data to correct the inherent drift of the ADC and amplifier, achieving offset voltage compensation without requiring offline standard resistance box calibration, ensuring experimental continuity. Sharing an ADC with the temperature measurement channel and switching via a channel selector simplifies circuit design while improving the consistency of calibration and temperature measurement.

[0074] 5. Precision Instrument Amplifier

[0075] The model selected is AD8422ARZ (rail-to-rail architecture). The input side is connected to the differential signal of terminal 2 of the four-wire PT100, and the output side is connected to the input of the multiplexer 7, which amplifies the weak voltage signal at both ends of the PT100 by 10 times.

[0076] 6. Reference Source

[0077] The model selected is ADA4528, which outputs a high-precision 2.5V reference voltage. It supplies a 24-bit Δ-Σ analog-to-digital converter 8 and a precision resistor network 4, respectively. Its maximum offset voltage is only 2.5uV and the offset voltage temperature drift is 0.015μV / °C.

[0078] 7. Multiplexer

[0079] The selected model is ADG709BRUZ (CMOS analog multiplexer). The input terminals are divided into two categories: ① the three sets of reference voltages + ground of the precision resistor network 4; ② the signal amplified by the precision instrumentation amplifier 5; the output terminal is connected to the analog input channel of the 24-bit Δ-Σ analog-to-digital converter 8; and the control terminal is connected to the address signal of the MCU11.

[0080] Achieve rapid, distortion-free switching between "self-test mode" and "temperature measurement mode": Upon device startup, it receives an MCU command to switch to resistor network 4. After completing the self-test, it switches to the temperature measurement channel. The channel switching flatness is up to 1.65Ω, avoiding signal distortion. Combined with precision resistor network 4, the same ADC can handle both self-test and temperature measurement, solving the complex design problem of requiring additional calibration circuits in traditional systems, simplifying the architecture while improving system integration.

[0081] The 8.24-bit delta-sigma analog-to-digital converter (ADC) uses the ADS124S08 (24-bit delta-sigma architecture), integrating 12 analog input channels, supporting single-ended / differential configurations, and featuring a built-in IDAC programmable constant current source. Signals are received via a multiplexer 7 (a resistor network reference voltage during self-test, and an amplified PT100 voltage during temperature measurement); the built-in constant current source output is connected to a precision reference resistor 3 in series with the PT100; the digital signal output is connected to an optocoupler isolation chip 9 via an SPI interface.

[0082] The constant current source features dynamic control, responding to MCU commands by outputting multiple current levels to achieve "dynamic multi-current scanning." Its 24-bit resolution, combined with reference source 6, ensures accurate digitization of weak voltage signals. It also integrates proportional measurement: receiving the voltage signal from precision reference resistor 3 via the REF0 / REF1 interface, it provides a benchmark for calculating the PT100 resistance value, eliminating errors caused by constant current source fluctuations.

[0083] 9. Optocoupler isolation chip

[0084] The model selected is ACSL6410 (multi-channel bidirectional optocoupler), which supports SPI communication isolation and is connected in series between the 24-bit Δ-Σ analog-to-digital converter and the MCU. The SPI digital signal of the ADC is transmitted to the MCU after being isolated by the optocoupler, and the control signal of the MCU is fed back to the ADC and the multiplexer 7 after being isolated by the optocoupler.

[0085] 10. Digital Power Supply

[0086] It supplies power to the digital circuit section (digital side of MCU11 and optocoupler isolation chip 9) and uses single-point grounding isolation, just like the analog power supply 1.

[0087] 11. MCU

[0088] The model selected is STM32F103ZET6, which receives the digital signal from the ADC through the optocoupler isolation chip 9.

[0089] Furthermore, it processes the self-test data of resistor network 4, calculates offset, gain, and linearity error, and stores correction parameters to compensate for ADC drift in real time during temperature measurement. It outputs control signals to multiplexer 7 (channel switching address) and 24-bit Δ-Σ analog-to-digital converter 8 (constant current source range control). It calculates the temperature change rate in real time and dynamically switches between ADC calibration mode and PT100 measurement mode based on this value, balancing response speed to sudden temperature changes with measurement accuracy in stable conditions. It receives voltage data corresponding to the three current ranges transmitted by the ADC, calculates and eliminates system errors in single-current measurements. Based on the above, it achieves closed-loop control of "self-test-calibration-temperature measurement," solving the problems of traditional systems requiring offline calibration and failing to detect faults in real time.

[0090] like Figure 2 As shown, based on the same concept as the methods in the above embodiments, this verification example also provides a PT100 temperature measurement method for multi-range current fitting resistance measurement and online ADC calibration, which is applied to... Figure 1 The circuit includes the following steps:

[0091] S1: ADC online calibration:

[0092] After the circuit is powered on, the control unit MCU first performs an online self-test on the high-precision ADC to ensure that the ADC's offset, gain, and linearity are all within normal limits before entering the PT100 measurement. This includes the following steps:

[0093] S1.1: The MCU controls the multi-channel analog switch to connect the differential input of the ADC to each reference voltage node of the ADC calibration reference resistor network in sequence, while disconnecting the PT100 measurement channel to avoid the temperature measurement signal from interfering with the calibration process.

[0094] S1.2: The calibration reference network consists of multiple precision resistors connected in series, forming multiple voltage nodes with fixed ratios. In this verification example, four proportional nodes are set up with ratios of 0.1. 0.5 0.8 and .

[0095] S1.3: The MCU controls the ADC to sequentially acquire the actual voltage values ​​of the four nodes mentioned above, denoted as M1, M2, M3, and M4 respectively; and calculates the corresponding theoretical voltage values ​​T1, T2, T3, and T4 based on the resistance ratio, where... , , , .

[0096] S1.4: To achieve higher calibration accuracy, this verification example uses multi-point least squares method to measure the actual gain of the ADC. With offset error The fitting model is as follows:

[0097]

[0098] By performing least squares calculations on the four reference points, we can obtain:

[0099]

[0100]

[0101] in, Let be the theoretical voltage value at the i-th calibration point. This corresponds to the measured voltage value; and These are the arithmetic mean of the theoretical voltage and the measured voltage, respectively.

[0102] Calculated and The predicted value for each theoretical node is obtained:

[0103]

[0104] Linear Deviation (LD) is defined as the maximum absolute error at each reference point, normalized to LSB (LSB (Least Significant Bit) is the smallest unit of resolution for an ADC):

[0105]

[0106] If the linear deviation exceeds the set threshold (e.g., 2-5 LSB), the MCU will prompt "ADC self-test failed" and prevent it from entering the subsequent temperature measurement stage.

[0107] S1.5: All subsequent PT100 measurement data must be based on... and After compensation, the corrected actual PT100 measurement data is obtained:

[0108]

[0109] S2: Single-range constant current rapid temperature measurement:

[0110] After the ADC self-test is completed, the MCU controls the channel switch to connect the ADC to the PT100 measurement channel, and outputs a single-level constant current (e.g., 0.5mA, which is used in this verification example, but the invention is not limited to this) through the ADC's built-in constant current source to perform initial measurement of the PT100 resistance. Specifically:

[0111] Select the PT100 measurement channel, and connect the PT100 and the reference resistor RREF in series in the constant current source output path; the ADC acquires the PT100 voltage signal at a sampling frequency of 1Hz.

[0112] The MCU calculates the resistance value of the PT100 based on the proportional measurement relationship; according to the IEC standard PT100 equation, the resistance value is converted into temperature data to obtain the real-time temperature T; the temperature change rate is calculated by the difference between the temperature data obtained from the previous and current measurements.

[0113]

[0114] in, The temperature data obtained in this measurement Temperature data obtained from the previous measurement; The interval for obtaining temperature data (i.e., the sampling period, is set to 1 second in this verification example).

[0115] S3: Pattern Detection:

[0116] This step is used to determine whether the current reaction temperature is in a rapidly changing range, in order to decide whether to proceed to the multi-current scanning measurement stage.

[0117] Specifically, if the absolute value of the temperature change rate is greater than 0.2℃ / s, the system determines that it is in the rapid change stage, and continues to maintain the single-level constant current rapid temperature measurement step in step S2; if the absolute value of the temperature change rate is less than or equal to 0.2℃ / s, and the condition is met in several consecutive samplings, the temperature is determined to be in the stable stage, and the process proceeds to step S4.

[0118] S4: Multi-range constant current resistance measurement:

[0119] Multiple constant current levels are sequentially applied to the PT100, and the corresponding voltage values ​​are collected to form a multi-point current-voltage characteristic dataset. This includes the following steps:

[0120] S4.1: Current range switching control:

[0121] The MCU controls the register of the ADS124S08 precision ADC to start its built-in IDAC constant current source, and switches the output level sequentially according to the preset current sequence I1=0.5mA, I2=1mA, I3=1.5mA. After the current output of each level stabilizes, the corresponding voltage signal is acquired.

[0122] S4.2: Corresponding voltage signal acquisition:

[0123] For each set constant current source output current The voltage drop across the PT100 is acquired through the four-wire differential input ports (AIN0 / AIN1).

[0124] Furthermore, the verification example in this application uses a proportional measurement method, and the specific circuit topology is as follows:

[0125] The PT100 resistor to be tested (denoted as...) ) and high-precision low-temperature drift precision reference resistor (denoted as A resistor (1.62kΩ, ±0.01%) is connected in series and jointly connected to the IDAC constant current source output circuit of the ADC chip. In this circuit, Flowing in sequence and Thanks to the PT100's four-wire connection and the Kelvin interface (REF0 / REF1) provided by the ADC chip, the lead resistance in the circuit is not included in the measurement range. The ADC's conversion reference voltage... Directly taken from precision resistors The stable voltage drop across the two ends. Based on the above circuit connection, define... and These are the voltages to ground at the analog input terminals AIN0 and AIN1 of the ADC, respectively. According to Ohm's law, the voltage relationships at each node can be expressed as:

[0126]

[0127]

[0128] The voltage drop V across PT100 DIFF It can be represented as:

[0129]

[0130] Based on the correspondence between digital and analog quantities converted by the ADC chip, the minimum voltage resolution V △ It can be represented as:

[0131]

[0132] From formulas (10) and (11), the digital voltage value after ADC sampling and conversion can be derived as follows:

[0133]

[0134] Introducing the gain from the preamplifier, equation (12) can be derived as follows:

[0135]

[0136] The voltage values ​​V1 (corresponding to 0.5mA) under current I1, V2 (corresponding to 1mA) under current I2, and V3 (corresponding to 1.5mA) under current I3 were recorded synchronously. During the acquisition process, the AD8422ARZ precision operational amplifier was used to amplify the signal with a fixed gain of 10.

[0137] Substituting the gain factor into formula (13), we can derive the digital measurement equations corresponding to the three current inputs under one current scan cycle. The resulting measurement result expressions for the three ranges are as follows:

[0138]

[0139]

[0140]

[0141] S5: Least squares fitting:

[0142] A linear model is constructed based on the aforementioned multi-point current-voltage feature dataset, and the optimal resistance estimate of PT100 is calculated using the least squares method. Specifically:

[0143] Establish a linear model and construct the least squares objective function based on Ohm's law and three levels of sampled data:

[0144]

[0145] in, This is the sum of squares of the residuals between the measured voltage value and the theoretical voltage value; The voltage analog quantity is collected and converted under the i-th constant current excitation (i.e., V1, V2, V3 obtained in step S4). R is the preset constant current value corresponding to the i-th level (i.e., 0.5mA, 1mA, 1.5mA); R is the estimated value of the PT100 resistance to be solved.

[0146] Take the first derivative of equation (17) with respect to R, and set the derivative to zero (satisfying the extremum condition). After simplification, the best-fit resistance is obtained. The calculation formula is as follows:

[0147]

[0148] The MCU takes the analog voltage values ​​V1, V2, and V3 obtained in step S4 and the current sequence values ​​I1=0.5mA, I2=1mA, and I3=1.5mA and substitutes them into equation (18) to obtain the optimal resistance estimate of PT100 obtained in this measurement when random noise is present.

[0149] After least-squares fitting, the sensitivity of the resistance measurement process to random noise, current source jitter, and ADC drift changes is significantly reduced. According to simulation and experimental data, the standard deviation of the fitted resistance in the verification example of this application is only 1 / 3 to 1 / 4 of that of the traditional single-range method, and the final temperature error is improved by about 70%–75%, which significantly improves the stability and reliability of the PT100 temperature measurement system in dynamic and noisy environments.

[0150] S6: Compensation and Temperature Conversion

[0151] The compensated resistance value is substituted into the standard PT100 resistance-temperature relationship formula to calculate the final temperature value. The MCU outputs the final temperature data and writes it to the data buffer or sends it to the host computer. After completing this temperature measurement cycle, the system returns to step S2 to judge the temperature change again, thus forming a dynamic and adaptive temperature measurement process.

[0152] like Figure 3As shown, by observing the entire temperature range from -45℃ to 200℃, the error curve (solid line) of the traditional single-stage constant current measurement ranges from 0.10℃ to 0.32℃, while the error curve (dashed line) of the multi-stage fitting method of this invention ranges from ±0.04℃ to ±0.05℃. It can be seen that this invention controls the multi-stage fitting error within the range of ±0.04℃ to ±0.05℃, reducing the overall error by about 70%. This verifies that the temperature measurement error of the multi-stage fitting method of this invention is significantly smaller than that of the traditional single-stage method, and has the advantage of greatly reducing the temperature measurement error across the entire temperature range.

[0153] like Figure 4 As shown, the multi-level fitted curves (circles) closely match the actual temperature curves (solid lines) within the overall temperature range. Specifically, within the magnified range of 100–104℃, the deviation is only within ±0.03–0.04℃, while the deviation of the traditional single-level method is around ±0.22℃. This verifies that the measurement results of the multi-level fitting method of this invention are closer to the actual temperature, thus improving the accuracy of temperature measurement.

[0154] Compared to the traditional single-stage constant current method, this invention employs a combination of online calibration and multi-stage constant current least squares fitting, significantly reducing the temperature measurement error of the PT100. Under the full temperature range of -45℃ to 200℃, the peak temperature error of traditional single-stage constant current measurement is approximately ±0.16℃; while using the method provided by this invention, the peak temperature error is reduced to between ±0.04℃ and ±0.05℃, representing a reduction of over 70%.

[0155] In summary, this invention, by combining online calibration with multi-stage constant current least squares fitting, features small temperature measurement error and high measurement accuracy across the entire temperature range, effectively optimizing the temperature measurement performance of the PT100.

[0156] Obviously, the above embodiments of the present invention are merely examples for clearly illustrating the present invention, and are not intended to limit the implementation of the present invention. For those skilled in the art, other variations or modifications can be made based on the above description. It is impossible to exhaustively list all the implementation methods here. All obvious variations or modifications derived from the technical solutions of the present invention are still within the protection scope of the present invention.

Claims

1. A PT100 temperature measurement circuit with multi-range current fitting resistance measurement and online ADC calibration, characterized in that, include: An ADC calibration reference resistor network is used to calibrate the ADC's offset error and actual gain online, and to monitor the ADC's linear deviation online. The ADC unit, with its input terminal optionally connected to the PT100 measurement channel or the resistor network, is used to ensure that the PT100 measurement and ADC calibration processes use a unified sampling path to eliminate channel bias. A multi-level constant current excitation unit is used to sequentially output multiple current levels to the PT100 to obtain multi-point current-voltage characteristic datasets. A channel switcher is used to switch the input path of the ADC between ADC calibration mode and PT100 measurement mode. The control unit is used to acquire the reference node voltage and perform online ADC calibration; In measurement mode, voltage values ​​corresponding to multiple constant current levels are collected, and the resistance value of PT100 is calculated based on the dataset.

2. The circuit according to claim 1, characterized in that, The ADC calibration reference resistor network consists of several precision resistors connected in series between the reference voltage source and ground, and is configured as a voltage divider circuit structure.

3. The circuit according to claim 1 or 2, characterized in that, When performing online ADC calibration, the control unit dynamically updates the ADC error compensation coefficient and subtracts the drift amount in real time from each frame of raw sampled data.

4. The circuit according to claim 1 or 2, characterized in that, In measurement mode, the control unit synchronously controls the scanning timing of multiple constant current sources, collects multiple sets of corresponding voltage values, and performs least squares fitting calculations; through a resistance-current linear model, it extracts the statistically optimal resistance estimate from multiple data points.

5. A PT100 temperature measurement method using multi-range current fitting resistance measurement and online ADC calibration, applied to the circuit as described in any one of claims 1 to 4, characterized in that, Includes the following steps: S1: Perform an online self-test on the ADC unit to ensure that the ADC's offset, gain, and linearity are all in normal condition before entering the PT100 measurement. S2: Switch to the PT100 measurement channel, output a single constant current and collect the corresponding voltage value. Determine whether to enter the PT100 measurement mode based on the temperature change rate: Apply multiple constant current levels to the PT100 in sequence and collect the corresponding voltage values ​​to form a multi-point current-voltage feature dataset. S3: Construct a linear model based on the obtained dataset, and calculate the optimal resistance estimate of PT100 using the least squares method; substitute the fitted resistance value into the resistance-temperature characteristic relationship of PT100 to obtain the final temperature value.

6. The method according to claim 5, characterized in that, S1 includes: The control unit controls the multi-channel analog switch to connect the differential input terminal of the ADC to each reference voltage node of the ADC calibration reference resistor network in sequence, while disconnecting the PT100 measurement channel. Set the voltage node ratio of the ADC calibration reference resistor network; control the ADC to sequentially acquire the actual voltage values ​​of the voltage nodes, and calculate the corresponding theoretical voltage values ​​based on the resistance ratio; Based on the calculation results, the multi-point least squares method is used to fit and compensate the actual gain and offset error of the ADC, while monitoring the linear deviation and issuing an early warning when it exceeds the limit.

7. The method according to claim 5 or 6, characterized in that, S2 includes: S2.1: Switch to the PT100 measurement channel, output a single-level constant current and collect the corresponding voltage value, and calculate the temperature change rate; S2.2: Determine whether the temperature change is stable based on the temperature change rate. If it is stable, proceed to step S2.

3. S2.3: Apply multiple constant current levels to the PT100 sequentially and collect the corresponding voltage values ​​to form a multi-point current-voltage feature dataset.

8. The method according to claim 7, characterized in that, Specifically, S2.1 is as follows: The control unit controls the channel switcher to select the PT100 measurement channel, and the ADC acquires the PT100 voltage signal at a fixed sampling frequency. The control unit calculates the resistance value of PT100 based on the proportional measurement relationship; According to the IEC standard PT100 equation, the resistance value is converted into temperature data to obtain the real-time temperature and calculate the rate of temperature change.

9. The method according to claim 8, characterized in that, Specifically, S2.3 is as follows: The multi-level constant current excitation unit is controlled to switch the output level sequentially according to the preset current sequence, and the corresponding voltage signal is collected after the current output of each level stabilizes. For each output current level, the voltage drop across the PT100 resistor is collected through the PT100 temperature measurement channel; By employing the proportional resistance measurement method and based on the correspondence between the digital and analog quantities converted by the ADC, the gain brought by the pre-amplifier is introduced to derive the digital voltage quantity after ADC sampling and conversion, thus forming a multi-point current-voltage characteristic dataset.

10. The method according to claim 5 or 9, characterized in that, S3 includes: S3.1: Establish a linear model, construct a least squares objective function based on Ohm's law and the multi-point current-voltage characteristic dataset, and calculate the best fitting resistance value for PT100; S3.2: Substitute the compensated resistance value into the standard PT100 resistance-temperature relationship formula to calculate the final temperature value; S3.3: The control unit outputs the final temperature data and writes it to the data buffer or sends it to the host computer; after completing this temperature measurement cycle, it returns to step S2 and performs temperature change judgment again to form a dynamic and adaptive temperature measurement process.

Images