Pressure sensors are among the most commonly used sensors in real life, and they are widely used in various industrial control, automotive electronics, and medical equipment. This article mainly provides a brief explanation of the production calibration of strain gauge Wheatstone bridge sensors, for reference only by pressure sensor engineers.
Sensor characteristics
Figure 1 shows a DC bridge. When the current output resistance is infinitely large, the output of the front bridge can be simplified as follows:
Figure 2 (left) shows the output characteristic curves of the xxx type pressure sensor . The bridge outputs were tested at different pressures in temperature environments of -40°C, 0°C, 25°C, and 85°C. In practical applications, we often expect the sensor output to be temperature-independent and linear, as shown in Figure 2 (right). Due to limitations in MEMS design and fabrication processes, material physical properties, etc., MEMS sensors themselves rarely meet these ideal requirements. In practical applications, a signal conditioning chip is often added to the back end of the MEMS for calibration.
Figure 2 shows that the main factors contributing to the sensor's suboptimal output are: zero-point drift (offset), sensitivity, and nonlinearity. Both zero-point drift (offset) and sensitivity (sensitivity) are affected by temperature, i.e., they are represented by temperature coefficients. These can be expressed by the following formulas:
Formula 1 represents the output characteristics of an ideal sensor , while the actual sensor characteristics need to be described by formulas 2, 3, and 4, respectively, for nonlinearity, zero-point drift, and sensitivity drift.
Vout: Sensor output after incorporating temperature drift and nonlinearity.
V0: Benchmark point for nonlinear fitting polynomial expansion
kn: nth-order nonlinear coefficient
Tstand: Polynomial-fitted temperature reference point
TCN: Nth-order temperature coefficient of zero-point drift (offset)
tsn: nth-order temperature coefficient of sensitivity
Introduction to Signal Conditioning Chips
Sensor signal conditioning chips typically provide amplification, calibration, and temperature compensation functions for data acquired by front-end sensors, thereby ensuring stable and reliable operation of sensor devices and good consistency between devices. Besides using integrated signal conditioning chips, traditional solutions can use external discrete circuits for sensor calibration and standardization, but this is inefficient and relies heavily on manual labor and experience, making it unsuitable for mass production. Using signal conditioning chips eliminates the tedious work of manual calibration and improves the reliability of sensor devices. Common bridge signal conditioning chip models include the ZSC31010 from ZMDI (Germany), the MAX1452 from Maxim Integrated (USA), and the NSA2300 from Suzhou Nanochip Microelectronics. This article will use the NSA2300 as an example to introduce the relevant explanations for using signal conditioning chips to calibrate sensors. The NSA2300 is a highly integrated, low-power, and high-precision sensor interface chip specifically designed for bridge sensors, providing sensor signal acquisition, amplification, and calibration. It can be used for pressure sensor conditioning, magnetic sensor conditioning, and various strain gauge sensor interfaces. The NSA2300 provides zero-point offset and temperature compensation, sensitivity and temperature compensation, and nonlinearity compensation for sensors without requiring additional external components. The NSA2300 also offers multiple temperature measurement modes, supports both I2C/SPI digital and analog outputs, and can perform single-wire digital communication (OWI) via a multiplexed analog output pin (AOUT).
Key performance indicators:
1. Ultra-wide operating voltage range: 1.8V~5.5V
2. Ultra-wide operating temperature range: -40℃~125℃
3. Simultaneously supports 24-bit ADC digital output and 12-bit DAC analog output.
4. Excellent noise performance: 600nV@OSR=1024X, Gain=32X (equivalent to input noise)
5. Calibration accuracy: 0.05%FSO (Simultaneously supports second-order temperature coefficient and third-order nonlinear calibration)
6. Ultra-fast conversion time: 2ms @ OSR=1024X
7. Supports sleep mode, significantly reducing the burden on the MCU.
8. Low-noise amplifier with 1X-128X variable gain
9. Supports sensor diagnostics and output clamping functions.
10. High-precision internal temperature sensor (absolute accuracy <0.5℃, resolution <0.01℃)
11. Supports various external temperature sensors (diodes, two-terminal and three-terminal thermistors, etc.)
12. Supports I2C/SPI serial communication interfaces
13. Supports single-wire programming (multiplexing analog output pins)
NSA2300 calibration principle:
The NSA2300 can calibrate zero-point drift (offset), sensitivity (sensitivity) to the second-order temperature coefficient and the third-order nonlinear coefficient, respectively.
